Vous êtes sur la page 1sur 368

Page 234: PB 481


PL ATE LE T S induced inhibitory mediators cGMP, stabilised cAMP and down-

stream kinase PKG-­PKA mediated VASP phosphorylation to regu-
late α IIbβ III activation which was counteracted by guanylyl cyclase,
PKG and PKA inhibitors.
Conclusions: CXCR7 can be utilised as an anti-­thrombotic drug tar-
PB001 | Platelet Chemokine Receptor
get without compromising physiological haemostasis.
CXCR7 Mediates an Anti-­thrombotic and
Anti-­thromboinflammatory Effect
M. Chatterjee1; M. Manke1; M. Cebo2; J. Rheinlaender3; PB002 | Ticagrelor Inhibits Platelet-­induced
A. Witte1; T. Schäffer3; M. Lämmerhofer2; O. Borst1;
M. Gawaz1 Formation of Neutrophil Extracellular Traps
I.C. Moschonas1; K.M. Hansson2; D. Stakos3; A.D. Tselepis1
University Clinic of Tübingen, Department of Cardiology and Cardiovascular
Diseases, Tübingen, Germany, 2University of Tübingen, Institute of 1
University of Ioannina, Atherothrombosis Research Centre/Laboratory of
Pharmaceutical Sciences, Tübingen, Germany, 3University of Tübingen, Institute
Biochemistry, Department of Chemistry, Ioannina, Greece, 2AstraZeneca,
of Applied Physics, Tübingen, Germany
Cardiovascular and Metabolic Diseases, Innovative Medicines and Early
Development Biotech Unit, Mölndal, Sweden, 3Democritus University of Thrace,
Background: Elevated platelet surface expression of the chemokine Cardiology Department, Alexandroupolis, Greece

CXCL12 and its receptors CXCR4-­CXCR7 in acute coronary syn-

drome influences functional recovery and prognosis. Background: Neutrophil extracellular traps (NETs) are filamentous
Aims: This study explores the anti-­thrombotic potential of CXCR7 structures consisting of decondensed chromatin and granular pro-
and its effects on thromboinflammation. teins that have antimicrobial properties and are also involved in
Methods: Flowcytometry, thrombinoscopy, thromboelastography, atherothrombosis. Ticagrelor is a potent reversible antagonist of the
bleeding time, flow chamber assay, intravital microscopy, mass spec- platelet ADP receptor P2Y12.
trometry, live imaging by scanning ion conductance microscopy. Aims: We investigated the possible role of platelet activation and the
Results: Deletion of CXCR7 from platelets in Pf4-Cre +CXCR7flox/flox effect of ticagrelor on NETs formation, in vitro.
mice led to enhanced thrombotic disposition without affecting Methods: Platelet-­
rich plasma (PRP) was prepared from whole
haemostasis. The pharmacological CXCR7-­a gonist, counteracted blood of healthy volunteers and adjusted to 2.5×108 platelets/mL.
GPVI-­PAR1-­P 2Y12 induced degranulation, α IIbβ III-­i ntegrin acti- Autologous neutrophils were isolated by double-­
Ficoll gradient,
vation, aggregation, ex vivo thrombus formation in healthy sub- placed on poly-­l-­lysine coverslips and incubated for 1 hour at 37°C,
jects, STEMI patients and mice, reduced the rate and extent of 5% CO2. PRP was incubated with/without 1.25 μM ticagrelor for
platelet interaction with collagen, fibrinogen, ROS generation, 5 minutes in an aggregometer under stirring and then activated with
the release of thromboinflammatory arachidonic acid, thrombox- 20 μM ADP for 5 minutes. The platelet suspension was then trans-
ane A2, 12-­H ETE, HHT. CXCR7-­a gonist countered activation in- ferred to neutrophils at a physiological number ratio of platelets/
duced externalisation of thrombogenic phosphatidylserine, thus neutrophils and incubated for 3.5 hours at 37°C, 5% CO2. In other
checked PRP but not plasma dependent (PPP) thrombin genera- experiments PRP was premixed with neutrophils and incubated
tion, or efficiency of clot formation, stabilisation and dissolution. with/without 1.25 μM ticagrelor for 5 minutes before activation in
In vivo CXCR7-­a gonist administration decreased platelet activa- situ with 50 μM ADP for 3.5 hours at 37°C, 5% CO2. Cells were then
tion, thrombus formation following carotid artery injury and fixed, permeabilized and stained with anti-­myeloperoxidase anti-
subsequent inflammatory platelet interaction with leukocytes, body and DAPI. The % NETs formation was evaluated by fluores-
without affecting tail bleeding time and coagulation parameters cence microscopy.
(prothrombin time, activated partial thromboplastin time). CXCR7 Results: PRP preactivated in the aggregometer increased NETosis by
ligation inhibited agonist-­induced intracellular calcium mobilisa- 106±32% (P=0.03) whereas in situ platelet activation with ADP in-
tion, phosphorylation of PLC-­γ-­S rc-­PKC-­PI3K-­A kt-­p 38MAPK. It creased NETosis by 63±22% (P=0.011). Ticagrelor inhibited NETosis

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in
any medium, provided the original work is properly cited, the use is non-­commercial and no modifications or adaptations are made.
© 2018 The Authors. Research and Practice in Thrombosis and Haemostasis published by Wiley Periodicals, Inc on behalf of International Society on Thrombosis
and Haemostasis.

Res Pract Thromb Haemost. 2018;2(Suppl. 1):1–368.  wileyonlinelibrary.com/journal/rth2  |  1


under both experimental conditions by 60±10% (P=0.028) and adhesion. The efficacy of PDE inhibition is enhanced by the vascular
77±15% (P=0.012), respectively. In the absence of platelets, neither endothelium. Combination of PDE3/5 inhibition with current anti-
ADP nor ticagrelor alone influenced NETosis. platelet medication might be the key in the regulation of acute and
Conclusions: ADP-­activated platelets induce NETosis, which is sig- prolonged thrombus activities.
nificantly inhibited by ticagrelor. Given the antimicrobial properties
of NETs and their implication in various pathophysiological conditions
including inflammation and atherothrombosis, the above results may PB005 | Effects of Aspirin and Rivaroxaban
represent a novel effect of ticagrelor. The clinical significance of this
Treatment on Murine Arterial Vessel Wall
effect remains to be established.
Remodelling and Thrombus Activity
M. Karel1; T. Mastenbroek1,2,3; M. Nagy1; D. Coenen1;
W. Chaouyâ1,4; A. Brouns5; J. Debets5; P. Leenders5; H.
PB003 | Differential Effects of PDE3 and -­5 van Essen5; R. van Oerle1; S. Heitmeier6; H. Spronk1; J.
Inhibitors on Thrombus Formation on Collagen Heemskerk1; M. Kuijpers1; J. Cosemans1
versus Inflamed Endothelium 1
Maastricht University, Biochemistry, Maastricht, the Netherlands, 2MERLN
Institute for Technology-­Inspired Regenerative Medicine, Department of Complex
D.M. Coenen1; H.J. Albers2,3; A.D. van derMeer3; J.M.E.M. Tissue Regeneration, Maastricht, the Netherlands, 3Regenerative Medicine
Cosemans1 Crossing Borders (RegMed XB), Maastricht, the Netherlands, 4Synapse BV,
Maastricht University, Department of Biochemistry, Cardiovascular Research Maastricht, the Netherlands, 5Maastricht University, Pharmacology & Toxicology,
Institute Maastricht (CARIM), Maastricht, the Netherlands, 2University of Maastricht, the Netherlands, 6Bayer Pharma AG, Wuppertal, Germany
Twente, BIOS Lab-­on-­a-­Chip Group, MESA+ Institute for Nanotechnology,
MIRA Institute for Biomedical Technology and Technical Medicine, Enschede, Background: The COMPASS clinical study investigated the effect of
the Netherlands, 3University of Twente, Applied Stem Cell Technologies Group,
MIRA Institute for Biomedical Technology and Technical Medicine, Enschede, the
combined therapy of ASA and 2.5 mg bid rivaroxaban on patients
Netherlands with coronary of peripheral artery disease. However little is known
about the effects of this treatment on the vasculature.
Background: Embolization of platelet-­
thrombi formed after ath- Aims: To study the effects of aspirin and/or rivaroxaban on throm-
erosclerotic plaque rupture may lead to acute myocardial infarcts bus activities and on injury-­induced vascular remodelling in mice.
and strokes. Despite antiplatelet treatment, the recurrence rate is Methods: C57BL/6 mice (n=12/group) were either treated with ve-
~30 percent, urging the need for optimized treatment. Our group hicle, aspirin (5 mg aspirin/kg/d) and/or rivaroxaban (12 mg/g chow).
showed that thrombi, formed ex vivo, remain active, secreting vari- Vascular injury was induced by temporary ligation of the carotid ar-
ous mediators (Mastenbroek et al. ATVB 2015). tery. Vascular stiffening was monitored for 2 weeks by non-­invasive
Aims: This study aims to examine the effect of suppressing plate- ultrasound imaging. Two weeks after ligation, platelet-­
let secretion, by elevating cAMP and/or cGMP, on key functional aggregates (PLA) and vascular changes were assessed. Whole
thrombus activities. blood thrombus formation and thrombus-­induced cleavage of dye-­
Methods: The effect of phosphodiesterase (PDE)3/5 inhibitors on quenched collagen was studied with microfluidics.
platelet α IIbβ3 activation and α-­ and δ-­granule secretion was meas- Results: Treatment with aspirin with(out) rivaroxaban decreased the
ured with flow cytometry. Microfluidics were used to study whole number of adhered platelets per PLA (P<0.05) in unstimulated blood
−1 and blood stimulated with stable ADP analogue or PAR4 peptide
blood thrombus formation at 1000 s over a collagen type I + tis-
sue factor surface or over a confluent monolayer of tumor necrosis (flow cytometry). Importantly, treatment with aspirin with(out) ri-
factor-­α treated human umbilical vein endothelial cells. varoxaban protected the arteries against ligation-­induced stiffening
Results: Cilostazol (PDE3i, 5 μM) and tadalafil (PDE5i, 10 nM) signifi- (P<0.05), while rivaroxaban alone showed a trend herein (P=0.08).
cantly inhibited platelet αIIbβ3 activation of collagen-­related peptide-­ Histological analysis indicated that ligation of the carotid artery re-
treated washed platelets (P=0.005 and P=0.03), whereas platelet sulted in local intima-­media thickening, which was reduced by treat-
α-­ and δ-­granule secretion were not significantly affected (P>0.05). ment with aspirin or rivaroxaban (P<0.01). Treatment of mouse blood
In whole blood, PDE3 (50 μM) or -­5 (100 nM) inhibition resulted in with aspirin and/or rivaroxaban did not alter the size of thrombi
smaller thrombi, although non-­significant. Cilostazol mainly affected formed ex vivo. Yet, interestingly, thrombus-­induced collagenolytic
the earlier phase of thrombus growth, while the effect of tadalafil activity ex vivo was reduced (P<0.05) upon treatment with aspirin
was more pronounced at a later time point. Furthermore, PDE3 inhi- or rivaroxaban.
bition significantly increased time to fibrin (P=0.03 vs. P=0.17 with Conclusions: This study provides new insight into the vascular-­
PDE5 inhibition). Strikingly, a ten-­fold lower dose of cilostazol (5 μM) directed effects of aspirin and/or rivaroxaban treatment after throm-
gave a strong decrease in size and compactness of platelet thrombi botic injury. Treatment with aspirin significantly antagonised, and
on inflamed endothelial cells. rivaroxaban showed potential to inhibit, vascular stiffening, intima-­
Conclusions: These data suggest that PDE3 and PDE5 are key en- media thickening, PLA formation and thrombus-­
zymes in thrombus growth and stability, but not in initial platelet activity.

PB006 | Association of PEAR1 Gene Variant

with Platelet Reactivity and Clinical Outcomes in
Patients Undergoing Stenting and Treated with
Aspirin and Clopidogrel
K. Xu1; N. Chan2; F. Wang1; L. Yang1; H. Zhu1,3; J. Li1,4;
J. Zhang1; Y. Fan1; J. Chen5; L. Xu6; L. Ying1,7; X. Hu1; S. Ye1;
C. Li1
The First Affiliated Hospital of Nanjing Medical University, Department of
Cardiology, Nanjing, China, 2McMaster University, Department of Medicine,
Hamilton, Canada, 3The Second People’s Hospital of Changzhou Affiliated
to Nanjing Medical University, Changzhou, China, 4Fuyang Fifth People’s
Hospital, Fuyang, China, 5Maanshan People’s Hospital, Department of
Cardiology, Maanshan, China, 6 Sir Run Run Shaw Hospital, Nanjing Medical
University, Nanjing, China, 7 The Affiliated Huai’an Hospital of Xuzhou Medical
University and Huai’an Second People’s Hospital, Department of Cardiology,
Huai’an, China

Background: Dual antiplatelet therapy (DAPT) with aspirin and

clopidogrel is the standard therapy for patients undergoing coro-
nary stenting. Genetic determinants of variable response to aspirin
and clopidogrel are incompletely characterized. Polymorphisms in
platelet endothelial aggregation receptor-­
1 (PEAR1, rs12041331)
were initially described as being important for platelet aggregation.
The prevalence of PEAR1 A/A homozygotes, its associations with
laboratory and clinical outcome in a Chinese population undergoing
stenting and DAPT, and the mechanism explaining any association
F I G U R E   1   Genotypes of PEAR1 (rs12041331) and platelet
are unclear.
reactivity. ADP-­and AA-­induced platelet aggregation were showed
Aims: To explore the prevalence of PEAR1 rs12041331 A/A ho-
in (A) and (B), respectively
mozygotes and its association with platelet aggregation and clinical
Methods: We recruited 2007 consecutive patients undergoing
stenting and DAPT. Arachidonic acid-­and ADP-­
induced platelet
aggregation (PL AA and PL ADP) were performed on the 5th day after
the procedure. Genotyping was performed using an improved mul-
tiplex ligation detection reaction technique. All patients underwent
a 30-­day follow-­up and data for MACE (cardiovascular death, non-
fatal myocardial infarction or ischemic stroke) were collected. This
study was approved by the local ethics committee and registered.
Informed consent was obtained.
Results: PL ADP was significantly lower in AA-­than non-­A A-­geno-
F I G U R E   2   Cumulative incidence of 30-­day MACEs in patients
types (P<0.001) but not PL AA . Independent of CYP2C19*2 poly-
undergoing stenting and DAPT stratified by PEAR1 rs12041331
morphism, 30-­day MACEs occurred significantly more in AA-­than genotype
in GG-­homozygotes or heterozygotes (adjusted HR, 3.13; 95% CI,
1.05-­9.31; P=0.041).
PB007 | Beta-­1,4-­galactosyltransferase 2
Conclusions: In the largest study to date, PEAR1 A/A homozygosity
c.909C>T Gene Variant Is Predictive of On-­
occurs in 14.8% of Chinese patients and is associated with a 3-­fold
increase in MACE risk after stenting. PL AA was unaffected and PL ADP clopidogrel Platelet Reactivity
was lowest in AA-­homozygotes. The discordant laboratory findings T. Belleville-Rolland1,2; N. Pallet3,4; A. Savalle3; M. Lejeune3;
raise the possibility that PEAR1 polymorphism might impact clinical L. Mauge1,5; S. Bertil1; M.-A. Loriot3,4; P. Gaussem1,2
outcome through mechanisms independent or distal to the inhibition AP-­HP, Hôpital Européen Georges Pompidou, Hematology, Paris, France,
Paris Descartes University, Inserm UMR-­S1140, Paris, France, 3AP-­HP, Hôpital
by aspirin or clopidogrel.
Européen Georges Pompidou, Clinical Chemistry, Paris, France, 4Paris Descartes
University, Inserm UMR-­S1147, Paris, France, 5Paris Descartes University, Inserm
UMR-­S970, Paris, France

Background: CYP2C19*2 loss-­of function variant influences clopi- Aims: To monitor the on-­treatment platelet reactivity (PR) in pa-
dogrel response but only accounts for a small part of the variability tients with ST segment elevation myocardial infarction (STEMI), to
of platelet reactivity, suggesting the involvement of other gene vari- find out prevalence of HTPR and LTPR, to reveal the prevalence of
ants. Recently, exome sequencing of coronary patients under dual ischemic and bleeding episodes during DAPT and their relation to
antiplatelet therapy identified a variant of the gene encoding beta-­ laboratory findings.
1,4-­galactosyltransferase 2 (B4GALT2), a platelet-­expressed galacto- Methods: 73 patients with acute STEMI after PCI and on DAPT were
syltransferase. Carriers of the B4GALT2 c.909C>T variant had lower included in the prospective monocentric study. PR was assessed
platelet reactivity, as assessed by the VerifyNow P2Y12 assay and using both the VASP phosphorylation analysis and light transmit-
light-­transmission aggregation, compared with non-­carriers, indicat- tance aggregometry (LTA) in 3 intervals: T1 (time of PCI), T2 (1 day
ing that B4GALT2 could influence clopidogrel sensitivity. after PCI) and T3 (30 days after PCI). Patients were observed during
Aims: To determine if B4GALT2 c.909C>T influences clopidogrel a period of 1 month with the aim to reveal thrombotic and bleeding
pharmacodynamics as assessed by vasodilator-­stimulated phospho- episodes.
protein (VASP) assay and ADP-­induced platelet aggregation in a non-­ Results: High variability in PR was found in clopidogrel treated
selected cohort of patients. patients in T2 and T3. Significant difference in PR was found
Methods: We retrospectively analyzed 353 patients under clopi- between clopidogrel and novel P2Y12 antagonists in T2 and T3,
dogrel (mono or bi-­antiplatelet therapy) addressed to our center while no significant difference between PR in prasugrel and tica-
between January 2010 and January 2017 to undergo genetic and grelor group was found. Prevalence of HTPR in clopidogrel treated
pharmacodynamic testing (ADP-­induced light-­transmission aggrega- group in T2 and T3 exceeded 45%. HTPR in prasugrel/ticagrelor
tion and VASP assay). All patients gave written informed consent. treated group was rare (5%). Stent thrombosis was diagnosed in
Results: Among the 353 patients screened, DNA was available for 12% of clopidogrel treated patients (all with HTPR). In spite of high
174 of them: 28% were under clopidogrel monotherapy, and 72% prevalence of LTPR in novel P2Y12 antagonist treatment no bleed-
under clopidogrel+aspirin. In individuals under dual antiplatelet ing episodes were found.
therapy, B4GALT2 c.909C>T was associated with a significant lower Conclusions: Results confirm a high effectiveness of PR monitor-
ADP-­induced platelet aggregation. Surprisingly, it did not influence ing in patients with acute STEMI undergoing PCI on DAPT. VASP
VASP assay. Otherwise, B4GALT2 c.909C>T and CYP2C9*2 were analysis has a high specificity to signal route of P2Y12 receptor of
independent genetic predictors of on-­treatment platelet reactivity. platelets thus it seems to be suitable for monitoring of response to
Conclusions: Our findings confirm that B4GALT2 c.909C>T influ- its antagonists.
ences on-­treatment platelet reactivity in a cohort of non-­selected
patients under clopidogrel-­
based antiplatelet therapy, indepen-
dently of the P2Y12 signaling pathway. B4GALT2 could influence
PB009 | Detection of Anthocyanins Effects to
platelet adhesion through posttranslational modification of integrins
Alleviating Thrombotic Risks in Comparison to
and/or other cell surface glycoproteins.
A. Gaiz1; A. Kundur2; S. Mosawy2; N. Colson2; I. Singh2
PB008 | High On-­treatment Platelet Reactivity 1
Almustansiriah University, National Center of Hematology, Baghdad, Iraq,
Griffith University/Menzies Health Institute of Queensland, Southport, Australia
(HTPR) on Clopidogrel but Not on Novel P2Y12
Antagonists in Patients with ACS Undergoing
Background: Platelet hyperactivity has a significant role in initiating
PCI – Single Centre Study vascular thrombosis and subsequent cardiovascular disease (CVD).
J. Staško1; R. Šimonová1; M. Samoš1; J. Sokol1; L. several studies demonstrate resistance to currently used antiplate-
Stančiaková1; I. Škorňová1; T. Simurda2; P. Kubisz1 let therapies such as aspirin (AS). The antioxidant activity of poly-
Comenius University in Bratislava, Jessenius Faculty of Medicine in Martin, phenolic compounds including anthocyanin (AC) has been shown to
University Hospital Martin, National Centre of Hemostasis and Thrombosis,
reduce platelet activity and reduce CVD morbidity.
Martin, Slovakia, 2Comenius University in Bratislava, Jessenius Faculty of
Medicine in Martin, National Centre of Hemostasis and Thrombosis, Martin, Aims: Comparing the antiplatelet effect of pure AC compounds with
Slovakia Aspirin.
Methods: Healthy human subject (n=50) were recruited in this study
Background: P2Y12 receptor antagonists together with acetylsalicylic and divided into two groups. Each participant consumed either
acid (ASA) are used in treatment of patients with acute coronary syn- 80 mg Aspirin/day or 320 mg of AC/day in the form of capsules for
drome (ACS) and secondary prophylaxis of thrombotic complications 28 days. Fasting blood samples were collected at baseline (before)
after percutaneous coronary intervention (PCI). High on-­treatment and after the treatment period. Flow cytometry was used to assess
platelet reactivity (HTPR) during dual antiplatelet treatment (DAPT) platelet activation by measuring platelet surface marker expression
represents a risk of thrombosis. On the other side low on-­treatment of CD41a and P-­
selectin. Platelet aggregation studies were per-
platelet reactivity (LTPR) can lead to bleeding. formed by stimulating platelets with ADP, collagen or arachidonic

acid. Haematological indices, lipid profile, glucose, uric acid, and high patients, there were 28 (54%) clopidogrel high responders with ADP,
sensitivity CRP were assessed. 15 (30%) low responders and 8 (16%) non-­responders. Mean platelet
Results: There was a significant suppressive effect of AC on the count was significantly lower in the cardiovascular group than in the
expression of P-­selectin. There was a substantial reduction of cerebrovascular group (210.680±73.830 μL vs. 280.530±65.860 μL,
ADP-­s timulated platelet aggregation. On the other side, Aspirin respectively) (P<0.05). Mean MPV and PDW were significantly
significantly reduced collagen-­
s timulated and arachidonic acid-­ higher in the cardiovascular group than in the cerebrovascular group
stimulated platelet aggregation assays, but it has an insignificant (MPV 10.9±0.8 /fL vs. 10.1±0.9 /fL; PDW 13.9±1.9 fL vs. 11.6±2.1
inhibitory effect on ADP-­
s timulated platelet aggregation and fL) (both P<0.05).
expression of P-­s electin. There were trends of lowering impact Conclusions: Our findings suggest that clopidogrel dose adjustment
of AC on total cholesterol and triglycerides, but they were not is necessary to prevent recurrences in patients with previous cardio-
significant. vascular events who are low responders clopidogrel.
Conclusions: The results show that AC may reduce platelet aggrega-
tion and activation as demonstrated by inhibition of ADP-­stimulated
platelet aggregation and P-­selectin expression. These results pro- PB013 | Assessment of Platelet Reactivity with
vide greater insight into the effect of AC and the possible mechanism
Multiple Electrode Impedance Aggregometry
by which AC can reduce platelet function. Hence, AC may act as a
complement to other anti-­platelet agents to reduce the occurrence
(MEA) in Patients Reseiving Ticargelor
of thrombotic events. D. Dineva; I. Paskaleva
National Heart Hospital, Sofia, Bulgaria

PB011 | Assessment of Clopidogrel Response Background: Monitoring platelet reactivity during P2Y12-­inhibitors

helps prevent stent thrombosis (ST) and/ or bleeding. High platelet
in Patients with Prior Atherothrombotic Events
reactivity (HPR) is associated with an increased risk for ischaemic
Z. Kaya1; P. Özen1; A.P. Nas1; H. Batur Çağlayan2; complications, while low platelet reactivity (LPR) is associated with
G. Aydoğdu Taçoy3; B. Nazlıel2
greater bleeding events.
Gazi University Medical Faculty, Pediatric Hematology Unit of the Department
Aims: The purpose of this study was to evaluate the effect of tica-
of Pediatrics, Ankara, Turkey, 2Gazi University Medical Faculty, Department of
Neurology, Ankara, Turkey, 3Gazi University Medical Faculty, Department of grelor on residual platelet reactivity.
Cardiology, Ankara, Turkey Methods: The platelet function was assessed with Multiplate imped-
ance aggregometry by ADP-­test in 91 patients on 2×90 mg/d tica-
Background: Clopidogrel is a prophylactic antiplatelet drug that has grelor therapy who underwent PCI. Blood was collected at the first
been successful at preventing recurrences in patients with athero- month after stent implantation. Adequate response to ADP receptor
thrombotic events. Abnormal platelet count, high MPV, and PDW in- blocking was defined as ADP-­test <450 AU, on the base of cut-­off
dicating platelet activation, are additional risk factors for thrombosis value, determined by ROC analysis (AUC 0.864 with 0.84 specificity
in patients with prior cardiac events. and 0.78 sensitivity). Low platelet reactivity was defined as ADP-­test
Aims: We retrospectively evaluated platelet tests and clopidogrel <150 AU.
response for 80 patients with prior cardiovascular or cerebrovascu- Results: A total of 91 patients were enrolled, 88 of them showed
lar events. adequate response on ticagrelor treatment (ADP 203±89 AU). We
Methods: All patients had received 75 mg/day clopidogrel for observed residual HPR in three patients (ADP 635±78 AU) and they
at least 7 days prior to platelet aggregation testing. Clopidogrel were switched on prasugrel 10 mg/d. Additional platelet inhibition
response was assessed by light transmission aggregometry of 52.8% was observed when switching them from ticagrelor to
l og Model 700) using ADP agonist. According to the prasugrel (663 AU vs. 313 AU).
manufacturer’s description, high responders to the drug were de- Minor bleeding complications like cutaneous hematoma, epistaxis,
fined as having platelet reactivity below 49%, low responders be- haemorrhoidal bleeding and hemoptoe were found in patients with
tween 50% and 61%, and non-­responders between 62% and 70%. lower values of ADP test (107±64 AU). Antiplatelet therapy in pa-
Platelet count, MPV and PDW were measured as a part of each tients with bleeding was switched to clopidogrel 75 mg/d. Reduction
patient’s complete blood count in a Sysmex , XN-­3 000 automated of platelet inhibition and decreasing of the bleeding was seen when
hematology analyzer. ticagrelor was switched to clopidogrel.
Results: There were 29 patients with prior cardiovascular events Conclusions: We found impaired response to ticagrelor 2×90 mg
(median age: 70 years) and 51 patients with prior cerebrovascular in 3 of 91 patients (3.29%). ADP-­test <150 AU was associated with
events (median age: 62 years). Of the cardiovascular patients, there increasing risk of minor bleeding. Tailored treatment with different
were 12 (42%) clopidogrel high responders with ADP, 10 (34%) low P2Y12 inhibitors leads to optimal level of ADP inhibition.
responders and 7 (24%) non-­responders. Of the cerebrovascular

PB014 | Results of Platelet Aggregation Testing patients taking antiplatelet therapy. Recovery time of platelet function
of 6-­7 days after drug withdrawal could be explained by the fact that
by Light Transmission Aggregometry using Drug
the function of inhibited enzyme or receptor recovers for about 10% a
Specific Agonists in Patients Taking Antiplatelet
day as a result of the entry of new platelets in circulation.
S. Margetić1; N. Vrkić1,2; B. Getaldic1; D. Ruzic Ferenec1; I. Vuga1
PB015 | Optimal Antiplatelet Therapy
Sestre Milosrdnice University Hospital Center, Department of Clinical Chemistry,
Zagreb, Croatia, 2Faculty of Pharmacy and Biochemistry University of Zagreb,
Zagreb, Croatia in Moderate to Severe Asymptomatic and
Symptomatic Atherosclerotic Carotid Stenosis:
Background: In the recent years platelet aggregation testing has ex- A Systematic Review of the Literature
panded in the assessment of the effectiveness of antiplatelet therapy.
S. Murphy1; R. Naylor2; J.-B. Ricco3; H. Sillesen4; S. Kakkos5;
Aims: To evaluate results of platelet aggregation by light transmis-
A. Halliday6; G. de Borst7; M. Vega de Ceniga8; G. Hamilton9;
sion aggregometry (LTA) using drug specific agonists in patients on D. McCabe10
antiplatelet therapy. 1
Adelaide and Meath Hospital, Department of Neurology, Dublin, Ireland, 2Glenfield
Methods: Platelet aggregation testing was performed in 107 patients Hospital, Department of Vascular Surgery, Leicester, United Kingdom, 3University
on antiplatelet therapy: ASA, ADP receptor inhibitor (clopidogrel or of Strasbourg, Department of Vascular Surgery, Strasbourg, France, 4Rigshospitalet,
University of Copenhagen, Department of Vascular Surgery, Copenhagen, Denmark,
ticagrelor) or dual therapy (ASA plus ADP receptor inhibitor) as well as 5
University of Patras Medical School, Department of Vascular Surgery, Patras,
in 27 patients in whom antiplatelet therapy was discontinued from 2 to Greece, 6University of Oxford, Nuffield Department of Surgical Sciences, Oxford,
8 days due to surgery. Inhibition of platelet aggregation was assessed United Kingdom, 7University Medical Center Utrecht, Department of Vascular
Surgery, Utrecht, the Netherlands, 8Hospital de Galdakao-­Usansolo, Department of
by ADP (2 μmol/L) agonist for ADP receptor inhibitor and by arachi-
Angiology and Vascular Surgery, Bizkaia, Spain, 9University College London Medical
donic acid (AA, 1 mmol/L) as agonist for acetylsalicylic acid (ASA) drug School, Department of Vascular Surgery, London, United Kingdom, 10School of
on Sysmex CS2500 analyzer using Hyphen BioMed (France) reagents. Medicine, Trinity College Dublin, Academic Unit of Neurology, Dublin, Ireland
Results: Platelet aggregation with drug specific agonists was signifi-
cantly inhibited in all patients, as shown in Table 1. Platelet inhibi- Background: Carotid stenosis patients are at risk of cerebrovascular
tion with ADP was statistically different (P=0.003) between patients events despite antiplatelet therapy. Data on prescribed antiplate-
treated with clopidogrel and ticagrelor. However, there were no dif- let regimens have not been comprehensively-­collated from trials to
ferences in AA or ADP inhibitions between patients treated with one guide optimal therapy in this population.
or both drugs (Table 1). Among all patients on ASA and ADP recep- Aims: To perform a comprehensive systematic review of the litera-
tor inhibitor alone and among those on dual therapy, only one patient ture on randomised trials of patients with extracranial moderate to
was poor responsive to AA inhibition and no one was poor responsive severe carotid stenosis treated with antiplatelet therapy.
to ADP inhibition. Among 27 patients with discontinued antiplatelet Methods: We searched Medline, Ovid, Embase, Web of Science and
therapy, platelet function was recovered to the values above the limit Google Scholar from 01/1988 to 03/2016 for randomised trials in
of the reference range (60%) 6th day for AA and 7th day for ADP ago- this patient population on any form of antiplatelet therapy in which
nist after drug withdrawal. cerebrovascular outcome events were reported.
Conclusions: Use of specific agonists for certain antiplatelet drugs con- Results: 22 studies were deemed eligible for inclusion. Data from
tribute to the better understanding of platelet aggregation inhibition in one RCT showed a non-­significant benefit from aspirin vs. placebo

TA B L E   1   Results of platelet aggregation testing using specific agonists by light transmission aggregometry (LTA) in patients on
antiplatelet therapy

ASA + ADP inhibitor

acetylsalicylic acid ADP clopidogrel 75 mg/ ADP inhibitor
Antiplatelet drug (ASA) 100 mg/day ADP inhibitor inhibitor P day ticagrelor 90 mg/day P

Agonist / N N=35 N=24 N=48 N=50 N=22

Arachidonic acid (AA) % aggregation
 Median 12 — 10 0.342 — — —
 95% CI for median 6-­18 — 6-­14 — — —
 Interquartile range (IQR) 6-­19 — 11-­20 — — —
Adenosine-­diphosphate (ADP) % aggregation
 Median — 29 30 0.539 32 22 0.003
 95% CI for median — 11-­4 0 24-­35 28-­4 0 9-­26 —
 Interquartile range (IQR) — 13-­39 22-­4 0 24-­41 12-­26 —

in asymptomatic carotid stenosis, but it is still reasonable to recom- carried out considering the results of complex of clinical data,
mend aspirin (81-­325 mg daily) for prevention of vascular events in platelet functional and molecular genetic tests. According to this
these patients. Low-­medium dose aspirin (81-­325 mg daily) is supe- approach all patients were divided into 5 groups with own regi-
rior to higher doses (>650 mg daily) at preventing recurrent vascular men of antiplatelet therapy: ASA 100 mg or 150 mg, clopidogrel
events in patients undergoing endarterectomy. Data from endovas- 75 mg, clopidogrel 75 mg+ASA 100 mg or 150 mg. After 3 months
cular treatment (EVT) trials support peri-­procedural treatment of of individual antiplatelet therapy all patients showed a signifi-
asymptomatic and symptomatic patients with 81-­325 mg of aspirin cant decrease in platelet reactivity from 58 (43-­75) to 28 (6-­58)
daily. The use of peri-­procedural aspirin-­clopidogrel in patients un- (p=0.01) and from 54 (41-­6 8) to 40 (25-­55) (P=0.05) for ADP and
dergoing EVT is based on one pilot trial, but appears safe. Short-­ collagen induced aggregation respectively. We have achieved a
term treatment with aspirin-­dipyridamole or aspirin-­clopidogrel are good clinical result. No limb amputation was performed within
equally effective at reducing micro-­embolic signals on transcranial 5 years of follow-­u p.
Doppler ultrasound in patients with ≥50% symptomatic carotid ste- Conclusions: One of the ways to improve the long-­term results of
nosis. There is insufficient evidence to recommend routine aspirin-­ reconstructive interventions in PAD might be the development of
clopidogrel combination therapy to reduce the risk of ‘recurrent the algorithm for an integrated approach to individual antiplatelet
clinical ischaemic events’ in patients with symptomatic moderate-­ therapy.
severe carotid stenosis.
Conclusions: This systematic review facilitates an evidence-­based
approach to ‘optimal antiplatelet therapy’ in carotid stenosis pa-
PB017 | Impact of Time-­to-­treatment on
tients. Future trials should randomise such patients to receive differ-
Freedom-­from-­relapse for Steroid-­Refractory
ent antiplatelet regimens.
Immune Thrombocytopenia
P. Vishnu1; W.A. Hammond2; E.M. Rodriguez3; Z. Li1;
C.E. Rivera1
PB016 | The Original Algorithm for Platelet 1
Mayo Clinic Florida, Jacksonville, United States, 2Baptist MD Anderson Cancer
Functional and Molecular Genetic Analysis to Center, Jacksonville, United States, 3University of Puerto Rico at Mayagüez,
Optimize Antiplatelet Therapy in Outpatients Puerto Rico, United States

with Peripheral Artery Diseases

Background: Splenectomy (S) or rituximab (R) are potentially cura-
O. Sirotkina1,2; N. Surint3; A. Topanova1; T. Vavilova1,3 tive options for patients with steroid-­refractory ITP (sr-­ITP) with
Almazov National Medical Research Centre, Saint-­Petersburg, Russian Federation,
good responses, but the long-­term outcome is not known and the
B.P. Konstantinov Petersburg Nuclear Physics Institute of National Research Centre
«Kurchatov Institute», Saint-­Petersburg, Russian Federation, 3First City Consultative
treatment order is uncertain.
and Diagnostic Center, Saint-­Petersburg, Russian Federation Aims: To evaluate freedom-­from-­relapse (FFR) after therapy with S
or R for sr-­ITP and the impact of time from diagnosis to 2nd line
Background: Inhibition of platelet aggregation is the main ap- treatment (TT) based on sequence of therapy, and the difference in
proaches to treatment and prevention of CVD. efficacy of R in patients with or without a prior S.
Aims: The development of original algorithm for laboratory and Methods: Patients age ≥18 years with sr-­ITP treated with S or R be-
molecular genetic analysis to optimize antiplatelet therapy in out- tween Jan 2002 and Dec 2015 at Mayo Clinic who underwent S or
patients with peripheral artery disease (PAD) underwented recon- R as 2nd line therapy. TT, FFR and response following 2nd-­and 3rd-­
structive surgery. line treatment with S or R were measured.
Methods: 105 patients (age 68±1) with PAD were included into the Results: We identified 237 patients; 122 had S and 115 received R
study. We analyzed the Leu33Pro GPIIIa, C807T GPIa, G36T P2Y12, for sr-­ITP. Patients in S group were younger at diagnosis compare to
CYP2C19*2, CYP2C19*3 and CYP3A5*3 polymorphism by PCR, R (median 49 v 60 years, P=0.001) and at time of 2nd-­line treatment
ADP/collagen induced platelet aggregation by impedance method. (median, 53 v 61 years, P=0.007). Most patients required 2nd line
Results: The distribution of appraised genotypes was not sig- therapy within 12 months of steroid refractoriness (65.4%). There
nificantly different from previously explored healthy donors. The were more patients with TT of ≥12 months in S group compared to
presence of 36T P2Y12 allele in homo or heterozygous state was R group (42.6 v. 26%; P=0.0075). Patients in S group had a higher re-
a risk factor for high platelet reactivity in patients on standard an- sponse (87.7% v 43.5%, P<0.001) and a higher 5-­year FFR (53.25% v
tiplatelet therapy: 19 (6-­32) in carriers GG and 83 (66-­91) in carri- 18.9%, P<0.001) compared to R. Patients treated with R after failure
ers GT/TT genotype (p=0.0006) for ADP induced aggregation and of S achieved similar response (88.5% vs. 82.8%, P=0.39) and 2-­y
34 (29-­4 8) and 63 (55-­74), respectively (P=0.003) for collagen in- FFR (67.31% vs. 58.13%, P=0.68) compared to those who had S for
duced aggregation. Moreover 16% of patients have polymorphism relapsed ITP after R. The 5-­year FFR after R was significantly longer
of CYP3A5 and CYP2C19 genes, associated with poor clopidogrel in patients with a prior S than in those with an intact spleen (43.51%
metabolism. Individual selection of antiplatelet therapy was vs. 18.70%, P<0.001). Multivariate model predicting FFR following

2nd line therapy was in favor of splenectomy (HR 0.33, 95% CI 0.22, TA B L E 1
0.50; P<0.001).
Acknowledgement of RAND Survey Participants-­Alphabetical
Conclusions: Sequence of S before R has a clinical benefit for treat-
ment of sr-­ITP. Use of R following failure of S appears to be as effec-
Suchitra Acharya (USA), Said Enayat (UK), Emmanuel Favaloro (AU),
tive as S following failure to R. Although the impact of time to 2nd
Michelle Lavin (Ireland), Will Lester (UK), Gilian Lowe (UK),
line treatment was significant, it is likely that the treatment type was Jonathan Miller (USA), Paquita Nurden (Fr), Shirin Ravavood (Iran),
the impetus for the different outcomes. Analia Sánchez-­Luceros (Argentina), Jerry Ware (USA)

PB018 | Guidance on the Diagnosis and

Management of PT-­V WD: A Project from the PB019 | PK Simulation of Avatrombopag-­
Platelet Physiology Subcommittee of the ISTH induced Increases in Platelet Counts with
M. Othman1,2; P. Gresele3 Redosing in Patients with Thrombocytopenia and
Queen’s University, Biomedical and Molecular Sciences, Kingston, Canada, Chronic Liver Disease
St Lawrence College, School of Baccalaureate Nursing, Kingston, Canada,
3 S. Hayes1; C. Farrell1; K. Aggarwal2; M. Vredenburg2;
University of Perugia, Department of Medicine -­Section of Internal and
Cardiovascular Medicine, Perugia, Italy
L.F. Allen2
ICON, Dublin, Ireland, 2Dova Pharmaceuticals, Durham, United States

Background: Only 55 patients with PT-­V WD have been reported

worldwide. Literature is scarce in providing information required to Background: Avatrombopag (AVA) is a novel oral thrombopoi-

establish standards for diagnosis and management of the disease. etin receptor agonist being developed to treat thrombocytopenia

Currently there are no clear guidelines/consensus and many patients in patients with chronic liver disease (CLD) scheduled to undergo

remain undiagnosed or undertreated. a procedure; it is administered orally as a 5-­day treatment course.

Aims: To generate guidance for a standard diagnostic and manage- Population pharmacokinetic (PK) and PK/pharmacodynamic (PK/

ment approach for PT-­V WD based on consensus expert views. PD) models have been developed that characterize the relationship

Methods: RAND based approach to obtain a formal consensus between AVA dosing and the induced platelet count increase.

among experts. A panel formed of 11 experts (9 countries, 5 conti- Aims: This model was used to simulate platelet count changes that

nents) was formulated. A series of 46 survey statements on history could be expected with a second course of AVA.

and clinical presentation (14), diagnosis (21) and management (11) of Methods: Simulations were conducted for 4 AVA repeat dosing regi-

PT-­V WD were prepared for the experts to review. Statements were mens. For all regimens, 60 mg AVA was dosed daily on Days 1-­5, and

rated by the experts from 1-­9 where 1 is completely inappropriate then a second 5-­day course of 60 mg AVA was started on either Day

and 9 is fully appropriate. Statements were classified as inappropri- 17, 24, 31 or 38. Baseline platelet counts between 20 and 50×109/L

ate (scores of 1-­3), uncertain (scores of 4-­6), or appropriate (scores were assumed. Each regimen was simulated 500 times in 100 sub-

of 7-­9). jects, and the median and 80% Prediction Intervals (PI) obtained.

Results: Experts have agreed largely on most statements related Results: The original PK/PD model (93 subjects) described the PK of

to basic history, clinical features and methodology for diagnosis AVA by a one-­compartment model with first-­order absorption and

and management approaches. Of these are: common bleeding elimination. AVA PK was dose-­independent and not significantly

symptoms and severity, family history, GP1BA gene mutation and affected by age, weight, race, gender, or liver function. Simulations

using RIPA mixing studies as essential lab tests with some disa- of redosing on Day 17 led to only a small increase in the maximum

greement on ristocetin cutoff. Few statements were judged un- platelet counts above the initial Day 1-­5 course (~6×109/L). Second

certain with respect to common bleeding symptoms in children, courses administered on Day 24, 31 or 38 resulted in maximum

the association with cancer, inflammation and bone pathologies platelet counts comparable to the initial Day 1-­5 course. For each

indicating research is required in these areas. Some uncertainty/ regimen, the median upper bound of the 80% PI (i.e., 90th percen-

ambiguity remained around the use of DDAVP, rFVIIa, or appro- tile) was less than 200×109/L.

priate perinatal monitoring. These ambiguous statements will be Conclusions: A redosing simulation of AVA courses on Day 24, 31

resolved in a second survey. Large agreement was made towards a or 38 resulted in median platelet count increases comparable to the

proposed algorithm for diagnosis and management of the disease. initial Day 1-­5 dosing regimen. Redosing at Day 17 led to only a small

Conclusions: A consensus around a standard diagnostic and man- increase in maximum platelet counts, and the 80% PI did not exceed

agement approach for PT-­V WD can be established and would be 200×109/L. These results suggest that AVA can be redosed as early

critical to improving the diagnosis and treatment of this bleeding as 12 days after completion of the first dosing regimen (Day 17)

disorder. without excessively increasing platelet counts.


PB020 | Exploratory Analyses of the Efficacy PB021 | Influence of Preoperative Platelet

of Avatrombopag versus Placebo from 2 Phase 3 Activity on Coagulation and Blood Loss in Large
Studies Using Alternate Baseline Platelet Count Orthopedic Surgery
Cohorts and an Alternate Secondary Efficacy I. Antropova1; B. Yushkov2; I. Shlykov3; E. Volokitina4
Endpoint 1
Ural State Medical University, the Ministry of Health of the Russian Federation,
1 2 2 2 Central Research Laboratory, Ekaterinburg, Russian Federation, 2Institute of
F. Poordad ; L. Allen ; K. Aggarwal ; M. Vredenburg ; Immunology and Physiology, Ural Branch of Russian Academy of Sciences,
W. Tian2; N. Terrault3 Ekaterinburg, Russian Federation, 3Ural Institute of Traumatology and Orthopedics
Texas Liver Institute, San Antonio, United States, 2Dova Pharmaceuticals, named after V.D. Chaklin, Ekaterinburg, Russian Federation, 4Ural State Medical
Durham, United States, 3UCSF, San Francisco, United States University, the Ministry of Health of the Russian Federation, Department of
Traumatology and Orthopedics, Ekaterinburg, Russian Federation

Background: Thrombocytopenia (TCP) is common in chronic

Background: During surgery, activation and secretion of platelets
liver disease (CLD), increasing in severity with worsening disease.
not only promote clot formation, which stop bleeding, but also con-
Avatrombopag (AVA) is a thrombopoietin receptor agonist that has
tribute to reparative processes in damaged tissues.
completed Phase 3 development as an alternative to platelet trans-
Aims: To evaluate the effects of preoperative platelet activity on
fusions for patients with TCP and CLD undergoing a procedure.
postoperative platelet activity, coagulation and general blood loss
Aims: This analysis evaluated AVA in cohorts with Baseline platelet
9 9 after total knee replacement (TKR).
counts (PC) alternately defined as <35×10 /L and ≥35×10 /L, and
9 Methods: The study included 35 patients undergoing TKR. In ac-
the proportion of patients who achieved both a PC ≥50×10 /L and
cordance with the preoperative level of β-­thromboglobulin (β-­TG),
increase of 20×109/L.
which is a specific protein of platelet alpha-­granules, all patients
Methods: Two randomized, double-­
blind, placebo (PBO)-­
were divided into 2 groups: low β-­TG (LTG-­group, β-­TG<25  U/ml) and
Phase 3 trials (ADAPT-­1 and ADAPT-­2) enrolled adults with CLD and PC
9 high β-­TG (HTG-­group, β-­TG≥25 U/ml). Characteristics of groups is
<50×10 /L undergoing a procedure. Patients were randomized 2:1 to
presented in Table 1.
AVA or PBO, and dosed orally for 5 days based on Baseline PC. For the
Platelet count, β-­TG level (“Asserachrom β-­TГ”), D-­dimer concen-
alternate analyses, patient data were divided into 2 Baseline PC cohorts:
9 9 tration (“Technozym D-­dimer ELISA”), thromboelastography (TEG
<35×10 /L and ≥35×10 /L. The primary efficacy endpoint remained the
5000), blood loss (intraoperatively and 24 hours after surgery) were
proportion of patients not requiring platelet transfusion or rescue pro-
determined. Blood samples were taken 1 or 2 days before operation,
cedure up to 7 days post-­procedure. An alternate secondary endpoint
9 after 30 minutes, at 1, 3 days after TKR. Comparison of the groups
analysis assessed the proportion of patients with a PC ≥50×10 /L and
9 was carried out using the Mann-­Whitney test. The study was ap-
an increase of ≥20×10 /L from Baseline during the study.
proved by a local ethics committee, and informed consent was ob-
Results: 435 patients were randomized in the 2 studies. AVA was
tained from each patient.
again shown to be superior to PBO in both cohorts in the number
Results: Total blood loss was significantly higher in LTG group
of patients not requiring a platelet transfusion or rescue procedure
9 9 (Table 1). At the end of operation, β-­TG activity and D-­dimer con-
(<35×10 /L: AVA-­ 63.4%, PBO-­ 26.7%; P<0.0001; ≥35×10 /L: AVA-­
centration were significantly higher in HTG group (Table 2).
83.0%, PBO 34.7; P<0.0001). The alternate secondary endpoint
Thromboelastography showed that reaction time (R) was signifi-
analysis demonstrated the superiority of AVA in the proportion of
9 9 cantly shorter and Alpha angle was significantly larger in HTG group
patients achieving a PC ≥50×10 /L and an increase of ≥20×10 /L:
9 9 at the end of TKR (Table 2). Platelet counts in LTG and HTG groups
<35×10 /L (AVA-­ 60.2%, PBO-­ 0%) and ≥35×10 /L (AVA-­72.0%,
did not significantly differ.
PBO 6.1%).
Conclusions: The preoperative platelet activity influences postop-
Conclusions: AVA was superior to placebo in increasing PC and re-
erative platelet activity, rate of clot formation, the amount of fibrin
ducing platelet transfusions in all evaluated Baseline PC cohorts.
and also affects the blood loss in large orthopedic surgery.

TA B L E   1   Characteristics of patients and surgery

Criterion LTG (n=17) HTG (n=18) P

Sex, (male / female)* 9/8 6/2 0.335

Age (years)** 62±9 60±8 0.606
Type of anesthesia 1/16 3/8 0.667
(general / regional)*
Total blood loss (ml)** 868±239 687±241 0.035

*Absolute number; Two-­tailed Fisher’s exact test; **Mean ± standard

­deviation; Mann-­Whitney test

TA B L E   2   Haemostasis parameters before and after total knee replacement

Alpha angle
Group β-­TG (U/ml) Platelets (×1000/µL) D-­Dimer (ng/ml) R (min) (degrees)

Before surgery LTG 19.8 [17.0; 22.0] 252 [186; 273] 145 [78; 169] 16.3 [12.9; 18.5] 37.4 [29.4; 43.7]
HTG 27.1 [26.2; 30.0]* 242 [214; 311] 98 [80; 123] 14.5 [11.8; 17.3] 38.6 [32.0; 49.1]
30 minutes LTG 38.2 [28.9; 65.0] 221 [182; 244] 398 [245; 708] 15.2 [12.0; 18.9] 38.9 [29.5 45.6]
HTG 78.6 [58.0; 105.1]* 210 [196; 285] 1203 [450; 1839]* 9.2 [8.3; 12.5]* 56.0 [50.3; 60.3]*
Day 1 LTG 32.1 [28.9; 59.4] 179 [156; 238] 739 [448; 1404] 15.2 [12.8; 20.3] 40.1 [34.8; 50.1]
HTG 47.0 [31.7; 70.7] 212 [184; 263] 1726 [905; 2398]* 12.0 [9.7; 15.8] 44.2 [39.6; 59.5]

Values expressed as median [interquartile range]; Comparisons between groups were assessed with Mann-­Whitney test, *P<0.05.

PB022 | Superiority of Avatrombopag to in PCs were also greater with AVA compared with PBO in both co-
horts (Cohort 1: AVA-­31.7×109, PBO-­1.8×109/L; Cohort 2: AVA-­
Placebo in Increasing Platelet Counts and
41.0×109, PBO 3.5×109/L). The most common TEAEs were pyrexia,
Reducing Platelet Transfusions in Patients
abdominal pain, nausea, and headache.
with Chronic Liver Disease-­associated Conclusions: AVA was superior to placebo in reducing the need for
Thrombocytopenia Undergoing Scheduled PLT transfusions, the proportion of patients achieving the target PC,
Procedures -­Pooled Analysis of 2 Randomized and increasing the PC.
Phase 3 Studies
F. Poordad1; L.F. Allen2; K. Aggarwal2; M. Vredenburg2;
N. Alkhouri1 PB023 | Atypical Presentations of RUNX1
Texas Liver Institute, San Antonio, United States, 2Dova Pharmaceuticals,
Durham, United States
Associated Familial Platelet Disorder with
Predisposition to Myeloid Malignancy (FPDMM)
Background: Thrombocytopenia (TCP) is common in patients with C. Ward; M.-C. Morel-Kopp; C. Tang; D. Rabbolini;
chronic liver disease (CLD) and complicates clinical management with W. Stevenson
these patients requiring multiple invasive procedures. Prophylactic University of Sydney, Kolling Institute of Medical Research, St Leonards, Australia
platelet transfusions are used to reduce bleeding risk, but have mul-
tiple associated safety risks. Background: RUNX1 pathogenic variants were the first to be
Aims: Avatrombopag (AVA), an oral thrombopoietin receptor ago- linked to myeloid malignancy, presenting as familial platelet de-
nist, was studied in 2 Phase 3 studies as an alternative to platelet fects. Recognition of RUNX1-­a ssociated FPDMM is hindered by
transfusions for patients with TCP and CLD. phenotypic heterogeneity including normal counts and variable
Methods: Two randomized, double-­blind, placebo (PBO)-­controlled bleeding. The condition is regarded as autosomal dominant, but
Phase 3 trials (ADAPT-­1 and ADAPT-­2) enrolled adults with CLD with incomplete penetrance; delays between diagnosis and leu-
and platelet count (PC) <50×109/L undergoing a procedure. Patients kaemia can exceed 50 years. The spectrum of associated cancers
were divided by Baseline PC into Cohort 1-­PC <40×109/L and extends beyond MDS and acute myeloid leukaemia to lymphoid
Cohort 2-­PC 40 to <50×10 /L, and randomized 2:1 to once-­daily malignancies.
oral AVA (Cohort 1-­6 0 mg; Cohort 2-­4 0 mg) or PBO for 5 days, with Aims: Our laboratory genotypes of suspected FPD cases as a human
the procedure 5 to 8 days after the last dose. The primary efficacy ethics approved research project.
endpoint was the proportion of patients not requiring a platelet Methods: We have recently identified four kindreds with RUNX1
transfusion or rescue procedure. Secondary endpoints assessed mutations and atypical clinical features.
the proportion of patients achieving the target PC (≥50×109/L) by Results: The first index case is a 22 year old male with thrombo-
Procedure Day, change in PC from Baseline to Procedure Day, and cytopenia (platelets 126×109/L), presenting with T lymphoblas-
safety. tic lymphoma associated with a large deletion encompassing the
Results: 435 patients were randomized in the 2 studies. AVA was RUNX1 promoter. His mother had ALL aged 29 and was in remis-
superior to PBO in both cohorts in the number of patients not re- sion (platelets 70); other family members developed acute leukae-
quiring a platelet transfusion or bleeding rescue (Cohort 1: AVA-­ mia. The second case (RUNX1 c.C884G, p.S295X) is a 43 year old
66.9%, PBO-­28.6%; P<0.0001; Cohort 2: AVA-­88.0%, PBO 35.8%; woman presenting with worsening thrombocytopenia and massive
P<0.0001). A greater number of AVA-­treated patients also achieved splenomegaly due to chronic eosinophilic leukaemia with an ac-
the target PC in both Cohort 1 (AVA-­68.1%, PBO-­5.5%; P<0.0001) quired KIT mutation. She responded to therapy including an alloge-
and Cohort 2 (AVA-­90.6%, PBO-­29.9%; P<0.0001). Mean increases neic SCT, but her mother died of acute myelomonocytic leukaemia
in her 70s. The third family (RUNX1 c.187delG, p.V63fs) is a father once daily for 5 days for patients with Baseline platelet counts be-
(aged 44) and son with easy bruising. The father’s platelet counts tween >40 and <50×109/L.
fell over time to 60×10 /L, whereas his 5 year old son has normal Conclusions: Phase 2 PK/PD data with AVA contributed to develop-
numbers; both their platelets are extremely small, with unrecordable ment of a predictive model for the selection of Phase 3 dosing regi-
MPVs. There is a distant relative with leukaemia but no recent cases mens, which achieved optimal efficacy while limiting the proportion
of malignancy. Finally, a fourth male (37) has been identified with mi- of patients achieving platelet counts >200×109/L and its associated
crothrombocytopenia associated with a reported RUNX1 pathogenic risk of thromboembolic events.
variant (c.G514A, p.G172R).
Conclusions: These cases illustrate the heterogeneity of FDPMM
including multi-­lineage malignancy and progressive thrombocytope- PB025 | Gray Platelet Syndrome Variant:
nia. Genotyping is essential for FDPMM diagnosis and monitoring
Combined Alpha Granule, Dense Granule, and
strategies are needed.
GPVI Deficiency
E. Salazar; P. Monroig-Bosque; L. Rice
PB024 | Population PK/PD Modeling Houston Methodist Hospital, Houston, United States

of Avatrombopag in Patients with

Background: Gray platelet syndrome (GPS) is a rare disorder char-
Thrombocytopenia and Chronic Liver Disease -­ acterized by platelet alpha-­granule deficiency. Thrombocytopenia,
Guided Phase 3 Dose Selection pale platelets on peripheral smear, progressive myelofibrosis, and
M. Nomoto1; Z. Hussein1; K. Aggarwal2; M. Vredenburg2; variable bleeding tendency may be present. NBEAL2 mutations are
L.F. Allen2 causative.
1 2
Eisai Pharmaceuticals, Woodcliff Lake, United States, Dova Pharmaceuticals, Aims: We present a case of GPS variant, characterized by predomi-
Durham, United States nant alpha granule deficiency, but with moderately decreased dense
granules and GPVI deficiency.
Background: Avatrombopag (AVA) is a novel oral thrombopoietin Methods: An extensive diagnostic workup for platelet function de-
receptor agonist being developed for the treatment of thrombocy- fects was performed.
topenia in patients with chronic liver disease (CLD) scheduled to un- Results: A 37 year old Mexican male presented to Houston
dergo an invasive procedure. Methodist Hospital with a history of progressive easy bruising and
Aims: Phase 2 population pharmacokinetic (PK) and PK/pharma- a previous episode of profuse hematuria and blood in stool. A prior
codynamic (PK/PD) data for avatrombopag were analyzed, and a appendectomy was complicated by major hematoma. Family history,
model developed to characterize the relationship between AVA including 2 daughters and 1 son, was unremarkable.
concentration and platelet count increase. This model was then Platelet aggregation studies showed decreased aggregation in re-
used to identify Phase 3 dosing regimens that maximized the pro- sponse ADP (14%), arachidonic acid (0%), collagen (0% and 5% for
portion of subjects who would achieve the target platelet effect, an low-­and high-­
dose collagen, respectively), low-­
normal aggrega-
increase to ≥50×109/L, while minimizing those with platelet counts tion in response to epinephrine (15%), and normal aggregation in
≥200×109/L. response to low-­and high-­dose ristocetin (1% and 78%, respec-
Methods: Population PK and PK/PD data with AVA were analyzed tively). Lumiaggregometry assessing ATP release revealed similar
using non-­linear mixed effects modeling (NONMEM). Two cohorts results. Platelet function analysis-­100 was prolonged in both the
of patients were sequentially dosed: Cohort A-­AVA was adminis- ADP and epinephrine cartridges (135 seconds and >300 seconds,
tered orally with a 100 mg loading dose followed by 6 days of 20 mg, respectively). Thromboelastograph showed a low-­normal maximum
40 mg, or 80 mg; Cohort B-­an 80 mg oral AVA loading dose was amplitude of 53.9 millimeters. Peripheral smear showed thrombo-
followed by 10 mg for 6 days or 20 mg for 3 days. cytopenia (90,000/microliter) with mostly pale platelets, although
Results: PK/PD model simulations showed that the achievement of some platelets were granular.
the targeted platelet count and a high response rate was importantly Transmission electron microscopy showed markedly decreased
affected by the Baseline platelet count; therefore, dosing of ava- alpha granules and moderately decreased dense granules. Flow cy-
trombopag for Phase 3 was best stratified by Baseline platelet count. tometry identified markedly decreased surface GPVI receptor (17%),
The simulations also demonstrated no advantage with respect to the and normal levels of GPIIb, GPIIIa, GPIX, GPIb-­alpha, and GPIa.
effect on platelet count response in regimens that included a loading Conclusions: Various platelet defects may be identified along with
dose. Based on these modeling simulations, the recommended opti- decreased alpha granules in patients with pale platelets on periph-
mal dosing regimens for Phase 3 were 60 mg once daily for 5 days eral smear. Detailed molecular studies are required for definitive
for patients with Baseline platelet counts <40×109/L, and 40 mg characterization and diagnosis.

PB026 | Bloody Diarrhea with mechanism: one states the existence of an immunologic process
and the second poses a non-­immunologic adherence mediated by
Thrombocytopenia in Male Children: Think of
Wiskott Aldrich Syndrome
Aims: Present a case-­report of PS in a patient diagnosed with chronic
D. Suri1; A. Kumar Jindal1; Rashmi1; J. Kumar Shandilya1; secondary immune thrombocytopenia (ITP).
P. Vignesh1; J. Ahluwalia2; A. Gupta1; A. Rawat1; Methods: A 59-­year-­old male patient was diagnosed in 2010 with
S. Singh1
ITP. The patient was previously diagnosed in 2004 with B-­cell non-­
PGIMER, Department of Pediatrics, APC, Chandigarh, India, 2PGIMER,
Hodgkin lymphoma and was in complete remission after chemother-
Department of Hemato Pathology, Chandigarh, India
apy and splenectomy since 2008.
Results: The complete blood count showed no alterations other
Background: Wiskott-­Aldrich syndrome (WAS) is a rare X-­linked pri-
than thrombocytopenia (56 G/L). Due to clinical history a blood
mary immunodeficiency disorder caused by mutation in WAS gene.
smear was analyzed. The blood smear obtained from EDTA anti-
Clinical manifestations vary from mild, isolated micro thrombocy-
coagulated blood revealed the presence of 4-­9 platelets attached
topenia as in X-­linked thrombocytopenia to full-­blown presentation
circumferentially to most neutrophils. No platelet aggregation
that can be complicated by life-­threatening hemorrhages, immu-
around other cells or isolated was seen. The estimated counting of
nodeficiency, atopy, autoimmunity, and cancer. Early recognition
platelets in the blood smear was 110 G/L. The observations were
and institution of hematopoietic stem cell transplant can improve
consistent through the time. Several laboratory tests were per-
formed to study the PS. Antiplatelets antibodies and cryoglobu-
Aims: To describe clinical presentation in children with XLT/WAS.
lins screening was negative. Sedimentation rate was persistently
Methods: A review of case records of all children diagnosed as XLT/
below 5 mm/hr. Antinuclear antibody investigation and direct and
WAS and registered in Pediatric Immune deficiency clinic from the
indirect Coombs tests were negative. Serum protein electropho-
year 2007 to 2017 was carried out.
resis and immunofixation show no alterations. Serology for her-
Results: A total of 42 patients with confirmed WAS gene muta-
pes simplex, Epstein-­B arr, hepatitis and human immunodeficiency
tions were analyzed. Incidence of bleeding manifestation before
virus, cytomegalovirus, toxoplasma, and Treponema pallidum was
diagnosis was seen in 90% of patients and the most common
presenting symptom was blood stained stools in 60% patients
Conclusions: PS constitutes a cause of pseudothrombocytopenia
followed by skin bleeds in 35% children. Three patients were ini-
and if not identified can lead to inappropriate treatment. Contrary to
tially diagnosed as cow’s milk protein intolerance, 2 children as
the literature, PS was maintained over the years. The identification
dysentery and 3 children as disseminated cytomegalovirus infec-
of PS was only possible due to blood smear examination since no flag
tion. Evidence of infections over follow up were found in 31 chil-
other than the associated thrombocytopenia was displayed by the
dren while eczema in 36 children. Microthrombocytopenia was
hematology analyzer.
an important clinical clue to the diagnosis and ascertaining mean
platelet volumes helped clinch the diagnosis. Reduced expression
of intracytoplasmic Wiskott Aldrich Syndrome Protein was seen in
only 57% children. PB028 | Coagulation Abnormalities in Dengue
Conclusions: We found bleeding from the GI tract in the form of and Dengue Hemorrhagic Fever Patients in
bloody diarrhea as the most consistent symptoms and were seen Pakistan
in 60% of the children. This is also probably the earliest symptom
J. Hassan; M. Borhany; M. Abid; N. Fatima; T. Shamsi
which should raise the suspicion of this disorder among the treating
National Institute of Blood Disease & Bone Marrow Transplantation, Karachi,

Background: Dengue fever is endemic in Pakistan with seasonal

PB027 | Pseudothrombocytopenia: Beyond rise in cases. Morbidities and mortalities are proportionately
Platelets Aggregates reported to be increasing in number and associated with dis-
seminated intravascular coagulation resulting in hemorrhagic or
A. Carmo1; N. Costa e Silva1; M. Gomes2
1 thrombotic manifestations in patients having deranged coagulation
Centro Hospitalar e Universitário de Coimbra, Clinical Pathology Department,
Coimbra, Portugal, 2Centro Hospitalar e Universitário de Coimbra, Clinical profiles.
Hematology Department, Coimbra, Portugal Aims: The aim of this study was to assess abnormality of coagulation
and anticoagulation parameters in dengue fever and its effect on the
Background: Platelet satellitism (PS) is the in vitro adherence patient’s outcome.
of platelets to leucocytes. It is observed on blood smears pre- Methods: This is an observational analytical study, conducted at
pared from blood collected in EDTA and usually it is absent with the National Institute of Blood disease Karachi during the year
citrate or heparin. Two hypothesis were posed to explain the 2016-­17. All age group patients were included in this study and

TA B L E   1   Comparison of CBC parameters at Baseline and Follow up PB030 | Acquired Platelet Function Defect
Follow up and Moderate Cutaneous Bleeding in a Patient
Baseline Mean Mean and on Treatment with a Leukotriene Receptor
and Standard Standard
Parameters Deviation Deviation P-­Value Antagonist for Asthma
Hemoglobin 14.14±2.37 13.49±1.81 0.000 M. Marchetti; T. Lerede; M. Testa; S. Brevi; S. Gamba;
Total Leucocyte 4.83±2.73 6.70±7.84 0.001
A. Falanga
Count Papa Giovanni XXIII Hospital, Immunohematology and Transfusion Medicine,
Bergamo, Italy
Platelet 48.65±40.30 88.8±69.26 0.000
Neutrophils 40.82±19.61 35.74±16.21 0.002
Background: Acquired defects of platelet function are rare hemor-
Lymphocytes 40.81±18.83 43.77±17.64 0.046
rhagic diseases. We described here the case of a 33-­year-­old woman
Hematocrit 42.64±6.19 40.57±2.94 0.000
with a history of allergic asthma referred to our ambulatory on
October 2017 for the persistent (about 2 years) and diffused pain-
less ecchymotic lesions on legs and arms. No anti-­platelet therapies
TA B L E   2   Mean values of coagulant and anticoagulant was taking and she denied any trauma. Family history for coagulopa-
parameters in patients presented with <50000 and >50000 Platelet thies was negative. Blood film, platelet count and coagulation values
count were normal.
Aims: To highlight this rare drug-­induced bleeding disorder.
Platelet count Platelet count
Parameters <50000 Mean±SD >50000 Mean±SD Methods: Platelet aggregation was studied in platelet rich plasma
(PRP) by light transmission aggregometry (LTA) using ADP, collagen,
Prothrombin time 14.62±0.379 14.55±0.354
adrenaline, arachidonic acid (AA) and ristocetin. The PFA-­100 was
Activated partial 40.14±0.834 39.70±1.235
thromboplastin time used with collagen/ADP and collagen/adrenaline cartridges. Platelet
membrane glycoproteins (GPIb, GPIIb/IIIa, GPIIIa), and P-­selectin
Fibrinogen 2.461±0.619 2.629±0.118
were measured by flow cytofluorimetry.
D-­Dimer 2.185±0.141 1.92 1±0.251
100 ® was found with
Results: Prolonged closure time by PFA-­
Fibrin monomers 17.25±2.226 14.83±2.859
epinephrine cartridge only. VWF was normal. Platelet GP and
Antithrombin III 90.84±2.210 88.91±3.131
P-­s electin expression were normal. Platelet agglutination by ris-
Protein C 68.90±1.718 71.47±2.745
tocetin was normal, while platelet aggregation was impaired with
Protein S 55.37±2.244 60.33±3.431 collagen (reduced, irreversible), adrenaline (reduced, irreversible),
ADP (reduced, reversible), and absent with AA (no ATP release).
This type of response was suggestive for “Aspirin-­like defect”, an
dengue status proven by serology. Statistical analyses were done inherited dysfunction of AA metabolism. However, the patient
by SPSS version 23. was on treatment with montelukast, a leukotriene receptor an-
Results: Total 200 patients were selected with mean age of tagonist, for her asthma. In the literature, we found a report of a
28.68 (±13.28) and male predominance (147/200). Mean platelet woman who during assumption of this drug showed similar signs of
counts at baseline were 49000 and platelets were transfused to bleeding that resolved after montelukast discontinuation. In order
76 (38%) patients. Bleeding was present in 44 (22%) patients with to exclude in our case a similar drug side-­effect, the treatment
a mean platelet count of 34000 (P-­value 0.005). Comparison of was withdrawal. After 10 days, the bruising resolved and platelet
CBC parameters at Baseline and Follow up is shown in Table 1. function by PFA 100 and LTA provided normal results. At today
dimers showed significant association (P-­
value 0.005) in she had no relapsed.
coagulation group where as Protein S showed significant as- Conclusions: The clinicians should be aware about this very rare
sociation (P-­
value 0.03) in anticoagulant group respectively, side-­effect of montelukast, taking also into account the high fre-
when categorized the patients in platelet counts <50 and >50 quency of its prescription.
(Table 2).
Conclusions: A significant association was found with coagulation
and anticoagulation parameters in patients presented with bleeding.
To our knowledge, this is the first study incorporating a spectrum
of coagulation and anticoagulation parameters in dengue fever in

PB031 | Pelvic Hematoma Secondary to assessment in these patients is seldom evaluated which is very im-
portant for their well-­being and existence.
Glanzmann’s Thrombasthenia: A Case Report
Aims: The aim of this study was to assess the health related quality
S.L. Hon1; J. Sathar1; V. Selvaratham1; F. Leow1; of life (HRQoL) found in GT patients; correlate with their phenotype
M. Ganesalingam2
and to explain clinical and psychosocial factors that affect their daily
Hospital Ampang, Hematology Unit, Selangor, Malaysia, 2Hospital Ampang,
activities and medical care.
Obstetrics and Gynaecology Unit, Selangor, Malaysia
Methods: 20 patients with GT were selected for the study after
obtaining informed and written consent. History was recorded in a
Background: Glanzmann’s thrombasthenia (GT) is an autosomal re-
structured data sheet and bleeding score was assessed in a question-
cessive bleeding disorder characterized by defect in platelet integrin
naire based on MCMDM-­1VWD Bleeding Questionnaire. HRQoL
α IIbß3 (previously known as GPIIb/IIIa). It is a rare genetic platelet dis-
assessment questionnaire (Likert scale type) was based on differ-
order characterised by deficient or dysfunctional α IIbß3, The clinical
ent aspects of their well-­being and functioning, their social status,
presentation is usually recurrent and spontaneous mucocutaneous
as well as their interpretations on medical care. Flowcytometry was
bleeding, which can range from mild to life-­threatening haemorrhage
performed on BD FACSCALIBUR (using CD 41, 61, 42a, 42b antibod-
independent of the type of defects.
ies BD Biosciences) to classify the GT patients.
Aims: This case illustrates a rare disease which can present with sig-
Results: GT type I had the highest average bleeding score and scored
nificant bleeding tendencies.
the highest points on the severity assessment for poor healthcare,
Methods: We described a case of GT presenting with concealed pel-
social deprivation, lack of education and self-­esteem followed by GT
vic hematoma resulting in obstructive uropathy and significant blood
type II. GT type III scored least points.
loss which required blood transfusions and surgical intervention.
Conclusions: Patient with GT type I were the worst affected need-
Results: A multidisciplinary meeting involving gynaecologist, anaes-
ing revision in medical care and improving social support and right
thesiologist was arranged. She had undergone surgical intervention
to education.
without any complications after a careful optimisation of her hemo-
globin level with parenteral iron and erythropoietin, and definite
treatment with Recombinant factor VIIa and antifibrinolytic agent
peri-­and post-­operatively. She remained transfusion free through- PB033 | Platelet’s Disorder in a Case of
out her admission. Chronic Lymphocytic Leukemia
Conclusions: With the appropriate measures, we have reduced the
C. Lodigiani; I. Quaglia; P. Ferrazzi; L. Librè; M. Bacci
possible post-­surgical complications and avoid unnecessary blood
Humanitas Research Hospital, Thrombosis and Haemorrhagic Diseases Center,
transfusion. The use of continuous combined hormonal therapy en- Rozzano, Italy
sures anovulation preventing further bleeding into the pelvis.

Background: Bleeding complications are frequent in patients with

Chronic Lymphocytic Leukemia (CLL) and tend to be related to
PB032 | Health Related Quality of Life thrombocytopenia or an acquired clotting factor inhibitor.
Assessment in Patients with Glanzmann Aims: A 65-­year-­old white man affected by CLL was admitted at our
Hospital for recurrent intracranial hemorrhages (ICH). He was inves-
Thrombasthenia from Selected Regions of
tigated for hemorrhagic diathesis and was diagnosed δ-­storage pool
disease. He started therapy with bendamustin 90 mg/m2 and rituxi-
M.Y.J. Siddiqi1; A. Naz1; A. Najmuddin2; A. Imran3; T.S. mab 375 mg/m2/28 days and he was also treated with antifibrino-
Shamsi1 lytic tranexamic acid (1 g/daily) to prevent bleedings. At the end of
National Institute of Blood Disease & Bone Marrow Transplantation,
the therapy the patient was revaluated.
Coagulation and Hemostasis, Karachi, Pakistan, 2Fatimid Foundation,
Coagulation and Hemostasis, Karachi, Pakistan, 3Chughtai Lab Lahore, Methods: We first screened all blood clotting assays; secondly we
Coagulation and Hemostasis, Lahore, Pakistan studied platelet’s function including aggregation to different ago-
nists, surface antigens in resting and activated platelets and Delta-­
Background: Glanzmann thrombasthenia (GT) is one of the com- granules contents by flow cytometry. Mixing assays were also
monest disorders among inherited platelet functional defects. It is performed adding plasma and serum of the patient to platelet’s rich
a familial disorder being autosomal recessive in character. The de- plasma of an healthy control.
fect is caused by mutations in the genes encoding ITGA2B or ITGB3 Results: Before starting therapy, all coagulation assays were normal;
located on chromosome 17, resulting in qualitative or quantitative platelet aggregation to ADP 10 μM, TRAP 10 μM and epinephrine
abnormalities of the expression of αIIb-­β3 integrin, a fibrinogen 10 μM were reduced (56%, 15%, 23% vs. 70%, respectively), to
receptor positioned on platelets. It is characterized by a bleeding collagen showed mild decrease, to ristocetin and arachidonic acid
diathesis which is variable. Incidence is increased in those territo- showed normal function before and during the therapy. Surface
ries where consanguineous marriages are practiced. Quality of life GpIbα, GpIIb and α-­granules contents were normal, we observed a

decrease in δ-­granules contents (45% vs. 70%) suggesting a diagno- PB035 | Dichotomous Cytosolic Mobilization
sis of δ-­storage pool. Mixing tests seem to exclude an autoimmune
of Calcium, Sodium and Potassium Ions during
mechanism. The assays performed at the end of the treatment,
Procoagulant COAT Platelet Formation
documented an improvement of platelets response to ADP, TRAP,
ephinephrine and δ granule’s contents (60%). A. Aliotta; D. Bertaggia Calderara; L. Alberio
Conclusions: Our analysis confirmed the clinical suspicion of a plate- Division of Hematology and Central Hematology Laboratory, CHUV, University
Hospital and University of Lausanne, Lausanne, Switzerland
lets disorder, as δ-­storage pool, related to CLL in a patient with ICH.
We documented an improvement in platelet function associated to
Background: The combined activation of platelets (PLTs) with colla-
an apparently clinical resolution of bleeding diathesis, subsequent
gen and thrombin induces the formation of procoagulant COAT PLTs,
to the treatment of the hematological malignancy. We would like to
which retain a coat of prohemostatic α-­granule proteins on their sur-
confirm these data after a longer follow-­up.
face. While the clinical relevance of COAT PLTs is increasingly rec-
ognized, dichotomous intracellular signaling pathways generating a
subpopulation displaying procoagulant activity instead of aggrega-
PB034 | Platelet CXCL14 Regulates Thrombotic tion endpoints are still not fully elucidated.
Function and Influences Clinical Outcome in Aims: As cytosolic ion fluxes play an important role in PLT activation,
Coronary Artery Disease here we propose a continuous flow cytometry kinetic monitoring of

A. Witte; D. Rath; M. Gawaz calcium, sodium and potassium fluxes during generation of proco-
agulant COAT PLTs.
Universitätsklinik Tübingen, Cardiology, Tübingen, Germany
Methods: PLTs were preloaded with different ion fluorescent indica-
tors: Fluo-­3 AM for calcium, Asante NaTRIUM Green-­2 AM for so-
Background: Platelets are the source of a variety of chemokines with
dium and Asante Potassium Green-­2 AM for potassium. Annexin-­V
immune-­modulatory mediators. Recently, we could show that plate-
co-­staining allowed to identify COAT PLT subpopulation. PLTs were
lets are a source of CXCL14 and release it upon activation. Platelet
simultaneously activated with thrombin and convulxin (collagen re-
derived CXCL14 induces monocyte migration and counteracts the
ceptor GPVI agonist) and each ion indicator fluorescence was con-
angiogenic effect of VEGF on HUVECs, which demonstrates its pro-­
tinuously acquired over time on an Accuri C6 flow cytometer.
inflammatory activity.
Results: Annexin-­V co-­staining revealed a dichotomous cytosolic
Aims: The current study investigated the thrombotic platelet func-
ion mobilization. Procoagulant COAT PLTs demonstrated high and
tions in global CXCL14 deficient mice. Furthermore, the surface ex-
sustained cytosolic calcium concentrations (up to micromolar range),
pression of CXCL14 on platelets was analyzed for prognostic impact
decreased sodium concentrations after a transient increase, and
in coronary artery disease patients.
important potassium efflux. In non-­COAT PLTs, calcium levels in-
Methods: Flow cytometry, immunoblot, luminescence aggregom-
creased transiently in the nanomolar range and afterwards declined
etry and ex vivo flow chamber assay were performed.
over time, sodium levels increased and stayed stable, and potassium
Results: In luminescence aggregometry experiments, platelets of
efflux was lower compared to COAT PLTs.
CXCL14 deficient mice showed decreased ATP release after activation
Conclusions: We demonstrated a characteristic dichotomous mo-
with thrombin, which indicates an impaired activation potential and
bilization patterns following PLT activation with thrombin and
an impaired release of dense granule content. In ex vivo flow chamber
convulxin. This work reveals peculiar ion flux kinetics during pro-
assay, CXCL14 deficient mice showed decreased thrombus formation
coagulant response, diverging from an initial common aggregating
on collagen coated surface under a high shear rate which could be com-
response. This method allows to investigate the differential modula-
pensated with supplementation of the blood with recombinant CXCL14.
tion of biological messengers leading to procoagulant response.
These results indicate that CXCL14 might have a regulatory impact on
thrombotic functions. A combined endpoint study of a patient cohort
of 80 CAD patients showed a better survival when assessed for major
adverse cardiac events (MACE; all-­cause death, myocardial infarction,
ischemic stroke) in patients with higher platelet CXCL14 surface ex-
pression on platelets than median, compared to the patient subgroup
with a lower CXCL14 surface expression level on platelets.
Conclusions: Current results show that CXCL14 might be involved
in regulating platelet function. An increased surface expression of
CXCL14 on platelets in CAD patients is associated with an adverse
clinical outcome. Given its previously described pro-­inflammatory
effects, CXCL14 may be an interesting target for new therapeutic

PB036 | Emerging Value of Platelet Function Conclusions: Emerging data from small-­medium sized studies indi-
cate that HTPR-­status may predict outcomes on antiplatelet ther-
Testing at Predicting the Risk of Recurrent
apy in CVD. Adequately-­sized, prospective, multicentre studies are
Vascular Events and Outcomes after TIA/
needed to determine whether altering antiplatelet therapy in pa-
Ischemic Stroke: A Systematic Review of the tients with HTPR on platelet function tests reduces the risk of recur-
Literature rent vascular events, functional outcomes or survival following TIA/
S.T. Lim1; S. Murphy1; I. Fernandez-Cadenas2; J. Montaner3; ischemic stroke.
V. Thijs4; L. Marquardt5; P. Kelly6; P. Bath7; G. Ford8; B.
Norrving9; D. Cox10; D. McCabe11; OATS-I
Trinity College Dublin, Dublin, Ireland, 2Institut de Recerca Vall d’Hebron,
Universitat Autònoma de Barcelona, Department of Neurology, Barcelona,
PB037 | Lysis of Blood Cells by Arachidonic
Spain, 3Neurovascular Research Lab, Vall d’Hebron Research Institute, Acid Cause ADP-­dependent Platelet Activation
Hospital Vall d’Hebron, Barcelona, Spain, 4Austin Health and Florey Institute
of Neuroscience and Mental Health, Department of Neurosciences, Victoria,
Responses in Platelet Function Tests using Whole
Australia, 5Asklepios Klinik Wandsbek, Department of Neurology, Hamburg, Blood
Germany, 6Mater University Hospital, Neurovascular Clinical Science Unit,
Dublin, Ireland, 7University of Nottingham, Stroke Trials Unit, Nottingham, S. Ramstrom1,2
United Kingdom, 8Oxford University Hospitals NHS Foundation Trust, Oxford, 1
Örebro University, School of Medical Sciences, Örebro, Sweden, 2Linköping
United Kingdom, 9Lund University, Faculty of Medicine, Lund, Sweden, University, Department of Clinical Chemistry and Department of Clinical and
Royal College of Surgeons in Ireland, Department of Molecular and Cellular Experimental Medicine, Linköping, Sweden
Therapeutics, Dublin, Ireland, 11Trinity College Dublin, Academic Unit of
Neurology, Dublin, Ireland
Background: The use of arachidonic acid (AA) to stimulate platelets
is considered the most specific approach to study aspirin treatment
Background: The value of testing for ‘high on-­treatment platelet re-
efficacy. However, very high concentrations of AA are used, and it
activity’ (HTPR) to predict outcomes in TIA/ischemic stroke patients
has been previously reported that AA can induce cell lysis. Several
on antiplatelet therapy is unclear.
clinical studies have reported decreased responses to AA in whole
Aims: The role of ex-vivo platelet function/reactivity testing at
blood tests in the presence of clopidogrel.
Aims: To investigate whether cell lysis could lead to unspecific ef-
fects contributing to platelet activation in situations where AA is
(a) the risk of recurrent vascular events and
used as platelet agonist.
(b) neurological outcomes following TIA or stroke.
Methods: Blood from healthy donors was collected after informed
consent in a procedure approved by the local ethics committee.
Methods: A systematic literature review was performed to collate
Sub-­samples were collected during AA-­induced platelet activation n
relevant data on the relationship between ex-vivo HTPR testing and
light transmission aggregometry (LTA) in citrated platelet-­rich plasma
recurrent vascular events or outcomes in ischemic cerebrovascu-
(PRP), and in two assays using hirudinized whole blood, multiple
lar disease (CVD) patients on antiplatelet therapy. We focused on
electrode aggregometry (MEA, Multiplate®) and flow cytometry.
data from commonly-­available whole blood platelet function analys-
The appearance of platelet and red blood cell fragments as well as
ers and platelet aggregometry. P<0.05 was considered statistically
activated platelets was detected using flow cytometry.
Results: We found that cell lysis, especially of red blood cells, does
Results: Sixty-­seven of 1168 articles were assessed; 23 met in-
occur at the concentrations of AA used in the clinically used tests
clusion criteria: PFA-­100 ® (N=9); VerifyNow® (N=8); Multiplate®
® (0.5-­1 mM) and that this induces ADP-­dependent aggregation re-
(N=1); Thrombelastograph Haemostasis Analyser (N=1); TOA-­
sponses. In flow cytometry, very limited platelet activation was
100 (N=1), and Aggregometry (N=5). Five studies found a
detected before reaching AA concentrations where cell lysis also oc-
higher risk (P<0.001), two found no increase in the risk of recurrent
curred, making it problematic to develop a reliable flow cytometry
stroke (P=1.0) between patients with vs. those without Aspirin-­
test using AA as platelet agonist.
HTPR, whereas seven were inconclusive. Aspirin-­HTPR was as-
Conclusions: We conclude that cell lysis and ADP release contrib-
sociated with more severe strokes at baseline (P=0.005; N=4),
ute to AA-­induced platelet responses, most markedly in whole blood
early deterioration (P=0.017; N=3), poorer outcomes (P=0.047;
assays. This finding could potentially explain some differences be-
N=2), larger infarct size (P<0.001; N=2), new DWI lesions (P=0.04;
tween studies comparing methods using whole blood and platelet
N=1) and higher mortality (P=0.038; N=1) after stroke. There
rich plasma and also how clopidogrel treatment could influence AA-­
was no relationship between Dipyridamole-­or Clopidogrel-­HTPR
induced aggregation results in previously published studies. Our
status and outcomes in 2 small studies (PFA-­100); Clopidogrel-­
findings highlight some issues with AA as reagent for platelet activa-
HTPR was associated with recurrent stroke in one study
tion, which also have an impact on how platelet activation assays
using AA should be interpreted.

PB038 | Assessment of Platelet Count Conclusions: This study showed that PRP platelet counts without
adjustment were ≥2X concentrated than WB and that count is criti-
Standardization in Platelet-­rich Plasma for
cal for evaluation of aggregation response and secretion. Platelet
Platelet Function Testing by Light Transmission
count adjustment played a greater role with lower dose agonists.
Aggregometry Thus, when evaluating platelet function, it is recommended that
1 1 1,2
M. Dixon ; J. Kotha ; L.K. Jennings platelet counts are standardized, especially if platelet reactivity may
CirQuest Labs, LLC, Memphis, United States, 2University of Tennessee Health be influenced by granule secretion.
Science Center, Internal Medicine, Memphis, United States

Background: The optical density of platelet-­rich plasma (PRP) is di-

PB039 | Kinetic Application of Flow Cytometry
rectly proportional to platelet concentration. As density increases,
platelet collisions increase resulting in a greater rate and extent of
for Patients with Unknown Bleeding Disorders:
platelet aggregation. Granule secretion is based on platelet density The Underestimated Value of Mepacrine
and serves to stabilize platelet aggregates. G. Manukjan1; O. Andres2; H. Schulze1
Aims: To evaluate if standardized PRP counts provide more con- 1
Institute of Experimental Biomedicine, University Hospital of Würzburg,
sistent platelet function data when compared to non-­standardized Würzburg, Germany, 2University Children’s Hospital, University of Würzburg,
Würzburg, Germany
counts and to establish parameters that ensure minimal ex vivo
platelet reactivity during processing.
Methods: Healthy donor blood (WB) with varying platelet counts Background: Many patients with bleeding diathesis do not obtain a

was obtained by routine venipuncture (n≥5). PRP was prepared by diagnosis even after expanded testing. After a coagulation disorder

centrifugation (150× g for 15 minutes); PPP was generated from is excluded, platelet defects are estimated mostly by aggregometry

residual blood (2500× g for 15 minutes). Platelet activation was as- or luminometry or fast and easy-­to-­use, but also limited point-­of-­

sessed by measuring nmoles ATP released. PRP platelet counts ad- care machines such as VerifyNow or PFA-­100. Flow cytometry has a

justed to 100-­300×106/mL using autologous PPP. Agonists TRAP (5, strong operating power allowing function testing of resting and acti-

10, 15 μM) or ADP (2, 5, 10 and 20 μM) were used. vated platelets at single cell level with only minute amounts of blood.

Results: PRP preparation resulted in a 2.2-­fold increase in platelet With this approach, platelet degranulation could easily be quanti-

density (neat PRP) when compared to WB. While maximal aggre- fied. While for alpha granules a good marker (P-­selectin; CD62P) is

gation (MA) response at higher ADP or TRAP concentrations was established, the release of dense granules and their content can only
not affected by platelet counts (neat -­250×10 /mL), response was be judged indirectly (i.e. by CD63) or by mepacrine.

highly dependent upon count at lower doses. Increased response Aims: We developed and evaluated a kinetic mepacrine assay, with
variability was noted at ≤200×10 /mL. Granule secretion was sig- which we measure its uptake into dense granules and release in

nificantly dependent upon platelet count even when there was no response to agonists over time. In contrast to endpoint measure-

difference in MA response. Marked differences were observed at ments we intended to obtain refined insight into the loading capac-
platelet counts <250×10 /mL. Granule secretion was not detected ity of platelet dense granules and the feasibility and dynamics of

due to sample processing. TRAP data are in Tables 1A and B. degranulation.

TA B L E   1 A   MA and Secretion is Dependent upon Platelet Count and Agonist Strength

TRAP (μM) Neat PRP Neat PRP 250×106/mL PRP 250×106/mL PRP P value P value

% MA nmoles ATP % MA nmoles ATP % MA nmoles ATP

15 90±4 3.53±0.80 87±2 1.64±0.43 0.278 0.003

10 86±4 3.46±0.89 81±5 1.32±0.38 0.134 0.002
5 84±6 3.08±0.89 30±31 0.34±0.67 0.009 0.001

TA B L E   1 B   MA and Secretion is Dependent upon Platelet Count and Agonist Strength

TRAP μM Neat PRP Neat PRP 150×106/mL PRP 150×106/mL PRP P value P value

% MA nmoles ATP % MA nmoles ATP % MA nmoles ATP

15 90±4 3.53±0.80 76±9 0.66±0.31 0.025 <0.001

10 86±4 3.46±0.89 48±22 0.31±0.34 0.010 <0.001
5 84±6 3.08±0.89 8±6 <0.01 <0.001 <0.001

Methods: Platelets were loaded with 5 μM mepacrine in whole using the Pf4-Cre transgene, respectively. Surface levels of the re-
blood and CD41-­positive events were measured by flow cytometry. ceptor tyrosine phosphatase CD148 were reduced similarly by 94-­
Upon platelet activation using 10 μM TRAP-­6, mepacrine release 95% using either strain. Direct comparison of Csk, Shp1 and CD148
was recorded over five minutes. The difference in MFI values indi- conditional KO mice generated using either deleter strain revealed
cated storage and release capacities. similar platelet phenotypes. However, additional inflammatory and
Results: We evaluated our kinetic approach on a cohort of 50 child- immunological anomalies were observed in Pf4-Cre-­generated KO
hood storage pool disease (SPD) patients and could corroborate its mice due to non-­specific deletion in other hematopoietic lineages.
specificity. We were able to subgroup SPD patients by either storage Conclusions: We demonstrate that the Gp1ba-Cre mouse is highly
(uptake), release or combined defects, opening more precise thera- efficient at deleting floxed genes in the MK lineage, with no evidence
peutic opportunities. Unexpectedly, we also detected dense granule of leakiness into other hematopoietic lineages, thus eliminating leu-
uptake defects in patients with GFI1B-­implicated Gray-­platelet like kocyte contributions to phenotypes.
disease (BDPLT17) or Wiskott-­Aldrich Syndrome (WAS), allowing to
connect platelet dense granule deficiencies with unrelated disease
Conclusions: We established an easy, quick and inexpensive assay PB041 | The Fluidity of Platelet Membrane
as an additional diagnostic module in flow cytometry-­based platelet
an Important Parameter for Platelet Function in
function tests.
Chronic Myeloproliferative Neoplasms Patients
V.M. Popov1,2; H. Bumbea3; I. Dumitru4; B.M. Matei5; C.
Oktaviani Matei5; M. Omer1; M. Andreescu6; O. Patrinoiu6;
F. Mihai6; T. Savopol7; A.M. Vladareanu8; E. Kovacs9; M.G.
PB040 | The Gp1ba-­Cre Transgenic Mouse: A Moisescu10
New Model to Delineate Platelet and Leukocyte 1
Colentina Hospital Bucharest, Hematology, Bucharest, Romania, 2University of

Functions Medicine and Pharmacy Carol Davila, Biophisics and Biotehnology, Bucharest,
Romania, 3UMF Carol Davila, Hematology, Bucharest, Romania, 4University
Z. Nagy1; T. Vögtle1; M.J. Geer1; S. Heising1; G. Di Nunzio1; Emergency Hospital, Hematology, Bucharest, Romania, 5UMF Carol Davila,
J. Mori1; R. Gareus2; G. Desanti3; A. Mazharian1; Y.A. Senis1 Biophysics and Biotechnology, Bucharest, Romania, 6Colentina Hospital
Bucharest, Bucharest, Romania, 7Colentina Hospital Bucharest, Biophysics and
University of Birmingham, Institute of Cardiovascular Sciences, Birmingham,
Cellular Biotechnology, Bucharest, Romania, 8UMF Carol Davila, Hematology
United Kingdom, 2The Jackson Laboratory, Bar Harbor, United States, 3University
SUUB, Bucharest, Romania, 9UMF Carol Davila, Biophysics and Cellular
of Birmingham, Institute of Microbiology and Infection, Birmingham, United
Biotechnology, Bucharest, Romania, 10UMF Carol Davila, Biophysics and Cellular
Biotechnology, Bucharest, Romania

Background: The Pf4-Cre transgenic mouse is widely used to gener- Background: Patients with chronic myeloproliferative neoplasms
ate megakaryocyte/platelet (MK/plt)-­specific knockout (KO) mouse (MPNs) had qualitative and quantitative modifications of plate-
models. However, several studies have challenged the specificity of let membrane receptors that are involved in alteration of platelet
this transgene, and raised the possibility that some of the pheno- function.
types associated with this model may be due to non-­specific dele- Aims: The aim of our study was to determine if changes of platelet
tion of floxed genes. membrane anisotropy (FA) could be correlated with alterations in ex-
Aims: In this study, we established the Gp1ba-Cre transgenic mouse pression of platelet receptors and reactive species production (ROS).
to produce MK/plt-­specific KO mice and directly compared it to the Methods: This retrospective study included 100 patients with MPNs
Pf4-Cre strain in terms of specificity and efficiency. as well as 10 controls. The determination of platelet membrane fluid-
Methods: Expression pattern of the transgenes in vivo was investi- ity was performed by fluorescence anisotropy measurements using
gated by crossing deleter mice with mT/mG double-­fluorescent Cre as marker 1-­(4-­trimethylammoniumphenyl)-­6-­phenyl-­1,3,5-­hexat
reporter mice and quantifying GFP+ haematopoietic cells by flow riene p-­toluenesulfonate (TMA DPH). Production of reactive spe-
cytometry. Protein ablation efficiency in platelets was tested by
cies (ROS) was examined using fluorescence method with DCFDA
crossing Gp1ba- or Pf4-Cre mice with Csk-, Shp1-, or CD148-floxed
and was assessed by aria under curve (AUC) in serial measurements
mice and assessed by a quantitative capillary electrophoresis-­based
during 900 seconds. Fluorescence anisotropy (FA) was analysed de-
pending ROS production and the expression of platelet membrane
Results: Over 98% of platelets from mT/mG;Gp1ba-Cre reporter mice
receptors measured by flow cytometry analyses. We examined the
were GFP+, demonstrating highly efficient Cre-­mediated gene exci-
flow cytometry markers of platelet adhesion (CD42a,CD42b), aggre-
sion in the MK lineage. Notably, no significant GFP expression was
gation (CD41, CD61) and CD36.
detected in erythrocytes or leukocytes from these mice. Protein lev-
Results: The expression of CD41/CD61 and CD42a/CD42b recep-
els of the tyrosine kinase Csk and tyrosine phosphatase Shp1 were
tors was lower in MPNs group compared with controls group and
reduced by 90-­93% using the Gp1ba-Cre transgene, and by 96-­99%
no statistical difference in expression of CD 36. The fluorescence

anisotropy of platelet membrane in MPNs group is higher than control PB043 | Functional Validation of microRNA-­
group, P=0.05 The CD36 expression had a positive correlation with
126-­3p as a Platelet Reactivity Regulator using
value of FA in MPNs patients group. (rho=0.75 -­95% CI 0.308-­0.925;
Human Progenitor Cells
P=0.005). Patients with MPNs had higher level of reactive species
production than controls group but was not observed any correlation A. Garcia1; S. Dunoyer-Geindre1; V. Zapilko1; S. Nolli1;
with FA.
J.-L. Reny1,2; P. Fontana1,3
University of Geneva, Geneva Platelet Group, Geneva, Switzerland, 2Geneva
Conclusions: MPNs patients have higher FA and low expression of
University Hospitals, Division of General Internal Medicine, Geneva, Switzerland,
CD41/CD61, CD42a/CD42b receptor. The expression of CD 36 is 3
Geneva University Hospitals, Division of Angiology and Haemostasis, Geneva,
higher in MPNs patients and was direct correlated with FA modi- Switzerland
fications. The correlation of expression of CD36 with ROS was not
observed and need to be analysed in a higher lot of patient Background: Platelets are an abundant source of microRNAs (miR-
NAs), which may play a role in the regulation of platelet function.
Some miRNAs, such as miR-­126-­3p, are pointed out as potential
biomarkers of platelet reactivity and recurrence of cardiovascular

PB042 | The Effect of FXa on Platelet events. However, the biological relevance of these associations re-
mains uncertain and functional validation of these candidate miR-
Activation, in vitro
NAs on human-­derived cells is lacking.
S. Papadaki; A. Tselepis Aims: To develop a flow-­based assay to monitor the functional im-
University of Ioannina, Atherothrombosis Research Centre/Laboratory of pact of candidate miRNAs on platelet-­like structures derived from
Biochemistry, Department of Chemistry, Ioannina, Greece
human progenitor cells.
Methods: CD34+-­derived megakaryocytes were transfected with
Background: Factor Xa (FXa) is a critical vitamin K-­dependent co-
miRNA or siRNA and were differentiated in platelet-­like structures.
agulation serine protease, which also exerts non-­haemostatic cel-
Platelet adhesion phenotype was assessed by perfusion of differ-
lular effects that may contribute to the development of various
entiated cells in a microfluidic system under a constant shear rate.
pathophysiological conditions. The main mediators of the cellular
Results: miR-­126-­3p transfection procedure does not affect mega-
responses induced by FXa are protease-­activated receptors -­1 and
karyocytes differentiation and platelet-­like structures production.
-­2 (PAR-­1 and -­2).
Aims: The aim of the present study was to investigate the effect of
FXa on platelet activation, in vitro.
Methods: Blood from healthy volunteers was collected and washed
platelets were prepared and adjusted at 250,000/μL. FXa-­induced
platelet aggregation (0.01-­0.1 nM) was evaluated using light trans-
mittance aggregometry. In addition, the effect of FXa (0.05 and
0.5 nM, incubation time: 5 and 20 minutes) on the activation of the
platelet-­integrin receptor α IIbβ3 (PAC-­
FITC) and the membrane
expression of P-­selectin (anti-­CD62P-­PE), was studied using flow
cytometry. The anti-­CD61-­PerCP antibody was used as a platelet
marker. The activation of αΙΙbβ3 and the expression of P-­selectin
was assessed as the percentage of PAC-­1/CD61+ and CD62P/CD61+
cells, respectively.
Results: FXa induced a robust platelet aggregation which reached
its maximum levels (76% and 81%) after 20 minutes of incubation at
concentrations 0.025 and 0.05 nM, respectively. Additionally, FXa
induced activation of α IIbβ3 and membrane expression of P-­selectin,
which reached the maximum after 20 minutes of incubation at a con-
centration of 0.5 nM (from 0.80±0.16% to 69.93±2.94% for PAC-­1/
CD61+ cells, P<0.05 and from 10.94±4.47% to 76.02±2.92% for
CD62P/CD61+ cells, P<0.05).
Conclusions: FXa induces a slow but strong platelet activation, thus
the direct oral anticoagulants that are used in clinical practice, apart
from their anticoagulant action, may also exert significant antiplate-
F I G U R E   1   Platelet-­like structures (white arrows) and CD34+-­
let effects.
derived megakaryocytes (black arrows) adhered on fibrinogen-­
coated channel

with an optimized automated LTA system using CE-­marked agonist

Methods: Platelet rich plasma from patients with suspected plate-
let disorders, von Willebrand disease or antiplatelet therapy have
been assessed using a wide range of agonist concentrations: ADP
(0.5-­10  μM), collagen (2 mg/μL), ristocetin (0.625-­
1.2 mg/mL) or
arachidonic acid (1 mM). We used the Siemens CS-­2500 analyzer
with a dedicated software and CE-­marked agonists from Hyphen
Biomed. The APACT-­
4 004 aggregometer (Elitech, France) was
used as the reference instrument with platelet agonists provided
from different manufacturers. Results were expressed as Maximal
Platelet Aggregation and correlation between the CS-­2500 and
APACT-­4 004 was analyzed using the Passing and Bablok regres-
sion test.
Results: Platelet aggregometry studies were performed in 49
F I G U R E   2   MiR-­126-­3p transfection increased the adhesion of patients. Maximal aggregation response with ADP (0.5-­10 μM),
platelet-­like structures on fibrinogen-­coated channel collagen (2 mg/μL), ristocetin (1.2 mg/mL) and arachidonic acid
(1 mM) agonists showed significant correlation between the
two aggregometers (P<0.001). We observed a more variable re-
Overexpression of miR-­126-­3p increased platelet-­like structures ad- sponse using an intermediate dose of 2.5 μM ADP. Moreover,
hesion compared to control (Figures 1 and 2). we also noted discrepancies with the low dose of ristocetin,
Moreover, miR-­126-­3p transfection was associated with downregu- showing excessive paradoxal agglutination with the CS-­2 500,
lation of ADAM9, a validated target of miR-­126-­3p, and of PLXNB2, suggesting that a lower ristocetin dose should be used with this
an actin dynamics regulator. Silencing PLXNB2 led to similar func- system.
tional results than miR-­126-­3p transfection. Conclusions: These data show that CS-­2500 is comparable to a
Conclusions: Taken together, using a flow-­based assay, we func- stand-­alone aggregometer. It has the advantages of a walk-­away
tionally validate miR-­
3p as a regulator of platelet reactivity technology and the use of CE-­marked reagents also permits the pos-
on platelet-­
like structures derived from human progenitor cells. sibility of an easier certification.
Moreover, PLXNB2 was identified as a platelet reactivity regulator
supporting the hypotheses that miR-­126-­3p modulates platelet re-
activity by downregulation of PLXNB2.
PB045 | Lack of Correlation between
Verifynow and TEG-­platelet Mapping in
PB044 | Assessment of Light Transmission Ventricular Assist Device Patients
Aggregometry on the Routine Coagulation E. Salazar
Analyzer Siemens CS-­2500 using CE-­marked Houston Methodist Hospital, Houston, United States

Agonists from Hyphen Biomed

Background: Ventricular assist device (VAD) patients must main-
V.-E. Brett1; B. Pougault2; A. Guy1; S. Castet3; Y. Huguenin3; tain a precarious coagulable state, in which the patient does not
A. Ryman1; C. James1; M. Fiore1,4
develop bleeding complications, and the ventricular device does
University Hospital of Bordeaux, Laboratory of Hematology, Pessac, France,
not thrombose. Bleeding is a frequent cause of hospital admis-
Siemens Healthcare, Saint-­Denis, France, 3University Hospital of Bordeaux,
Haemophilia Center, Bordeaux, France, 4University Hospital of Bordeaux, sion in ventricular assist device patients, and pump thrombosis
Reference Center for Platelet Disorders, Bordeaux, France can have devastating consequences. Typically, patients are anti-
coagulated with warfarin with routine monitoring of INR. Aspirin
Background: Light transmission aggregometry (LTA) is still consid- is often used as an antiplatelet agent, although in some patients,
ered as the “gold standard” for platelet function assessment but, as a P2Y12 inhibitors, such as clopidogrel can also be used. Several as-
completely manual technology, it is labour intensive. This challenge says are available for assessment of antiplatelet effect, including
can be overcome by performing platelet aggregometry in an auto- verifynow, teg platelet mapping (TEG-­PM), and traditional platelet
mated method on a routine coagulation analyzer. aggregation, among others.
Aims: We aimed to compare and correlate results obtained from Aims: We sought to assess the correlation between TEG-­PM and
a traditional manual LTA solution realized in our Reference Center Verifynow results in ventricular assist device patients as a first step

in determining which assay may be beneficial for guiding antiplatelet for FC using ADP as agonist. Using PAR1-­AP, P-­selectin expression
therapy in VAD patients. was little affected, but a gradual decrease in fibrinogen receptor acti-
Methods: From 12/1/2016-­
12/1/2017, antiplatelet testing with vation (PAC-­1 median fluorescence) down to 50% was seen as platelet
TEG-­PM and Verifynow was run in parallel on all VAD patients pre- counts decreased. This was due to the contribution of paracrine ADP
senting to the outpatient clinic. Aspirin effect was assessed in 151 stimulation, as this effect was eliminated by addition of apyrase.
patients, with some on repeated visits, for a total of 553 measure- Conclusions: Both LTA and IA are strongly influenced by sample plate-
ments. P2Y12 inhibitor effect was assessed in 38 patients, with let count. The effect on FC was less pronounced and most likely due
some on repeated visits, for a total of 107 measurements. Verifynow to concentration-­dependent effects on paracrine stimulation via ADP.
results in platelet reaction units were plotted against TEG-­PM re-
sults in percent inhibition. Linear regression analysis was performed
PB048 | Comparison of PAC-­1 and α-­fibrinogen
to determine correlation.
Results: Aspirin effect assessment showed very poor correlation
Antibody Binding to Platelet Subpopulations
between the two tests with an R 2 of 0.0856. P2Y12 inhibitor ef- A.L. Södergren1; S. Ramström2,3
fect assessment also showed very poor correlation with an R 2 of 1
Linkoping University, Department of Clinical and Experimental Medicine,
0.0015. Linköping, Sweden, 2Linkoping University, Department of Clinical Chemistry and
Department of Clinical and Experimental Medicine, Linköping, Sweden, 3Örebro
Conclusions: Antiplatelet effect measured with two common anti-
University, School of Medical Sciences, Örebro, Sweden
platelet monitoring devices did not correlate in VAD patients.

Background: Fibrinogen binding potential in platelets can be evalu-

ated in different ways. The PAC-­1 antibody recognizes the active
PB047 | Influence of Sample Platelet Count conformation of the fibrinogen receptor, α IIbβ3, while an α-­fibrinogen
on Three Common Platelet Function Tests: A antibody (αFib-­AB) can evaluate the amount of bound fibrinogen.

Comparative Study The procoagulant platelet subpopulation, exposing phosphatidylser-

ine (PS+), is reported to have reduced PAC-­1 binding, attributed to
N. Boknäs1,2; A.S. Macwan3; A.L. Södergren3; S.
downregulation of α IIbβ3. Coated platelets, a type of PS+ platelets,
have however been reported to have surface bound α-­granule pro-
Linkoping University, Department of Haematology and Department of Clinical and
Experimental Medicine, Linköping, Sweden, 2Monash University, Australian Centre for
teins, such as fibrinogen. Despite previous reports, a clear side-­by-­
Blood Diseases, Melbourne, Australia, 3Linkoping University, Department of Clinical side presentation of the results is lacking, which may explain some of
and Experimental Medicine, Linköping, Sweden, 4Linkoping University, Department of the confusion in the field.
Clinical Chemistry and Department of Clinical and Experimental Medicine, Linköping,
Aims: To compare the binding of PAC-­1 and αFib-­AB to PS+ and
Sweden, 5Örebro University, School of Medical Sciences, Örebro, Sweden
PS− platelets.
Methods: Platelets in citrated whole blood were obtained from
Background: Platelet function tests (PFTs) are indispensable for diag-
healthy volunteers after verbal consent as approved by the local
nosing platelet function disorders (PFD). Previous studies report im-
ethics committee. Platelets were activated in the presence of fibrin
paired analytical precision of PFTs when platelet concentrations fall
polymerization inhibitor GPRP, Annexin V (binding PS), and PAC-­1 or
below 100×109/L, which is often the case in PFDs. As the analytical
an αFib-­AB and evaluated with flow cytometry. Agonists were cross-­
principle of flow cytometry (FC) is unrelated to platelet count, FC has
linked collagen-­related peptide (CRP-­XL) and thrombin or PAR1 and
been brought forward as a preferable method in this setting. However,
-­4 activating peptides.
to date no study has compared the impact of low platelet counts on
the results of different PFTs side-­by-­side.
Aims: To compare the influence of sample platelet count on the results of
three common PFTs; light transmission aggregometry (LTA), whole blood
impedance aggregometry (IA; Multiplate) and FC, using two common
platelet agonists: PAR1-­activating peptide (AP, 32 μM) and ADP (6.5 μM).
Methods: Hirudinized blood was collected from healthy volunteers
after verbal consent. The procedure was approved by the local eth-
ics committee. Blood was separated into red blood cells, platelet-­
rich (PRP) and platelet-­poor plasma by centrifugation. Blood and PRP
with platelet counts of 200, 100, 50 and 10×109/L were created by
combining the components in different proportions.
Results: For LTA, test results were reasonably stable at sample plate-
let counts of 100-­200×109/L, but heavily affected at counts below
F I G U R E   1   Binding of PAC-­1 to platelets with (PS+) or without
100×109/L. For IA, a linear decline in the area under the curve (AUC)
(PS−) phosphatidylserine exposure after activation. Data are
was observed for the entire test range. Minor effects were observed pooled, n=6

Methods: Blood from full-­term neonates was collected at 24 hours

of life via heel stick. Umbilical cord blood samples were obtained
immediately after clamping post-­C aesarean delivery, in accordance
with Institutional Review Board-­approved protocols. Glass bottom
dishes were coated with poly-­L-­lysine, seeded with agonist-­treated
or vehicle whole blood, stained with a platelet dense granule marker
and imaged. Platelet dense granule release was measured as light
output. Flow cytometry was performed using vehicle or treated
blood labeled for CD62P and PAC-­1, together with markers for total
platelet glycoprotein and receptor expression. Thrombin genera-
tion was measured for constant shear stress-­exposed isolated cord
F I G U R E   2   Binding of αFib-­AB to platelets with (PS+) or without platelets. Annexin V and CD62P expression was measured in shear-­
(PS−) phosphatidylserine exposure after activation. Data are
exposed cord plasma and whole blood.
pooled, n=6
Results: Our results suggest that neonatal granule secretion was un-
responsive to protease-­activated receptor activation but not P2Y1/
Results: PAC-­1 bound all activated PS− platelets. When PS+ plate-
P2Y12, despite neonates having significantly reduced P2Y1/P2Y12
lets were formed, PAC-­1 binding became dichotomous with reduced
receptor expression. Impaired dense granule secretion was rescued
binding to the PS+ platelets. However, PAC-­1 binding in most PS+
following the addition of exogenous ADP. Sheared cord blood plate-
platelets was still higher than in resting control platelets, Figure 1.
lets exhibited similar thrombin generation trends, but lower annexin
The αFib-­AB bound both PS+ and PS− activated platelets, Figure 2.
V and CD62P expression compared to adult platelets.
Activation including thrombin further increased the αFib-­AB binding
Conclusions: Our results suggest impaired granule trafficking, mo-
to all platelets. Samples with thrombin and αFib-­AB, were enriched
bilization, and secretion, but comparable thrombin generation, in
in debris, possibly consisting of fibrin monomers detected by the
neonates. While assessment of neonatal platelet function remains
challenging, this study offers further insights and methods for inves-
Conclusions: PAC-­1 binding was dichotomous with reduced binding
tigating their response under static and shear conditions.
in PS+ platelets, but it was not lacking. αFib-­AB bound both PS+ and
PS− platelets and binding increased upon stimulation with thrombin.
Thus a reduction in PAC-­1 binding to PS+ platelets does not neces-
sarily mean a reduction in fibrinogen binding. PB051 | On the Potential Use of
Morphometrics as a Biomarker for Platelet
PB049 | Assessment of Neonatal Platelet S. Lickert1; M. Kenny2; V. Vogel1; I. Schoen2
Granule Trafficking and Shear-­induced Platelet 1
ETH Zurich, Health Sciences and Technology, Zurich, Switzerland, 2Royal College
Activation in Neonatal Peripheral and Cord Blood of Surgeons in Ireland, Molecular and Cellular Therapeutics, Dublin, Ireland

A. Ngo1; J. Sheriff2; K. Jones3; A. Rocheleau1; A. Mitrugno1;

Background: Platelets roll, adhere and contract at sites of vascular
J. Aslan1; S. Worthington4; A. Cox4; M. Recht4; M. Nieman5;
L. Malone6; A. Zigomalas2; W. Bahou6; D. Bluestein2; O. injury. These distinct functions are closely linked to cellular mechan-
McCarty1; K. Haley4 ics and based on an active actomyosin machinery and adhesion re-
Oregon Health & Science University, Biomedical Engineering, Portland, United ceptors which together orchestrate the generation and transmission
States, 2Stony Brook University, Biomedical Engineering, Stony Book, United of mechanical forces. Recent work showed that platelet contractility
States, 3Oregon State University, Chemical Engineering, Corvallis, United States,
testing could aid the diagnosis of elusive platelet-­related bleeding
Oregon Health & Science University, The Hemophilia Center, Portland, United
States, 5Case Western Reserve University, Pharmacology, Cleveland, United phenotypes [D.R. Myers et al., 2016, Nature Materials 16: 230-­35].
States, 6Stony Brook University, Division of Hematology, Stony Brook, United Our own investigation of morphological structure-­function relation-
States ships suggested that cytoskeletal textures could serve as an indirect
alternative to measuring platelet contractility [S. Lickert et al., 2017,
Background: Neonatal platelets are characterized as hyposensitive ISTH Congress, PB1283].
compared to adults, yet they display efficient hemostasis through Aims: The aim of this study was to establish a direct link be-
unclear mechanisms. Furthermore, neonatal platelet mechanotrans- tween specific platelet cytoskeletal architectures and platelet
duction remains poorly understood. contractility.
Aims: This study aims to investigate platelet function through gran- Methods: We fabricated PDMS-­based micropost array detectors
ule secretion, as well as to elucidate differences in shear-­specific re- [J.L. Tan et al., 2003, PNAS 100: 1484-­9] to measure traction forces
sponses between neonatal and adult platelets. exerted by spread human platelets on fibrinogen. In parallel, we per-
formed a morphometric analysis of the platelets’ cytoskeleton. The

system was validated with platelets from healthy consenting volun- risk factors for cardiovascular disease. Smoking has also been shown
teers. The work was conducted in accordance with the declaration to induce a state of hypercoagulability, thus influencing the normal
of Helsinki. platelet physiology and hemostasis.
Results: Experiments revealed a correlation between morphologi- Aims: This study was undertaken to assess the association of smok-
cal changes and alterations in cellular contractility. In line with the ing duration with platelets indices.
cytoskeletal architecture, traction forces were highly sensitive to Methods: Platelet count (PC), mean platelet volume (MPV) and plate-
perturbations of integrin αIIbβ3 binding. A bipolar morphology was let distribution width (PDW) were determined by automation (sys-
only observed on ligands binding to αIIbβ3 and were accompanied mex kx 21N) from the sample of forty two consenting male smokers
by a polarized traction profile. and forty two non-­smokers in Calabar, Nigeria. Platelet aggregability
Conclusions: These preliminary results indicate a tight relation- was estimated colorimetrically by modified O’Brien’s method while
ship between platelet contractility and cytoskeletal morphology in Relative Plasma Viscosity (RPV) was Determined by Reid and Ugwu
human platelets. They support the potential usage of morphomet- method. Subjects were from 18 to 35 years of age. Statistics was
rics as an easy-­to-­access diagnostic marker for certain biomechani- done with student’s t-­test, ANOVA and Pearson correlation using
cal platelet defects. This work was funded by RCSI Dublin (I.S.) and SPSS software version 19.
ETH Zurich (S.L., V.V). Results: The platelet aggregability, MPV, PDW and RPV of smok-
ers were found to be significantly higher than that of non-­smokers
(P<0.05). Smokers with pack-­year ≤4 years had platelet parameters
that were not significantly different from that of the non-­smokers
PB052 | Impact of Cigarette Smoking on
(P>0.05) while smokers with smoking park-­year >4 years had an in-
Platelet Count, Mean Platelet Volume, Platelet
creased PC, platelet aggregability, MPV, PDW and RPV that were
Aggregability and Platelet Distribution Width of significantly higher than that of the non-­smokers (P<0.05). Positive
Male Cigarette Smokers in Calabar, Cross River correlation was observed between PC and smoking pack-­
State, Nigeria (r=0.9495, P<0.05), MPV and pack-­year (r=0.9475, P<0.05).

I.M. Okafor 1,2 1

; I. Ngusha ; U. Udo 1 Conclusions: This study has shown that smokers tend to have in-
1 creased PDW, MPV, platelet aggregability and RPV compared to
University of Calabar, Haematology Unit, Department of Medical Laboratory
Science, Calabar, Nigeria, 2University of Calabar Teaching Hospital, non-­smokers. The findings from the present study along with evi-
Haematology, Calabar, Nigeria dences from literature enables us to suggest that smokers should be
monitored for PC, PDW, MPV and platelet aggregability which will
Background: Cigarette smoking affects the normal hemostasis by be helpful in evaluating the altered coagulation status and hemosta-
influencing the coagulation pathways. Habitual smoking is one of the sis in cigarette smokers.

TA B L E   1   Comparison of platelet count, mean platelet volume and platelet distribution width of smokers and non smokers

Parameters Smokers (n=42) Non-­Smokers (n=42) P-­value Remarks

Platelets count (×109/l) 229.55±46.87 209.64±60.17 0.009 Not-­significant

MPV (femtolitre) 10.0±0.97 9.50±0.68 0.001 Significant
PDW (%) 13.20±2.17 11.93±2.04 0.001 Significant
Platelet aggregation 0.069±0.013 0.061±0.026 0.004 Significant
Relative Plasma Viscosity 1.84±0.167 1.74±0.123 0.003 Significant

TA B L E   2   Comparison of platelet count, mean platelet volume and platelet distribution width of smokers with smoking pack ≤4 years and

Parameters ≤4.0 pack years (n=25) >4.0 pack years (n=17) Non smokers (n=42) P-­value Remarks

Platelets count 218.52±55.84 241.18±35.78* 209.64±60.17 0.004 Significant

MPV (femtolitre) 9.80±0.79 10.20±1.14* 9.50±0.68 0.0005 Significant
PDW (%) 12.50±2.29 13.69±2.16* 11.93±2.04 0.0005 Significant
Platelet aggregation 0.065±0.014 0.073±0.009* 0.061±0.026 0.001 Significant
Relative Plasma 1.79±0.10 1.94±0.29* 1.74±0.123 0.004 Significant

PB053 | Platelets Role in the Plasma Clotting is PB054 | Stability of Platelet Function in

Not Limited to Providing a Lipid Surface Healthy Individuals over a One-­year Period
A. Balandina1,2; N. Petrunina3; E. Koltsova1,2; D. Huskens1,2; J. Konings1,2; Y. Sang1,2; B. de Laat1,2;
F. Ataullakhanov1,2,3 M. Roest1,2; H. Kelchtermans1,2
1 1
Center for Theoretical Problems of Physicochemical Pharmacology, Moscow, Maastricht University Medical Center, Synapse Research Institute, CARIM,
Russian Federation, 2National Center of Pediatric Hematology, Oncology and Maastricht, the Netherlands, 2Maastricht University Medical Center, Department
Immunology, Moscow, Russian Federation, 3Moscow Institute of Physics and of Biochemistry, CARIM, Maastricht, the Netherlands
Technology, Dolgoprudny, Russian Federation

Background: Platelet function testing is used for diagnosing bleed-

Background: Activated platelets accelerate blood coagulation by ing disorders, monitoring anti-­platelet therapy and the prediction of
providing procoagulant membranes for the surface-­dependent reac- thrombosis. To achieve clinical applicability, platelet function should
tions. However, the integral effect of platelets on coagulation is a be relatively stable within one individual.
subject of debate. Aims: To determine the stability of platelet function in healthy indi-
Aims: We investigated the role of platelets in the homogeneous and viduals over a 1-­year period.
reaction-­diffusion in vitro experimental models. Methods: Blood was collected from 12 healthy individuals once a
Methods: Clot formation in platelet-­free plasma (PFP) supplemented month during 1 year. Flow cytometry based whole blood platelet
with phospholipid microvesicles (PL) or platelet rich plasma (PRP) activation testing was performed to quantify αIIbβ3 activation and
was activated with the identical amounts of tissue factor (TF) either P-­selectin expression in response to ADP, thrombin receptor peptide
homogeneously distributed (final concentration 5 pM, homogeneous (TRAP) and collagen-­related peptide (CRP). Anti-­mouse Ig κ particles
model) or immobilized on the surface (surface density 100 pmole/ were included to standardize results. The local medical ethical board
m2, spatially heterogeneous model). Fibrin clot growth and thrombin evaluated the protocol and all participants gave written informed
generation were observed using videomicroscopy in plasma samples consent.
of six healthy donors. Results: Over a period of 12 months, the intra-­individual variation in
Results: In homogeneous model in PRP (2×105/μl) thrombin peak agonist-­induced αIIbβ3 activation and P-­selectin expression on plate-
maximum was reached after 10±3 minutes with amplitude of lets ranged between 11% and 21% (except for ADP induced P-­selectin
124±30 nM. The peak amplitude decreased simultaneously with expression 58%). Significant differences were demonstrated between
the decrease in platelet concentration, but a second peak ap- different months (Figure 1), however, the pattern proved to be random
peared later than the main one, disappearing in PFP. The decrease and different for αIIbβ3 receptor activation and P-­selectin expression,
in platelet count caused the decrease in second peak amplitude and for the individual triggers TRAP, CRP or ADP.
and elongation of the time to reach it. Addition of PL (4μM) into
PFP led to increase in the peak amplitude from 25±9 to 102±29nM
(less than in PRP with 2×105/μl of platelets, P<0.05) and decrease
in peak time from 8±2 to 5±2 minutes. In spatially heterogene-
ous model in PRP (2×105 /μl) a propagating thrombin wave with
the amplitude of 119±37 nM was formed. Its amplitude decreased
with platelet count decrease. In PRP with platelet count 0.4×105/μl
and lower the wave disappeared and spatial thrombin distribution
was similar to the exponential decay. The amplitude of thrombin
wave in PFP with PL was 63±23 nM and was significantly different
from PRP (P<0.05).
Conclusions: The effect of PL on thrombin generation is qualitatively
different from the effect of platelets i.e. platelets role in the plasma
clotting is not limited to providing a lipid surface.

F I G U R E   1   Monthly variation in platelet function. Differences

were analysed by Friedman test with Dunn’s post-­hoc test: *P<0.05,
**P<0.01, ***P<0.001

Conclusions: Our data suggest that platelet activation within one

healthy individual over a 1-­year period is sufficiently stable to allow its
usage as a clinical tool. Although significant differences were observed
in platelet function between different months, no clear trend was de-
tected as variation differed per agonist and activation marker. Therefore,
these differences are most likely not of physiological relevance.

PB055 | The Canine Platelet Secretome (CAPS):

Proteomic Analysis of Thrombin-­stimulated
S.E. Cremer1,2; J.L. Catalfamo2; A.T. Kristensen1; R.A.N.
Goggs3; M.B. Brooks2
University of Copenhagen, Department of Veterinary Clinical Sciences,
Copenhagen, Denmark, 2Cornell University, Department of Population Medicine F I G U R E   1   SDS-­PAGE for canine platelet proteomics. Gels (12%)
and Diagnostic Sciences, Ithaca, United States, 3Cornell University, Ithaca, were loaded with 5 µg reduced protein per lane and stained with
United States Colloidal Coomassie Blue

Background: Domestic dogs share the same environment as their into sections and in-­
gel trypsin digested. Tryptic peptides were
owners and represent a valuable animal model to study naturally-­ separated by nano-­LC (liquid chromatography) followed by tandem
occurring human disease. Proteomics holds promise for discovery mass-­spectrometry (MS/MS). Proteins were identified with Sequest
of cancer biomarkers, however comparative platelet proteomics are software searching a canine database. Identified proteins had mini-
lacking. mally 1 unique peptide and were sorted based on +/-­ ≥2 peptides. A
Aims: To establish a protocol for shotgun proteomic identification ratio of ≥2 for MS1 abundance (activated/control) was used to iden-
and quantification of proteins released from agonist-­activated ca- tify proteins released.
nine platelets. Results: 1,126 proteins were identified (Table 1). 634 proteins
Methods: Washed platelets were isolated from ACD-­A anticoagu- (56.3%) were classified as canine platelet thrombin-­stimulated CAPS
lated blood from a hound mixed-­breed dog by serial centrifugation. proteins. For 22.6% of identified proteins, the MS1 abundance was
Stirred, washed platelets (1.0×109/mL) were stimulated with saline too low to permit accurate quantification.
or 50 nM gamma thrombin at 37°C for 6 minutes. Protease inhibi- Conclusions: A proteomics protocol was established and we defined
tors were added and followed by centrifugation to remove cells a subset of canine platelet proteins released following in-vitro throm-
and debris. The supernatant was spun at 50,000× g to yield soluble bin activation that was similar to that reported for human platelets.
and pellet (microparticle) fractions. The former was concentrated. Future studies will characterize agonist-­
dependent and disease-­
Protein concentration was measured in both fractions. SDS-­PAGE associated canine platelet secretomes.
was used to separate proteins present in soluble and pellet frac-
tions (Figure 1). For shotgun proteomic analysis, the gel was divided

TA B L E   1   Proteins identified in saline (control) and thrombin-­stimulated platelets

MS1 Abundance Ratio (Releasate/Control)

Protein ≥2 <2 and >1 ≤1 Solely Identified Total

Total number 634 (56.3 %) 155 (13.8 %) 83 (7.4 %) 254 (22.6 %) 1126 (100.0 %)
 Unique to soluble 267 (42.1 %) 99 (63.9 %) 36 (43.4 %) 123 (48.4 %) 525 (46.6 %)
 Unique to pellet fraction 121 (19.1 %) 42 (27.1 %) 47 (56.5 %) 111 (43.7 %) 321 (28.5 %)
 In both fractions 246 (38.8 %) 14 (9.0 %) 0 (0.0 %) 20 (7.9 %) 280 (24.9 %)
Identified with ≥ 2 461 (72.7 %) 118 (76.1 %) 52 (62.7 %) 42 (16.5 %) 673 (59.8 %)
 Unique to soluble 179 (38.8 %) 82 (69.5 %) 25 (48.1 %) 20 (47.6 %) 306 (45.5 %)
 Unique to pellet fraction 85 (18.4 %) 25 (21.2 %) 27 (51.9 %) 19 (45.2 %) 156 (23.2 %)
 In both fractions 197 (42.7 %) 11 (9.3 %) 0 (0.0 %) 3 (7.1 %) 211 (31.4%)

PB056 | Immunofluorescence Analysis on a PB057 | Mean Platelet Volume Positively

Blood Smear – Validation of Stability of Blood Correlates with Uterine Volume after
Smears during Storage Menopause, Except Myomas Presence
K. Althaus1,2; L. Lex1; J. Wesche1; A. Greinacher1 B. Bajteková1,2; S. Wimmerová1; V. Rusnáková1,3
1 1
Universitätsmedizin Greifswald, Institut für Immunologie und Slovak Medical University, Faculty of Public Health, Institute of Biophysics
Transfusionsmedizin, Greifswald, Germany, 2Universitätsklinik Tübingen, Informatics and Biostatistics, Bratislava, Slovakia, 2GynAc, s.r.o., Clinic of
Transfusionsmedizin ZKT, Tübingen, Germany Obstetrics and Gynaecology, Šaštín-­Stráže, Slovakia, 3Trnava University, Faculty
of Health Sciences and Social Work, Trnava, Slovakia

Background: Many inherited platelet disorders are associated

with distinct morphological alterations in expression or distribu- Background: The size of the uterus varies in the course of life and

tion of platelet receptors, granule proteins, or cytoskeletal pro- individual differences exist also after the cessation of ovarian ac-

teins. These changes in platelet morphology can be detected by tivity. Leiomyoma is frequent benign uterine tumor. Mean platelet

immunofluorescence microscopy using patient blood smears. volume (MPV) has been indicated as a marker of platelet reactivity.

However, the preanalytic requirements for immunofluorescence Aims: To explore, if there was any association of blood parameters

assessment of platelet morphology on a blood smear are not well with female menopausal uterus volume (MUV) in cohort of 178

defined. middle-­European women; 40-­74 years old (median 58).

Aims: A systematic approach to identify optimal conditions to pre- Methods: Retrospective analysis. Tested parameters of blood count:

pare blood smears for immunofluorescence analysis of platelet Hemoglobin, Erythrocyte count, Mean corpuscular volume (MCV),

morphology. Platelet count, Mean platelet volume, White blood cell count (WBC),

Methods: From three healthy donors anticoagulated blood (EDTA, Absolute neutrophil and lymphocyte count. Also body weight (BV),

Citrate, Hirudin) were obtained with two technical replicates per body mass index (BMI), age, 25 OH D vitamin and hs-­C-­reactive pro-

donor. Blood smears were prepared within 4 hours after drawing. tein (CRP) serum level. Nonparametric correlation (Spearman′s rho)

Air dried blood smears were stored at room temperature and then was used for assessment, in triple way:

fixed and stained on days 0, 1, 3 or 7 after preparation, using an- A: postmenopausal women without uterine leiomyoma (n=70).

tibodies against GP IIbIIIa,GP IbIXa granula markers (P-­

vWF, Thrombospondin, CD63, Lamp1, Lamp2), and cytoskeleton
structures (NMMIIa, alpha-­and ß1 tubulin, and filamin A). Intensity
and distribution of markers were evaluated by two independent
Results: Morphological markers were most stable when EDTA-­
anticoagulated blood was used for preparation of blood smears.
Expression of GPIIbIIIa and GPIbIX was stable during storage of air
dried blood smears with a slight reduction in intensity after 3 days.
Alpha and dense granule markers remained stable until day 3 with
a slight reduction in staining intensity after Day 3. NMMIIA was
stable until day 7. Unexpectedly, ß1-­Tubulin and filamin A were
diffusely distributed in platelets when stained at day 0, but pre-
sented in a ring structure when the blood smear was stored for at
least 24 hours.
Conclusions: For immunofluorescence assessment of platelet
markers in patients with suspected hereditary platelets air dried F I G U R E   1   MPV correlation with Menopausal Uterus Volume
blood smears from EDTA-­b lood should be used, which are stable (MUV), for women without myomas
for 3-­7 days for shipment before they are fixed in a specialized

F I G U R E   2   Spearman’s correlation coefficient for MUV

correlated to age, body weight and parameters of blood count
B: postmenopausal women as a whole, regardless of the myoma
presence (n=178).
C: postmenopausal women with ultrasound-­verified uterine myoma
Results: A). In group with intact uterus a positive correlation be-
tween MPV and menopausal uterine volume was observed (rs 0.415
P=0.023; Figure 1).
B). For postmenopausal women generally, MPV and MUV were only
borderline associated (rs 0.184 P=0.066). MUV was positively cor-
related with BV (rs 0.214 P=0.004), but not BMI.
C). Myomatous postmenopausal women (n=108): Menopausal uter-
ine volume was faintly positively associated with absolute Neutrophil
count (rs 0.247 P=0.046), but negatively correlated with CRP serum
level (rs −0.601 P=0.030) and 25 OH D vitamin at winter (rs −0.374
P=0.005). Myomatous women were significantly heavier than those
without myomas (P=0.007). Menopausal uterine volume in all groups
was negatively correlated with age (rs −0.448 P<0.001; Figure 2). F I G U R E   1   Correlations for agonist responses. Numbers and
Conclusions: In our study a positive correlation between MPV and colors represent correlation coefficient
uterine volume after menopause was observed. However this did
not apply to women with uterine leiomyoma.
correlates more strongly with secretion of other agonists than to ADP
aggregation). Finally, comparison of WBILA with LTA showed that for
the 7 patients with both studies, two had normal WBILA after border-
PB058 | Clinical Experience Evaluating Platelet line LTA results and two had congruent findings. The remaining pa-
Function by Whole Blood Impedance Lumi-­ tients had specimen problems with LTA but successfully had WBILA.

Aggregometry in a Pediatric Population Conclusions: Our data highlight the demographic features associ-
ated with platelet abnormalities and the patterns seen in WBILA.
A. Obstfeld1; B. Doshi2; M.P. Lambert3
WBILA performs acceptably in comparison to LTA in a pediat-
Children’s Hospital of Philadelphia, Pathology and Laboratory Medicine,
ric population, but with fewer cancellations due to pre-­a nalytic
Philadelphia, United States, 2Children’s Hospital of Philadelphia, Pediatrics,
Philadelphia, United States, 3Children’s Hospital of Philadelphia, Pathology and issues.
Laboratory Medicine and Pediatrics, Philadelphia, United States

Background: Light Transmission Aggregometry (LTA) is used in con- PB059 | Functional Activity of Platelets in
firming congenital platelet function defects (PFD) and most expe-
Patients with Kasabach-­Merritt Syndrome
rience and literature relates to this method. Less data concerning
the performance of whole blood impedance lumi-­
aggregometry E. Gluhanyuk1; A. Ignatova1; A. Fedotov1; A. Karelin1; D.
Nikolaeva1; E. Orekhova1; L. Khachatryan1; M. Panteleev1,2
(WBILA) is available, especially in pediatrics where the lower blood
National Medical Research Centre of Pediatric Hematology, Oncology and
volume requirements affords an advantage.
Immunology named after Dmitry Rogachev, Moscow, Russian Federation,
Aims: To summarize the performance of a WBILA system at a single 2
Lomonosov Moscow State University, Moscow, Russian Federation
tertiary care center serving pediatric patients.
Methods: A Chrono-­
Log model 700 aggregometer was used. Background: Kasabach-­
Merritt syndrome (KMS) is a rare life-­
Demographics, testing indications, and results were collected over ap- threatening condition characterized by combination of vascular
proximately 1 year. 115 cases were included. Summary statistics and as- tumors and consumption thrombocytopenia and coagulopathy. We
sociations were calculated. Results were compared to LTA when available. hypothesized that apart from massive platelet destruction in situ the
Results: Ages were distributed bimodally with peaks at 33 months and platelet intactness in peripheral blood may be disrupted.
15 years with an average of 8.7 years (SD=5.5 years). The most fre- Aims: To evaluate the platelet functional state and its dynamics dur-
quent indication for testing was a history of bleeding (78%), most com- ing treatment in KMS patients.
monly epistaxis (29%) and easy bruising (26%). The rest had a family Methods: 4 KMS patients at the age of 1-­3 months were examined
history of bleeding or a genetic predisposition for a PFD. Of the demo- till the age of 8 months. Patients received metronomic chemo-
graphic features, only platelet count and morphologic abnormalities therapy, propranolol, heparin, prednisolone if needed. Platelets
were associated with an abnormal WBILA result (P<0.01 and P<0.05). were automatically counted and their functional activity was esti-
Secretion and aggregation responses of each agonist correlated with mated by flow cytometry with mepacrine and antibody to CD61,
one another (Figure 1), however correlation of secretion and correla- PAC1, CD42b and annexin V before and after activation by CLP and
tion of aggregation between agonists was stronger (e.g. ADP secretion SFLLRN. Since there is no reference data for children of this age we
28      |
proposed to consider 2 children with vascular tumor but without Biofouling due to platelet aggregation was assessed through moni-
KMS as a control. toring epifluorescence when DIOC6 labelled whole blood was per-
Results: Therapy resulted in a decrease in tumor size, consumption fused through pump manifolds. Platelet activation was assessed
thrombocytopenia and coagulopathy. We revealed that difference through FACS analysis of PAC-­1 binding, P-­Selectin expression and
in all parameters between activated and non-­activated platelets Phosphatidyl Serine exposure.
differed in KMS patients compared to control group demonstrat- Results: We demonstrate an in-­line micropump which may be used
ing the reduced ability of platelets to be activated. Mepacrine, in pumping or mixing applications requiring haemocompatibility with
PAC1, CD42b and annexin V changed significantly (P≤0.05 for regards to platelets. We demonstrate a valving iteration that mini-
each parameter) after treatment, resulting in an increase in abil- mises platelet reactivity, biomarker loss or biofouling, while main-
ity to be activated. To analyze correlation between platelet count taining capacity as an in-­line pump-­mixer.
(PC) and cytometric parameters observations were divided into Conclusions: This study describes the development and char-
two cohorts depending on PC: <50×10 9/l (A) and >50×10 9/l (B). acterisation of an in-­
line elastomeric microfluidic micromixer
In group A the correlations between PC and expression of CD62p which may be utilised in applications requiring haemocompat-
2 2
(R =0.83, P<0.01), CD61 (R =0.64, P<0.01) after activation, and ibility, such as in high-­
t hroughput anti-­
platelet drug screening
PC and degranulation degree of dense granules (R 2=0.71, P<0.01) systems.
were revealed. In group B we found the correlation between PC
and expression of CD42b both before (R 2=0.68, P<0.05) and after
(R 2=0.88, P<0.01) activation.
PB061 | Flow Cytometric Analysis of Platelet
Conclusions: KMS apparently induces alterations in platelet func-
Function in Patients on Antiplatelet Therapy and
tional state, with a tendency to reduction during therapy.
Suspected Thrombocytopathy
D. Huskens1,2; M. Roest1,2; L. Florin3; B. de Laat1,2; K.
PB060 | Characterisation of a 1
Maastricht University Medical Center, Synapse Research Institute, CARIM,
Haemocompatible Dual in-­line Reciprocating Maastricht, the Netherlands, 2Maastricht University Medical Center,
Micropump and Mixer for on-­chip Assessment of Department of Biochemistry, CARIM, Maastricht, the Netherlands, 3Ghent
University, Department of Clinical Chemistry, Microbiology and Immunology,
Experimental Anti-­platelet Agents Ghent, Belgium

R. Brazilek1; C. Szydzik 2; F. Akbaridoust3; M. Knoerzer2;

I. Marusic3; H. Nandurkar1; A. Mitchell2; J. Hamilton1; Background: Light transmission aggregometry (LTA) is still the most
W. Nesbitt1,2 used test for the identification and diagnosis of platelet function de-
Monash University, Prahan, Australia, 2RMIT University, Melbourne, Australia, fects (PFD). Disadvantages of LTA include its laborious nature, the
The University of Melbourne, Melbourne, Australia
large volumes of blood required and the relative insensitivity to small
changes in platelet function.
Background: Microfluidic technologies offer unique possibilities for Aims: We investigated if the flow cytometry-­based whole blood
improvements in clinical platelet diagnostics. However, the use of platelet activation test (WB-­PACT) correlates with LTA or could be
blood with such systems presents significant challenges, particularly used as a complimentary test.
due to biomarker loss and biofouling due to platelet aggregation. The
impact of advancements in microfluidic automation, such as micro-
pumps and microvalves, must be carefully considered in the devel-
opment of haemocompatible devices.
Aims: This study aims to explore the development and charac-
terisation of a haemocompatible elastomeric microvalve that can
act as a dual pump-­m ixer for antiplatelet drugs, and to determine
the design considerations for microscale active mixing in whole
Methods: Through application of novel injection-­
moulding fabri-
cation methods, we fabricated v-­shaped and straight valve-­gated,
actuated reciprocating micropumps. Analysis
of platelet-­
like particle motion through Micro-­
Particle Image
Velocimetry and Computational Fluid Dynamic analysis was used to
determine the optimal pumping geometry from a shear and platelet
F I G U R E   1   Correlation of WB-­PACT with LTA for determining
distribution perspective. Platelet-­sensitive biomarker loss was as-
platelet function. The Spearman correlation coefficient and P value
sessed via ELISA measurement of VWF and Fibrinogen adsorption.
are depicted

Methods: WB-­PACT quantifies αIIbβ3 activation and P-­selectin ex- Aims: We used flow cytometry assay for the platelet functional anal-
pression in response to 3 agonists and VWF binding to platelets in ysis in high platelet-­related hemorrhagic risk-­associated disorders.
response to ristocetin. In total, 153 patients (32 on dual antiplatelet Methods: Platelet surface membrane antigen composition and
therapy (acetylsalicylic acid and ADP receptor antagonist), 121 sus- functional response to physiological agonists were determined.
pected for bleeding diathesis) were tested. For WB-­PACT param- Activation was with collagen-­related peptide plus SFLLRN. Levels
eters, the 2.5th, 10th and 25th percentiles were determined in 48 of CD42b, CD61, CD62P, PAC1, annexin V binding, and mepacrine
healthy donors to score the patient samples (0-­2.5%: 3, 2.5-­10%: 2, release were determined.
10-­25%: 1 and >25%: 0). For LTA, maximal aggregation, disaggrega- Results: Pediatric patients hemorrhagic syndrome of unclear origin
tion and prolongation of the lagtime in response to 4 agonists and and excluded coagulopathy (n=32), preterm neonates (33-­34 gesta-
ristocetin was scored. Patients with at least one parameter lower tion weeks, n=10) and term neonates (>37 gestation weeks, n=10),
than the 10th percentile measured with WB-­PACT in healthy donors pediatric (n=44) and adult (n=32) patients with chronic immune
or at least one value deviating from the normal range measured with thrombocytopenia (ITP) and adult patients with chronic lympho-
LTA were diagnosed with PFD. A bleeding score (BS) was calculated cytic leukaemia (CLL, n=17) before and on ibrutinib treatment were
with the ISTH/SSC bleeding assessment tool (Rodeghiero et al., JTH, analysed. A direct relationship between the clinical index of hemor-
2010). rhage and the integrin activation degree and dense granules release
Results: A moderate correlation between LTA versus WB-­PACT was (P<0.05) was revealed for pediatric patients with hemorrhagic syn-
found with a R of 0.63 (Figure 1). In total, 70 patients were diag- drome. All platelet functions without exception (specifically, lack of
nosed with PFD by both tests (κ=0.43, Table 1). BS were recorded all granules, of their release and integrin activation) were profoundly
for 18/21 and 14/21 patients who were diagnosed with PFD accord- (by 50-­70%) impaired in newborn, with greater defects in pre-­term
ing to WB-­PACT only and LTA only, respectively. Interestingly, 6/18 newborn. Patients with CLL had initially impaired response of plate-
patients with PFD according to WB-­PACT had a BS>3. For LTA, only lets and ibrutinib additionally inhibited it, which corrrelated with
1/14 patients with abnormal platelet function had a BS>3. bleeding risk. In chronic ITP, our data indicate that platelets are pre-­
activated, large and slowly reacting; these changes are associated
with bleeding independently of platelet count, and can be used for
Conclusions: In comparison with LTA, flow cytometric analysis of
risk stratification.
platelet function by WB-­PACT is of additional value to determine
Conclusions: Functional platelet flow cytometry parameters corre-
late with bleeding risk in hematological disorders and other acquired
platelet dysfunction conditions.
TA B L E 1 Diagnosing PFD with WB-­PACT and LTA

PB063 | Study of Leukocyte-­platelet
0 1 Total
Aggregates by Flow Cytometry in Patients with
LTA 0 41 21 62
Coronary Artery Disease
1 21 70 91
Total 62 91 153 A. Ermakov1; O. Sirotkina2; T. Vavilova2; L. Gaykovaya1
North-­western State Medical University Named After I.I. Mechnikov, Saint-­
Petersburg, Russian Federation, 2Almazov National Medical Research Centre,
Saint-­Petersburg, Russian Federation
PB062 | Characterization of Platelet Function
with Flow Cytometry for Bleeding Risk Background: Platelets are not only participants of hemostasis, but also
Evaluation in Hematological Disorders directly related to immunity, the course of inflammatory reactions and
the repair of damaged tissues. A characteristic reaction of platelets to
A. Ignatova1; E. Suntsova1; D. Polokhov1; V. Ptushkin2;
inflammation is the formation of leukocyte-­platelet aggregates (LTA),
E. Nikitin2; P. Zharkov1; D. Fedorova1; A. Maschan1; G.
Novichkova1; M. Panteleev1 which occurs as result of their activation and leads to the launch of
Dmitry Rogachev National Research Center of Pediatric Hematology, Oncology
a cascade of pro-­inflammatory reactions. The appearance of LTA can
and Immunology, Moscow, Russian Federation, 2Botkin City Clinical Hospital, serve as a diagnostic sign of the inflammatory response of the body, in
Moscow, Russian Federation particular in acute myocardial infarction.
Aims: To study the LTA content by flow cytometry in patients who
Background: Combined decrease of platelet count and platelet func- underwent acute myocardial infarction (AMI).
tion is a frequent complication in autoimmune diseases, hematologi- Methods: 66 patients (71% male, 29% female, age 59±11) with AMI
cal malignancies and other hematological disorders. Its relationship who received dual antiplatelet therapy (clopidogrel 75 mg+ASA
to bleeding is poorly characterized. 100 mg) and 37 healthy donors (41% male, 59% female, age 48±13)
as a control group were included into the study. LAT was evaluated

using fluorescently labeled monoclonal antibodies CD61-­

FITC, thrombopathies (Clopidogrel; n=6, Aspirin; n=14 or dual therapies; n=18).
CD62P-­PE and CD45-­PC5 specific to GPIIb-­IIIa, P-­selectin of plate- Results using Bland & Altman and correlation tests or ROC curves showed
lets and to the common leukocyte antigen, respectively, after cells perfect correlation between these reagents and references.
activation by 10 mkM ADP. Particular events were counted on the Conclusions: This new range of agonists have shown the same ef-
flow cytofluorimeter CYTOMICS FC-­500 (Beckman Coulter, USA). fectiveness as the existing references in the diagnosis of patient with
Results: In the group of patients with AMI a statistically significant acquired or inherited platelet disorders.
(P<0.0001) increase in LTA in the blood was observed in compari-
son with the donor group -­30.4% (10.9-­107.1) and 6.7% (3.9-­10.5)
respectively. Also in patients with AMI, after activation of plate- PB065 | Usefulness of Optical Aggregometry
lets 10 mkM ADP, the amount of GPIIb-­IIIa 8.8MFI (7.9-­13.2) and
and Flow Cytometric Studies as a
P-­selectin 9.7% (1.3-­15.9) was significantly lower (P<0.05) than in
healthy donors -­18.1MFI (11.2-­22.5) and 22.7% (6.1-­32.4), respec-
Complementary Tools in Patients with Acquired
tively, that indicated an positive antiplatelet effect in the presence of Platelet Defects (APD)
an active inflammatory process. M.F. Alberto1; M. Asensio1; E.I. Bermejo1; M. Agazzoni1;
Conclusions: LTA analysis is an objective indicator reflecting the L. Romero1; A. Sanchez-Luceros1,2
state of the immune system with damaged vascular bed, which can 1
Hematological Research Institute, National Academy of Medicine, Hemostasis
be used to assess the severity of the disease and to characterize the and Thrombosis, Caba, Argentina, 2IMEX-­CONICET-­ANM, Hemostasis and
Thrombosis, Caba, Argentina
stabilization of the pathological process in therapy.

Background: Acquired nature of bleeding disorder is suggested by

a negative personal and family bleeding history. In many cases, di-
PB064 | Evaluation of a New Range of Platelet agnosis could be difficult to elucidate. In APD, the association of
Agonists for the Diagnosis of Inherited or functional and non functional methods, including mixing assays, may
Acquired Platelet Dysfunctions contribute to diagnosis.
1 2 2 3 Aims: To evaluate the usefulness of optical aggregation (OA) and
M.-C. Alessi ; O. Georgelin ; M. Ibrahim ; S. Harvel ;
S. Beaufils3; E. Maurer3; P. Ohlmann3 flow cytometry (FC) to characterize APD.
Faculté de médecine de la Timone, INSERM_S1062 NORT, Marseille, France, Methods: Patient 1 (P1): a 4-­year-­old boy with sudden spontane-
Faculté de médecine de la Timone, INSERM UMR_S 1062 NORT, Marseille, ous bruising, petechia, epistaxis and gums bleeding. P2: a 65 years
France, 3Agro-­Bio/Stago, Research & Development, La Ferté Saint-­Aubin, France old man with a recent diagnosis of multiple myeloma (monoclonal
IgG 2.4 g/dL), presenting spontaneous severe bilateral epistaxis.
Background: Diagnosis of thrombopathies (inherited or acquired) P3: a 31-­year-­old woman on a surgery plan with precedent of PTI
are performed by light transmission aggregometry (LTA) using spe- on childhood. Hemostatic routine tests;VWF:Ag;VWF:RCo;FVIII
cific agonists targeting the major receptors (thrombin: PAR-­1; ADP: ;FI;OA on platelet rich plasm (PRP) and mixing studies using ADP,
P2Y1 & P2Y12; epinephrine: α2A; collagen: GPVI) and arachidonic ADR, arachidonic acid, collagen and ristocetin; expression of GPIb,
acid signaling pathways (thromboxane: TxA 2). GPIIb, GPIIIa, direct and indirect platelet associated immunoglobulin
Aims: The aim of this study was to evaluate the performance of a new (PAIgG, PAIgM) and mepacrine uptake (MU) measured by FC; were
range of agonists (Arachidonic Acid, ADP, Collagen, Epinephrine and performed.
TRAP-­6) by determining reference intervals in healthy donors and diag- Results: Results are shown on Tables 
1 and 2. P1: showed
nosing of platelet dysfunction in comparison with reference agonists. Glanzmann′s thrombasthenia like pattern on OA; no inhibitory ef-
Methods: LTA (aggregometer TAV8, SD-­
Medical) was performed fect of patient′s poor platelet plasm (P′s PPP) upon a normal PRP
using citrated platelet-­rich plasma (cPRP). Platelet aggregation was and elevated levels of both PAIgG -­PAIgM on direct assay. P2: had
induced by different agonists (Agro-­Bio a Stago brand and reference exclusive absence of ristocetin aggregation (up to 2 mg/mL) that not
CE marked agonists) at different concentrations (low and high): ADP normalized when VWF was added; inhibition of normal PRP ristoce-
(2 and 10 μM), Arachidonic acid (1 mM), Collagen (2 and 10 μg/mL), tin aggregation (1.2 mg/mL) induced by P′s PPP and normal levels of
Epinephrine (5 and 25 μM) and TRAP-­6 (10 and 50 μM). Statistical PAIgG-­PAIgM. P3: showed exclusive absence of collagen aggrega-
analyzes were carried out using the MedCalc® software. tion; aggregation of normal PRP induced by P′s PPP (without agonist)
Results: Reference intervals (min, max) were defined on 82 healthy donors and elevated levels of PAIgM. All of them presented normal expres-
(50 women, 32 men with platelet count between 152-­402 G/L and ranged sion of GPIb, GPIIb, GPIIIa and MU.
from 56% to 100% for all agonists regardless of concentration, except for Conclusions: We presented three patients with different clinical
ADP to 2 μM. The performance of agonists was evaluated in the diagnosis context where the combination of functional studies as OA and the
of inherited thrombopathies (d-­storage pool disease; n=2, CalDAG-­GEFI evaluation of platelet glycoproteins and PA immunoglobulins by FC
and Kindlin-­3 deficiency; n=3), Glanzmann′s diseases were mimicated was the key to reach a diagnosis of APD, necessary to guide to the
with anti-­GPIIb-­IIIa (Abciximab; n=5) and in the diagnosis of acquired most appropriate treatment.

TA B L E  1

Platelet Bleeding Factor
Volume Time (Ivy Prothrombin Activated Partial Fibrinogen VIII VWF:Ag
Platelet Count (7.5-­11.5 1.5-­4.5 Time Thromboplastin (200-­ (70-­120 (50-­150 VWF:RCo
(150-­450.109/L) fL) min) (70-­120%) Time (37-­48 seg) 400 mg/dL) %) %) (50-­150 %)

PATIENT 150 6.8 not tested 95 39 265 140 142 113

1 (P1)
PATIENT 177 8.8 4.5 65 58 (normal: 42, 500 32 61 60
2 (P2) normal + patient:
PATIENT 106 9.8 >9.0 106 43 300 100 98 76
3 (P3)

TA B L E  2

Direct platelet Indirect platelet

associated associated
immunoglobulin immunoglobulin Expression
Mixing Test of GPIb,
(P´s PPP + IgG (<11.0 IgM (<6.0 IgG (<11.0 IgM (<6.0 GPIIb, Mepacrine
Aggregation pattern Normal PRP) MFI) MFI) MFI) MFI) GPIIIa uptake

Patient 1 Absent with: ADP, ADR, No 12.6 23.0 4.9 5.8 NORMAL NORMAL
(P1) arachidonic acid, collagen. inhibitory
Primary aggregation with effect
Ristocetin 1.5 mg/mL detected.
Patient 2 Normal with: ADP, ADR, Inhibitory 2.1 5.9 3.3 5.9 NORMAL NORMAL
(P2) arachidonic acid, collagen. effect over
Absent with: ristocetin up ristocetin
to 2 mg/mL. Do not aggregation
correct with VWF
Patient 3 Normal with: ADP, ADR, Aggregation 8.0 9.4 10 10.5 NORMAL NORMAL
(P3) arachidonic acid, without
ristocetin. Absent with: agonist
collagen up to 2 μg/mL.
Biphasic pattern (atypical)
with collagen 8 μg/mL

PB066 | Platelet Function in Polycythemia Methods: We obtained platelet responses to arachidonic acid (ASPI
test), adenosine diphosphate (ADP test) and thrombin receptor ac-
Vera and Primary Myelofibrosis Using Multiple
tivating peptide (TRAP test) and ROTEM curves in 14 patients with
Electrode Aggregometry
PMF and 23 patients with PV. Thirty-­one patients were treated with
N.S. Nytofte1; O.B. Pedersen2; H.C. Hasselbalch3 aspirin and two patients were treated with an ADP receptor antago-
Næstved Sygehus, Center for Thrombosis and Anticoagulation, Næstved, Denmark, nist. Cytoreductive treatment but not antiplatelet treatment were
Næstved Sygehus, Department of Clinical Immunology, Næstved, Denmark,
3 paused for at least 1 week before the MEA tests.
Zealand University Hospital, Department of Hematology, Roskilde, Denmark
Results: Neutrophil count, JAK V617F mutation allele burden and
ROTEM parameters of clot strength but not platelet count cor-
Background: The myeloproliferative neoplasms (MPNs) polycythemia
related with the ASPI test. In the multiple regression analysis, the
vera (PV) and primary myelofibrosis (PMF) are clonal diseases associ-
ASPI test correlated with neutrophil count, treatment with aspirin
ated with thrombosis and bleeding. Antiplatelet treatment is often used
and clot strength. The ADP test correlated with platelet count, neu-
in order to prevent thrombosis. Responsiveness to antiplatelet treat-
trophil count and ROTEM parameters of clot strength. The TRAP
ment can be monitored using multiple electrode aggregometry (MEA).
test correlated with the neutrophil count, the JAK2 allele burden
However, few studies have evaluated the use of MEA in PV and PMF.
and ROTEM parameters of clot strength. In the multiple regression
Aims: The aim of this study was to identify determinants of platelet
analysis, only the neutrophil count and ROTEM parameters of clot
function using MEA in patients with PV and PMF.
strength remained statistically significant.

Conclusions: Neutrophil count, clot strength and platelet count are

associated with platelet function. Clot strength predicted MEA tests
possibly due to platelet activation causing both increased clot stabil-
ity and adherence to the MEA electrodes. MEA should be used with
caution in MPNs.

PB067 | Platelet to Lymphocyte Ratio was

Significantly Lower for Slovak Postmenopausal
Women Born from November to April, Except
those with Leiomyoma
B. Bajteková1,2; S. Wimmerová1; V. Rusnáková1,3 F I G U R E   1   Distribution of PLR, Fe and WBC is not the same
Slovak Medical University, Faculty of Public Health, Institute of Biophysics across categories Winter/Summer SOB, moreover is influenced by
Informatics and Biostatistics, Bratislava, Slovakia, 2GynAc, s.r.o., Clinic of the leiomyoma presence
Obstetrics and Gynaecology, Šaštín-­Stráže, Slovakia, 3Trnava University, Faculty
of Health Sciences and Social Work, Trnava, Slovakia

Background: Several studies have confirmed that season of birth

(SOB) affects the incidence of some tumors. In our previous study,
the relative risk of uterine myomas was 2.2 times lower (P=0.006) for
women born in the Winter SOB (November-­April) than for women
with the Summer SOB (May-­October). Myoma is a benign uterine
tumor with incidence up to 40%.
Aims: Test the hypothesis that distribution of blood count param-
eters, platelet to lymphocyte ratio (PLR), neutrophil-­to-­lymphocyte
ratio (NLR), serum iron (Fe) and calcium is the same across categories
“Winter-­SOB” and “Summer-­SOB”. F I G U R E   2   The mean and median values of some tested
Methods: Retrospective analysis. Cohort of 211 Slovak postmeno- parameters, depending on the presence of myoma and season of
pausal women, born in 1941-­76, screened by vaginal ultrasonogra- birth (SOB)
phy and split into four groups according to the SOB and occurrence
of uterine myoma. Tested parameters of blood count: Hemoglobin, increases with age. So in studies involving postmenopausal women it
Erythrocyte count, Mean corpuscular volume, Platelet count, Mean seems to be wise to consider myoma impact and SOB.
platelet volume (MPV), White blood cell count (WBC), Absolute
neutrophil and lymphocyte count. Null hypothesis were tested by
Mann-­Whitney U test, in 3 modes: PB068 | Mild Thrombocytopenia and Bleeding
A: postmenopausal women without uterine leiomyoma (n=78), SOB
Diathesis in Mice Lacking the PLT/MPV-­
Winter to SOB Summer.
B: just women with myoma (n=133), SOB Winter to SOB Summer.
associated F-­BAR Protein PACSIN2
C: all postmenopausal women together, SOB Winter to SOB Summer. N. Eaton1; C. Drew1; J. Wieser1; W. Kahr2; M. Plomann3;
Results: A: Platelet to lymphocyte ratio was significantly lower for H. Falet1
postmenopausal women born in winter season (P=0.020). However Blood Research Institute, Milwaukee, United States, 2Hospital for Sick Children,
University of Toronto, Toronto, Canada, 3University of Cologne, Cologne,
this did not apply to women with uterine myoma. Alike, serum iron
was lower for postmenopausal women born in summer just in case
they have no myoma (P=0.022; Figure 1).
Background: Exomechip analysis has associated a single nucleo-
B: Women with myomas born in the summer, had significantly
tide variant in the gene encoding the F-­BAR protein PACSIN2 with
higher WBC versus myomatous women with winter SOB (P=0.011;
altered platelet count and mean platelet volume. We have previ-
Figure 2).
ously shown that PACSIN2 associates with cell membrane GPIbα
C: No difference in this case.
in human and mouse platelets and is concentrated in mid-­stage
Conclusions: Distribution of PLR, Fe and WBC is not the same across
mouse megakaryocytes (MKs) in a well-­defined invagination of the
categories Winter and Summer SOB, moreover is influenced by the
cell membrane reminiscent of the initiating demarcation membrane
presence of uterine leiomyoma. The incidence of myoma is high and
system (DMS).

Aims: Here we investigated the role of PACSIN2 in platelet produc- with LASSBio-­788 once daily, totalizing 15 days. The animals were
tion and hemostatic function. euthanized by cervical dislocation under anesthesia. Blood samples
Methods: In vivo and in vitro platelet parameters were determined were collected and the platelets were isolated to biochemical and
in Pacsin2-­null mice. molecular analysis. Data were analyzed using one-­
way ANOVA,
Results: Pacsin2-­null mice developed a mild thrombocytopenia with P<0.05 (CEUA/UFF 856/16).
slightly enlarged and shallow platelets. Pacsin2-­null platelets sur- Results: The hypercholesterolemic diet increased serum lipids levels
vived normally in vivo, and expressed normal levels of major plate- in the HC group. LASSBio-­788 reduced serum lipids. Furthermore,
let membrane glycoproteins on their surface. Pacsin2-­null mice had the hypercholesterolemic diet increased TXA 2 and a decrease of
a mild bleeding diathesis, as evidenced by significantly prolonged cAMP and cGMP. LASSBio-­788 inhibited these effects. NF-­kB sign-
tail bleeding, and had delayed in vivo thrombus formation follow- aling events, including PKC-­α , iNOS, P-­selectin and CD40L increased
ing FeCl3 mesenteric vessel injury. The functionality of Pacsin2-­null protein expression, IκB-­α degradation, eNOS, PKA and PKG de-
platelets in whole blood was tested in flow chamber experiments creased protein expression were observed in hypercholesterolemic
using the VenaFlux platform, where binding to a collagen-­coated rats. These signaling events were attenuated by chronic treatment
surface was measured at arterial and venous shear conditions. with LASSBio-­788.
Compared to control platelets, Pacsin2-­null platelets covered 25% Conclusions: Our study demonstrate for the first time that
of the collagen-­coated surface at high shear, but had normal adhe- LASSBio-­788 possesses potent antiplatelet activity, which may in-
sion at low shear. Pacsin2-­null platelets expressed P-­selectin and ac- volve activation of the eNOS/NO/GC/cGMP/PKG pathway, resulting
tivated the integrin αIIbβ3 normally in response to stimulation with in inhibition of the NF-­κB/PLC/PKC/TXA 2 cascade, and, finally, inhi-
thrombin and the GPVI-­specific collagen-­related peptide. Analysis bition of platelet aggregation. Our results indicate the LASSBio-­788
of bone marrow sections revealed normal MK numbers and ploidy compound as a potential drug for the treatment of cardiovascular
in Pacsin2-­null mice. However, the DMS appeared to be less well de- disorders such as atherosclerosis.
fined in Pacsin2-­null MKs, where distinct platelet territories were not
readily visualized.
Conclusions: The data indicate that the F-­BAR protein PACSIN2
PB070 | New Clinicopathologic Finding in
plays a critical role in platelet hemostatic function at high shear and
a Patient with Acquired Amegakaryocytic
in platelet production.
Thrombocytopenia Treated with Thrombopoietin
Receptor Agonist Therapy
PB069 | A Novel Role of LASSBio-­788 in R. Smith1; S. Zandi1; Z. Liederman1; M. Xu1,2; G. Lim2;
H. Ghaffar2; H. Ni1,2,3; M. Sholzberg4
Inhibiting NF-­κB Mediated Signaling in Platelet 1
University of Toronto, Toronto, Canada, 2St. Michael’s Hospital, Toronto,
of Hypercholesterolemic Rats: Would Be New Canada, 3Canadian Blood Services Centre for Innovation, Toronto, Canada, 4Li
Antiplatelet Agent? Ka Shing Knowledge Institute, St. Michael’s Hospital, Division of Hematology/
Oncology, Division of Laboratory Medicine and Pathobiology, Toronto,
N.A.V. Motta1; G.F. Lima1; E. Barreiro2; A. Kümmerle3; F.C.F. Canada
Brito1; Laboratorio de Farmacologia Experimental (LAFE)
Universidade Federal Fluminense -­UFF, Niteroi, Brazil, 2Universidade Federal Background: We present a case of a 73-­year old woman with iso-
do Rio de Janeiro, Rio de Janeiro, Brazil, 3Universidade Federal Rural do Rio de
lated idiopathic thrombocytopenia given an initial diagnosis of
Janeiro, Rio de Janeiro, Brazil
immune thrombocytopenia (ITP). After brief improvement with
corticosteroids and intravenous immunoglobulin, she experienced
Background: Circulating platelets become hyperactive under hyper-
multiple relapses unresponsive to standard therapies includ-
cholesterolaemic conditions, leading to the development and pro-
ing splenectomy. Bone marrow evaluation showed an absence
gression of atherosclerosis. LASSBio-­788 is a thienylacylhydrazone
of megakaryocytes thus the diagnosis was amended to acquired
derivative with anti-­atherogenic properties in an animal model of hy-
amegakaryocytic thrombocytopenia (AAT). Romiplostim, a second
percholesterolemia. Although the molecular mechanism underlying
generation thrombopoietin receptor agonist (TPO-­R A) was initi-
its antiplatelet effect remain unclear.
ated, and her disease remains in complete remission over 4 years
Aims: The aim of this study was to investigate the signaling pathway
since its initiation.
involved in the antiplatelet effects of LASSBio-­788.
Aims: To describe our experience with Romiplostim in a patient
Methods: Adult male Wistar rats (150-­200 g) were randomly divided
with refractory AAT, explore the pathogenesis, and evaluate the
into four groups: control group (C) and positive control (C+788) fed
bone marrow stem and progenitor cell composition after 4 years of
standard chow diet, hypercholesterolemic diet group (HC) and diet
group + compound LASSBio-­788 (HC+788) fed a hypercholester-
Methods: Patient serum was tested for anti-­
platelet antibod-
olemic diet. At 31° diet day, was performed the chronic treatment
ies. Phenotypic analysis of bone marrow stem and progenitor

NOD/SCID mouse model was used to analyze the impact of glycan

patterns on survival of human platelets.
Results: Sera from ITP patients and healthy donors were studied.
In the LBA different patterns of glycan modification on PLT surface
were observed. 30% sera induced high ECL binding and 60% sera
caused a significant increase in PNA binding compared to heathy
F I G U R E   1   Phenotypic analysis shows expansion donors. Of note, 30% sera showed strong decrease in RCA bind-
of CD34lowCD45low population comprised mostly of ing. Interestingly, a subgroup of ITP-­sera was able to modify glycan
megakaryocyte-­biased progenitors (CD71+cMPL+ cells)
pattern on megakaryocytes surface (3 fold increase in RCA-­binding
compared to control). The injection of AAbs induced an accelerated
compartments using 14 colour flow cytometric technique, while on
clearance of human PLTs from the circulation in vivo. The destruc-
TPO-­R A was performed.
tion of human PLTs by ITP-­A Abs was largely reduced but not com-
Results: At the onset of her diagnosis, we identified anti-­platelet
pletely prevented by a specific neuraminidase inhibitor that blocks
antibodies directed against glycoprotein Ib (GPIb), but not against
glycan changes on PLT surface (platelet survival after 5 hours: 60%
GPIIb/IIIa. Flow cytometry showed expansion of a megakaryocyte-­
vs. 40%).
biased progenitor population within the bone marrow compared to
Conclusions: Our results indicate that AAbs from ITP patients are
normal controls (Figure 1).
able to induce changes in glycan patterns on MKs and PLT surfaces.
Conclusions: We have isolated antibodies against GPIb, which may
The mechanism of antibody-­mediated modification of glycan pat-
represent a novel etiopathological mechanism in AAT as the anti-
terns seems to contribute to PLT destruction as well as to interfere
body may interfere with hepatic TPO generation, as shown by Xu
with PLT production from MKs.
et al. ISTH 2017. Moreover, phenotypic analysis of bone marrow,
while on TPO-­R A, revealed changes in the architecture of hemat-
opoietic hierarchy thus demonstrating plasticity of bone marrow
responsiveness in a condition on the bone marrow failure spectrum.
Our experience adds support to the growing literature for TPO-­R A
therapy in cases of refractory AAT. Further studies are warranted to PB072 | ThromboGenomics: High Throughput
understand the significance of the observed phenotypic changes in Sequencing in Patients with Bleeding and Platelet
the bone marrow. Disorders
K. Downes1; K. Freson2; K. Gomez3
University of Cambridge, Department of Haematology, Cambridge, United
PB071 | Antibody Mediated Glycan Kingdom, 2Center for Molecular and Vascular Biology, Department of
Modification on Platelets and Megakaryocytes: Cardiovascular Sciences, Leuven, Belgium, 3Haemophilia Centre and Thrombosis
Unit, Royal Free London NHS Foundation Trust, London, United Kingdom
A Potential Role in Platelet Destruction in
Autoimmune Thrombocytopenia Background: Individuals with bleeding and platelet disorders (BPDs)
I. Marini; R. Jouni; F. Rigoni; T. Bakchoul present as a clinically diverse group of patients with symptoms rang-
Zentrum für Klinische Transfusionsmedizin Tübingen gGmbH (ZKT), Tübingen, ing from life threatening hemorrhage to mild but persistent bleeding
Germany symptoms and thrombocytopenia with non haematological clinical
manifestations. Laboratory investigations frequently fail to provide
Background: Immune thrombocytopenia (ITP) is a bleeding dis- an explanation for the symptoms or to identify those at significant
ease caused by autoantibodies (AAbs) directed against PLT gly- risk of future bleeding. In addition, inherited thrombocytopenia
coproteins (GP). Recently, an Fc-­independent platelet clearance often cannot be distinguished from acquired disorders.
via Ashwell-­Morell receptors, which recognize glycan changes on Aims: We have developed a High Throughput Sequencing (HTS) ap-
platelet surface, has been proposed as a new mechanism of ITP in proach to provide a molecular diagnosis for patients suspected of
mice. having an inherited BPD caused by variants in one of the 70 ISTH-­
Aims: In this study we investigated the effects of AAbs from SSC approved Tier 1 BPD genes.
ITP patients on the glycan pattern of human megakaryocytes Methods: To date, over 1000 patients have been referred for test-
(MKs) and PLTs and the subsequent effects on their survival ing. Laboratory and clinical phenotypes were recorded using Human
in vivo. Phenotype Ontology (HPO) terms. Variants were called and prioritised
Methods: MKs were differentiated from CD34+ cells after cultiva- based on minor allele frequency, predicted impact and presence in the
tion with thrombopoietin. Megakaryocytes and platelets were in- Human Gene Mutation Database. Variants present in the 70 known BPD
cubated with ITP-­sera and the modification in glycan pattern was genes were reviewed by a multi-­disciplinary team (MDT) where patho-
assessed by flow cytometry using two lectins: RCA, ECL and PNA. genicity was assigned to variants using ACMG and ACGS guidelines.

Results: Pathogenic variants were identified in approximately half of pa- pronounced cytosolic calcium signaling and thus to a significant
tients where 40% were novel. Using sequencing read depth large-­scale increase in the probability of mitochondrial calcium overload and
pathogenic deletions were identified in approximately 3% of patients. platelet necrosis.
Conclusions: These results demonstrate the success of HTS ap- Conclusions: WAS platelets spontaneously expose PS via a
proaches in providing a genetic diagnosis for patients with well-­ mitochondria-­dependent necrotic mechanism that can be explained
defined inherited platelet or coagulation defects. Furthermore by purely geometric consideration. This could contribute to the de-
extensive replication has been demonstrated for recent BPD gene velopment of thrombocytopenia in WAS.
discoveries, for example RNU4ATAC, SLFN14, RASGRP2, ETV6 and
ACTN1 and for novel modes of inheritance, for example GP1BB auto-
somal dominant thrombocytopenia. All MDT-­reviewed variants and PB074 | Septic Conditions Modulate the Level
appended HPO terms are shared in ClinVar, supporting the interna-
of miRNAs in Platelets and Megakaryocytes via
tional effort to improve reference catalogues.
Altered Dicer Level
Z. Fejes1; B. Szilágyi1; T. Orosz1; Z. Czimmerer2; S. Póliska2;
G. Nagy3; J. Kappelmayer1; B. Nagy Jr1
PB073 | Mitochondria-­dependent Necrosis Is 1
University of Debrecen, Faculty of Medicine, Department of Laboratory
Promoted in Wiscott-­Aldrich Syndrome Platelets as Medicine, Debrecen, Hungary, 2University of Debrecen, Faculty of Medicine,
a Result of the Abnormal Surface-­to-­Volume Ratio Department of Biochemistry and Molecular Biology, Genomic Medicine and
Bioinformatics Core Facility, Debrecen, Hungary, 3University of Debrecen,
S. Obydennyi1,2; E. Artemenko1; A. Sveshnikova1,2,3; Faculty of Medicine, Department of Anesthesiology and Intensive Care,
A. Ignatova1; G. Novichkova1; A. Maschan1; A. Shcherbina1; Debrecen, Hungary
M. Panteleev1,2,3
Dmitry Rogachev National Research Center of Pediatric Hematology, Oncology Background: We have previously described decreased levels of
and Immunology, Moscow, Russian Federation, 2Center for Theoretical Problems activation-­
dependent platelet miRNAs contributing to elevated
of Physico-­Chemical Pharmacology, Moscow, Russian Federation, 3Moscow State
University named after M.V. Lomonosov, Moscow, Russian Federation
platelet activation is sepsis. However, the regulatory mechanisms
of altered platelet miRNAs among these conditions are largely
Background: Wiskott-­Aldrich syndrome (WAS) patients suffer from
Aims: In this study, we examined whether Dicer level is altered in
bleeding caused by microthrombocytopenia of unclear origin. Some
platelets of septic patients that modulate miRNA expression. In par-
of the previous reports found WAS platelets to have increased phos-
allel, Dicer level was analyzed in platelets and megakaryocytes (MKs)
phatidylserine (PS) expression.
after lipopolysaccharide (LPS) treatment in vitro.
Aims: To investigate dynamics and mechanisms of PS exposure in
Methods: Dicer1 mRNA levels were investigated in leukocyte-­
WAS platelets.
depleted platelet samples obtained from 10 individuals suffering
Methods: Whole blood of 19 WAS patients and 14 healthy donors
from sepsis and in 14 age-­and gender-­matched healthy controls.
was collected into sodium citrate under a protocol approved by the
We also quantified miR-­223 and miR-­26b with target P2RY12 and
institutional ethics committee. Platelet function including granule
SELP mRNAs, respectively, by RT-­qPCR in patient samples as well
release, integrin activation, and procoagulant activity was investi-
as in stimulated platelets by LPS (O55:B5, up to 3 μg/mL) or PAR1
gated before and after stimulation using flow cytometry. Real-­time
agonist TRAP (30 μM) for 30 minutes. To prove the effect of altered
cytosolic calcium dynamics, mitochondrial membrane potential and
Dicer level of MKs in sepsis, miRNA and Dicer levels were analyzed
PS exposure were investigated in single fibrinogen-­bound platelets
in MEG-­01 and K562-­MK cells after the treatment with LPS (100 ng/
using confocal microscopy. Computer systems biology model of
mL) for 24 hours.
platelet calcium homeostasis was used for the data analysis.
Results: Septic conditions induced inflammatory response in plate-
Results: WAS platelets responded normally to potent dual-­agonist
lets and MKs showing elevated miR-­155 in both ex vivo and in vitro
stimulation, but had abnormally high PS exposure when stimulated
experiments. Patient platelet samples and both LPS-­treated MK
via PAR1 only. Fibrinogen-­immobilized WAS platelets spontaneously
cultures showed decreased Dicer1 mRNA levels resulting in low-
lost mitochondrial membrane potential followed by PS exposure
ered miR-­223 and miR-­26b with upregulated P2RY12 and SELP
(20.1% vs. 3.6% in healthy donors). Necrostatin-­1, calpeptin, Z-­VAD-­
mRNA levels. Interestingly, short-­term LPS stimulation elevated
FMK did not affect this, while cyclosporin A reduced PS+ fraction.
Dicer1 mRNA level with higher miRNAs in platelets compared to
The number of mitochondria was decreased in WAS platelets and
untreated samples, while TRAP did not markedly affect these RNA
was a reliable predictor of spontaneous necrosis: 50% of WAS plate-
lets with <4 mitochondria exposed PS and 19% for ≥4 mitochondria.
Conclusions: Abnormal Dicer levels are generated by septic milieu
Systems biology analysis revealed that the smaller platelets of WAS
that alter the levels of platelet and MK miRNAs in sepsis. However,
patients would have higher cytosolic inositol-­triphosphate concen-
distinct regulatory mechanisms seem to function in platelets after
tration caused by higher surface-­to-­volume ratio. This leads to more

short-­and long-­
term septic triggers, which need to be further PB076 | Bleeding Diathesis Associated with
Severe Acquired Platelet Dysfunction and
This study was supported by “ÚNKP-­17-­3 New National Excellence
Overproduction of Cyclic AMP in Two Patients
Program of the Ministry of Human Capacities” (Z.F.).
with Hematologic Malignancies
A. Lecchi1; E.A. Femia1; S. La Marca1; A. Artoni1; F. Onida2;
PB075 | Evidence of a Dominant Negative F. Peyvandi1; M. Cattaneo3
A. Bianchi Bonomi Hemophilia and Thrombosis Center, Department of Internal
Effect of Gain-­of-­Function ITGB3 Mutations over Medicine Fondazione IRCCS Ca’ Granda -­Ospedale Maggiore Policlinico, Milano,
Loss-­of Function Mutations in Variant Glanzmann Italy, 2University of Milan, BMT Center Fondazione IRCCS Ca’ Granda Ospedale
Maggiore Policlinico, Hemato-­Oncology Department, Milano, Italy, 3Medicina II,
Thrombasthenia Azienda Ospedaliera Santi Paolo e Carlo, Dipartimento di Scienze della Salute,
L. Bury1; E. Zetterberg2; E. Leinoe3; E. Falcinelli1; Università degli Studi di Milano, Milano, Italy

A. Marturano1; G. Manni1; A.T. Nurden4; P. Gresele1

University of Perugia, Department of Medicine, Perugia, Italy, 2Skaane Background: The pathophysiology of impaired platelet function in
University Hospital, Department of Haematology, Coagulation Unit, Lund, acquired disorders is often poorly characterized.
Sweden, 3Copenhagen University Hospital, Rigshospitalet, Department of
Aims: We describe a severe bleeding diathesis in two unrelated pa-
Haematology, Copenhagen, Denmark, 4Hopital Xavier Arnozan, French National
Reference Centre for Platelet Disorders, Pessac, France tients with hematologic malignancies, associated with severe plate-
let abnormalities, including overproduction of inhibitory cyclic-­AMP
Background: The autosomal dominant form of GT, associated with (cAMP).
platelet dysfunction, macrothrombocytopenia and bleeding, is caused Methods: Patient 1 was a 74-­year-­old woman with severe mucocu-
by gain-­of-­function (GOF) variants of ITGA2B or ITGB3 that generate a taneous bleeding of recent onset, mild macrocytic anemia and
constitutively activated αIIbβ3. GOF ITGB3 variants exert a dominant ef- thrombocytopenia (100×109/L), who was diagnosed with chronic
fect over normal ITGB3, leading to the clinical picture of autosomal dom- myelomonocytic leukemia few months later.
inant GT. No information, however, is available on the impact of GOF Patient 2 was a 77-­year-­old man with myelodysplastic syndrome
variants when a concomitant loss-­of-­function (LOF) variant is present. with excess blasts. He had mucocutaneous bleeding of recent onset,
Aims: We aimed to explore the effect of GOF ITGB3 mutations over despite a platelet count persistently over 100×109/L. Personal and
LOF mutations starting from the characterization of a family with a family history for abnormal bleeding was silent in both patients.
novel form of autosomal dominant GT caused by co-­inheritance of Prothrombin time (PT), activated partial thromboplastin time (aPTT),
two ITGB3 variants. von Willebrand Factor (vWF) antigen and RiCof, the concentrations
Methods: Platelet α IIbβ3 expression and activation were assessed of platelet fibrinogen, ADP, ATP, serotonin, cAMP and the serum
by flow cytometry, DNA variants by Sanger sequencing. Mutant β3 levels of thromboxane B2 were measured by standard techniques.
expression vectors were generated and transfected into CHO cells Platelet aggregation and secretion were tested by lumiaggregometry.
expressing α IIb. α IIbβ3 expression, activation and internalization were The platelet expression of CD41/61 and CD42b and the vasodilator-­
assessed by flow cytometry, adhesion on fibrinogen by spreading stimulated phosphoprotein (VASP) phosphorylation were tested by
assay, FAK phosphorylation by Western blotting. flow cytometry.
Results: We diagnosed autosomal dominant GT in a young female Results: PT, aPTT, vWF, CD41/61 and CD42b were normal.
with long-­
lasting mucocutaneous bleeding and found compound Agonists-­
induced platelet aggregation and ATP secretion were
heterozygosity for two novel ITGB3 variants: R786W and N331S. severely impaired. Serum TxB2 and the platelet content of ADP,
The autosomal-­
dominant GT phenotype co-­
segregated with the ATP, serotonin and fibrinogen, were markedly decreased. Baseline
N331S variant among family members. Expression studies revealed platelet cAMP levels were mildly elevated (pt1: 21; pt2: 15; healthy
that R786W was a LOF variant causing reduced cell spreading and controls: 3.-­14 pmoles/109plts) and were markedly increased by
FAK phosphorylation on fibrinogen. N331S was a GOF variant and, PGE1 (pt1: 473; pt2: 325; healthy controls: 23-­97 pmoles/109plts).
of note, when co-­expressed with normal β3 or R786W, exerted a ADP normally inhibited PGE1-­induced cAMP increase and VASP
dominant-­negative effect producing impaired αIIbβ3 surface expres- phosphorylation.
sion, increased internalization, PAC-­1 binding and FAK phosphoryla- Conclusions: To the best of our knowledge, these are the first pa-
tion under resting conditions and accelerated spreading. tients with acquired bleeding diathesis associated with severe plate-
Conclusions: We show that a GOF mutation of β3 can exert its let dysfunction and overproduction of platelet cAMP.
negative effect even when associated with a LOF
β3 mutant on the second allele, giving rise to the clinical pic-
ture of autosomal dominant GT of platelet dysfunction and

PB077 | In-­silico Prediction of Molecular Mass calculated by umbrella sampling method with mass center distance
between both molecules at every 0.5 Å.
Distance Dependent Changes in Binding Energies
Results: Most energetically stable structure of VWF bound with
between von Willebrand Factor and Platelet
wild type and G233V mutant GPIbα was achieved at 27.2 Å and
Membrane Glycoprotein (GP) Ibα in Wild Type 27.7 Å, respectively. Stable binding structure of VWF with wild type
and G233V Mutant and G233V GPIbα did not differ markedly (panel A)The values of
S. Goto1; N. Tamura2; K. Ayabe3; H. Yabushita3; H. Oka3; the potential of mean forces (PMF) generated between wild-­type
S. Goto3 VWF and GPIbα in wild type and G233V mutant are shown in Figure
Keio University School of Medicine, Cardiology, Tokyo, Japan, 2Tokai University Stable s… panel B. The binding energy had a single peak of 20.0 kcal/
School of Medicine, Kanagawa, Japan, 3Tokai University School of Medicine, mol at mass centers distance of 27.2 Å in wild-­t ype GPIbα and VWF,
Tokyo, Japan
while it showed 2 distinct peaks consisting of a 20.0 kcal/mol peek at
27.7 Å and another 5.3 kcal/mol peek at a longer distance of 50.0 Å
Background: Platelet adhesion under blood flow is mediated exclu-
for G233V GPIbα and VWF.
sively by binding of platelet membrane glycoprotein (GP) Ibα with
Conclusions: Our results suggest that the apparent “gain of function
von Willebrand factor (VWF). Single amino acid substitution from
phenotype” of G233V GPIbα mutation may depend on 2 energeti-
Gly to Val at position 233 of GP1bα is known as a cause of platelet-­
cally stable structure of GPIbα binding with VWF in G233V mutant.
type von Willebrand diseased due to enhanced binding of GP1bα
to VWF. However, detailed molecular basis for this phenotype of
G233V mutant of GPIbα for the interaction with VWF is still to be
elucidated. PB078 | Molecular Diagnosis of Inherited
Aims: To unravel the molecular mechanism for the apparent “gain of Platelet Disorders by Analysis of the MYH9 and
function” phenotype of G233V mutation. the Genes Encoding the GB1b-­IX-­V Complex in
Methods: Molecular dynamics (MD) calculation was conducted to an Irish Cohort
predict the binding characteristics of N-­terminus domain of GPIbα
R. Campbell; S. Mc Namara; L. Dempsey; M. Byrne; K. Ryan;
and A1 domain of VWF. The NAno Molecular Dynamics (NAMD)
B. White; N.M. O’Connell; C. Keenan
software using Chemistry at Harvard Macromolecular Mechanics
National Coagulation Centre, St James’s Hospital, Dublin, Ireland
(CHARMM)-­22 force field database were applied for wild-­t ype and
G233V mutation in GPIbα binding with wild type VWF. The Binding
Background: Congenital thrombocytopenias are a group of rare
energy between G233V mutant of GPIbα with wild-­t ype VWF were
disorders resulting in thrombocytopenia, giant platelets, and often
comorbidities. The presentation of common symptoms can lead
to misdiagnosis and inappropriate treatment. The MYH9 gene and
the genes encoding the GB1b-­IX-­V complex are both associated
with similar clinical phenotypes caused by two distinct disorders,
related disorders and Bernard Soulier Syndrome respec-
tively. Molecular analysis may provide a definitive diagnosis in these
patients and identify inheritance patterns.
Aims: Here we describe a cohort of patients with macrothrombo-
cytopenia investigated for variants within the MYH9, GP1Ba, GP1Bb
and GP9 genes.
Methods: After informed counselling and consent, direct re-­
sequencing was performed (TA cloning on one patient).
Results: 37 patients were referred for molecular analysis. Initial
targeted analysis of MYH9 detected pathogenic variants in 11 of
32 patients (34%) in exons 2, 31, 39 and 41. No variant was de-
tected in 20 patients (63%). Eight were further analysed for the
remaining MYH9 coding regions and no variant detected. The
pathogenic variant in exon 2 was a complex rearrangement pre-
dicting p.Val59_Glu60ins9. This patient had a severe phenotype
consistent with variants within the head domain of the MYH9
protein product NMMHC-­IIA. Analysis of GP1Ba, GP1Bb and GP9

F I G U R E 1   Stable structure and potential of mean force in six patients detected a GP1Ba variant in three members of the
calculated for wild-­t ype and G233V GP1b-­alpha and wild-­t ype same family (exon 2; p.Gly249Val), associated with pseudo-­

VWD. The same variant in GP9 (exon 3; p.Asn61Ser) was detected PB081 | CD8 T Cell-­mediated Platelet
in three patients from two families.
Clearance Does Not Occur via Traditional
Conclusions: Targeted MYH9 analysis to regions commonly associ-
Cytotoxic Protein Killing
ated with pathogenic variants proved successful in 30% of patients.
The use of NGS technology would greater increase the likelihood C. Maier; S. Patel; A. Bennett; S. Stowell
of detecting a pathogenic variant in this group. We have introduced Emory University School of Medicine, Department of Pathology and Laboratory
Medicine, Atlanta, United States
integrated reporting to include platelet morphology, clinical pheno-
typing, immunophenotyping and genetic data to comprehensively
Background: Poor survival of transfused platelets (plt) remains a
diagnose inherited platelet disorders.
significant problem, especially in multiply transfused patients with
immune-­mediated refractoriness. While patients with alloantibod-
ies can be managed with matched (ie antigen-­negative) plt, patients
PB079 | Investigation of Polymorphic with T cell-­mediated refractoriness remain difficult to treat, as this
Variations in Glutathione S Transferase Genes etiology of plt clearance has only recently been identified and its
among Immune Thrombocytopenia Patients in mechanism is unknown.
Pakistan Aims: We aimed to further characterize the CD8 T cell/allogeneic
plt interaction and to specifically investigate the role of the typical
S.N. Mukry; A. Arshad; A. Shahni; T.S. Shamsi
perforin/granzyme cytotoxic protein pathway.
National Institute of Blood Disease and Bone Marrow Transplantation,
Department of Post Graduate Studies & Research, Karachi, Pakistan Methods: CD8 T cells from OTI transgenic mice, OTI/perforin
knock-­out mice (OTI/prf−/−), or WT mice were co-­cultured with WT

Background: Immune thrombocytopenic purpura (ITP) is an autoim- or mOva plt for 24 hours and analyzed by flow cytometry. In vivo T

mune clinical syndrome characterized by a low platelet count, which cell-­mediated plt clearance was modeled by transfusion of mOva or

is the result of increased platelet destruction and insufficient plate- WT plt directly into OTI, OTI/prf−/−, and WT recipients, or by trans-

let production [1]. The pathophysiology of ITP is heterogeneous and fusion into WT mice adoptively transferred with OTI, OTI/prf−/−, or

complex. Recently, several evidences have suggested that oxidative WT splenocytes. Plt survival was measured and significance deter-

stress plays an important role in the pathogenesis of autoimmune mined when P<0.05 by one-­way ANOVA with multiple comparisons

diseases, and it is also involved in the development of ITP [2, 3]. by Tukey’s test.

Glutathione S transferases are phase II antioxidant or detoxifying Results: mOva plt are more adherent to OTI and OTI/prf−/− CD8 T cells

enzymes which conjugates reduced form of glutathione to xeno- then to WT CD8 T cells, and CD8 T cells from OTI and OTI/prf−/− mice

biotics for detoxification and thereby control oxidative imbalance. express the activation marker CD69 after co-­culture with mOva, but

Aims: To investigate the possible role of polymorphic variations of not WT, plt. Transfusion of mOva plt into either OTI or OTI/prf−/− mice

phase II antioxidant enzyme GSTT1 and GSTM1 genes in the patho- results in equally significant clearance. Specifically, mOva plt survival

genesis of ITP. in OTI or OTI/prf−/− recipients is significantly less (<30%) at 24 hours

Methods: Multiplex PCR for GSTT1 and GSTM1 was performed than in WT recipients (>60%). Adoptive transfer experiments show

using previously designed primers in compliance with the declara- similar results, whereby recipients receiving OTI or OTI/prf/− spleno-

tion of Helsinki. cytes have significantly decreased mOva plt survival.

Results: A total of confirmed 197 cases (99 females and 98 males) Conclusions: We conclude that minor antigen mismatched plt ad-

of ITP with mean age of 35 (median) range (3-­8 0) and 167 normal here to and activate CD8 T cells, resulting in plt clearance that is

healthy individuals of same sex and age as control were analyzed. independent of the traditional cytotoxic killing pathway via perforin

GSTM1 null deletion was significantly more frequent in ITP patients and granzyme.

(49.23%) as compared to controls (32.54%). The ITP patients have

low frequency of GSTT1 null deletion (5.58%) as compared to con-
trols (13.60%). Double deletion was observed in ITP and controls PB082 | Human Platelets Labeled at Multiple
both with almost similar frequency i.e. 9.18% and 8.88% respec- Biotin Densities: A Promising Approach for
tively. These overall differences in frequencies were statistically
Monitoring in vivo Platelet Survival in Clinical
significant (P=0.005).
Conclusions: This study has allowed determining the frequency of
GSTM1 and GSTT1 polymorphism in a sample of our patients. A sig- C. Ravanat; V. Heim; A. Pongérard; A. Dupuis; C. Gachet
nificant increase has also been observed in control group which sug- Université de Strasbourg, INSERM, EFS Grand Est, BPPS UMR-­S 1255, FMTS,
Strasbourg, France
gest they might be susceptible to disease occurrence. GSTM1 null
deletion was more frequent and significant in ITP patients (P=0.007).
Background: Biotin (Biot) is an alternative to radioactive blood
These findings might provide a new hypothesis for the initiation of
cell tracers. It allows one to concurrently track in vivo multiple cell
abnormal autoimmunity and a possible novel therapeutic approach.
TA B L E   1   Results for in vitro functions of biotinylated human platelets (Mean±SEM)

Biot-­sulfo-­NHS (µM) 0 4.5 45 180

Mean Fluorescence Intensity 0.7±0.3 3.7±0.2 35±1.2 168±27

Quantitation of P-­selectin expression 7086±631 6244±709 6156±868 3040±492
(60  µM-­TRAP)
Amplitude of aggregation (5 µM-­ADP) 100 97±5 51±6 4±1.6
Amplitude of aggregation (2.5 µg/ 100 101±2 88±6 36±21
mL-­collagen) %
Amplitude of aggregation (1 mM-­ 100 101±0.7 96±3 80±9
arachidonic acid) %

populations labeled at different Biot densities. Multi-­labeled red PB083 | Quantitative Alteration of Vesicle
blood cells have been safely transfused in man to assess their sur-
Associated miRs and RISC Complex Reflect
vival (Mock, 2014). However, only one study in man has described
Platelet Concentrate Molecular Pattern during
the tracking of a Biot-­platelet population (Stohlawetz, 1999).
Aims: Our aim was to label human platelets (PLTs) with different
Storage as Platelets Activity Model
levels of Biot to determine whether biotinylation affects their func- V. Kishenko1; K. Kondratov1; A. Fedorov1; O. Sirotkina1,2
tion. This was a preliminary study to explore the feasibility of multi-­ 1
Almazov National Medical Research Centre, Saint-­Petersburg, Russian
labeled Biot-­PLT production for clinical trials. Federation, 2Pavlov First Saint Petersburg State Medical University, Saint-­
Petersburg, Russian Federation
Methods: PLTs in Tyrode’s albumin (TA) were treated with Biot-­
sulfo-­NHS (0, 4.5, 45 and 180 μM), washed twice and re-­suspended
Background: Platelet concentrate (PC) storage is accompanied by
(3×108/mL) in TA/PPPcit (v/v). Biot densities were controlled by
permanent release of platelet (PLT) derived membrane vesicles. PLT
FACS analysis with streptavidin-­PE labeling and PLT functions by
miRs associate with microparticles and form complexes with RISC-­
assessing parameters such as secretion, glycoprotein (GP) integrity
associated proteins (Ago2 and GW182).This process serves as a
and aggregation.
platelets activity model and molecular composition of PLT derived
Results: Various Biot-­
PLT populations could be distinguished by
vesicles – as biomarkers of platelet reactivity.
FACS and all displayed swirling. Using TRAP, P-­selectin externaliza-
Aims: To study molecular pattern in PLT and PLT derived vesicles at
tion was stable from 0 to 45 μM Biot labeling but slightly lower for
different stages of PC storage.
180 μM Biot. GPIb and GPIIbIIIa expression was normal for all Biot-­
Methods: Six pulled PC from non-­smoking man 20-­30 years old,
PLTs. PLTs aggregability using arachidonic acid was well preserved
without hematologic disease, 0(I)Rh+ were included in the study.
in all Biot-­PLTs, whereas biotinylation impaired the response to ADP
Samples were collected on 2nd and 7th day of storage. Fractions
and collagen in a dose dependent manner with an aggregation of the
of PLT and membrane vesicles were obtained by differential cen-
Biot180-­PLTs almost abolished (Table).
trifugation at 5000 g, 16000 g (16P) and 100000 g (100P). Levels of
Conclusions: The main PLT functions tested, i.e., secretion, GP integ-
miRs (142, 223, 223*, 21, 126, 199, 15, 451, 221, 145, 326, 210) were
rity and aggregation in response to the strong agonist arachidonic
quantified by qRT-­PCR; proteins Ago2 and GW182 from RISC com-
acid, were fully preserved for all the Biot densities used. In contrast,
plex were analyzed by Western blot.
Biot labeling affected the response to ADP and collagen. We con-
Results: In a series of analyzed molecular components miR142
clude that multi-­Biot labeling would be suitable to trace PLTs in clini-
and miR223 were the most abundant in PLT. Level of miR21 was
cal trials provided low Biot densities (<45 μM) are used when full PLT
moderate, level of other targets, including, miR126 was low. The
function is critical. This approach should be helpful to evaluate new
absolute quantity of miRs in PC fractions was different on stor-
PLT products and fulfill the need for non-­radioactive tracers.
age days 2 and 7 with increasing on day 7: miR221 and miR21 -­in
100P fraction (P=0.03 in both), miR15 -­in 16P fraction (P=0.03).
MiR223, miR126 increased in both 16P (P=0.03; P=0.03) and 100P
(P=0.03; P=0.03) fractions; miR451 level increased in PLT (P=0.03),
100P (P=0.03) and decreased in supernatant (P=0.03) fraction. PLT
as well as 16P and 100P fractions contained Ago2 and GW182 pro-
teins. Levels of these proteins increased in 16P and 100P fractions
during storage.
Conclusions: PLT-­
derived vesicles contain RISC complex protein
components and abundant miR species. Their quantity changes in

the course of PC storage reflects platelet activation process via mo- PB089 | Compensatory Effect of Fibrinogen
lecular mechanism. Supported by RFBR 16-­0 4-­01142.
in Patient with Bone Marrow Aplasia, Septic
Shock and Severe Thrombocytopenia Guided by
Thromboelastometry: A Case Report
PB084 | Improving Transfusion Safety by
T. Crochemore; F. Savioli; C. Pessoa; A. Resende;
Implementing Platelet Bacterial Screening in a
E. Casaroto; L. Bernz; R. Narciso; F. Souto
Resource-­limited Country Hospital Leforte Morumbi, Intensive Care Unit, Sao Paulo, Brazil
F. Karim; B. Moiz
Aga Khan University Hospital, Karachi, Pakistan Background: Platelet transfusion is a common practice to prevent
spontaneous bleeding, or bleeding due to invasive procedures or
Background: Bacterial contamination of platelets remains a persis- surgery, in patients with hematologic diseases and thrombocyto-
tent problem in modern transfusion medicine. In a developing coun- penia. Transfusion of allogeneic blood components is associated
try like ours, introduction of bacterial screening was a significant with increased mortality and worse (worsened?) clinical outcome.
challenge due to cost constraints. We developed a cost effective The clot strength is accessed (assessed?) by thromboelastometry
protocol for bacterial detection in platelet concentrates. and determined by the interaction between platelets and fibrino-
Aims: To implement platelet bacterial screening method so as to im- gen. The compensatory effect of high levels of fibrinogen on clot
prove blood safety with regard to bacterial infections in a resource strength in patients with thrombocytopenia has been demonstrated
limited country. in different clinical settings such as in dilutional coagulopathy. We
Methods: The study was conducted at blood bank of Aga Khan report the case of a patient with severe thrombocytopenia related
University Hospital, Karachi, Pakistan from August 2016 to to bone marrow aplasia whose thromboelastometry showed the clot
December 2017. Platelet concentrates [PC] were prepared from strength compensated by the increase in plasma fibrinogen levels as
platelet rich plasma. PC were screened for aerobic bacterial con- an acute phase reactant of septic shock.
tamination by culture method [Bactec 9240, BD Bactec]. A volume Aims: –
of 2-­3 ml was collected from each PC on day 2 post-­collection. A Methods: We reported a case of a 62 years old caucasian Brazilian
sample pool made from five individual PC was added under sterile female admitted in the ICU with bone marrow aplasia. She devel-
conditions to a pediatric blood culture bottle. Identity of the pool oped septic shock being necessary orotracheal intubation due
and samples was ensured using barcoded labels. All inoculated bot- to respiratory insufficiency and antibiotic therapy. A bronchos-
tles were incubated for 5 days. If a pool was positive for bacterial copy was requested but she presented severe thrombocytope-
growth then pool resolution was performed by doing culture of sam- nia (1000/mm3). As her fibrinogen levels were 1050 mg/dl due to
ple from individual platelet unit to identify the contaminated con- septic shock, we performed a thromboelastometry showing a nor-
centrate. Initial assessment for any growth was done after 24 hours mocoagulable state. The bronchoscopy showed signs of bilateral
of incubation. PC with initial negative results (in first 24 hours) were alveolar hemorrhage, presence of organized clots in the inferior
removed from ‘quarantine’ and placed in ‘inventory’ and dispensed lobe segment, but without active bleeding. 7 days after she was
as “Negative-­to-­date”. extubated.
Results: Total number of PC tested during study period was 28,718 Results: –
(5743 pools). Confirmed positive culture was reported in one case Conclusions: The use of thromboelastometry as a tool used to diag-
(0.003%) where Streptococcus species was detected. No case of nose coagulation disorders has made it possible to avoid prophylac-
septic transfusion reaction was reported during study period. False tic transfusion of platelet concentrate unnecessarily. We believe that
positive culture was reported for 89 pools where no growth was thromboelastometry could be safer and more effective at predicting
seen on pool resolution. bleeding risk than platelet counts.
Conclusions: Successful implementation of bacterial screening of
platelets is a milestone in enhancing transfusion safety in a develop-
ing resource limited country. PB090 | A Novel Murine Model of Immune
Thrombocytopenia Induced by Immunized CD41
Knockout Mice
X. Li; S. Wang; X. Zuo; M. Hou; J. Peng
Qilu Hospital of Shandong University, Jinan, China

Background: Immune thrombocytopenia (ITP) is the most common

autoimmune bleeding disorder. Semple, et al. has reported a mu-
rine model of ITP in which splenocytes from CD61 knockout (KO)

mice immunized against CD61+ platelets were transferred into se- Conclusions: CD41-­KO mice have no placental defects or postna-
vere combined immunodeficient (SCID) mouse recipients (CD61 ). tal hemorrhage, and no mortality are observed in our murine model
However, CD61 is also expressed on endothelial cells, leading to of ITP using immunized CD41-­KO donors. This study develops an
a high bleeding morbidity (80%) of SCID recipients. Moreover, equally efficient murine ITP model and overcomes the disadvan-
CD61-­KO donors are presented with placental defects (25%) and tages induced by CD61-­KO donors.
postnatal hemorrhage (13%).
Aims: To develop a novel murine model of ITP using immunized
CD41-­KO mice. PB091 | Polyphosphate Is Concentrated in the
Core of Platelet Aggregates
(1) CD41-KO mice were transfused weekly with 10 8 platelets
of wild type C57/BL6 mice for 4 consecutive weeks. Then C. Clark; S. de Maat; A. Barendrecht; C. Maas
spleens were removed and prepared into splenocyte University Medical Center Utrecht, Clinical Chemistry and Haematology, Utrecht,
the Netherlands
(2) SCID mice were subjected to 200 cGy total body irradiation, and
Background: Platelets contain polyphosphate (polyP), a charged
injected intraperitoneally with 5×10 4 splenocytes.
polymer that acts on coagulation and fibrinolysis. We recently re-
(3) Platelet counts and serum antibody levels were recorded weekly.
ported that platelets retain polyP on their surface after activation
in complex with calcium. The resulting polyP particles can trig-
(1) None of the CD41-KO mice were presented with placental
ger contact activation, suggesting that contact activation should
defects. No bleeding occurred in embryos or pups.
always follow platelet activation. From clinical studies, there is
(2) SCID mice transferred with non-immunized CD41-KO spleno-
little evidence for this pathway. Furthermore, there is no bleed-
cytes or equal volume of PBS served as controls. All the mice
ing diathesis in persons with contact factor deficiencies. It was
developed thrombocytopenia at day 7 (Figure 1). Platelet counts
previously reported that platelet aggregates consist of a tightly
increased gradually in controls (Figure 1A), but remained at
packed core region and a loosely adherent shell region, composed
low levels within 21 days in the SCID mice receiving spleno-
of unactivated platelets. We hypothesize that polyP is mainly lo-
cytes from immunized CD41-KO mice (Figure 1B). No mortality
cated in the inner core of platelet aggregates, preventing continu-
was observed due to bleeding diathesis.
ous contact activation.
(3) No serum antibody was detected in controls, while high levels
Aims: To investigate the distribution of polyP throughout platelet
of serum anti-platelet CD41 IgG antibodies were present in
the morbid mice.
Methods: We studied the spatiotemporal distribution of polyP
(4) Few mature megakaryocytes were seen in bone marrow of
(tracked with SYTOX orange) during aggregate buildup by perfusing
morbid mice (Figure 2B).
citrated whole blood through a laminar-­flow chamber over collagen/
tissue factor-­coated cover glasses. At fixed time points, layered im-
ages were captured and analysed with a deconvolution algorithm for
reconstruction of 3D images.
Results: A polyP-­rich core develops in the center of platelet aggre-
gates, surrounded by a polyP-­poor outer shell over time (Figure 1).
Perfusion of ADP over preformed aggregates rapidly converts the
shell to polyP positive and increases polyP release within the core
region. This phenomenon can be recapitulated on monolayers of un-
activated platelets, suggesting that polyP secretion and retention is
F I G U R E   1   Platelet counts differ after transfer of splenocytes.
The data are expressed as mean platelet counts±SEM over time dependent on strong and synergistic triggering of platelet activation,
(days). *P<0.05 as may occur in thrombotic states.

F I G U R E   2   Histologic section of bone marrow from (A) a control F I G U R E 1   PolyP in an aggregate core after 10 minutes
SCID mouse, (B) a morbid SCID mouse. The yellow arrows point to perfusion visualized with SYTOX orange, orthogonal view, and 3D
megakaryocytes projection (red=aggregate outline)

Conclusions: Our findings indicate that only a part of the platelet Conclusions: Human and mouse platelets responded differently to
polyP is secreted within an aggregate. Under these conditions, polyP BjV. Our results suggest that BjV induces activation, aggregation,
is shielded from the plasma environment, which may limit contact secretion and protein synthesis in human and mouse platelets.
activation until the thrombus is ruptured or the shell becomes ex- Support: CAPES, CNPq and FAPESP.
posed to soluble platelet activators.

PB093 | Effect of Anorexigenic Peptide

PB092 | Effects of Bothrops jararaca Venom Obestatin on Platelet Aggregation at White Rat
on Platelet Aggregation, Secretion and Protein with Overweight in vitro Experiments
synthesis M. Golubeva; E. Khirazova
J.G. Rosa; M.L. Santoro Moscow State University named after M.V. Lomonosov, Biology Faculty, Moscow,
Instituto Butantan, Lab. Fisiopatologia, São Paulo, Brazil Russian Federation

Background: Victims from Bothrops snakebites manifest clinical Background: It is known that in people suffering from overweight,

symptoms such as hemorrhage, blood incoagulability and throm- there are serious violations in the system of hemostasis, which can

bocytopenia. Platelets are involved in the development of systemic lead to the development of thrombotic complications. The leading

bleeding evoked by B. jararaca venom (BjV), and they may contribute role in these processes belongs to platelets.

to other complications resulting from snakebites. Aims: The aim of this study was to examine the effect of the anorexi-

Aims: To evaluate the effect of BjV on platelet aggregation, granule genic peptide obestatin on platelet aggregation (PA) in experiments

secretion and protein synthesis in human and mouse platelets. in vitro in the platelet rich plasma (PRP) of healthy animals and ani-

Methods: All procedures involving human and mice were approved mals with obesity.

by the respective Ethics Committees. Washed mouse and human Methods: In vitro studies, the effect of obestatin on ADP-­induced

platelets were pretreated with 1 mM puromycin and cycloheximide (15 μM) PA in the PRP of healthy rats and animals with obesity was

(protein synthesis inhibitors) or vehicle for 1 hour. Platelet aggrega- studied. To create a model of alimentary obesity, animals were kept

tion and ATP secretion was then stimulated with thrombin (0.1 U/ for 2 months on a high-­calorie diet. Peptide was added to PRP sam-

mL) or BjV (24 μg/mL) in a platelet aggregometer and compared. ples in doses of 100, 300, 500, 800, or 1000 nmol/kg, Control sam-

Protein expression of von Willebrand factor (vWF), β-­actin, GAPDH ples contained appropriate volume of saline. Data were statistically

and thromboxane synthase (TX) from mouse platelets was evaluated processed.

by Western blotting. Results: It was shown enhancement of PA in obese animals in com-

Results: The extent of BjV induced platelet aggregation was less in- parison with healthy rats was revealed. Adding of anorectic pep-

tense in human than in mouse platelets. Granule secretion was simi- tide obestatin to plasma significantly decreased PA in plasma from

lar between BjV and thrombin in mouse platelets, and did not involve healthy rats as well as in plasma from obese animals. It was shown

protein synthesis. However, human platelet secretion stimulated by that the peptide in healthy animals in all concentrations of 1.5,

BjV was dependent on protein synthesis. Preliminary experiments 4.5, 7.5, 12 and 15 nmol/ml caused a decrease in PA, the great-

demonstrated that vWF expression was reduced in mouse platelets est effect was observed with a concentration of obesity of 7.5,

activated by BjV, in either the presence or absence of protein synthe- 12 and 15 (P<0.001). Obesity caused a significant increase in PA

sis inhibitors. vWF expression induced by thrombin was less intense in the PRP of rats. After incubating the PRP from these animals

in the presence of protein synthesis inhibitors. Under pretreatment with obestatin at all concentrations of the peptide, a decrease in

with protein synthesis inhibitors, mouse platelets activated by BjV PA in comparison with the control 85% saline) occurred, with a

did not modify β-­actin expression, while the protein expression of maximum reduction in PA on 26.7% by the concentration of the

GAPDH and TX showed a discrete reduction. peptide was 4.5 nmol/ml.

Conclusions: Obestatin was considered as a potential weight low-
ering drug. Since the pathological conditions associated with over-
TA B L E   1   Human and mouse platelets preincubated with protein
weight are often accompanied by increased blood clotting, PA
synthesis inhibitors showed a decrease in the extent of aggregation
inhibition with obestatin was regarded as a positive effect.
Platelet aggregation (%)

Without protein synthesis With protein synthesis

inhibitors inhibitors

Thrombin BjV Thrombin BjV

Human 60.9±3.4 23.10±2.2 12.5±3.6 6.5±7.6

Mouse 71.8±4.7 57.7±1.9 33.1±8.1 40.2±11.1

PB095 | Janus Kinase 3 (JAK3)-­dependent JAK3 could play a decisive role in arterial thrombosis underlying

Signaling Is Critical to Ca2+-­dependent Platelet myocardial infarction or ischemic stroke.

Activation and Thrombus Formation

M. Manke1; S. Geue1; P. Münzer1; B. Walker-Allgaier1;
D. Eißler1; F. Lang2; M. Gawaz1; O. Borst1
PB096 | An Intermediate State of Integrin
University Clinic of Tübingen, Tübingen, Germany, 2University of Tübingen, αIIbβ3 Mediates Platelet Aggregation under
Tübingen, Germany Disturbed Flow
Y. Chen1; L. Ju2; S. Jackson2; C. Zhu1
Background: Adhesion of platelets to subendothelial collagen after 1
Georgia Institute of Technology, Biomedical Engineering, Atlanta, United States,
plaque rupture results in an increase of cytosolic Ca2+ concentra- 2
University of Sydney, New South Wales, Australia
tion ([Ca2+]i) followed by Ca2+-­dependent platelet activation. Platelet
activation is essential for the development of thrombotic vascular Background: Platelets are sensitive to force in disturbed flow
occlusions. Janus kinase 3 (JAK3) is a member of the Janus family often induced by vessel stenosis or medical device intervention,
protein-­tyrosine kinases which is strongly expressed in platelets. resulting in occlusive thrombosis. However, the molecular mecha-
The inhibition of JAK3 in human platelets decreased pseudopod nism underlying the ‘biomechanical thrombosis’ is incompletely
formation, degranulation and aggregation. Although STAT3, a down- understood.
stream target of JAK3, plays a crucial role in Ca2+ signaling and plate- Aims: We aimed to investigate the integrin activation states and func-
let aggregation less is known about the impact of JAK3 in platelet tion in biomechanical thrombi, and define its triggering mechanism.
activation and thrombus formation. Methods:
Aims: The present study explored the role of JAK3 in Ca2+-­ 1) A stenosed microfluidic device was used to mimic a stenosed
dependent platelet activation and thrombotic vascular occlusions in artery. Human whole blood was perfused through such a de-
JAK3-­deficient (jak3−/−) mice. vice to form thrombus under high shear flow. Platelets were
Methods: Ca2+ measurements, luminescence aggregometry and stained with conformation-specific monoclonal antibodies to
flow chamber of washed jak3−/− and jak3+/+ platelets; tail bleeding report the conformational state of the integrin α IIbβ3.
time and in vivo thrombus formation in bone marrow chimeric jak3−/− 2) Dual biomembrane force probe (dBFP) was used to characterize
and jak3+/+ mice. αIIbβ3 activation by GPIb mechano-signaling by separating the li-
Results: Fura-­2-­AM spectrofluorometric Ca2+ measurements re- gand engagement of the two receptors in space and time.
vealed a blunted increase of [Ca2+]i in jak3−/− platelets as compared Results: Using the stenosed microfluidic device, we found that the
to jak3+/+ platelets after stimulation with thrombin or collagen-­ ‘biomechanical thrombosis’ is mainly supported by platelet integrin
related peptide (CRP). In addition, the activation-­dependent in- α IIbβ3 with an extended-­closed headpiece (EC) conformation (Panel
creases of dense granule secretion (reflected by ATP-­release) and A), which is distinct from the extended-­open headpiece (EO) inte-
aggregation were abrogated in jak3−/− platelets. In vitro thrombus grins found in the ADP-­triggered biochemical thrombi (Panel B).
formation on a collagen-­coated surface was diminished, and throm- Using dBFP on single platelets, we demonstrated that EC integrins
botic vascular occlusion after FeCl3-­induced injury in vivo was have an affinity (Panel C) and bond lifetime (Panel D) for ligands
blunted in jak3−/− chimeric mice, but no prolongation of bleeding that are intermediate between the well-­recognized inactive and ac-
time was observed. tive (ADP-­triggered) states. This intermediate state is inducted by
Conclusions: The present study reveals JAK3 as a regulator of Ca2+-­ mechano-­signaling of GPIb via a pathway distinct from that of chem-
dependent platelet activation, dense granule secretion and ag- ical signaling of ADP receptors. This inside-­out mechano-­signaling
gregation. JAK3-­deficiency results in decreased in vitro and in vivo also potentiates the outside-­in mechano-­signaling capacity of α IIbβ3,
thrombus formation, but does not affect bleeding time. Therefore, promoting its transition from the intermediate to the active state,
resulting in mechanical affinity maturation.
Conclusions: Our work mapped the landscape of integrin α IIbβ3 ac-
tivation in live platelet, and elucidated the different requirements
α IIbβ3 upregulation in the development of biomechanical and bio-
chemical thrombi.

F I G U R E   1   Representative images of in vivo thrombus formation

after FeCl3-­induced injury

PB097 | In vivo Imaging Reveals the Regulation

of Thrombus Formation by ERK and PKA Activity
T. Hiratsuka1; T. Sano2; M. Matsuda3
Kyoto University, Graduate School of Medicine, Department of Drug Discovery
Medicine, Division of Pathology, Kyoto, Japan, 2University of British Columbia,
Faculty of Medicine, Department of Urologic Sciences, Vancouver, Canada,
Kyoto University, Graduate School of Medicine, Department of Pathology and
Biology of Diseases, Kyoto, Japan

Background: Dynamic features of thrombus formation have been

visualized by conventional video wide field microscopy or confo-
cal microscopy in live mouse. However, due to technical limitations
precise spatio-­temporal regulation of intracellular signaling molecule
activities, which have been extensively studied in vitro, remains elu- F I G U R E   2   In vivo visualization of ERK and PKA activation at the
bottom of thrombus
sive in vivo.
Aims: By two-­photon excitation microscopy of transgenic mice ex-
PB098 | Role of Protease-­activated Receptor 4
pressing Förster resonance energy transfer (FRET) biosensors for
extracellular signal-­
regulated kinase (ERK) and protein kinase A
in Regulating Platelet Dense Granule Release
(PKA), we visualized ERK and PKA activities during thrombus forma- T. Chu1,2; R. Rigg2; L. Healy3; A. Ngo2; A. Mitrugno2;
tion in laser-­injured subcutaneous arterioles. J. Aslan2; A. Gruber2,4; C. Lindsley5,6; M. Duvernay5;
Methods: We have already reported transgenic mice encod- H. Hamm6; O. McCarty2,3,4
ing the FRET biosensors which visualize ERK and PKA activities. Johns Hopkins University, Department of Biomedical Engineering, Baltimore,
United States, 2Oregon Health & Science University, Department of Biomedical
Subcutaneous arterioles with a diameter of 10-­30 μm of these mice Engineering, Portland, United States, 3Oregon Health & Science University,
were subjected to observation. We injured endothelium of arterioles Department of Cell, Developmental & Cancer Biology, Portland, United States,
by laser beam and induced thrombus formation. Thus, we visualized Oregon Health & Science University, Division of Hematology and Medical
Oncology, Portland, United States, 5Vanderbilt University School of Medicine,
spatio-­temporal regulation of ERK and PKA activity.
Department of Pharmacology, Nashville, United States, 6Vanderbilt Center for
Results: When a core of densely packed platelets was developed, Neuroscience Drug Discovery, Nashville, United States
ERK activity was increased from the basal region near the injured
arterioles. PKA was activated at the downstream side of an un- Background: Human platelets express PAR1 and PAR4, G-­protein cou-
stable shell that overlays the core of platelets. A MEK inhibitor pled receptors (GPCRs) that are activated by proteolytic cleavage to
suppressed platelet tethering and dislodged platelet aggregates, initiate intracellular signaling. PAR1 and PAR4 are cleaved by overlap-
indicating that ERK activity is indispensable for both initiation and ping but distinct sets of proteases, leading to diverse functional out-
maintenance of the thrombus. A cAMP analog inhibited platelet puts. For instance, the granulocyte releasate cathepsin G only cleaves
tethering but failed to dislodge the preformed platelet aggregates PAR4, suggesting that PAR4 may play a unique role in mediating cross-
suggesting that PKA can antagonize thrombus formation only in talk between granulocytes and platelets.
the early phase. Aims: Investigate the role of PAR4 activity on platelet dense granule
Conclusions: In vivo imaging of transgenic mice expressing FRET release in the context of platelet-­granulocyte interactions.
biosensors will open a new window to visualization of the spatio-­ Methods: Human washed platelets were isolated and treated with the
temporal changes of signaling molecule activities during not only agonists human α-­thrombin, human cathepsin G, or thrombin receptor
thrombus formation but also other hematological disorders. activator peptide 6 in the presence of small-­molecule PAR4 or PAR1
inhibitors. To assay dense granule secretion, luminescence of an ATP-­
luciferase reaction was measured by spectrophotometry.
Platelet-­granulocyte interactions were assessed by electron micros-
copy and whole blood flow cytometry.
Results: Either PAR4 inhibitor VU0652925 or VU0661245 partially
impaired platelet dense granule release stimulated by thrombin
and fully blocked dense granule release stimulated by cathepsin G,
whereas the PAR1 inhibitor SCH79797 did not impair dense granule
release by either agonist. A combination of SCH and VU0652925 re-
sulted in increased impairment of dense granule release by thrombin
in comparison to VU0652925 alone. In human or baboon whole blood,
F I G U R E   1   Mode of action of the FRET biosensor for ERK or platelet-­granulocyte interactions were stimulated by PAR4 activity
PKA and specifically blocked by PAR4 inhibition.

Conclusions: Our results demonstrate that PAR4 small-­molecule in- PB100 | New Regulatory Mechanisms of RhoA
hibitors impair platelet dense granule release by thrombin and cath-
in Platelets
epsin G. These initial results suggest a functional role for platelet
PAR4 cleavage in platelet-­granulocyte interactions. PAR4 may be a
S. Comer1,2; Z. Nagy3; S. Gambaryan4; U. Walter5;
R. Zahedi6; K. Jurk5; A. Smolenski1,2
therapeutic target for inhibition in diseases where platelet-­leukocyte
UCD Conway Institute, UCD School of Medicine, University College Dublin,
interactions play a pathogenic role.
Dublin, Ireland, 2Irish Centre for Vascular Biology, Royal College of Surgeons
in Ireland, Dublin, Ireland, 3Centre for Cardiovascular Sciences, University of
Birmingham, Birmingham, United Kingdom, 4Sechenov Institute of Evolutionary
Physiology and Biochemistry, Russian Academy of Sciences, St Petersburg,
PB099 | Modulation of Smooth Muscle Cell Russian Federation, 5Center for Thrombosis & Hemostasis, Universitätsklinikum
Phenotype by Platelet Adhesion to Endothelial der Johannes Gutenberg-­Universität Mainz, Mainz, Germany, 6Jewish General
Hospital, Proteomics Centre, Lady Davis Institute, Montréal, Canada
Cells in Type 2 Diabetes Mellitus
M. Luo; M. Zeng; Y. Lin; D. Fang; J. Wu Background: RhoA, a Rho family small GTPase, acts as a bridge be-
Southwest Medical University, Luzhou, China tween extracellular signals and cytoskeletal restructuring of plate-
lets during activation. Vascular endothelium-­derived prostacyclin
Background: Vascular abnormalities in Type 2 diabetes mellitus (DM2) (PGI2) and nitric oxide (NO) interfere with platelet activation via
are often associated with the transition of vascular smooth muscle cell cAMP/cGMP-­dependent protein kinases (PKA/PKG). RhoA has been
(VSMC) phenotype from differentiated (contractile) to proliferative proposed as a direct PKA/PKG target but other proteins might con-
(synthetic) as characterized by VSMC-­specific phenotypic markers. tribute to PKA/PKG-­mediated RhoA regulation.
Further, platelet-­endothelial cell (EC) interactions play a major role in Aims: To identify Rho guanine nucleotide exchange factors
the pathogenesis of vascular complications in DM2. (RhoGEFs) and Rho GTPase activating proteins (RhoGAPs) that con-
Aims: While both platelets and ECs appear to contribute to pheno- vert PGI2/NO signals into RhoA inhibition.
typic switching of VSMCs, the precise mechanisms by which platelet Methods: Platelet phosphoproteome screening presented RhoGEF2
interaction with ECs affects VSMC phenotype, are unclear. and RhoGAP6 as candidate PKA/PKG substrates. Further protein
Methods: Using co-­
culture cell models, we aimed to determine analysis included mass spectrometry, mutagenesis studies, pulldown
whether diabetic platelet adhesion to, or interaction with, ECs influ- assays, PhosTag and SDS-­PAGE/Western blotting in human platelets
enced proliferation, migration, and phenotypic switching of VSMCs and transfected cells.
using quantitative RT-­PCR, western blotting, scratch assays, CCK-­8, Results: We show that RhoGEF2 (also called GEF-­
H1) and
gel zymography and fluorescence microscopy. RhoGAP6 are expressed and targeted by PKA/PKG in human
Results: Our results demonstrated that a significantly greater num- platelets. RhoGEF2 phosphorylation on serine 886 attenuates
ber of DM2 platelets adhered to ECs compared to control platelets. RhoA activation and stimulates RhoGEF2/14-­3-­3 complex forma-
DM2 platelet interactions with ECs correlated significantly with the tion. RhoGEF2 binds specifically to 14-­3-­3 β and η but not ε or γ
down-­regulation of mRNA and protein levels for VSMC-­specific con- isoforms. RhoGAP6 is phosphorylated on multiple sites within its
tractile phenotypic markers, including SM-­MHC, Smoothelin-­B, SMA C-­terminal region. RhoGAP6 strongly inhibits RhoA and cell adhe-
andSM22a. In contrast, VSMC-­specific synthetic phenotypic mRNA sion. When co-­t ransfected, the ability of RhoGAP6 to inhibit RhoA
and protein levels e.g. SMemb, PCNA, CCND1 and CCND2 were up-­ supersedes the ability of RhoGEF2 to activate RhoA.
regulated. Moreover, DM2 platelet interactions with ECs enhanced Conclusions: This data shows two new RhoA regulatory proteins
TGF-­β1-­induced PI3K/Akt/MMP9 signaling, leading to enhanced ge- in platelets, RhoGEF2 and RhoGAP6, which are novel targets for
latinolytic activity and migration and proliferation of VSMCs. PGI2/NO-­induced control of platelet functions. This work high-
Conclusions: Collectively, these findings illustrate that under diabetic lights the complexity of RhoA signalling and supports the crucial
conditions, interactions between platelets and ECs modulate the transi- role GEFs and GAPs play in mediating cyclic nucleotide effects in
tion of VSMCs between a contractile and synthetic phenotype, conceiv- platelets.
ably contributing to the development of cardiovascular complications..

PB101 | Platelet CLEC-­2 Ligation Offers Critical PB102 | Identification of Platelet Intracellular

Enhancement for ADP and Thrombin-­induced Signaling Pathways Controlling the Reversible
Activation via Synergy between Intracellular and Irreversible Platelet Aggregation
Signalling Pathways of CLEC-­2 and G-­protein D. Morozova1; A. Demianova2; M. Canault3; M. Panteleev2;
Coupled Receptors M.C. Alessi3; A. Sveshnikova4
Lomonosov Moscow State University, Faculty of Basic Medicine, Moscow,
A. Martyanov1,2,3; M. Panteleev1,2,3; A. Sveshnikova1,2,3
Russian Federation, 2Center for Theoretical Problems of Physicochemical
Lomonosov Moscow State University, Physics Faculty, Moscow, Russian Pharmacology, Russian Academy of Sciences, Moscow, Russian Federation, 3Aix-­
Federation, 2Center for Theoretical Problems of Physicochemical Pharmacology, Marseille University, Inserm, Inra, C2VN, Marseille, France, 4Lomonosov Moscow
Russian Academy of Sciences, Moscow, Russian Federation, 3National Medical State University, Faculty of Physics, Moscow, Russian Federation
Research Centre of Pediatric Hematology, Oncology and Immunology named
after Dmitry Rogachev, Cell Hemostasis and Thrombosis, Moscow, Russian
Federation Background: Platelet integrin αIIbβ3 activation is a major response
to stimulation leading to aggregation. Although reversible and irre-

Background: CLEC-­2 is a platelet receptor with a proposed role in versible platelet aggregation have been known for a long time, func-

maintaining blood vessel integrity promoting leukocyte extravasa- tional states of integrin molecules corresponding to these scenarios

tion from microvessels during inflammation. However, standalone and signaling pathways that regulate them still remain unclear.

CLEC-­2 induced platelet activation has a significant lag phase and Aims: To investigate reversible and irreversible binding of fibrinogen

altogether weak platelet responses. Thus why CLEC-­2 ligation could to activated platelets and signaling mechanisms governing transition

be of significance in some conditions remains elusive. of platelets between these different activation states.

Aims: To analyze the origin of the weak responses of platelets to Methods: Blood was collected from healthy donors under a proto-

CLEC-­2 stimulation as well as a model setting in which these re- col approved by the institutional Ethics Committee with written in-

sponses could be critical. formed consent. The state of platelet integrins was assessed by either

Methods: We constructed a comprehensive 3-­

dimensional aggregometry or binding of fibrinogen to platelets in flow cytometry

reaction-­diffusion computational model, which describes CLEC-­2 , and fluorescence microscopy. Multicompartmental computational

P2Y1, P2Y12, PAR1, PAR4 ligation, receptor diffusion in the mem- model of platelet integrin activation was devised to describe ex-

brane and receptor-­

induced signalosome assembly. Activation perimental data. The model included intracellular signaling cascades

of calcium-­
f luorophore loaded platelets by fucoidan, ADP and of PAR1, PAR4, P2Y1, P2Y12, GP-­VI and αIIbβ3 receptors, calcium

thrombin was studied by aggregometry, flow cytometry and TIRF signaling module, phosphoinositide signaling. GTP-­bound Rap1b was
microscopy. considered as the main trigger for integrin αIIbβ3 activation.
Results: Based on modelling results it can be claimed that both Syk Results: Weak and reversible binding of fibrinogen to platelets was
and SFK kinases phosphorylate CLEC-­2 cytoplasmic domain. CLEC-­2 observed under conditions of activation through a single receptor
induced cytosolic Ca2+ spiking was predicted by model and con- without secondary activation and outside-­in signaling. Reversible
firmed by TIRF microscopy. Flow cytometry assay, as well as TIRF fibrinogen binding correlated with reversible platelet aggregation.
microscopy, in line with the model prediction, proved signalling in- In the absence of PI3K activation, an elevated intracellular calcium
terplay between CLEC-­2 and ADP/thrombin (Gq and Gi) signalling level correlated with reversible integrin activation. Activation of
pathways: platelet cytosolic calcium responses to low doses of ADP PI3K without calcium also induced weak reversible fibrinogen bind-
or thrombin dramatically increase in the presence of CLEC-­2 agonist. ing. High-­level-­fibrinogen binding correlated with clustering of inte-
Surprisingly there are two merging points of the signalling cascades, grins, shown by fluorescence microscopy. Binding of fibrinogen to
those are PI3K and PLC activity. clusters was irreversible and occurred whenever two different sign-
Conclusions: Slow CLEC-­2 induced platelet activation occurs due aling pathways were activated.
to the need of signalosome assembly. In presence of ADP or low Conclusions: Irreversible platelet aggregation is a consequence of
concentrations of thrombin it can be significantly enhanced via syn- platelet integrin αIIbβ3 clustering, and requires both calcium and
ergetic PI3K and PLC activation by Gq, Gi and tyrosine-­kinase signal- phosphoinositide signaling.

ling pathways. This synergy makes CLEC-­2 role critical for vessel wall
protection in inflammation.
PB103 | S100A9 Activates Platelets via RAGE
S. von Ungern-Sternberg; K. Posavec; D. Heinzmann;
M. Gramlich; M. Gawaz; P. Seizer
University Clinic of Tübingen, Tübingen, Germany

Background: S100A9 is a cytosolic protein, which belongs to

the S100 family and can form heterodimers with S100A8. During

inflammatory conditions S100A9 is secreted into the extracellular

space where it can induce pro-­inflammatory properties. S100A9 is
able to induce the migration of inflammatory cells, including mono-
cytes. However, S100A9 is elevated in platelets and serum of pa-
tients with acute coronary syndrome. S100A9 is a ligand of several
surface receptors, including TLR, RAGE and CD147.
Aims: The aim of this study was to investigate the effect of S100A9
in platelet activation via different receptors.
Methods: Flow cytometry, dynamic adhesion model.
Results: Human monocytes were analyzed for their surface expres-
F I G U R E   1   Green dotted line:negative control (no FAM). Pink
sion of active alpha IIb beta 3 in a time dependent manner. Within
line: FAM-­labelled collagen without S-­citalopram. Gray area: FAM-­
15 minutes there is a significant increase in platelet activation. labelled collagen with S-­cit
Whereas S100A9 do not lead to an increase in P-­selectin expres-
sion and thus is not involved in the degranulation of alpha granules.
(2) Collagen-induced cytosolic calcium increase with/without
An activation of murine platelets with S100A9 leads to an increase
S-citalopram, R-citalopram or the metabolites, was measured
of CD147 surface expression and to increase of active alpha IIb beta
with fluorescent indicator fura-2 and a cytosolic ion analyzing
3. However, the activation of platelets seems to be dependent on
the presence of RAGE because in RAGE knockout mice there is no
(3) Expression of CD62P and activation of integrin αIIbβ3 after stim-
increase in the active form of alpha IIb beta 3 on the platelet surface
ulation with collagen or convulxin with/without S-citalopram
and thus no platelet activation. Nevertheless, in the dynamic adhe-
were measured by flow cytometer.
sion assay over recombinant S100A9 antibodies against CD147 and
(4) Binding of fluorescently labeled collagen with/without
RAGE were able to reduce the rolling of platelets. Possible due to
S-citalopram were measured with a flow cytometer.
the lack of P-­selectin expression, S100A9 is not involved in the for-
mation of platelet-­monocyte aggregate formation.
(1) All SSRIs suppressed platelet aggregation induced by collagen
Conclusions: The results of the current study show that S100A9
significantly, but did not suppress platelet aggregation by the
activates platelets via the receptors RAGE and CD147. S100A9 is
other agonists.
involved in the integrin dependent activation of platelets but does
(2) The both citalopram isomers and the metabolites suppressed
not increase the surface expression P-­selectin.
collagen-induced cytosolic calcium increase.

(3) S-citalopram suppressed collagen/convulxin-induced CD62P ex-

PB104 | Antidepressant Suppresses Platelet pression and activation of integrin αIIbβ3.
Activation via Collagen Receptor (4) S-citalopram suppressed binding of fluorescently labeled colla-
gen (Figure 1).
N. Sugita1; H. Hirakata2; T. Murai1
Conclusions: SSRIs may suppress platelet activation by blocking col-
Kyoto University, Graduate School of Medicine, Psychiatry, Kyoto, Japan, 2Kyoto
University, Graduate School of Medicine, Emergency Medicine, Kyoto, Japan
lagen receptors.

Background: Selective serotonin reuptake inhibitors (SSRIs) are a

class of most frequently prescribed antidepressants. Recent epide- PB105 | Induced Apoptosis-­like Changes
miological studies have shown that SSRIs may increase the risk of Caused by Sirtinol, a Pharmacologic Inhibitor of
abnormal gastrointestinal bleeding and may prevent ischemic heart Sirt1/2, in Human Platelets
disease events. However, the mechanism is unknown.
R.L. Mallick1; S. Kumari2; S.N. Chaurasia2; D. Dash2
Aims: The aim of our study is to elucidate the mechanisms of SSRIs’
Kathmandu University/Birat Medical College & Teaching Hospital, Biochemistry,
effects on human platelets.
Biratnagar, Nepal, 2Banaras Hindu University/Institute of Medical Sciences,
Methods: We purchased pharmaceutical ingredients of S-­ Biochemistry, Varanasi, India
citalopram, paroxetine, fluvoxamine, sertraline, which are mar-
keted SSRIs for clinical use, and those of R-­citalopram (the optical Background: Sirtuins (like Sirt1 & Sirt2) regulate the facts related to
isomer of S-­citalopram) and metabolites of both citalopram iso- cellular life-­span. They determine the balance between apoptosis, cell
mers. Each material was resolved in platelet rich plasma (PRP) taken survival & cell proliferation. Sirtinol, a cell-­permeable six-­membered
from healthy volunteers and incubated for 90 minutes at 37°C. lactone ring, is a pharmacologic dual inhibitor of Sirt1/Sirt2.
(1) Platelet aggregation induced by ADP, collagen, arachidonic Platelets are enucleate cells that are derived as cytoplasmic buds
acid, epinephrine or STA 2, was measured with a light trans- from precursor bone marrow megakaryocytes. These circulating
mission aggregometer. cells express Sirt1/2.

Aims: As it is unclear whether sirtuins have a role in platelets, we Methods: 85 patients with ITP who have just been diagnosed not
aim to; received treatment, 85 age and sex matched healthy controls were
• Investigate their regulatory roles in apoptosis-like events and Results: Mean age of patients and controls were 45,72 (22-­87)
• Evaluate the biological effect of sirtinol in isolated human blood and 47,68 (20-­72) years respectively. Mean platelet count in the
platelets. patient and control group were 18.647/mm3 (1.000-­92.000) and
253.800 (103.000-­424.000) respectively. Mean TRAF6 levels in
Methods: the patient and control groups were 2348,51 (522,24-­6913,29)
• Isolation of platelets from fresh human blood by differential and 25,57 (6,5-­911) respectively. In patients with ITP, patients
centrifugation, who have presented with major bleeding (gastrointestinal, nerv-
• Flow cytometric analysis of mitochondrial transmembrane po- ous system vb) and patients with antinuclear antibody positivity
tential (ΔΨm), measurement of reactive-oxygen species (ROS) & have significant higher TRAF6 levels (P values 0.000 and 0.000
annexin binding assay respectively).
• Western blot analysis. Conclusions: Working as an activator of Transforming growth factor
β activated kinase 1 (TAK1), a key player in the cellular responses
and cascades, TRAF6 deficiency has been related with various
Results: disorders involving essential processes such as the maturation of
• The 10% SDS-PAGE followed by incubation of a PVDF membrane dendritic cells, lymph node organogenesis, central tolerance of T
(on which electrophoretically separated proteins were trans- cells. Increases in TRAF6 as an immune stimulant, have never been
ferred) with antibodies demonstrated expression of Sirt1 & Sirt2 demonstrated in hematological diseases. In our study, we observed
in human platelets, a dramatic significance in TRAF6 serum levels in patients with ITP
• Washed platelets, pre-incubated with sirtinol and preceded by in- regardless of platelet levels. Bleeding and antinuclear antibody posi-
cubation with the lipophilic cation JC-1 (2 μM) provoked progres- tivity was significant in patients with high TRAF6 levels. Our study,
sive dissipation of platelet ΔΨm with increasing concentrations, though with a limited sample size, will lead to further studies regard-
• Pre-treatment with sirtinol resulted in dramatic rise of annexin V ing the importance of immunologic pathways in the pathogenesis
binding in dose-dependent manner; indicative of apoptosis-like of ITP.
events in cells,
• Isolated platelets, pre-treated with sirtinol followed by incubation
with DCFH-DA dye, revealed a dose-dependent change in ROS. B LE E D I N G D I S O R D E R S

Conclusions: Here we demonstrate that Sirt1 and Sirt2 are ex-

pressed in enucleate platelets. Further sirtinol enhances the annexin
V binding and ROS production but drop in mitochondrial transmem-
PB107 | Patients with Aortic Stenosis Have von
brane potential that are the aspects related to apoptosis. On the
basis of above, it is tempting to suggest that sirtinol enhances the
Willebrand Factor Abnormalities and Aberrant
apoptic-­like changes in human platelets. Angiogenesis in BOEC
S.N. Selvam1; M. Bowman2; M. Inglis3; J. Grabell2; L. Casey2;
L. Thibeault2; A.M. Johri2,3; P. James1,2
PB106 | Tumor Necrosis Factor Receptor 2
Queen’s University, Pathology and Molecular Medicine, Kingston, Canada,
Queen’s University, Medicine, Kingston, Canada, 3Queen’s University,
Associated Factor 6 (TRAF6) in the Pathogenesis Biomedical and Molecular Sciences, Kingston, Canada

of Immune Thrombocytopenia
Background: Patients with aortic stenosis (AS) may bleed due to
V. Asoglu; E. Umit; M. Demir
acquired von Willebrand syndrome and experience gastrointestinal
Trakya University Faculty of Medicine, Hematology, Edirne, Turkey
bleeding from angiodysplasia. Von Willebrand factor (VWF) is a neg-
ative regulator of angiogenesis and loss of the largest multimers has
Background: The immune system is controlled by multiple mecha-
been suggested to play a role in the development of angiodysplasia
nisms that regulate tolerance and immunity. TRAF6, is a member of
but the mechanism is unclear.
TRAF family which are intracellular proteins that transmit signals
Aims: To examine the defects of VWF in plasma from patients with
from cells to TNF receptors and by this mechanism, participate in
AS and the subsequent impact on angiogenesis using patient-­derived
the regulation of the immune system by experimental models.
blood outgrowth endothelial cells (BOECs).
Aims: In our study, we aimed to evaluate the role of TRAF6 signaling
Methods: The study was approved by Queen’s University eth-
in the pathogenesis of immune thrombocytopenia (ITP).
ics board. Twenty-­eight patients with severe AS undergoing valve

PB108 | High-­dose Dexamethasone Alters

the Increase in IL-­16 Level in Adult Immune
X. Zuo; X. Wang; X. Li; L. Li; M. Hou; J. Peng
Qilu Hospital, Shandong University, Department of Hematology, Jinan, China

Background: Adult primary immune thrombocytopenia (ITP) is an

autoimmune-­mediated hemorrhagic disorder which has a Th1 domi-
nant profile. Several reports have demonstrated that interleukin-­16
(IL-­16) plays an important role in autoimmune diseases by regulation
of recruitment and activation of CD4+ cell at sites with Th1 polariza-
tion. Similarly, the immune dysfunction in ITP patients may be at-
tributed to IL-­16, suggesting the potential prognostic utility of IL-­16
in ITP. However, the roles of IL-­16 in ITP remain unknown.
F I G U R E   1   (A) VWF:Ag, (B) VWF:GPIbM, (C) VWF:CB and (D)
FVIII:C of controls and patients prior to (AS-­A), 3-­5 days (AS-­B) and Aims: The aim of this study is to investigate the roles of IL-­16 and the
6 months after surgery (AS-­C) effects of high-­dose dexamethasone (HD-­DXM) on IL-­16 expression
in ITP.
TA B L E   1   VWF and Ang-­2 levels in BOECs’ media and lysates Methods: In this study, we investigated the differences in IL-­16 ex-
pression in the bone marrow and peripheral blood of ITP patients and
VWF:Ag (Mean±SD Ang-­2:Ag (Mean±SD
IU/mL) ng/mL) healthy controls by ELISA. Furthermore, we detected changes in blood
IL-­16 expression before and after the 4-­day HD-­DXM therapy. The
mRNA expression of pro-­IL-­16, caspase-­3 and T-­bet in ITP patients with
Media Lysates Media Lysates
active disease and healthy controls were quantified by real-­time PCR.
Controls (n=10) 31±11 549±158 7±4 12±6
Results: In patients with active ITP, bone marrow supernatant and
Aortic Stenosis 44±13 478±114 36±25 70±31
plasma IL-­
16 levels increased (P<0.05) compared with those of
-­A (n=5)
healthy controls (Figure 1). In the meantime, gene expression of pro-­
P value 0.099 0.411 0.206 0.075
IL-­16 and caspase-­3 in bone marrow mononuclear cells (BMMCs) and
pro-­IL-­16, caspase-­3 and T-­bet in peripheral blood mononuclear cells
replacement and 10 healthy controls were enrolled. Blood was (PBMCs) of ITP patients were increased (P<0.05) relative to those of
obtained from patients prior to, 3-­5 days, and 6 months after sur-
gery. At each time point, plasma VWF/FVIII levels, VWF:CB, and
VWF:GPIbM were determined. VWF multimer patterns were also
analyzed. BOECs were isolated and evaluated for VWF and angi-
2 (Ang-­
2) storage and secretion. Angiogenic profiles of
BOECs was assessed by proliferation and tubule formation.
Results: VWF levels were elevated in patients with AS compared
to controls and significantly increased immediately after surgery
(Figure 1A; P<0.0001). This was associated with significantly
higher VWF:GPIbM (Figure 1B) and collagen binding (Figure 1C)
after surgery (P<0.05). FVIII levels were also increased (Figure 1D;
P<0.001). VWF multimer analysis indicated a loss of high molecu-
lar weight multimers prior to surgery. Normal multimer profiles
were restored shortly after surgery. BOECs isolated from five pa-
tients prior to surgery had no difference in VWF and Ang-­2 stor-
age or release compared to controls (Table 1). Forty-­eight hours
after seeding, AS BOECs were almost twice as proliferative as con-
trols (P<0.01) however, tubule formation in Matrigel did not vary
after eight hours.
Conclusions: Quantitative and qualitative VWF abnormalities asso- F I G U R E   1   Concentrations of IL-­16 in active ITP and controls
ciated with increased BOEC proliferation are observed in patients (A,B), and in active ITP before HD-­DXM, after HD-­DXM and in
healthy controls (C)
with AS.

F I G U R E   2   mRNA expression in active ITP patients and healthy

controls. Correlation between plasma and mRNA expression levels
of IL-­16 in active ITP

F I G U R E   1   PGRN levels increased in plasma of ITP patients

healthy controls. HD-­DXM treatment reduced the plasma IL-­16 level
and decreased after treatment. PGRN correlated negatively with
and down-­regulated mRNA expression of IL-­16, caspase-­3 and T-­bet platelet count of ITP patients
in PBMCs (P<0.05). The plasma IL-­16 levels were positively correlated
with pro-­IL-­16, caspase-­3 and T-­bet mRNA expression (Figure 2).
Conclusions: In summary, the abnormal level of IL-­16 plays an impor-
tant role in the pathogenesis of ITP. Importantly, regulating Th1 po-
larization associated with IL-­16 by HD-­DXM therapy may represent
a novel insight for immune modulation in ITP.

PB109 | Progranulin Facilitates the Increase of

Platelet Count in Immune Thrombocytopenia
X. Zuo; Y. Yu; X. Li; M. Hou; J. Peng
Qilu Hospital, Shandong University, Department of Hematology, Jinan, China

Background: Immune thrombocytopenia (ITP) is an autoimmune

hemorrhagic disease characterized by excessive platelet destruc-
tion. Progranulin (PGRN) is emerging as a critical immune mediator
involved in a variety of autoimmune disorders. However, its role in F I G U R E   2   PGRN facilitated the increase of platelet count in ITP
mouse models (A, C). The pretreated rPGRN protected against anti-­
ITP remains unclear.
CD41-­induced ITP (B)
Aims: The aim of this study is to investigate the roles of PGRN in ITP.
Methods: The ELISA was used for determining the plasma levels
investigated in two kinds of ITP murine models, passive-­transfer
of PGRN in ITP patients and controls, as well as in patients before
model and active severe combined immunodeficiency (SCID) mod-
and after treatment. The binding of PGRN on platelets was ana-
els. Further, we explored whether PGRN functioned by affecting the
lyzed by immunofluorescence. The role of PGRN in ITP in vivo was
number of T regulatory cells (Tregs) by flow cytometry.
      51 |
Results: The plasma levels of PGRN were significantly elevated in ITP
patients (n=52) compared to healthy controls (n=40), and the level
of PGRN declined in patients after receiving treatment. There was a
negative correlation between plasma PGRN levels and platelet count
of ITP patients (Figure 1). The PGRN on platelets were not detected by
immunofluorescence, suggesting no direct interaction between PGRN
and platelets. The pretreated recombinant PGRN (rPGRN) effectively
protected against anti-­mouse CD41-­induced ITP in a passive-­transfer
murine model (Figure 2B). Administration of rPGRN increased platelet
counts in SCID mice with active ITP (Figure 2C). Furthermore, PGRN
deficiency decreased platelets (Figure 2A) and impaired proliferation
of Tregs in the passive transfer ITP model. The data suggest that PGRN
may exert a protective role in ITP by promoting Treg proliferation.
Conclusions: Our study identifies PGRN as a new regulator involved
in the pathogenesis of ITP.PGRN facilitates the increase of platelet
count, which may be associated with the positive regulation of Tregs.
This suggests the therapeutic efficacy of PGRN in ITP models and
provides a potential strategy for the treatment of ITP.

PB110 | The Aberrant Expression of

F I G U R E   1   The expression of five miRNAs in CD19+ B cells of
microRNAs in B Lymphocytes of Patients with ITP patients and healthy controls
Immune Thrombocytopenia
was found. Bioinformatics analysis indicated that the highest scoring
X. Jia; B. Zuo; Y. He functional pathways associated with ITP were cell cycle and thyroid
Jiangsu Institute of Hematology, The First Affiliated Hospital of Soochow hormone signaling pathway.
University, Suzhou, China
Conclusions: The aberrant expression of miR-­3162-­3p and let-­7b-­5p
were verified in CD19+ B cells of ITP patients, which may contrib-
Background: Immune thrombocytopenia (ITP) is an acquired autoim- ute to ITP pathogenesis by targeting cell cycle and thyroid hormone
mune disease characterized by a transient or persistent decrease in signaling pathway.
platelet count. The pathogenesis of ITP has proven complicated, in-
volving CTL-­mediated platelet lysis, disturbed T helper cell response
balance, dysregulated secretion of cytokines, etc. Among all of these
pathological factors, autoreactive B cells that produce platelet antigen
PB111 | A Novel Model using Protein C-­to-­INR
specific autoantibodies still take center stage. Altered expression of mi- Ratio to Predict Short-­term Mortality in Cirrhosis
croRNAs (miRNAs) have been reported in patients with ITP. However, A. Patil; C. Bihari
the levels of miRNAs in B cells and their effects on ITP remain unclear. Institute of Liver and Biliary Sciences, Clinical Hematology, New Delhi, India
Aims: To investigate the expression profiles of five disease-­related
miRNAs (miR-­144-­3p, -­142-­3p, -­3162-­3p, -­320c and let-­7b-­5p) in Background: Chronic liver disease is characterized by decreased
CD19+ B cells of ITP patients and healthy controls. Target prediction levels of both procoagulant and anticoagulant factors. Conventional
and pathway enrichment analysis were performed. coagulation tests cannot accurately assess risk of coagulopathy as
Methods: The peripheral blood samples of 21 ITP patients and 21 they measure only procoagulant pathway. Protein C a physiological
healthy controls were obtained. The CD19+ B cells were isolated anticoagulant has predictive and therapeutic roles in sepsis. Plasma
from PBMCs using immunomagnetic microbeads. The miRNA levels hypercoagulability has been shown to cause progression of fibrosis
were detected by quantitative real-­time PCR and calculated using causing end-­organ dysfunction.
the comparative cycle threshold method. Aims: To assess relationship between natural anticoagulants, protein
Results: The expression of five miRNAs in CD19+ B cells of ITP C (PC), protein S (PS) and anti-­thrombin III (AT) with MELD, Child-­
patients and healthy controls were evaluated. The levels of miR-­ Pugh (CP) and hepatic decompensation.
3162-­3p and let-­7b-­5p were significantly increased in ITP patients Methods: PC, PS and AT3 activity were analyzed in 598 cirrhotics
compared to healthy controls (P=0.0341 and P<0.0001, respectively) and compared with 232 non-­cirrhotic controls. CP and MELD scores
(Figure 1). No significant difference for miR-­144-­3p, -­142-­3p and were used to assess severity of cirrhosis.
-­320c was found. The association between miRNAs and peripheral Results: Cirrhotics had lower PC (P<0.001), PS (P<0.001) and
platelet count in ITP patients was also analyzed and no significance AT (P<0.001) than non-­cirrhotic liver disease. PC:INR ratio were

significantly lower in cirrhotics (30.89±11.1) compared to non-­

cirrhotics (57.99±12.64); P<0.001. Patients who died (n=182) at
3 months had significantly lower PC:INR ratio (15.9±6.08) compared
with those who remained alive (52.9±19.9); P<0.001. Using multivar-
iable analyses, low PC:INR ratio (P<0.001) and high MELD (P=0.038)
were found to be independent predictors of mortality. A predictive
model with PC:INR ratio had better discrimination (AUROC=0.84)
than a model using MELD alone (AUROC=0.62) to predict 3-­month
mortality. Also low PC correlated with liver dysfunction and in-
flammation: INR (r=−0.72, P<0.001), bilirubin (r=−0.620, P<0.001),
albumin (r=0.539, P<0.001), creatinine (r=−0.417, P<0.001), ferritin
(r=−0.68, P=0.035), procalcitonin (r=−0.79, P=0.01) and ESR (r=0.56,
P<0.001). PC (but not INR or MELD) was independent predictor of F I G U R E   1   CD4 and CD8 T cell frequencies
hepatic encephalopathy (P=0.002) and ascites (P=0.048).
Conclusions: In cirrhosis low PC correlates with fibrosis, liver dys-
function, inflammation and sepsis and is an independent predictor
of 3-­month mortality. This score using PC:INR ratio and MELD im-
proves risk prediction of short mortality over MELD alone.

PB112 | Increased Cytotoxic Potential

of CD8 T Cells in Patients with Immune
J.R. Vrbensky1; I. Nazy1,2; N. Ivetic1; R. Clare1; A.M. Jaffer1;
J.G. Kelton1,2,3; D.M. Arnold1,2
McMaster University, Michael G. DeGroote School of Medicine, Hamilton, F I G U R E 2   Surface expression of CD107a on CD3+CD8+ cells.
Canada, 2McMaster University, McMaster Centre for Transfusion Research
(MCTR), Hamilton, Canada, 3McMaster University, Pathology and Molecular
Medicine, Hamilton, Canada T cells (r=0.77). The proportion of CD8 T cells expressing CD107a in
ITP (19±13%) was comparable to TC (15±10%; P=0.41) and higher than
Background: Immune thrombocytopenia (ITP) is an autoimmune HC (11±6%; P=0.04). 6 ITP patients showed elevated CD107a expres-
bleeding disorder characterized by a platelet count <100×10 /L and 9 sion after anti-­CD3 (red line, 2SD above HC mean; Figure 2) and 3 of
antibody-­mediated platelet destruction. CD8 T cells may also be in- these 6 patients had an elevated CD8 cell frequency.
volved, but they have not been well characterized in ITP. Conclusions: Consistent with the heterogeneity of ITP, a subset of
Aims: Measure the frequency and cytotoxicity of CD8 T cells in ITP. patients displayed increased CD8 T cell frequency and cytotoxicity.
Methods: Peripheral blood was collected from ITP patients (n=21), This will contribute to thrombocytopenia if platelets or megakaryo-
non-­immune thrombocytopenia patients (TC; n=18 with splenomeg- cytes are targeted.
aly, liver disease, or familial thrombocytopenia) and healthy controls
(HC; n=23). Mononuclear cells were isolated by density gradient cen-
trifugation. CD4 and CD8 T cell frequency was determined via flow PB113 | Gene Expression of Vanin-­1 (VNN-­1) in
cytometry. T cells were stimulated for 5 hours with 5 μg/mL anti-­CD3 Egyptian Adults with Immune Thrombocytopenia
in the presence of anti-­CD107a. Surface expression of CD107a was
(ITP): A Single Center Experience
measured on CD8 T cells via flow cytometry, indicating the release
of cytotoxic granules containing perforin and granzyme. Granzyme B D. El Demerdash1; R.E. Hussein2; N. Abo Elsoad3; M.
Mattar1; A. Ayad1
(GrB) in the culture supernatant was measured by enzyme immuno-
Cairo University, Kasr Al Ainy School of Medicine, Internal Medicine, Clinical
assay. Mann-­Whitney P < 0.05 were considered significant.
Hematology Unit, Cairo, Egypt, 2Cairo University, Kasr Al Ainy School of
Results: Median platelet counts for ITP, TC, and HC groups were 36, 47, Medicine, Biochemistry, Cairo, Egypt, 3Tanta University, Clinical and Chemical
and 200×109/L, respectively. ITP T cells consisted of 62±15% CD4 and Pathology, Cairo, Egypt
36±15% CD8 cells, which was similar to the TC group (71±11% CD4
and 28±11% CD8), and the HC group (63±10% CD4 and 34±9% CD8). Background: The Distinguishing of self-­
limited acute immune
4 ITP patients had an elevated CD8 frequency (red line, 2SD above HC thrombocytopenia (ITP) from chronic ITP identify potential targets
mean; Figure 1). CD107a expression was <1% on resting CD8 T cells. for therapeutic interventions. Overproduction free radical in the
After anti-­CD3 stimulation, GrB levels correlated with CD107a on CD8 absence of adequate antioxidant defense may cause the generation

of neo-­antigenic determinants. This may ultimately contribute to decision making and record clinical reasoning. Management was
epitope spreading and the abrogation of self-­
tolerance. Vanin-­
1 crossed-­referenced against electronic clinical records by three indi-
(VNN1) has been proven to be an oxidative stress sensor, which viduals; to confirm indication for PCC, anticoagulant, additional PCC
regulates endogenous glutathione levels. required, complications, length of stay and mortality at 30 days.
Aims: Find an association between overexpressed VNN1 in blood Results: Over a 7-­month period PCC was issued for patients receiv-
and ITP pathogenesis as well as progression to chronic ITP. ing Xa-­DOAC in 34% of cases (30/89), as off-­label Octaplex® 50 IU/
Methods: We investigated 40 Egyptian patients with immune kg up to 3000 units. Issue was associated with bleeding in 73%
thrombocytopenia as well as 10 healthy controls (HC) through his- (22/30) and pre-­operatively in 27% (8/30). No additional PCC was
tory, physical examination, laboratory tests including CBC, reticu- given. There was a trend towards lower 30 day mortality in patients
locyte counts, ESR, PTT, PT, virology markers; CMV IgM, EBV IgM, on Xa-­DOAC issued PCC for bleeding; VKA 15/32 (47%) versus Xa-­
HCVAb, HBsAg and HBcAb, ANA, Lupus anticoagulant, anticardi- DOAC 7/22 (32%) (P=0.14), although mortality in those with intrac-
olipine, H pylori antigen in stool and TSH and response to therapy. erebral haemorrhage was higher; VKA 8/19 (42%) versus Xa-­DOAC
Vanin 1 gene expression by RT-­PCR. 6/11 (55%) (P=0.27). There was no thrombosis or DIC noted in pa-
Results: 40 ITP patients with mean age of 34.233±13.06 years. 6 tients on Xa-­DAOC receiving PCC.
of them were males and 34 females. With mean PLT count at pres- Conclusions: The issue of PCC for patients with life-­threatening
entation of 18.633±7.748×10 3 cells/cmm at presentation. Vanine 1 bleeding receiving a Xa-­DOAC is a frequent occurrence. Despite
gene expression was significantly higher in ITP cases (8.115±5.348) being a non-­specific reversal agent, real-­life clinical data supports
transcripts/1 μg total RNA as compared with HC (1.08±0.171) non-­inferiority to PCC use in patients receiving a VKA. Mortality in
transcripts/1 μg total RNA (P<0.001), there was no difference be- both groups remains high.
tween chronic and acute ITP patients as well as responder and non-­
responder ITP patients. Also, no significant correlation was found
between vanine 1 expression and other studied parameters.
PB116 | A 9 Year National Study of Patients
Conclusions: Our observations suggest that pathogenesis of ITP in
with Acquired Haemophilia A
Egyptian patients with ITP could be related to altered Vanin 1 ex-
pression levels but no relation to the chronicity of the disease or the D. Thakkar1; C. McDermott2; C.L. Robertson3; R.C. Tait2; C.
Bagot2; J.R. Rodgers1; H.G. Watson3; M.M. Khan3; R. Kerr4;
response to therapy. However, these results should be interpreted
J. Ward4; J. Craig5; J.A.M. Anderson1
with caution and further studies with larger samples sizes are re- 1
Royal Infirmary of Edinburgh, Department of Haematology, Edinburgh, United
quired to confirm these findings.
Kingdom, 2Glasgow Royal Infirmary, Department of Haematology, Glasgow,
United Kingdom, 3Aberdeen Royal Infirmary, Department of Haematology,
Aberdeen, United Kingdom, 4Ninewells Hospital and Medical School, Department
of Haematology, Dundee, United Kingdom, 5Raigmore Hospital, Department of
PB115 | Real-­life Bleeding Risk of Anti-­factor Haematology, Inverness, United Kingdom

Xa Direct Oral Inhibitors; Does Prothrombin

Complex Concentrate Work? Background: Acquired Haemophilia A (AHA) is a rare bleeding disor-
der with high mortality. Clinical manifestations vary from mild to life
J. Graham1; P. Gupta1; R. Sivers2; B. Badugama1; D. Sutton1;
threatening bleeding and treatment includes haemostatic manage-
D. Chandra1
1 ment and immunosuppressive therapy (IST).
University Hospitals of North Midlands, Clinical Haematology, Stoke-­on-­Trent,
United Kingdom, 2University Hospitals of North Midlands, Laboratory Sciences, Aims: To profile our management practices and outcome for AHA
Stoke-­on-­Trent, United Kingdom patients in Scotland.
Methods: Retrospective analysis of patients with AHA man-
Background: Outside of clinical trials, the management of life-­ aged across Scotland from January 2008-­D ecember 2017 was
threatening bleeding in patients receiving anticoagulation with an performed. Data obtained from hospital records included demo-
anti-­factor Xa direct oral inhibitor (Xa-­DOAC) (apixaban, rivaroxa- graphics, clinical presentation, laboratory parameters, manage-
ban, edoxaban) is currently dependent on non-­specific measures. ment and outcome.
National guidelines suggest consideration of non-­activated, 4-­factor
prothrombin complex concentrate (PCC); with data supporting cor-
rection of in vitro clotting studies in healthy human subjects. TA B L E   1   Laboratory parameters at Diagnosis
Aims: This audit of clinical practice at a large UK university teaching
Laboratory Number of
hospital aimed to assess the frequency of PCC usage, efficacy and parameter patients Median value Range
complications in a real-­life cohort of patients receiving anticoagula-
aPTT (seconds) 72 72 36.4-­>210
tion with a Xa-­DOAC.
factor VIII (iu/ml) 73 0.03 0-­0.3
Methods: PCC issue by the transfusion laboratory was recorded
Inhibitor titre (BU/ml) 70 10.6 <0.4-­1154
prospectively using specific flowcharts designed to aid laboratory

TA B L E   2   Information on management of bleed Conclusions: The demographic profile of our cohort was similar to that
reported from the UKHCDO study (Collins P et al.); the number of pa-
Number of
Characteristic patients [n (%)] tients receiving haemostatic treatment and IST was in keeping with
the UKHCDO study and the EACH2 registry report (Knoebl P et al.)
Information available for haemostatic 65
Haemostatic management received 45 (69.2)
 rVIIa alone 6 (13.3) PB117 | Is the Profile of Patients with
 FEIBA alone 23 (51.1) Intracranial Hemorrhage Changing?
 Both Bypassing agents sequentially 7 (15.5) S. Reeves1; C. Gray2; J. Blundell2
 rFVIII alone 4 (8.8) 1
University of Exeter, Truro, United Kingdom, 2Royal Cornwall Hospital, Truro,
 rFVIII with one of the Bypassing agents 5 (11.1) United Kingdom

Procedural intervention for management of 8 (12.3)

bleeding Background: Intracranial hemorrhage (ICH) is a devastating compli-
Red cell transfusion 24 (36.9) cation of oral anticoagulation therapy (OAC).Vitamin K (VK) and pro-
thrombin complex concentrates (PCC) reverse warfarin but this may
not translate into good outcomes. Most Direct Oral Anticoagulants
Results: Of the 73 AHA patients, 3 were post-­
partum (age (DOACs) have no specific antidote but PCC is used in life threatening
37 years). For the remaining patients M:F ratio was 4:5, hemorrhage.
median age 77 years (35-­
91.3). Mean duration of follow-­
up ICH increases with age. It is reduced with DOACs compared to VK
30.35 months (0 days-­155.7 months). 37 patients were alive at last antagonists (VKAs). In Cornwall these are increasingly used in pref-
follow-­up. erence to VKAs.
The most common bleeding category at diagnosis was skin and Aims: Identify the number of prescriptions for OACs in Cornwall be-
soft tissue (59%) then genitourinary (15.1%). The most common tween 2015 and 2017 and the OAC patients were taking at the time
nature of bleeding was spontaneous (57.5%) then prolonged of ICH.
bleeding post-­
t rauma (18.1%); one patient had intracranial Establish demographics, immediate and 6 month mortality of those
haemorrhage. Laboratory parameters at diagnosis are as in presenting to the Royal Cornwall hospital (RCH) with ICH over this
Table 1. period.
Information on management of bleed and inhibitors was available Review time from CT (computerized tomography scan) request to
for 65 patients. Of these 69% received haemostatic management VK administration and PCC issue.
(Table 2) and 97% received IST (one or combination of steroid, cyclo- Methods: A prospective database of anticoagulation related ICH
phosphamide, rituximab, iv immunoglobulin, mycophenolate mofetil, from September 2015 to September 2017 was created. This in-
azathioprine, cyclosporin). Of the 67 patients whose remission sta- cluded demographics, INR, CT time, VK, PCC administration, imme-
tus was known, 76% achieved remission. Of the non-­remitters 3 are diate and 6 month mortality. Prescription data was obtained from
alive, on treatment and follow-­up. Mean duration to achieve remis- our commissioning group.
sion was 16.6 weeks (0.5-­198). Results: The number on OACs has increased -­the number treated
Among the deceased patients, death was attributable to AHA re- with a DOAC has increased and Warfarin prescribing has fallen.
lated bleeding in 13.8%, immunosuppression in 8.3% patients. Cause Despite increased use of DOACs the number of ICH is stable, 59 in
of death was unknown in 33.3% patients. year 1, 58 in year 2 (see table).

TA B L E   1   All anticoagulants prescribed and number of intracranial bleeds

Anticoagulant Warfarin Sinthrome Apixaban Rivaroxaban Dalteparin Dabigatran Total

Number of bleeds September 45 1 5 4 3 1 59

Number of bleeds September 35 0 4 16 3 0 58
Total number of prescriptions 12,402 8 558 2,996 408 469 16,841
issued September 2015
Total number of prescriptions 11,853 11 1,264 4,418 395 473 18,414
issued August 2016
Total number of prescriptions 10,906 7 2,438 5,754 369 550 20,024
issued August 2017

68% were male. 3% under 60, 49% between 60 and 80 and 48% were PB120 | Noonan Syndrome and Bleeding
over 80. Of the total number 32% died on that admission and a further
Disorders: An Overview of Current Diagnostic
23% died within 6 months. Only 29% on warfarin had an INR >4.0.
In total 59% were given PCC and VK. 19% of these were on DOACs.
Mean time to PCC was 103 minutes and mean time to VK adminis- P.E.G. van derPas1; E.J. Huisman2; I.H.I.M. Hollink1; C.M.
tration was 95 minutes.
Zwaan3; C.H. van Ommen4; M.H. Cnossen2
Erasmus University Medical Center, Rotterdam, the Netherlands, 2Erasmus
Conclusions: Despite a therapeutic INR and timely PCC and VK mor-
MC-­Sophia, Pediatric Hematology, Rotterdam, the Netherlands, 3Erasmus MC-­
tality is high supporting the importance of ongoing risk/benefit as- Sophia, Pediatric Oncology, Rotterdam, the Netherlands, 4Erasmus MC, Pediatric
sessments in individuals. Hematology, Rotterdam, the Netherlands
We anticipate more ICH with DOAC’s as they are prescribed more
to older people. Background: Noonan Syndrome (NS) is a genetically defined syndrome
There may be a role for measuring drug levels so expensive reversal characterized by a range of physical-­and developmental symptoms
agents are used appropriately. which require surgical interventions. However, NS is also associated
with a variety of bleeding disorders. Current guidelines advise only
screening by APTT, PT, FXI preoperatively or at 5 years of age.

PB118 | Hospital Admissions due to Bleeding Aims: To investigate diagnostic management of bleeding disorders
in children with NS.
Events under Direct Oral Anticoagulants
Methods: Retrospective cohort study.
V. Grilo1; A.G. Fonseca2 Results: A total of 32 children with NS were identified. Genetic mu-
Hospital Garcia de Orta, Internal Medicine, Almada, Portugal, 2Hospital Garcia tations were known in 30 children. A cardiac abnormality was de-
de Orta, Almada, Portugal
tected in 27/32 (84%), most often pulmonary stenosis (23/27).
Hemostatic testing was performed in 31/32 children. Screening by
Background: The novel oral anticoagulants (NOAC) are an alterna-
APTT, PT and FXI was performed in 22 of 32 children (69%), in three
tive to vitamin K antagonists for long-­term treatment and prevention
only FXI was tested. Three of these 25 had a mild FXI deficiency, with
of venous thromboembolic events (VTE) and stroke in non-­valvular
two <0.4 U/L. In addition, nine patients had FIX levels below 0.6 U/L,
atrial fibrillation (AF). Despite having a favorable safety profile, clini-
7 of which were <0.5 U/L. Moreover, primary hemostatic defects were
cal relevant bleeding complications is still a concern.
frequently observed. One third of patients had mildly decreased von
Aims: We aimed to characterize bleeding events requiring hospital
Willebrand levels, only reflected in PFA-­values. Platelet aggregation was
admission in patients on NOAC.
performed in only 5 of the 32 patients and abnormal in X (60%). A bleed-
Methods: A retrospective study consisting on the analysis of 1 year
ing tendency was reported in 7/32 patients (22%), especially bruising and
hospital admissions due to bleeding event while on NOAC anticoag-
epistaxis. A surgical intervention was performed in 23 patients (72%).
ulation was conducted at a tertiary hospital (2017); 51 patients were
Perioperative bleeding was reported in 4 of these 23 patients. Nine of
included. Based on the clinical files, we characterized epidemiologic
them required a total of 12 red cell, platelet and plasma transfusions.
data, patient comorbidities, the bleeding event and its management.
Conclusions: Current guidelines in the Netherlands are followed in
Thrombotic and bleeding risks were scored using CHA2DS2-­VASc
majority of cases. However, FXI deficiency was rarely reported in
and HAS-­BLED for AF, Riete Registry Bleeding Score for VTE. We
NS and often not clinically relevant. In contrast, mild FIX deficiency
used IBM SPSS Statistics version 23. Chi-­Square, Wilcoxon tests
and primary hemostatic defects were detected more frequently. We
were used; significance level was 0.05.
propose an update of guidelines with and inclusion of hemostatic
Results: The median age of patients was 81 years (IQR 13), 52.9%
testing of FIX, von Willebrand factors and platelet function.
were men. The main indication for NOAC use was AF (90.2%), with
a CHA2DS2-­VASc score ≥2 in 97.8%, and a HAS-­BLED score ≥3 in
69.9%. NOAC dosing was adequate in 34 patients (66.7%). Major
bleeding occurred in 37 patients (82.4%); 14 (27.5%) had intracra- PB121 | Long-­Term Follow-­up of Children on
nial bleeding. Major bleeding events had lower hemoglobin levels Prophylaxis for Severe Factor VII Deficiency in
at admission (median 12.7 mg/dL, IQR 3.2, vs. 8.6 mg/dL, IQR 6.9, Lebanon
P<0.001). In 28 patients (54.9%) a new underlying diagnosis was
R.A. Farah1; H.Z. Farhat2; D. Buthiau3; M. Giansily-Blaizot3
made. In-­hospital mortality was 25.5%; median admission GFR was 1
Saint George Hospital University Medical Center, Department of Pediatrics,
significantly lower than median baseline GFR (60 mL/min/m2, IQR Beirut, Lebanon, 2LAU-­Rizk Hospital University Medical Center, Department
44.5 vs. 66.5 mL/min/m2, IQR 29.5 P=0.001). of Laboratory Medicine, Beirut, Lebanon, 3CHU de Montpellier, Laboratory of
Hematology, Montpellier, France
Conclusions: In an elderly, high-­hemorrhagic risk population, there
is a need for better surveillance of GFR and control of bleeding risk
factors, as major bleeding events may be potentially fatal. A clinical Background: Congenital Factor VII deficiency is the most com-

relevant bleeding may trigger a new diagnosis. mon autosomal recessive bleeding disorder. Severe forms are

associated with missense and consensus splice site mutations. The complication in 27 (100%), 5 (18.5%), 5 (18.5%), 3 (11.1%), and 1
F7:c.291+1G>C splice site mutation, seen in consanguineous mar- (3.7%) of the patients respectively. Trp187Arg (exon 4, C.559T>C)
riages, has resulted in high prevalence of severely affected FVII-­ mutation was the only FXIIID mutation which was found in all pa-
deficient patients in Lebanon. tients. History of central nervous system (CNS) bleeding was de-
Aims: To evaluate the efficacy, safety, impact on morbidity and mor- tected in 13 (48.1%) patients. There were no significant association
tality of long-­term prophylaxis in a series of Factor VII-­deficient pa- between CNS bleeding history and sex, familial history of FXIIID,
tients in Lebanon. or other clinical presentations. Also there was no significant differ-
Methods: We present a case series of 5 Lebanese children with ence in mean age of patients with CNS bleeding (3.4±0.9 days) and
severe factor VII deficiency, all carrying the F7:c.291+1G>C splice without CNS bleeding (2.9±0.7 days). However, a near significant
site mutation in homozygosity. Patients were followed closely and threshold difference was found between the groups regarding the
bleeding episodes recorded and classified according to the Tosetto mean number of suspicious death in family (1.8±0.5 and 0.7±0.1 re-
bleeding score. The occurrence of antibodies to FVII was assessed spectively, P=0.05).
using standard methods. Conclusions: Relatively high rate of ICH in neonates with FXIIID
Results: 5 children from 4 families presented with GI bleeds (n=4) necessitates early consideration of appropriate factor replacement
or a CNS bleed (n=1) within two months of birth (range 1-­4 0 days). therapy in these patients. So, we recommended a diagnostic algo-
Without prophylaxis, 2 families had lost affected children due to rithm for detection of FXIII deficiency in early life.
severe bleeding. Prophylaxis with recombinant Factor VII was im-
plemented using 25-­30 μg/kg, 3 times per week. Follow-­up ranged
from 1 to 11 years. No antibodies to FVII were detected. No major
PB123 | Effectiveness of Pediatric Bleeding
bleeds were reported with prophylaxis, however additional re-
Questionnaire in Hemorrhagic Disorders
combinant Factor VII was used in case of trauma or minor bleeds.
Two surgical procedures were carried out during prophylaxis: a A.C. Ramirez Cazares1; L.M. Nuño Vazquez1; J.E. Colunga
Pedraza1; D. Gomez Gonzalez1; M. Quiroga Treviño1;
ventriculo-­peritoneal shunt placement and a port-­a-­cath insertion.
L. Villarreal Martinez2
Recombinant Factor VII was administered prior to and after surgery. 1
University Hospital ‘Dr Jose Eleuterio Gonzalez’, Monterrey, Mexico, 2University
No bleeding was seen.
Hospital ‘Dr Jose Eleuterio Gonzalez’, Hematology, Monterrey, Mexico
Conclusions: 5 individuals with the F7:c.291+1G>C mutation in ho-
mozygosity were found, enabling the largest series of Lebanese fac-
Background: The study of hemorrhagic symptomatology can be par-
tor VII-­deficient pediatric patients to be given long-­term prophylaxis
ticularly difficult in children where questionnaires adapted to pedi-
and follow-­up. Prophylaxis impacted positively survival and quality
atric patients such as the Pediatric-­Specific Bleeding Score (PSB)
of life. It showed safety and efficacy in controlling bleeding episodes.
prove their usefulness.
Aims: Determine the effectiveness of the PSB questionnaire in
bleeding disorders.
PB122 | Clinical Presentations and Diagnostic Methods: There were included all patients with abnormal bleed-
Algorithm of Neonatal FXIII-­deficiency ing in the last 6 months referred to the hematology service of the
1 2 1 2 University Hospital “Dr. José Eleuterio González”. They were initially
M. Naderi ; N. Cohan ; G. Miri-Aliabad ; M. Karimi
evaluated with the PSB and those with a positive result were addi-
Zahedan University of Medical Sciences, Genetics of Non Communicable
Disease Research Center, Zahedan, Islamic Republic of Iran, 2Shiraz University
tionally assessed with specific coagulation tests.
of Medical Sciences, Hematology Research Center, Shiraz, Islamic Republic of Iran Results: A total of 26 children were analyzed, whose median age
was 7 years old (10 months to 15 years), none of them with family
Background: Clinical specifications of FXIII deficiency (FXIIID) in history of hemorrhagic disease. In order of frequency clinical man-
neonates of very young has not been studied previously. ifestation were: epistaxis (61.5%), metrorrhagia (11.5%), epidural
Aims: In present study, we assessed clinical picture and the rate of hematoma (7.7%), dental bleeding (7.7%), excessive bleeding from
intracranial hemorrhage (ICH) in Iranian neonates with FXIIID who trauma (3.8%), surgical bleeding (3.8%), prolonged aPTT (3.8%).
presented with bleeding diathesis in the early days of their lives. The laboratory tests reported alterations in 8 male patients (30%):
Methods: A total of 27 neonates presented with bleeding presenta- 5 hemophilia A mild to moderate confirmed by genetic mutations,
tions early in life and diagnosed as a FXIIID were evaluated as a study 1 factor XI deficiency, 1 positive lupus anticoagulant, and 1 pro-
group during March 2010 to March 2015. All clinical and demographic longated PT and aPTT. Median bleeding score in children with
data were recorded by a designed questionnaire. Either cryoprecipi- hemorrhagic disease was 8.5 (range 3-­13) and in the healthy group
tate or factor XIII concentrate was started as a first-­line treatment was 4 (range 1-­11). No patients with less than 3 points showed
strategy and prophylactic therapy was given to all patients. alterations in tests.
Results: Umbilical cord bleeding (UCB), Delayed detachment of um- Conclusions: The PSB provides a scoring system and lays a ground-
bilical stunt, seizure, hematoma, and ecchymoses were concurrent work. This could be particularly helpful in limited resource countries
      57 |
with difficult access to advanced laboratory techniques. Despite Von

11(29.7%) / 5(13.5%)
Willebrand Disease (VWD) is the most prevalent inherited bleed-
ing disorder reported in the literature, no case was diagnosed. The

/ 21(56.7%)
13(35.1%) /
24(64.8%) /

limitation of this study is the number of patients recruited, however

9(24.3%) /

according to other research regarding Hispanic population also with

small groups of patients there seems to be a tendency to a lower

incidence of VWD.


5 (13.5%)
PB124 | Surgical Management of Rare Factor




N. Eroglu1; E. Erduran1; A. Bahadır1; H. Saruhan2

FXI Deficiency
Karadeniz Technical University, Pediatric Hematology Oncology, Trabzon,

6 (16.2%)
Turkey, 2Karadeniz Technical Universitiy, Pediatric Surgery, Trabzon, Turkey



Background: Rare factor deficiencies (RFD) include fibrinogen de-
fects and factor (F) II, FV, combined FV and FVIII, FVII, FX, FXI, and

FX Deficiency
FXIII deficiencies.

4 (10.8%)
Aims: We report demographic features and surgical approaches of



RFD. We retrospectively evaluated 37 patients with RFD. Patient


characteristics, age at presentation and the causes of surgical inter-

FVII Deficiency
vention, preparation for surgery, preoperative replacement thera-
pies were recorded. Surgical interventions were classified as major

11 (29.7%)
and minor operations.



Methods: Surgical intervention was implanted in 27 surgical inter-

ventions of 37 RFD patients. 64.8% of the patients were male and
35.2% female. Parental consanguinity was determined in 48.6% of

1 (2.7%)
cases. Patient information is presented in Table 1. Of the surgical

interventions, 18 (66.6%) were major and 9 (33.3%) were minor.

Information of the patients who underwent major and minor surgery
is presented in Table 2.
FV Deficiency

Results: FFP was used for individuals with factor V, factor XI and
5 (13.5%)

factor XIII deficiencies and for patients with factor X deficiency.



Tranexamic acid was used as the antifibrinolytic agent. FFP was used

for individuals with factor V, factor XI and factor XIII deficiencies and
for patients with factor X deficiency; in FVII deficiency, r FVII a; in
Fibrinogen +


fibrinogen deficiency, fibrinogen concentrate was given. Tranexamic

2 (5.4%)

acid was used as the antifibrinolytic agent. Post-­surgical haemosta-



TA B L E   1   Patients’ Demographic Characteristics


sis of the patients were determined by physical examinations, whole


blood count and haemostasis tests.

Conclusions: Minor surgeries, were performed successfully with
1 day replacement therapy and 3 days of antifibrinolytic support,
whereas it is not possible to describe a general treatment scheme

for major surgeries. We believe that the above procedures should

3 (8.1%)



be applied for the treatment and in preparation for surgery of RFDs.


We wish to emphasize that pre-­operative screening is important be-

cause of the late diagnosis of asymptomatic cases since the clinical
<5/5-­3 0/30-­50%
Sex: Male/Female

findings of RFDs are not obvious.

<1/1-­5/>5  year

Factor Activity

Grade: 1/2/3
First Bleeding

58       |

PB125 | Assesment of Thirty Six Patients with Rare

iv10  mg/kg/+/-­/+/-­/-­/-­

po 25 mg/kg/d -­/-­/_
po 25 mg/kg/d -­/+
Factor Deficiency: Single-­centre Results from Turkey

iv 10 mg/kg/ +/+
Tranexamic Acid

po 25 mg/kg/d
iv 10 mg/kg/ -­
M. Söker1; H. Uzel1; K. Öncel1; S. Söker2

Dicle University, Pediatric Hematology and Oncology, Diyarbakır, Turkey, 2Dicle
University, Histology and Embriology, Diyarbakır, Turkey


Total Number

Background: Rare factor deficiencies (RFDs) are coagulation factor

deficiencies that ranging prevalence from 1/500.000-­1/2.000.000
of Doses

50 mg/kg+15 ml/kg 24 h+12 h 14+28

and predominantly autosomal recessive inheritance. It constitutes

3-­5% of hereditary coagulation factor deficiencies.


TA B L E   2   Major Surgeries+Minor Surgeries F: Female, M: Male, F:C: Factor coagulant activity, FFP: Fresh frozen plasma, rFVIIa: Recombinant factor VIIa

Aims: In this study, we aimed to evaluate the demographic features and


clinical findings of patients who were followed up and treated with RFDs.
4-­6  h

12 h

12 h
24 h

24 h

Methods: A total of 36 patients, aged between 0 and 16 years, who

kg-­3 0 U/kg-­30 U/

had been treated and followed-­up with RFDs were included in the
kg-­3 0  U/kg-­3 0  U/
15 ml/kg-­30 U/

study. Demographic characteristics, clinical follow-­up results and

Therapy Dose

laboratory data of patients were recorded.

30 mg/kg
50 mg/kg

15 ml/kg

15 ml/kg

15 ml/kg
15 ml/kg
Results: RFDs were diagnosed in 36 (35%) patients of the 108 patients
who were followed up and treated with coagulation disorders. 20 of

the patients were female and 16 were male. The age of admission to
concentrate + FFP

the hospital is ranged from 2 days to 16 years. Familial consanguinity

Therapy Used

was present in 75.6% and family history was 16.6% of the patients. 11





(32.4%) patients were diagnosed before the age of one. 14 patients

were diagnosed with FX deficiency (38.8%), eight patient had FVII defi-



ciency (22%), four patient had fibrinogen deficiency (11%), three patient
had FXI deficiency (8.3%), two patient had FV deficiency (5.5%), two pa-
74/10/10/15 mg/
32%/ 22%/ 28%/

Not Determined
20%/ 7%/ 8%

tient had FXIII deficiency (5.5%), one of them had combined deficiency
10 mg/dl+52%


of factors associated with vitamin K (Factor II, FVII, FIX, FX deficiency)

10 mg/dl

(2.7%), one of them had FXII deficiency (2.7%), and one of them had


combined deficiency of FV and FVIII (2.7%). Mucocutaneous bleedings


(50%) were seen as most common admission symptoms at the time of

Operation Deficiency




diagnosis. Intracranial haemorrhage (ICH) was seen in five patients with


inherited FX deficiency and in two patients with FXIII deficiency.



Conclusions: Early diagnosis and treatment is very important in the

case of RFDs due to the severe bleeding complications such as ICH.

RFDs are more common in place where familial consanguinity than

Age of



the general population. It′s a necessity that informing parents about


this and perform family screening earlier.





PB126 | Anti-­fviii igg4 Antibodies in Patients with

Circumcision + Adenoidectomy/ Myringotomy

Circumcision + Undescended Testis Surgery/

Hemophilia A without Inhibitors during Factor VIII

Humerus Fracture Surgery/ Circumcision

Replacement: Results from the Hemfil Cohort Study

Intracerebral Hematoma Evacuation +
Developmental Hip Dysplasia Surgery

Adenoidectomy / Adenoidectomy/
Adenoidectomy + Myringotomy/

L.M. Moura de Oliveira1; D. Goncalves Chaves2; L. Lemos

Jardim1; A.C. Simoes e Silva1; M.A. Portugal Santana3; C.
Carm-Deelder4; S. Meireles Rezende1
Minor Surgeries Dental

Minor Surgeries Dental

Minor Surgeries Dental

Universidade Federal de Minas Gerais, Faculdade de Medicina, Belo Horizonte,
Post-­op ASD Surgery

Brazil, 2Fundação Hemominas, Research Area, Belo Horizonte, Brazil, 3Fundação

Hemominas, Ambulatory, Belo Horizonte, Brazil, 4Leiden University Medical
Major Surgeries


Center, Leiden, the Netherlands


Background: Patients with haemophilia A (PHA) can develop in-

hibitors after exposure to factor (F) VIII due to stimulation of the

immune system, which is still not fully understood. Previous stud- immuno-­beads array. The difference was analyzed statistically, and
ies have demonstrated strong association between the presence the receiver operating curve (ROC) was constructed to assess the
of inhibitors and anti-­FVIII-­IgG4) antibodies. However, studies diagnostic value.
targeting antibody response during FVIII replacement are still Results: Compared with normal pregnant women and healthy con-
lacking. trols, the plasma levels of auto-­antibodies against glycoprotein
Aims: This study aimed to evaluate the levels of FVIII-­IgG4 in the GPIX and GPIb were significantly increased in pregnant women
plasma of PHA throughout FVIII replacement therapy. with thrombocytopenia (P<0.05), while the levels of other 3 auto-­
Methods: PHA were included at the moment of diagnosis. Plasma antibodies were similar. The positive rate of auto-­antibodies in
samples were collected at different exposure days (ED) to FVIII for pregnancy women with thrombocytopenia was higher than that in
inhibitor screening and by Bethesda and FVIII-­IgG4 by ELISA. Plasma normal pregnant women (59.3% vs. 31.8%, P<0.01). ROC analysis
pool of 20 healthy age-­matched boys was used as negative control showed that the auto-­antibody anti-­G PIX had a high diagnostic
(NC). Results were expressed as optical density (OD) and positive value in pregnant women with thrombocytopenia over normal
IgG4 was considered when the level was higher than the NC OD pregnant women with the area under the curve value of 0.779
plus two standard deviations. Wilcoxon matched-­pairs signed-­ranks (P<0.001), the sensitivity and specificity were 76.3% and 100%,
test was used to compare the IgG4 OD at the moment of diagnosis respectively.
(T0) and at >75ED (median, 113.0; IQR 79.0-­129.3) ED. The study Conclusions: The plasma profile of platelet auto-­antibodies in pa-
was approved by Ethics Committees and patients signed a written tients with thrombocytopenia in pregnancy was different from that
informed consent. in ITP patients. The auto-­antibody anti-­GPIX may be served as a bio-
Results: Plasma samples (n=28) from 14 severe (FVIII<1%; n=12) and marker of the disease.
moderately-­severe (FVIII=1%-­2%; n=2) PHA without inhibitors were
included. Levels of anti-­
IgG4 increased progressively from
T0 reaching its highest in the strata >75DE (T0 median, 0.108 OD; PB128 | Lifetime Bleeding History with
IQR 0.035-­0.174 vs. >75ED median, 0.041 OD; IQR 0.026-­0.053,
Modified ISTH-­BAT Survey in 2,339 Individuals
Establishes Population Prevalence Estimates
Conclusions: Our results suggest that anti-­
IgG4 increases
throughout the course of FVIII replacement therapy in PHA without M.-H. Chen1; B. Rodriguez1; C. Song1; A. Lachapelle1; C.
Wallace de Melendez1; P. Armstrong2; M. Chan2; T. Warner2;
A. Johnson1
National Heart Lung and Blood Institute’s The Framingham Heart Study,
Population Sciences Branch, Division of Intramural Research, Framingham,
United States, 2Blizard Institute, Queen Mary University of London, Barts and the
PB127 | Plasma Profile of Platelet Auto-­ London School of Medicine and Dentistry, London, United Kingdom
antibodies in Patients with Thrombocytopenia in
Pregnancy Background: The ISTH Bleeding Assessment Tool (BAT) is a tool to
screen individuals with potential platelet and bleeding disorders.
W. Shen1; H. Qin2
Such survey instruments are typically applied in clinical settings.
Jiangsu Institute of Hematology, The First Affiliated Hospital of Soochow
University, Suzhou, China, 2The First Affiliated Hospital of Soochow University,
There is limited information on the prevalence of bleeding symptoms
Department of Obstetrics and Gynaecology, Suzhou, China and conditions in the general population.
Aims: We aim to assess lifetime history and prevalence of bleed-
Background: Immune thrombocytopenia occurs when antibodies ing symptoms in >3,800 individuals in the Framingham Heart Study.
targeting specific glycoproteins (GPs) on the platelet surface lead Modified BAT scores and individual symptoms will be assessed for
to their destruction. Thrombocytopenia, defined as a platelet count association with genetics, risk factors, demographics and detailed
<150×109/L, is second only to anemia as the most common hemato- platelet and plasma assays.
logic abnormality encountered during pregnancy. Immune thrombo- Methods: We adapted the ISTH-­
BAT for self-­
administration by
cytopenia (ITP) is the most common cause of a platelet count below computer tablet, using skip-­logic to facilitate the survey. Positive
50×109/L detected in the first and second trimesters. answers trigger further questions on clinical actions, frequency and
Aims: To evaluate plasma levels of platelet auto-­antibodies in pa- attributes of events. To the ISTH-­BAT we added questions assessing
tients with thrombocytopenia in pregnancy and assess the clinical family history of major bleeding, and allowed free text entries for
significance. comments. The survey was translated into Spanish.
Methods: A total of 54 pregnant women with thrombocytopenia, Results: Of 2,339 respondents, n=2,144 were of European an-
44 normal pregnant women and 28 women with ITP, as well as 31 cestry and n=195 were of non-­
European ancestry. The sam-
healthy women were enrolled in this study. The plasma levels of ple was 53.5% female. The mean age was similar between men
platelet auto-­
antibodies against 5 platelet glycoproteins (GPIX, (54.8 years) and women (54.6 years). The modified BAT scores
GPIb, GPIIIa, GPIIb and P-­selectin) were detected by flow cytometric were positively skewed: 0 (81.9% of samples), 1 (13.1%), 2 (3.4%),

≥3 (1.6%). The maximum observed BAT score was 7. Those with Conclusions: Data on women with congenital FXIIID are scarce.
BAT scores ≥2 were more often women (80.3%). The mean age Menorrhagia was the third most common bleeding symptom. No
of those with BAT scores ≥2 (n=117) was significantly higher successful delivery was reported without prophylaxis. Prophylaxis
than those with scores <2 (n=2,222) (58.4 years vs. 54.6 years, treatment is a standard goal for management of women with severe
P<0.003). Responses for the modified BAT sub-­components are congenital FXIIID leading to decreased gynecological problems, ob-
summarized in the Table. stetric complications and successful delivery.
Conclusions: By late middle age, >5% of the population had signifi-
cant clinical bleeding events. Higher age associates with BAT score
≥2 consistent with more potential bleeding challenges in older in- PB130 | First Look at My Life Our Future
dividuals. As expected women have higher BAT scores. Consistent
Carrier Data from a Pilot Site
with some prior work, ~10% of women have a history of menorrhagia
requiring clinical intervention. J. Puetz1; L. Johnson2; C. Hugge1; P. Buchanan3
Saint Louis University, Pediatric Hematology/Oncology, St. Louis, United States,
SSM Health Cardinal Glennon Children’s Hospital, St. Louis, United States,
Saint Louis University, Center of Health Outcomes Research, St. Louis, United
PB129 | A Cohort Study on Women with States

Congenital Factor XIII Deficiency in Iran

Background: My Life Our Future (MLOF) was a genotype initiative
M. Naderi1; M. Karimi2; S. Haghpanah2; G. Miri-Aliabad1; H.
that offered free F8 or F9 genotyping for people with hemophilia liv-
ing in the United States. Subject enrollment began in 2013. In 2015
Zahedan University of Medical Sciences, Genetics of Non Communicable
Disease Research Center, Zahedan, Islamic Republic of Iran, 2Shiraz University
the program was expanded to include potentially affected female
of Medical Sciences, Hematology Research Center, Shiraz, Islamic Republic of Iran family members (carriers). New participant enrollment ended in
December 2017. As a pilot site for carrier detection, we were among
Background: Congenital factor XIII deficiency (FXIIID) is associ- the first to submit and receive data.
ated with life-­t hreatening hemorrhage underscoring timely diag- Aims: By reviewing data from a carrier pilot site we will demonstrate
nosis and management. Bleeding symptoms might render more what types of hemophilia carrier data are available from the MLOF
complicated in affected women than male due to pregnancy, and initiative and potential hypothesis that can be tested.
menstruation. Methods: We reviewed data from our center for all potentially
Aims: To evaluate a cohort study on women with congenital FXIIID affected female family members. The Bleeding Score (BLS) was
in Iran. obtained using the ISTH BAT. Each participant had a single fac-
Methods: A Cohort study was conducted on all registered women tor activity (FA) measurement using a one-­s tage assay. The study
with congenital FXIIID in southern Iran from 2007 to 2017. All was approved by the Institutional Review Board of Saint Louis
patients were severe with FXIII activity and antigenicity of less University and all participating subjects signed informed consent/
than 2%. Prophylaxis was given with dose of 10 U/kg every four assent. Only subjects who consented for participation in the re-
weeks. It was also scheduled 10-­20 U/kg every two weeks during search repository were included in this study. Statistical analysis
pregnancy. included independent samples t-­tests and chi-­square tests at a sig-
Results: A total of 241 women were investigated with median nificance level of 0.05.
age of 10 years (age range 1 day-­4 6 years).Thirty-­four percent Results: Sixty-­seven subjects had genotype data available. The ages
were in childbearing age (15-­4 0 years) and 89% had positive ranged from 1 to 89 years, and 97% of participants were White, non-­
history of first degree relative. The most common bleeding Hispanic. Among all participants, there was no correlation between
symptoms were umbilical cord bleeding (85%) and ecchymosis BLS and FA (r=−0.241, P=0.055).
(60%). Menorrhagia (33%), gum bleeding and post dental ex-
traction bleeding (6%), intracranial hemorrhage (15%) and ovu-
TA B L E   1   All Enrolled Subjects
lation bleeding (10%) were the other symptoms. One hundred
thirty four miscarriages were recorded in patients who weren’t Bleeding Score Factor Activity (IU/dL)
on prophylaxis with the frequency of one to 11 times per case.
None of 53 patients (18 cesarean section and 35 normal vagi-
n (SD) P-­value n Mean (SD) P-­value
nal delivery) had miscarriage or bleeding during pregnancy and
Carrier 43 6.5 (7.6) 0.11 42 81 (47.5) 0.001
post-­p artum while on prophylaxis but one patient developed
Non-­ 23 3.6 (5.0) 23 166 (109.1)
post-­p artum thrombosis. Forty eight surgeries were performed
covering with FXIII concentrate without any episode of signifi-
Age ≥18 46 6.7 (7.5) 0.006 45 107 (87.3) 0.67
cant bleeding except one post-­o perative portal vein thrombosis.
Age <18 20 2.6 (4.0) 20 117 (81.2)
None of them developed inhibitor.

TA B L E   2   Carriers Only

Bleeding Score Factor Activity (IU/dL)

n Mean (SD) P-­value n Mean (SD) P-­value

Hemophilia A 34 6.6 (8.2) 0.91 33 81 (52.2) 0.95

Hemophilia B 9 6.2 (5.0) 9 80 (25.9)
Age ≥18 37 7 (7.9) 0.23 36 77 (33.7) 0.62
Age <18 6 3 (4.3) 6 99 (100.3)
Factor Activity <40 6 12.5 (5.5) 0.04 6 31 (7.4) 0.004
Factor Activity >40 37 5.5 (7.5) 36 89 (46.2)
Known Status (mean age 26 8.2 (8.9) 0.04 26 72 (28.3) 0.24
Unknown Status (mean 17 3.9 (3.8) 17 93 (67.3)
age 37)

SD, standard deviation; IU, international units; dL, deciliter; Known/Unknown Status, carrier status known/unknown prior to enrollment.

Conclusions: Our data is consistent with prior studies showing a It was based on 2 tests: An evaluation of the validity of the
higher BLS in carriers compared to non-­carriers, although in this co- questionnaire construction and an assessment of the con-
hort it did not reach statistical significance. However, our data sug- cordance validity of the Arabic version of the PBAC with the
gests that prior knowledge of the carrier status may bias the BLS MCMDM-VWD.
determination. An analysis of the complete data set is necessary to
confirm this discovery. Results: Our patients were comparable in terms of socio-­
demographic characteristics. In terms of construct validity, the mean
score of the first half of menstruation with Ar_PBAC was 85 while
the mean score of the 2nd half of menstruation was 38 with a statisti-
PB131 | Translation and Validation of the
cally significant difference (P=0.001). In terms of validity of concord-
Pictorial Blood Loss Assessment Chart (PBAC) in
ance, the 2 conditions were validated. The mean scores of Ar_PBAC
Arabic in themenorrhagia group were statistically higher than that of the
O. Kaabia1,2; F. Letaief1; A. Brahim2,3; H. Skouri2,4; M. Bibi1 control group with an average score of 212 versus 64 (P=0.001). A
Farhat Hached Teaching Hospital, Gynecology and Obstetrics, Sousse, Tunisia, strong correlation to the MCMDM-­V WD questionnaire′s menorrha-
Faculty of Medicine, Sousse Tunisia, Sousse, Tunisia, 3Farhat Hached Teaching gia item score was found with a Spearman correlation coefficient
Hospital, Anesthesia and Resuscitation, Sousse, Tunisia, 4Sahloul Teaching
r=0.85 and P<0.05.
Hospital, Bio-­Hematology, Sousse, Tunisia
Conclusions: This Arabic version of the PBAC score is reliable to
quantify the menstruation of Arabophone women.
Background: The awareness of the burden of menstrual disorders
on women′s health led to a better quantification of the abnormal
menstrual bleeding by certain scales. But no version adapted to the
arabophone patients has not yet been validated. PB132 | Diagnosis Value of Screening Tools
Aims: The Pictorial Blood Loss Assessment Chart (PBAC) is an anno- for Hemostasis Disorders in Women Consulting
tated menstrual calendar of iconographic representations of the abun- for Prolonged Menstruation with Normal Pelvic
dance of menstrual flow. Its score is used to standardize the abundance
of menstrual blood loss. The aim of this work was to translate, adapt
and then validate this questionnaire for the arabophone women. O. Kaabia1,2; F. Letaief1; H. Abbessi3; I. Ghachem3; A.
Brahim2,4; S. Hadhri3; H. Skouri2,3
Methods: This work consisted in:
Farhat Hached Teaching Hospital, Gynecology and Obstetrics, Sousse, Tunisia,
Faculty of Medicine, Sousse, Tunisia, 3Sahloul Teaching Hospital, Laboratory of
- The Arabic translation of the content of the PBAC, Hematology and Blood Bank, Sousse, Tunisia, 4Farhat Hached Teaching Hospital,
- The adaptation of the coins used in the original questionnaire re- Anesthesia and Resuscitation, Sousse, Tunisia
specting the diameters,
- The deletion of the column which concerns the hygienic tampons, Background: Menorrhagia is a health problem that affects
- The statistical validation of the new questionnaire on a case- women quality of life. It remains unexplained in 50% of cases.
control study with a sample of 30 genitally active women (15 The awareness of the burden of menstrual disorders on
women complaining of menorrhagia and 15 with normal flow). women′s health led to a better understanding of their causes

and several tests have been developed to assess and quantify Hemostasis disorders were diagnosed in 63% of cases with a PBAC
menorrhagia. score of ≥100. ROTEM analysis approach had no significant added
Aims: The aim of this study was to determine the diagnosis value value to assess non organic menorrhagia.
of screening tools for hemostasis disorders in women consulting for Conclusions: The screening for hemostasis disorders in patients with
prolonged menstruation with normal pelvic exam. non organic menorrhagia is more efficient by the association of the
Methods: This is a case-­control pilot study including 40 cases with usual history taking with a PBAC score >100. We do not recommend
non organic menorrhagia and 20 controls who are volunteers with the use of the ROTEM analysis when screening for hemostasis dis-
no history of abnormal bleeding. We recorded the menstrual hab- orders in these women.
its and the hemorrhagic symptoms by two standardized question-
naires: PBAC and MCMDM-­V WD. All blood samples were taken
during the follicular phase of the menstrual cycle. Platelet aggrega- PB133 | Clinical Screening for Underlying
tion and ATP release were performed using different agonists: ADP,
Bleeding Disorder in Nigerian Women
Collagen, Arachidonic Acid and Epinephrin. Ristocetin has been also
used to assess platelet agglutination. Levels of coagulation factors, T. Nwagha1; E. Ugwu2; H. Okoye3; A. Ugwu3; Bleeding
Disorder in Nigerian Women
including factor XIII, have been assessed by Sysmex 2100CS ana-
University of Nigeria Teaching Hospital, Haematology & Immunology, Enugu,
lyzer. Thromboelastography study has been performed by ROTEM
Nigeria, 2University of Nigeria Teaching Hospital, Obstetrics and Gynaecology,
analysis. Ituku Ozalla, Nigeria, 3University of Nigeria Teaching Hospital, Haematology &
Results: A MCMDM-­DWD score of ≥ 4 detected only 52% of he- Immunology, Ituku Ozalla, Nigeria
mostasis disorders but it had a 100% specificity for VWD. The VWF
Ag level was significantly correlated with the MCMDM-­V WD score Background: Bleeding disorders in women can be detrimental to
with a Spearman coefficient r=−0.43 (P<0.05). their physical, psychological and social well-­
being with negative
impact on their quality of life. Given the scope of this problem lit-
tle is known about the prevalence of bleeding disorders in Nigerian
Aims: To evaluate the prevalence of perceived bleeding symptoms
in Nigerian women and the usefulness of a simple clinical screening
Methods: A cross-­
sectional observational survey of women be-
tween the ages of 16 and 50 years from 5 states in South east geo-
graphical zone in Nigeria using a structured pre validated 8-­question
questionnaire clinical screening tool. We distributed questionnaires
to women of all works of life (n=1,800). We received 1524 (85%)
Results: The mean age of the women was 26 (10.6) years.
Prevalence of bleeding symptoms was found to be 24.6%. Eleven
percent of women reported a positive family history of bleeding
disorder. There was a significant association between having a pos-
itive family history of bleeding disorder and experiencing bleeding
symptoms (AOR 0.12, 95% CI 0.06-­0 .22 P<0.0001). Commonest
bleeding symptom is menorrhagia, 141 (9.3%) had past history, 426
F I G U R E 1   Correlation between MCMDMVWD scores and
levels of VWF Ag (28%) had experienced at least flooding or passage of clots during

TA B L E   1   Diagnosis value of the screening tests

Sensitivity (%) Specificity (%) PPV (%) NPV (%)

PBAC ≥ 100
 Hemostasis disorders 75 13 63 20
 VWD 100 23 11 100
 Platelet dysfunction 75 13 63 20
EXTEM and/or INTEM abnormality
 Hemostasis disorders 10 88 50 44
 VWD 33 91 25 94
 Platelet dysfunction 10 89 50 47

their monthly cycle and 351 (23%) requiring resuscitation with TA B L E   1   Prevalence of hemostasis disorders in Tunisian Women
blood component therapy. with non organic menorrhagia
Conclusions: There is a high prevalence of perceived bleeding symp- Women with Healthy
toms among women with menorrhagia as the commonest bleeding non o­ rganic volunteers
symptoms. This tool is a cost effective easy tool to routinely identify ­menorrhagia (40 (20
cases) controls) P
women with bleeding symptoms needing further haemostatic or ob-
stetric evaluation. Hemostasis disorders 21 10 0.85
Platlets fonction 20 9 0.58
Von Willebrand disease 3 1 1
PB134 | Hemostasis Disorders in Tunisian
Coagulation factors 2 0 0.5
Women with Non Organic Menorrhagia: A Case deficit
Control Pilot Study
O. Kaabia1,2; F. Letaief1; H. Abbessi3; A. Brahim2,4; PB135 | Improving the Gynaecological
I. Ghachem3; S. Hadhri3; H. Skouri2,3
Management of Women with Bleeding or Thrombotic
Farhat Hached Teaching Hospital, Gynecology and Obstetrics, Sousse, Tunisia,
Faculty of Medicine, Sousse, Tunisia, 3Sahloul Teaching Hospital, Laboratory of Disorders by Multi-­professional Out-­patient Care
Hematology and Blood Bank, Sousse, Tunisia, 4Farhat Hached Teaching Hospital,
Anesthesia and Resuscitation, Sousse, Tunisia
K.M. Musgrave1,2; T.T. Biss1; J. Speedie3; D. Mansour4
Newcastle upon Tyne Hospitals NHS Foundation Trust, Department of
Haematology, Newcastle upon Tyne, United Kingdom, 2Newcastle University,
Background: The awareness of the burden of menstrual disorders Institute of Cellular Medicine, Newcastle upon Tyne, United Kingdom, 3NHS
especially menorrhagia on women′s health and quality of life led to Greater Glasgow and Clyde, Department of Sexual Health, Glasgow, United
a better understanding of the causes of abnormal menstrual bleed- Kingdom, 4Newcastle upon Tyne Hospitals NHS Foundation Trust, Department of
Women’s Services, Newcastle upon Tyne, United Kingdom
ing. Recently, several papers have investigated the role of hereditary
haemorrhagic disorders in non organic menorrhagia but none have
Background: Bleeding disorders and the treatment of thrombosis
investigated this issue in Tunisia.
complicates the management of gynaecological problems.
Aims: To enlighten the Tunisian situation facing this health problem,
Aims: To review the introduction of a multi-­
the aim of this study was to calculate the prevalence of hemostasis
haematology-­g ynaecology clinic at a UK teaching hospital.
disorders (platelet function abnormalities, Von Willebrand disease
Methods: Electronic medical records, for all women attending the clinic,
and/or coagulation factor deficits) in Tunisian women with non or-
were retrospectively reviewed from Jan 2016 to Dec 2017 (n=42).
ganic menorrhagia.
Results: Twenty-­seven cases with a bleeding history were identified
Methods: This is a case control pilot study including 40 cases with
(Von Willebrand Disease, 13; factor deficiency, 3; platelet disorder,
non organic menorrhagia and 20 controls who are volunteer healthy
3; dysfibrinogenaemia, 1; undefined disorder, 7). Median age was 28
women from the medical and paramedical staff with no history of
years (range 13-­53). 25 attended due to heavy menstrual bleeding
menorrhagia or any bleeding disorder. Controls were matched to
(HMB). 16 underwent venepuncture; 4 were iron deficient, 2 of whom
cases by age and by the menstrual cycle phase at the time of sam-
were also anaemic. Interventions included: oestrogen-­containing con-
pling (follicular phase).
traceptive pill (in 8); levonorgestrel-­releasing intrauterine system (4);
Platelet aggregation and ATP release were performed using dif-
progestogen-­only pill (5, 3 at higher than the usual dose); progestogen
ferent agonists: ADP, Collagen, Arachidonic Acid and Epinephrin.
intramuscular injection (3, 2 given 8-­weekly instead of 12-­weekly);
Ristocetin has been also used to assess platelet agglutination. Levels
tranexamic acid (3); combined hormonal vaginal ring (1); northisterone
of coagulation factors, including factor XIII, has been assessed by
(1); progestogen implant (1); oestrogen replacement (1). Multiple inter-
Sysmex 2100CS analyser. Thromboelastography study has been
ventions were required in 3.
performed by ROTEM analysis.
Fifteen cases with a thrombotic history were reviewed (2 inherited
Results: The prevalence of hemostasis disorders in Tunisian women
thrombophilia, 2 acquired, 5 single episode of thrombosis, 6 recur-
with non organic menorrhagia was 52.5% (95.2% platelet function
rent thrombosis). 12 were anticoagulated and 1 took an antiplate-
abnormalities, 14.3% VWD and 9.5% Factor X deficiency); however
let agent only. Anticoagulation included: vitamin K antagonists, 4 (3
there was no significant difference with controls. The prevalence of
with target INR ≥3); DOAC, 6; and low molecular heparin with anti-
VWD (type 1 or 2) was 7.5% in cases.
platelet agent, 1. HMB affected 12. Venepuncture was performed
Four menorrhagia cases had two concomitant hemostasis disorders;
in 11; 2 were anaemic, 1 was iron-­deficient. Interventions included:
3 cases of VWD and one case of coagulation factor deficiency had
progestogen-­only pill (8, 5 on the higher dose); progestogen injection
an associated platelet function abnormality.
(3 given 8-­weekly); levonorgestrel-­releasing intrauterine system (2).
Conclusions: Hemostasis disorders are prevalent in Tunisian women
Follow-­up data were available for 20 cases; all women reported an
with non organic menorrhagia.
improvement in symptoms. No surgical interventions were required.

Conclusions: This strongly suggests that joint working between hae- PB137 | Thrombocytopenia in Pregnancy: The
matology and gynaecology improves the care of women with bleed-
Role of Hemostasis Laboratory for Diagnosis and
ing and thrombotic disorders.
Management of Acute Fatty Liver of Pregnancy
I. Cuomo; M. Scannapieco; G. Rescigno
“Umberto I” Hospital ASL Salerno, Haemostasis and Thrombosis Unit, Nocera
PB136 | A Woman’s Life-­long Bleeding Inferiore, Italy
Condition: Diagnostic Dilemma and Expert
Opinion Background: Acute fatty liver of pregnancy (AFLP) represents a rare
R. Winikoff ; M. Othman 2,3 pregnancy complication. The clinical picture of AFLP is character-
1 ized by an early multi-­organ involvement.
University of Montreal, Haemophilia Treatment Centre at CHU Sainte-­Justine,
Montreal, Canada, 2Queen’s University, Biomedical and Molecular Sciences, Aims: We want to emphasize the great importance of coagulation lab
Kingston, Canada, 3St Lawrence College, School of Baccalaureate Nursing, testing for diagnosis and management of AFLP.
Kingston, Canada
Methods: We report a case of a 28 year old pregnant at 40 weeks of
gestation, admitted to emergency department for jaundice, abdomi-
Background: A 47 year old woman with lifelong menorrhagia inter-
nal pain and nausea.
fering with quality of life. No pelvic or gynaecological pathology was
Results: Entry laboratory tests: Hb 14 mg/dl, WBC 17,000/mm3, PLT
identified. Despite elaborate investigations, a cause for the bleeding
195,000/mm3, Total Bilirubin 7.15 mg/dl, Direct Bilirubin 5.28 mg/dl,
was not identified. Menorrhagia was eventually controlled by endo-
ALT 141 U/l, AST 191 U/l, Total Protein 6.5 g/dl and Albumin 3.2 g/
metrial ablation.
dl, Urea 33 mg/dl and Creat. 1.6 mg/dl, CRP 10.1 mg/l. PT Ratio
Aims: To present a diagnostic dilemma, obtain expert opinion and
1.62 and aPTT Ratio 1.31, Fibrinogen 144 mg/dl, DDimer 6121 ng/
investigate the possibility of underlying genetic bleeding disorder.
ml DDU, ATIII 12%, Factors Activity: VII 45%, IX 54%, X 40%, VIII
Methods: Full review of history, clinical and laboratory data was
220%, vWF Activity 400%, PC 35%, freePS 35.3%. A probable diag-
conducted together with genetic analysis using NGS.
nosis of AFLP was done following the Swansea’s criteria (Goel A. et
Results: The patient had ISTH-­BAT score of 12 with a strong fam-
al. Gut 2011). Additional lab test for differential diagnosis: absence
ily history of bleeding. She experienced serve menorrhagia and mu-
of schistocytes at peripheral Blood Smear, absence of Proteinuria,
cocutaneous bleeding and had two vaginal deliveries and several
Haptoglobin 60 mg/dl excluded Hemolysis and HELLP Syndrome,
surgeries without bleeding complications. All haemostasis investi-
normal blood pressure values excluded preeclampsia. Moreover se-
gations were normal including PT, APTT, fibrinogen, platelet count,
rology negative for hepatitis A, B, E and CMV, EBV excluded viral
MPV, factors V, VII, XIII, VWD profile, platelet aggregation in re-
infection. After urgent delivery, the AFLP patients usually improve in
sponse to ADP, collagen, ristocetin, epinephrine and arachidonic acid
24-­48 hours (Gami N. et al Int J Reprod Contracept Obstet Gynecol
and no spontaneous platelet aggregation or aggregation in response
2013) but in this case, the clinic and laboratory tests show a multi-­
to 0.5 ml ristocetin. The only abnormality was prolonged PFA-­100
organ failure. The reason of severe anemia, despite continuous blood
closure time to ADP but normal to EPI. The patient is known for au-
transfusions, is the widespread gastrointestinal bleeding found dur-
toimmune thyroiditis, hypermobility, dopamine-­secreting neck para-
ing an abdominal exploratory surgery resolved with administration
ganglioma, heterozygote MTHFR C677T; low ceruloplasmin, normal
of rFVII.
calcium and parathyroid hormone, and elevated 1,25 vitamin D with
Conclusions: Diagnosis and successful management of AFLP is still a
low 25 D. The patient’s 19 years old daughter had ISTH BAT score
challenge. Coagulation profile help to disclose the hepatopathy pat-
of 6 and similar bleeding and laboratory data. The patient reported
tern key for diagnosis and management of AFLP.
Omega 3, Aloes vera, Gluten and Phytates worsen menorrhagia. The
patient was treated with Tranexamic acid and iron therapy which
have only partially improved menorrhagia.
Conclusions: Despite advances in bleeding assessment tools and
PB138 | Quantitative Systems Pharmacology
laboratory investigations, unexplained lifelong menorrhagia remains (QSP) to Predict Thrombin Generation
a diagnostic dilemma precluding optimal therapies at times. The re- with Concomitant AT Lowering and Factor
sults of genetic analysis is hoped to identify a diagnosis in this par- Replacement or Bypassing Agents
ticular case and highlight its role in similar cases.
G. Sridharan1; S. Tsour1; K. Qian1; J. Liu1; Q. Lu2; C. Benson3;
K. Madigan1; S. Huang1
Alnylam Pharmaceuticals, Cambridge, United States, 2Sanofi, Bridgewater,
United States, 3Sanofi Genzyme, Cambridge, United States

Background: Fitusiran is a once monthly, subcutaneously adminis-

tered investigational RNAi therapeutic targeting antithrombin (AT)

to improve thrombin generation and promote hemostasis in patients Collagen type I messenger ribonucleic acid (mRNA) and the
with hemophilia A and hemophilia B with or without inhibitors. amount of fibrosis were quantified. Statistical analysis (Kruskal-­
Patients on fitusiran experiencing a bleed may require treatment Wallis followed by Scheffé multiple comparisons) was performed
with factor replacement or bypassing agents (BPA); thus, a deeper by Analyse-­it Software, ltd.
understanding of the relationship between factor/BPA dosing with Results: The mean TAA intake during the whole period was 38 mg/
AT lowering and thrombin generation is needed. kg/day and 24 mg/kg/day for HEMO-­
TAA and CTL-­
TAA respec-
Aims: To simulate the peak thrombin potential (PTP) that would be tively. Hemophilia mice (HEMO-­TAA) were exposed to a significantly
achieved with factor or BPA dosing in the context of AT lowering. (P<0.001) higher dose per body weight than the treated control mice
Methods: We developed a modeling framework that integrates fac- (CTL-­TAA). Nonetheless, in the TAA-­treatment group, the hemo-
tor pharmacokinetics with a quantitative systems pharmacology philic mice showed significantly less liver fibrosis (figure 1, P<0.001)
(QSP) model describing the coagulation cascade (Nayak S, et al. CPT and reduced collagen type 1 expression compared to their controls
Pharmacometrics Syst. Pharmacol, 4: 396-­4 05.) to simulate the in vitro (P=0.04).
PTP that would result with concomitant treatment of either factor or Conclusions: Our results bring experimental evidence that defi-
bypassing agents and AT lowering. Using Matlab and the Simbiology ciency in coagulation factor VIII protects from fibrosis and thereby
Toolbox®, we simulated PTP for doses of FVIII and FIX (standard support that the bleeding disorder is causally involved in the associa-
and extended half-­life), rFVIIa, and FEIBA under various time-­course tion between hemophilia and reduced liver fibrosis.
regiments with simultaneous AT lowering.
Results: Our modeling of concomitant AT lowering and dosing of
BPAs, Factor VIIa and FEIBA, reveals a correlation between throm-
PB140 | Geno-­and Phenotype Correlate in
bin generation observed from in vitro spiking experiments and that
Porcine Model of von Willebrand Disease Type 1
of QSP modeling. For a variety of extended half-­life factors, a time-­
course dosing regimen keeps the peak thrombin potential within the
and Type 3
normal healthy range for the full range of AT levels observed in pa- H. Allerkamp1; S. Lehner1; M. Ekhlasi-Hundrieser1; C.
tients on fitusiran (10-­25% residual). Detering1; C. Wermes1; C. Pfarrer2; M. von Depka1
Conclusions: The integration of the QSP model and plasma factor Werlhof Institut, Hannover, Germany, 2University of Veterinary Medicine,
Institute for Anatomy, Hannover, Germany
pharmacokinetic models adds to the data supporting clinical guid-
ance for dosing of standard and extended half-­life factors and by-
Background: von Willebrand disease (VWD) affects not only co-
passing agents for patients receiving fitusiran on a clinical trial.
agulation, but also angiogenesis. Up to 20% of VWD patients show
angiodysplasia. However, there is a lack of appropriate models to
investigate the influence of VWD on angiodysplasia in vivo.
PB139 | Does Hemophilia Slow Down the Aims: To characterise the correlation of geno-­and phenotype of
Development of Liver Fibrosis? porcine VWD type 1 and 3 as a model for investigating angiodys-
1 2 3 3 plasia in vivo.
M.-A. van Dievoet ; I. Leclercq ; C. Hermans ; C. Lambert ;
Y. Horsmans3; M. Jacquemin4; S. Eeckhoudt1 Methods: Blood of mature pigs with a naturally occurring mutation
Cliniques Universitaires Saint-­Luc, Department of Laboratory Medicine, in the VWF gene was analysed. Genotype was confirmed by molecu-
Brussels, Belgium, 2Université Catholique de Louvain, Laboratory of lar genetic analyses as described by Lehner et al. (2017) as heterozy-
Gastroenterology, Brussels, Belgium, 3Cliniques Universitaires Saint-­Luc,
gous for the mutation (het, type 1) and homozygous (hom, type 3).
Department of Internal Medicine, Brussels, Belgium, 4Universitair Ziekenhuis
Leuven, Department of Laboratory Medicine, Brussels, Belgium Gene analysis revealed a 12.3 kb duplication resulting in an almost
entire loss of the mature protein. Healthy pigs served as reference.
Background: Until the discovery of hepatitis C virus (HCV) and the Antigen, activity and multimeric pattern of von Willebrand factor
introduction of viral inactivating procedures, almost all patients (VWF) and factor (F) VIII were analysed.
treated with clotting factor products before 1989 were infected Results: The het animal produces 49% of the amount of duplication
with HCV. Unexpectedly, recent observations suggested that liver RNA product of the average of 12 measured hom pigs (101.1±16.0%).
complications of HCV were less severe in subjects with hemophilia All blood parameters show a decrease from the healthy to the hom
A than in other patients. animals. The het animal ranges in between. VWF antigen is affected
Aims: We investigated the difference in liver fibrosis in a chemically-­ to a greater extent than VWF activity. Decrease of FVIII activity has
induced liver cirrhosis model between factor VIII deficient mice and been shown for the hom animal only. The multimeric pattern of the
normal controls. het animal is comparable to human plasma but with less prominent
Methods: Liver fibrosis was induced by thioacetamide (TAA) in bands. The hom animal shows neither high weight nor intermediate
both normal (CTL) and factor VIII-­
d eficient (HEMO) mice and weight multimers.
compared to control groups receiving only water (H2O). After Conclusions: The decreasing activity of VWF and FVIII matches
25 weeks the mice were sacrificed and liver biopsies were taken. observation in human patients. Because of the good correlation

of geno-­and phenotype these pigs are a promising in vivo model. PB142 | Recombinant Expression of F9
Studying and breeding of affected pigs is simplified since the genetic
Nonsense Mutations and Fix Pharmacokinetics
mutation occurs naturally. This large animal model gives the possibil-
in Hemophilia B
ity to investigate VWD in general and its impact on angiogenesis at
an appropriate scale as well as angiogenic factors in all organs. A. Branchini1; M. Ferrarese1; M. Pinotti1; F. Bernardi1;
M. Morfini2
University of Ferrara, Department of Life Sciences and Biotechnology, Ferrara,
Italy, 2Italian Association Haemophilia Centers (AICE), Milan, Italy
PB141 | Pharmacokinetics and Safety of
Subcutaneous rIX-­FP Administration: Results Background: The distribution of Factor IX (FIX) in plasma, suben-
from a Phase 1 Trial dothelial and extravascular fluids and the variability of pharmacoki-
netics (PK) in Hemophilia B (HB) (Haemophilia 2016; 22:537-­42)
I. Pabinger1; T. Lissitchkov2; J. Roberts3; W. Seifert4; S. Puli3;
J. Ye3; Y. Li3 support the investigation of secreted FIX mutated molecules as co-
1 variates. We hypothesized that the residual amounts of mutated FIX
Medical University of Vienna, Clinical Division of Haematology and
Haemostaseology, Department of Medicine I, Vienna, Austria, 2Specialized molecules, including those truncated, could modulate FIX PK.
Hospital for Active Treatment (SHAT) ‘Joan Pavel’, Department of Hemorrhagic Aims: To correlate the expression of a broad spectrum of F9 non-
Diathesis and Anemia, Sofia, Bulgaria, 3CSL Behring, King of Prussia, United
sense mutations and their secreted recombinant FIX (rFIX) lev-
States, 4CSL Behring, Marburg, Germany
els with the PK outcomes of Nonacog alfa, which has never been
Background: Current factor IX (FIX) products are administered via
Methods: Nonsense mutations from patients (FIX<1%) were ex-
intravenous (IV) injections, which require repeated venous access
pressed in HEK293 cells with vectors bearing the premature F9 stop
and are burdensome for patients. Subcutaneous (SC) administration
codons (PTC). rFIX in the medium w[G1] as evaluated by ELISA and
may improve quality of life, particularly in patients with poor venous
Western blotting analysis.
access. rIX-­FP is a recombinant fusion protein linking FIX with albu-
Results: Eleven nonsense rFIX[M1] variants (p.R75X, p.L103X,
min. rIX-­FP has shown favorable safety and hemostatic efficacy in
p.R162X, p.W240X -­t ga/tag-­, p.R294X, p.R298X, p.Y330X,
clinical trials. An exploratory phase 1 study is investigating SC dosing
p.Q370X, p.R379X, p.R384X), collectively present in 70%
of rIX-­FP in hemophilia B patients (NCT02053792).
(324/469) of HB patients with PTCs in the F9 variant database
Aims: To investigate the pharmacokinetics and safety of single and
(www.factorix.org) were characterized [AB2]. In six of them
repeat SC doses of rIX-­FP.
(>50%, underlined) we detected secreted truncated molecules. For
Methods: Male hemophilia B patients (≥18 years old; FIX activity
the p.W240X normal levels (90-­120%) of truncated protein were
≤2%) enrolled in a phase 3b trial were eligible for the study. Patients
present. For the p.R162X we also detected traces of full-­length
received rIX-­FP as a single SC dose (25 or 50 IU/kg) or a repeat dose
rFIX. FIX PKs in single patients with the p.Q370X, p.R379X and
every three days for 15 doses (25 IU/kg). The first SC dose was ad-
p.R384X mutations indicated terminal (44.3, 57.5 and 37.8 hours;
ministered 14 days after the last rIX-­FP IV dose. For each dose SC
mean 46.5 hours) and beta half-­life (78.9, 41.9 and 34.6 hours,
rIX-­FP was administered in up to four injection sites (≤4 mL per site).
respectively; mean 51.8 hours) values higher than mean values
FIX activity was analyzed pre-­dose and then at regular intervals after
(terminal half-­life 29.7 hours; beta half-­life 37.3 hours) found in 11
administration until dose 15.
patients with missense mutations
Results: 14 patients have been enrolled. Three patients receiving re-
Conclusions: Truncated FIX molecules could flow to other compart-
peat doses of SC rIX-­FP had peak (Cmax) median (range) FIX activity
ments and compete with infused rFIX for binding to extracellular
levels of 6.1% (4.2-­12.0) following the 1st dose and 11.4% (6.0-­14.6)
receptors, which in turn could modify rFIX bioavailability. Our data
following the 2nd dose. Median pre-­dose (trough) values prior to the
suggest that an expression platform for patient’s mutations would help
1st dose were 3.6% (2.2-­11.6) and prior to the 15th dose were 6.2%
FIX PK analysis aimed at optimizing an individualized therapy in HB.
(5.9-­11.5). There were no serious adverse events (SAEs) reported,
no inhibitor development and no systemic hypersensitivity. Injection
site reactions were reported in five patients, most of which were
mild in severity. One patient withdrew following an injection site he- PB143 | Engineering Regulatory T Cells to
matoma of moderate severity. Block Inhibitor Formation
Conclusions: SC administration of 25 IU/kg with the usual rIX-­FP IV
D. Scott; J. Yoon; M. Abdeladhim; K. Parvathanen;
formulation every 3 days for 15 doses resulted in FIX activity trough A.-H. Zhang; Y. Kim
levels >5%, consistent with a mild hemophilia B phenotype. No pa- USUHS, Medicine, Bethesda, United States
tients experienced SAEs, systemic hypersensitivity or developed
inhibitors. Background: The immune response to therapeutic FVIII is a major
problem that blocks the efficacy of this critical protein. Currently, T

regulatory cells (Tregs) have be suggested to have potential for the analyses of FIX activity, pharmacokinetic and pharmacodynamics
treatment of adverse immune responses to biotherapeutics such as parameters of our hyperactive FIX variants, as well as evaluation
inhibitors to FVIII. of their efficiency and safety in AAV based gene transfer will be
Aims: To develop specific Tregs to induce tolerance to FVIII. presented.
Methods: To increase efficacy and specificity of Tregs, we previ- Conclusions: Our novel hyperactive FIX variants harbor increased
ously engineered human and mouse T cells to express chimeric an- activity compared to FIX-­wildtype and FIX-­Padua and reduction of
tigen receptors (CARs) by expressing either T-­cell receptors (TCR) collagen-­IV-­binding increases their bioavailability. Together, these
from hemophilia subjects or specific single chain fragments (scFv), in modifications promise enhanced efficiency of AAV based gene ther-
collaboration with the Königs lab in Frankfurt. apy for hemophilia B and allow reduction of the vector dose to avoid
Results: All of those engineered specific Treg cells can actively sup- therapy related toxicity.
pressed effector anti-­F VIII responses in vitro and in vivo! Recently,
we expanded this approach to create antigen-­specific cytotoxic T cells
as well as Tregs expressing FVIII domains (B-­cell antibody receptors PB145 | Population Pharmacokinetic
or BARs). Such BARs are able to kill FVIII-­specific hybridomas and
Analysis of Recombinant Fusion Protein Linking
LPS-­activated normal B cells in vitro. BAR Tregs actively suppress the
anti-­F VIII response in vivo via a direct mechanism.
Coagulation Factor IX with Recombinant Albumin
Conclusions: These results have potential to curb anti-­F VIII inhibitor (rIX-­FP) in Adult and Pediatric Patients with
formation in patients. Severe Hemophilia B
Y. Zhang1; J. Roberts1; T. Yuraszeck1; W. Seifert2; A.
Feussner2; A. Brainsky1; E. Santagostino3; J. Sidhu4
PB144 | Novel Hyperactive FIX Variants and 1
CSL Behring, Clinical Pharmacology and Early Development, King of Prussia,
United States, 2CSL Behring GmbH, PRD, Marburg, Germany, 3IRCCS Ca’ Granda
their Potential for Haemophilia B Gene Therapy Foundation, Maggiore Hospital Policlinico, Milan, Italy, 4CSL Limited, Clinical
Pharmacology and Early Development, Parkville, Australia
A.-K. Urbanowitz1; P. Milanov1; P. Quade-Lyssy1; D. Abriss1;
E. Seifried2; J. Schütttrumpf3; J. Schwäble 4
Institute of Transfusion Medicine, Molecular Therapy, Frankfurt am Main, Background: rIX-­FP, recombinant factor IX (FIX) fusion protein, was
Germany, 2Institute of Transfusion Medicine, Institute Director, Frankfurt am developed to extend the half-­life of FIX in hemophilia B patients.
Main, Germany, 3Biotest AG, Management, Dreieich, Germany, 4Institute of The population pharmacokinetics (PK) has been evaluated with the
Transfusion Medicine, Molecular Therapy Group Leader, Frankfurt am Main,
addition of PK data from a phase 3b study including extended rIX-­FP
dosing regimens of 100 IU/kg every 21 days for adults and 75 IU/kg
every 14 days for pediatric patients.
Background: Adeno-­associated-­virus (AAV) based gene transfer has
Aims: To characterize the population PK of rIX-­FP, assess the covari-
shown potential to provide sustained Factor IX (FIX) activity at low
ates such as demographic and clinical factors that are potential de-
levels in patients with severe hemophilia B, preventing spontane-
terminants of rIX-­FP PK variability and to simulate FIX activity-­time
ous bleeding and reducing FIX consumption. However, in AAV based
profiles for different dosing regimens.
gene therapy, dosing is limited by capsid-­specific cytotoxic immu-
Methods: A total of 116 hemophilia B patients (FIX ≤2%), aged 1 to
nity. Recently, application of the hyperactive FIX-­Padua mutant in-
70 years, were included. Patient blood samples were collected and
creased therapeutic efficiency of this approach, reaching mean FIX
FIX activity determined using the one-­stage clotting assay. A pop-
activities of 33.7% in hemophilia B patients.
ulation PK model was developed in NONMEM 7 software, where
Aims: To improve efficiency of AAV based gene therapy for hemo-
fit and prediction-­
corrected visual predictive check
philia B towards normal FIX activity, we developed novel FIX vari-
(pcVPC) used for model evaluation. Exogenous FIX activity levels
ants based on the FIX-­Padua mutant, with enhanced gain of function
were simulated for various dosing scenarios.
and modified collagen-­IV binding.
Results: FIX activity levels were appropriately described using a
Methods: Recombinant FIX variants were purified via anion ex-
2-­compartment population PK model. The final model precision was
change chromatography and tested in vitro for their clotting activity
confirmed by the goodness-­of-­fit and pcVPC. Median exogenous
and collagen-­IV-­binding. In-­vivo analyses include determination of
(rIX-­FP) simulated trough FIX activity is approximately 5% in patients
pharmacokinetic and pharmacodynamics parameters of the FIX vari-
≥12 years receiving 100 IU/kg every 21 days, and is approximately
ants and determination of their efficiency and safety in AAV based
5% for pediatric patients (6 to <12 years) and 3% for ages <6 years
gene transfer.
with both pediatric age groups receiving 75 IU/kg every 14 days.
Results: The activity of our novel hyperactive FIX variants was up
Conclusions: rIX-­FP PK was well characterized using a population PK
to 12 fold higher compared to the wildtype FIX protein and 1.5 fold
model approach. Dosing of rIX-­FP can be predicted using this model
higher compared to FIX Padua. Collagen-­IV-­binding was significantly
to produce FIX activity-­time profiles for different dosing schedules.
reduced in one of the new FIX-­variants and first in vivo experiments
The population PK simulations suggest that the majority of adult and
showed increased bioavailability of the respective mutant. Detailed

pediatric Hemophilia B patients receiving rIX-­FP on 21-­day and 14-­ CD61+ and CD62P+ were significantly increased in patients versus
day extended interval dosing regimens, reach sufficient FIX activity controls at both study sites (Figure). There were no correlations be-
levels to transition from severe to a mild or moderate hemophilia tween age at first joint bleed or annualized FVIII consumption and
classification. plasma EV levels.
Conclusions: EVs from platelets, leukocytes, erythrocytes or en-
dothelial cells do not appear to be involved in the clinical pheno-
type heterogeneity of patients with SHA. The finding that CD61+/
PB146 | Plasma Extracellular Vesicles (EVs)
CD62P+ EVs from activated platelets are increased in patients with
and Their Correlation with Clinical Phenotype
SHA warrants further investigation.
in Pediatric Patients with Severe Hemophilia A
PB147 | Pharmacokinetics of Recombinant
M. Rand1; S. Israels2; H. Wang1; E. McMillan-Ward2; V.
Blanchette1; R. Nieuwland3 Fusion Protein Linking Activated Factor VIIa to
Hospital for Sick Children, University of Toronto, Toronto, Canada, 2University Human Albumin (rVIIa-­FP) and Eptacog Alfa in
of Manitoba, Winnipeg, Canada, 3University of Amsterdam, Amsterdam, the
Hemophilia Patients with Inhibitors
J. Roberts1; C. Joch2; J. Ye3; S. Puli1; J. Mahlangu4;
Background: The biological factor(s) underlying bleeding severity E. Santagostino5
heterogeneity in patients with SHA are unknown; EVs released from CSL Behring, Clinical Pharmacology & Early Development, King of Prussia,
United States, 2CSL Behring, Clinical Development, Marburg, Germany,
blood and/or endothelial cells may play a role. 3
CSL Behring, Global Clinical Safety and Pharmacovigilance, King of Prussia,
Aims: To characterize plasma EVs in SHA and determine whether United States, 4Haemophilia Comprehensive Care Centre, Charlotte Maxeke
plasma levels correlate with bleeding phenotype. Johannesburg Academic Hospital and Faculty of Health Sciences, University of
the Witwatersrand and NHLS, Johannesburg, South Africa, 5Maggiore Hospital
Methods: Using a dedicated EV flow cytometer (A60-­Micro, Apogee;
Policlinico, Angelo Bianchi Bonomi Hemophilia and Thrombosis Center, IRCCS Ca’
detects single EVs ≤170 nm), levels of EVs expressing CD61, CD62P, Granda Foundation, Milan, Italy
both CD61 and CD62P, CD62E, CD45, CD235a and exposing phos-
phatidylserine (PS) were determined in citrated-­
plasma samples Background: Recombinant fusion protein linking recombinant acti-
from boys (≤18 years) with SHA (factor (F)VIII ≤1%) in 2 pediatric vated factor VII to human albumin (rVIIa-­FP) is a therapeutic option
hemophilia treatment centres. Controls were healthy adult males. designed to prevent and treat bleeding events in hemophilia patients
Bleeding severity was defined by age at first joint bleed and annual- with inhibitors. rVIIa-­FP was designed to reduce infusion frequency
ized FVIII consumption (U/kg), averaged over 3 years. The study was compared to current rVIIa treatments.
approved by the institutional Research Ethics Boards and informed Aims: To compare single-­dose pharmacokinetics (PK) of rVIIa-­FP to
consent was obtained from all subjects. eptacog alfa (rFVIIa) and assess safety of rVIIa-­FP in hemophilia pa-
Results: 20 boys with SHA and 20 controls were enrolled from Site tients with inhibitors.
1, and 22 boys and 20 controls from Site 2. There were no differ- Methods: In a Phase II/III multicenter, open-­label, dose escalation
ences between patients and controls in the concentration of EVs study of rVIIa-­FP 24 patients in Blocks A and B underwent single-­
from platelets (CD61+), platelets and endothelial cells (CD62P+), dose PK evaluation with a single injection of 90 (n=6) or 270 μg/kg
endothelial cells (CD62E+), leukocytes (CD45+) or erythrocytes (n=6) rFVIIa followed by a single injection of 1500 μg/kg rVIIa-­FP
(CD235a+), or EVs exposing PS. However, EVs expressing both in Block A or single dose of rVIIa-­FP of either 1500 μg/kg (N=4) or
750 μg/kg (N=8) in Block B. FVIIa activity was assessed using a vali-
dated StaClot® assay. PK parameters included plasma half‐life, area
under plasma concentration time curve (AUC) and maximum plasma
concentration (Cmax). Safety was assessed by number and type of
adverse events.
Results: For Block A, mean baseline-­
corrected FVIIa activity at
24 hours was higher for rVIIa-­FP (11.0 IU/ml) compared to rFVIIa
(0.06 IU/ml [90 μg/kg] and 0.11 IU/ml [270 μg/kg]). rVIIa-­
FP had
an extended half-­life of 8.5 hours, more than two-­fold higher than
rFVIIa (3.0 [90 μg/kg] and 2.6 hours [270 μg/kg]) and produced a
larger AUC0-t (847 vs. 112 [90 μg/kg] and 325 [270 μg/kg] IU/h/
mL). Cmax for rVIIa-­FP was 75.2 IU/mL, in between that of the 90
(44.1 IU/mL) and 270 μg/kg (131 IU/mL) rFVIIa groups. No inhibi-
tors to native FVII and no antidrug antibodies occurred. None of the
F I G U R E 1   CD61+/CD62P+ EVs in controls (C) and patients (P)
with SHA in the 2 study sites. *P=0.032, **P=0.004 by ANOVA study subjects developed serious adverse events.

In Block B, the average half-­life of rVIIa-­FP was 9.14 or 9.30 hours PB149 | APTT Reagent-­specific
and clearance was 7.28 and 5.95 mL/hr/kg. On average, Cmax and
Underestimation of N8-­GP (Turoctocog Alfa Pegol)
AUC0-t increased 2-­fold with 2-­fold increase in the dose. Overall, the
Activity Owing to Attenuated Thrombin Activation
PK of rVIIa-­FP in blocks A and B were comparable.
Conclusions: rVIIa-­
FP has an improved PK profile compared to E. Persson; T. Foscolo
rFVIIa and was well tolerated. Novo Nordisk A/S, Måløv, Denmark

Background: Many activated partial thromboplastin time (APTT) re-

agents give an accurate recovery of N8-­GP activity, but some silica-­
PB148 | Circulating Biochemical Markers of
based reagents underestimate it.
Early Joint Bleeding: Validation Study in Humans
Aims: To elucidate the mechanism behind N8-­
GP activity
X. Song; C. Enockson; L. Fogg; L. Boggio; M. Simpson; underestimation.
N. Hakobyan Methods: An assay was designed to mimic the contact activation and
Rush University Medical Center, Chicago, United States clotting phases of an APTT-­based clotting assay, but instead quanti-
fying factor Xa (FXa) amidolytic activity as a measure of factor VIIIa
Background: Early detection of joint bleeding in patients with severe (FVIIIa) cofactor activity. The FVIII source was either NovoEight®
hemophilia can prevent further hemarthrosis and ensuing destruc- (N8), Advate® or N8-­GP. Thrombin and factor IXa were optionally
tive changes. The discovery and application of biochemical markers added to the contact activation phase mixture (while omitting the
in addition to clinical and imaging assessment would greatly enhance phase itself) to measure contact activator effects on the rate of
the evaluation of early joint damage and may be used for the devel- thrombin activation of various forms of FVIII.
opment of new therapeutic agents. Results: Initially, we employed COATEST® SP FVIII to show that
Aims: The objective of this translational research was to evaluate in the same FVIIIa cofactor activity could indeed be achieved from the
hemophilia subjects circulating biochemical markers of joint bleed- same unit level of N8, N8-­GP and Advate®. If APTT-­SP (or two other
ing previously identified in our mouse studies. underestimating APTT reagents, TriniCLOT aPTT HS and STA-­
Methods: This project was approved by the Rush University IRB and Automate) replaced buffer in this assay, the rate of FXa generation
all subjects provided informed consent. in the presence of 0.6 IU/ml N8-­GP was the same as that obtained
Plasma (n=37) from severe hemophilia subjects was analyzed in 3 with a significantly lower concentration of N8, thus resembling the de-
groups. Control: no bleeding, target joints or synovitis; Synovitis: gree of underestimation of N8-­GP activity observed in a clotting assay.
target joints and/or synovitis, no assessed joint bleeding; Bleeding: With another silica-­based reagent, Pathromtin, 0.6 IU/ml N8-­GP and
blood in the joint. Joints were assessed by clinical and/or imaging N8 generated FXa at the same rate. Extending the contact activation
evaluation. Sixteen commercially available proteins were analyzed phase with APTT-­SP (90-­180-­300 s) resulted in a gradually increasing
using quantitative Luminex multiplex immunoassay. Because of difference in clotting time between N8-­GP and N8, and a similar pat-
small sample size, both the Effect Size (ES, Cohen’s d) and P value tern was observed in the FXa activity assay. In addition, with APTT-­SP,
were calculated. thrombin activation of N8-­GP appeared to be slower than that of N8.
Results: In Bleeding compared to Control, the expression of EGF, Conclusions: We have established analyses that assess FVIIIa co-
CSF2, IL4, IL13, FGF2, and MIP-­1α was significantly decreased with factor activity as FXa output rather than clot formation. This allows
very large effect size (ES=1.25 to 2.18), while expression of KC and mechanistic studies of the influence of APTT reagents on N8-­GP ac-
CSF3 was elevated (ES=0.80 to 1.54). Expression of EGF, CSF2, IL4, tivation and has unveiled attenuated thrombin cleavage as the cause
and CSF3 was different in Bleeding compared to Synovitis (ES=0.53 of N8-­GP activity underestimation.
to 1.17), and in Synovitis compared to Control (ES=0.20 to 0.53).
Since the Synovitis group might have subclinical joint bleeding, these
data may indicate a higher sensitivity of these biomarkers to joint
PB150 | Comparison of rFIX Utilization and
Bleed Rates in US Hemophilia B Patients on rIX-­
Conclusions: This pilot translational study identified a detectable
circulating biomarker signature associated with the presence of
FP and Their Prior rFIX Drug
blood in the joint and validates our results from animal research. M. Escobar1; C. Leissinger2; S. Yan3; G. Maro 4;
Once corroborated in larger scale studies, these plasma biomarkers G. Krishnarajah3
will be useful as a surrogate marker to monitor joint bleeding in he- University of Texas Health Science Center and the Gulf States Hemophilia
and Thrombophilia Center, Houston, United States, 2Tulane University Medical
mophilia subjects.
Center, New Orleans, United States, 3CSL Behring, King of Prussia, United States,
Adivo Associates, San Francisco, United States

Background: The three most commonly used drugs for recombinant

coagulation Factor IX (rFIX) replacement therapy in Hemophilia B

patients in the US include two long-­acting products, rIX-­FP (recombi- activity levels for each 12 hour time interval were simulated using
nant fusion protein linking human FIX with human albumin) and rFIXFc a previously published population pharmacokinetic model. A Cox
(recombinant fusion protein linking human FIX with a human Fc frag- proportional hazards model was used to relate the time-­to-­bleed
ment), and the standard-­acting nonacog alfa. Real world utilization and event data and various measures of FVIII exposure (>1%, >2%, >3%
outcomes need to be considered to understand how these products and >5%), as well as different dosing regimens (prophylaxis vs. on-­
compare in clinical practice. demand) to compare the risk of bleeding.
Aims: Determine real-­world FIX consumption and annual bleeding Results: Patients with FVIII levels >1% had a 74% reduction in bleed-
rate (ABR) in patients on rIX-­FP, and compare to prior drugs rFIXFc ing risk compared to patients with FVIII levels <1%. Patients who
or nonacog alfa. maintained FVIII activity level >1% for 6 months or 1 year decreased
Methods: De-­identified clinical patient chart information was ob- the relative bleeding risk by 71% and 92%. Maintaining thresholds
tained from US healthcare providers (hematologists and Hemophilia >2%, >3% and 5% further reduced the relative risk by up to an ad-
Treatment Centers) for 145 patients currently treated with rIX-­FP. ditional 10%. The prophylaxis regimen showed a 91% reduction in
ABR was calculated by annualizing the number of bleeds reported relative risk compared with on-­demand regimen.
over a period of 8 weeks to 1 year. Prescribed dosing and infusion Conclusions: An rVIII-­SingleChain treatment regimen maintaining FVIII
frequency were used to calculate FIX prophylaxis consumption for levels >1% over time is an effective approach to control bleeding events.
rFIX-­FP and their prior FIX drug used. The relative bleeding risk is reduced by 92% when the FVIII activity level
Results: Of the 145 patients, 84 (median age, 25 years; age range: 2-­ is maintained >1% for over a year. Higher FVIII levels further reduce
81; 82.1% severe, 17.9% mild or moderate) were treated on prophy- the relative bleeding risk slightly. Prophylaxis regimens provide greater
laxis both with rIX-­FP and their prior FIX drug. In a subset of these protection from bleeding than the on-­demand regimen. This analysis
patients with bleed information, ABR (±SD) was 0.7±1.0 on rIX-­FP ver- quantitatively supports the rationale for optimized treatment regimens
sus 8.9±9.6 on rFIXFc (n=12), and 1.5±5.8 on rIX-­FP versus 4.5±5.9 in adult hemophilia A patients by maintaining FVIII levels above 1%.
on nonacog alfa (n=17). Among the 37 rIX-­FP patients who previously
used rFIXFc, median rIX-­FP consumption (38.3; mean±SD: 43.1±17.5)
was 1.6 times lower compared to rFIXFc (60.6; mean±SD: 67.4±26.5; PB152 | Whole F8 and VWF Gene Sequencing
n=37). For the 29 patients previously using nonacog alfa, median rIX- using Next-­generation Sequencing for Mutation-­
­FP consumption (44.7; mean±SD: 47.1±14.3) was 2.5 times lower negative French and Canadian Hemophilia A Patients
than with nonacog alfa (109.9; mean±SD: 107.7±38.4; n=29).
F. Lassalle1,2; L. Jouan3; L. Swystun1; J. Gauthier4; C.
Conclusions: This study suggests that use of rIX-­FP for prophylaxis
Zawadzki2; J. Goudemand2; S. Susen2; G.-E. Rivard5; D.
leads to improved bleed control and lower consumption when com- Lillicrap1
pared to rFIXFc and nonacog alfa. 1
Queen’s University, Department of Pathology and Molecular Medicine, Kingston,
Canada, 2Lille University Hospital, Department of Hematology and Transfusion, Lille,
France, 3Sainte-­Justine University Hospital Research Center, Integrated Centre for
Pediatric Clinical Genomics, Montréal, Canada, 4Sainte-­Justine University Hospital,
PB151 | Identifying Efficacious Thresholds for Molecular Diagnostic Laboratory and Division of Medical Genetics, Department of
Pediatrics, Montréal, Canada, 5Sainte-­Justine University Hospital, Department of
Bleeding Risk Reduction in Relation to Factor VIII Hematology/Oncology, Montréal, Canada
Levels in Hemophilia A Patients Receiving rVIII-­
SingleChain Background: The identification of the molecular defect responsi-
1 2 1
P. Zhang ; C. Fosser ; J. Roberts ; A. Brainsky ; Y. Li ; 1 1 ble of hemophilia A (HA) is an important component in optimizing
W. McKeand3; J. Sidhu4 disease diagnosis. Despite the analysis of all exons of F8, 5’/3’ UTR
1 2
CSL Behring, King of Prussia, United States, Cytel, Cambridge, United States, regions and search for copy number variants, the pathogenic variant
CSL Behring, Clinical Pharmacology, King of Prussia, United States, 4CSL is still missing in about 5% of HA patients. In the last several years,
Behring, Parkville, Australia
Next-­Generation Sequencing (NGS) has provided the capabilities of
sequencing the complete F8 and VWF genes.
Background: There have been extensive discussions regarding the
Aims: The aim of this study is to characterize intronic mutations in
optimal FVIII level for prevention of bleeding events in severe he-
the F8 gene of “mutation-­negative” French and Canadian HA pa-
mophilia A (FVIII<1%).
tients, by using a whole gene sequencing method performed by NGS.
Aims: To assess the threshold levels of FVIII associated with significant
Methods: We developed a small panel containing whole F8 (195 kb)
reduction of bleeding risk, evaluate potential determinants of efficacy and
and VWF genes (188 kb). The SureSelect system was used for target
quantify the relative risk of bleeding episodes based on FVIII activity levels.
enrichment. We first analyzed 10 mutation-­negative HA patients and
Methods: rVIII-­SingleChain is a single chain recombinant FVIII. Data
6 controls (3 HA and 3 VWD) with known pathogenic variants. NGS
for up to 2 years from 147 adults with 715 bleeding events from a
was performed on a MiSeq sequencer. Data was analyzed with a bio-
Phase III study were utilized in the analysis, including all recorded
informatic pipeline according to the Genome Analysis Toolkit best
bleeding episodes after the first infusion of rVIII-­SingleChain. FVIII
practices. The effect on splice sites was predicted with Alamut® Visual.

Results: NGS data revealed 5 deep intronic candidate F8 variants (one Results: Seventy three patients used rFIXFc during the study period,
recurrent) in 5 HA patients. Patient 6 had an exonic variant in VWF 36 of whom met the inclusion criteria (34 severe, 2 moderate (0.01
described as VWD disease-­causing, that might lead to his reclassifica- and 0.02 IU/ml). The med. (IQR) age was 21 (12-­47) years and med.
tion. Two variants causing gene conversion in VWF have been found (IQR) post-­switch follow-­up was 21 (14-­29) weeks. Med. IF and CFC
in patient 7. Patient 8 and related patients 9 and 10 didn’t have an declined post-­switch by 46% and 42%, respectively (P≤0.01) (Table),
identified variant. We also found 5 variants that don’t change splice CFC reducing in all patients and IF in 35/36. Follow up is currently
sites so weren’t classified as potentially pathogenic. The average se- too short to evaluate the post-­switch annual bleed rate (ABR).
quence coverage depth was 20-­50X. Some variants had a coverage Conclusions: This direct within-­patient comparison of rFIX and rFIXFc
<20X threshold and they were all confirmed with Sanger sequencing. prophylaxis, demonstrates a significant reduction in IF and CFC using
Conclusions: This preliminary NGS run has corroborated that HA a PK driven dose-­restricted protocol. An analysis of weekly rFIXFc
might be caused by deep intronic variants, by discovering 4 new can- prophylaxis in the B-­LONG trial showed a 60% reduction in overall
didate variants. Future sequencing will optimize the depth of coverage CFC relative to rFIX, but similar post-­switch rFIXFc consumption. Our
for this assay. RNA and minigene studies are in progress to confirm the data, based on a larger cohort, potentially better reflects real-­world
deleterious effect of the intronic variants on F8 mRNA splicing. outcomes. The objective of dose-­restriction and cost-­neutrality was
only partially realised, however. Extended follow-­up will allow better
characterisation of these patterns and an analysis of ABR.

PB153 | Within-­patient Comparison of
Treatment Patterns before and after Switching
to rFIXFc: A Report from the UK National PB154 | Arthropathy and Joint Replacement
Haemophilia Database Surgery in Patients with Hemophilia A (PwHA) in
the US
M. Scott1,2,3; H. Xiang2; P. Collins4,5,6; R. Liesner2,7; C. Hay2,8,9
Manchester Royal Infirmary, Clinical Haematology, Manchester, United A. Patel; E. Yang; N. Engmann; R. Ko; K. Raimundo
Kingdom, 2UK National Haemophilia Database, Manchester, United Kingdom, Genentech Inc., South San Francisco, United States
University of Manchester, Institute of Cancer Sciences, Manchester, United
Kingdom, 4University Hospital of Wales, Cardiff, United Kingdom, 5UK National
Haemophilia Database, Cardiff, United Kingdom, 6University of Cardiff, Background: Little is known about the real-­world prevalence of ar-
School of Medicine, Cardiff, United Kingdom, 7Great Ormand Street Hospital, thropathy and joint replacement surgery rates in PwHA.
Department of Haematology, London, United Kingdom, 8Manchester Royal Aims: To assess the prevalence of arthropathy in PwHA and its rela-
Infirmary, Manchester, United Kingdom, 9University of Manchester, Department
of Vascular Sciences, Manchester, United Kingdom
tionship to joint replacement surgery in the US.
Methods: This was a retrospective, US claims analysis using the
MarketScan® Research Commercial (private insurance) and Medicaid
Background: Prophylaxis with rFIXFc in patients with haemophilia B
(public insurance) Databases. An algorithm (Lyons JL et al. 2017) was
(HB) may permit a reduction of infusion frequency (IF) and clotting
used to identify PwHA between 01/01/06 -­09/30/15 based on HA
factor consumption (CFC) relative to standard rFIX, whilst improving
diagnosis (ICD-­9: 286.0), treatment, demographics and resource use.
bleeding outcome. However, real-­world studies directly comparing
Index date was defined as the first date of HA diagnosis or treatment.
these products do not exist.
Patients were followed from index date to end of enrollment or analy-
Aims: To compare IF and CFC in patients using FIX prophylaxis be-
sis. PwHA with inhibitors were identified by ≥1 claim for a bypassing
fore and after switching to rFIXFc using the restricted, revenue-­
agent. HA arthropathy included arthropathy (ICD-­9: 710-­719) with
neutral, NHS-­England protocol.
a HA diagnosis in the same claim, or hemophilic arthropathy (ICD-­9:
Methods: rFIXFc was introduced in the UK from September 2016
713.2). Replacement surgery involving ankle, elbow, shoulder, hip or
using a PK-­optimised protocol. This used extended half-­life to opti-
knee joints was included. Patient characteristics, and rates of arthrop-
mise dose and dose-­interval to aim for cost-­neutrality. We analysed all
athy and joint replacement surgery were examined.
non-­trial patients switching from rFIX to rFIXFc prophylaxis, 01/09/16
Results: A total of 2,908 PwHA (Commercial N=2,285, Medicaid
-­ 30/09/17, with ≥12-­months pre-­switch and ≥4-­weeks post-­switch
N=623) were identified. PwHA on Medicaid were younger than those in
HT-­compliant data. IF and CFC were calculated for both periods and
Commercial plans with a mean age of 15.7 years (y) (SD=14.4 years) and
compared using summary statistics and Wilcoxon signed rank test.

TA B L E 1   Infusion frequency and clotting factor consumption pre- and post-switch to rFIXFc

Infusion Frequency (infusions/wk) FIX Consumption (IU/kg/wk)

Median IQR Range Median IQR Range

rFIX (pre-­switch) 1.98 1.53-­2.23 1.06-­7.05 79.00 52.00-­97.00 36.00-­416.00

rFIXFc (post-­switch) 1.09 0.95-­1.17 0.34-­2.33 46.00 31.00-­56.00 10.00-­162.00
Change −46% −42%
72       |
TA B L E   1   Hemophilia A arthropathy and joint replacement surgery by age between January 1, 2006 and September 30, 2015
(MarketScan® Research Databases)

Total, N=2,908a Truven Commercial, N=2,285 Truven Medicaid, N=623

Age group HA arthropathy, Joint replacement HA arthropathy, Joint replacement HA arthropathy, Joint replacement
(in years) n (%) surgery, n (%)b n (%) surgery, n (%)b n (%) surgery, n (%)b

Total, n 928 47 675 42 253 5

≤5 152 (16.4%) 0 (0.0%) 96 (14.2%) 0 (0.0%) 56 (22.1%) 0 (0.0%)
6-­12 189 (20.4%) 1 (2.1%) 124 (18.4%) 0 (0.0%) 65 (25.7%) 1 (20.0%)
13-­17 112 (12.1%) 0 (0.0%) 78 (11.6%) 0 (0.0%) 34 (13.4%) 0 (0.0%)
18-­24 133 (14.3%) 4 (8.5%) 112 (16.6%) 4 (9.5%) 21 (8.3%) 0 (0.0%)
25-­35 159 (17.1%) 12 (25.5%) 119 (17.6%) 10 (23.8%) 40 (15.8%) 2 (40.0%)
>35 183 (19.7%) 30 (63.8%) 146 (21.6%) 28 (66.7%) 37 (14.6%) 2 (40.0%)
a b
Includes total number of PwHA identified from Truven Commercial and Medicaid Databases. Includes joint replacement surgery following a HA ar-
thropathy diagnosis.

21.9 y (SD=15.4 years), respectively, and >99.5% patients were male. Methods: Data were collected under the patient authorization for the
Mean follow-­up of these patients was 2.1 years and 2.5 years (me- ATHNdataset at 3 timepoints, baseline (Jun 2016) then 9-­month in-
dian=1.6 years and 1.7 years), respectively. Inhibitor status was present tervals, (Mar 2017, Dec 2017). Patients with moderate or severe he-
in 41 (6.6%) Medicaid and 68 (3%) Commercial PwHA. Among PwHA, mophilia A or B without active inhibitors were included in the analysis.
31.9% had evidence of arthropathy related to HA. Of those, 5.1% had Results: Reported use of EHL FVIII rose from 9% of patients at baseline
evidence of joint replacement surgery (additional results in the table). to 21% as of 12/2017. The absolute number of patients on EHL FVIII
Conclusions: About 1/3rd of PwHA had evidence of arthropathy re- ppx nearly tripled, Figure 1A. Use of EHL FIX for ppx exceeded SHL at
lated to HA and 5% of those patients underwent joint replacement 52%. Absolute increase in EHL use varied by age group, Figure 1B&C.
surgery, which can have clinical and health resource use implications. Marked regional variability is observed in EHL adoption. Great Lakes
Future work should assess predictors of joint damage and associated region has the highest proportion of users at 63% EHL FVIII and 72%
health resource use in PwHA. EHL FIX. The rate of increase is highest in the Southeast. Median dose
and infusion frequency for SHL products were FVIII 40 IU/kg/dose,
156 annual infusions and FIX 67 IU/kg/dose, 104 annual infusions
PB155 | Extended Half-­life Factor Concentrate compared to EHL, FVIII 40 IU/kg/dose, 104 annual infusions, and FIX
Use for Patients with Moderate and Severe 75 IU/kg/dose, 52 annual infusions. On-­demand EHL use has grown,
Hemophilia among ATHN-­affiliated U.S. but its contribution to overall usage in minimal.
Conclusions: EHL FVIII use is increasing, but most remain on SHL.
Hemophilia Treatment Centers
EHL FIX now predominates for ppx. Despite broad regional variation,
S.E. Croteau1; D. Cheng2; A.J. Cohen3; C.E. Holmes4; P.A. a steady increase in EHL use is occurring nationally. While median IU/
Kouides5; L.J. Raffini6; M. Silvey7; C. Thornburg8; A.P.
kg and infusion frequency is less annually for prescribed EHL ppx, de-
Wheeler9; E.J. Neufeld10
1 creasing treatment burden, total cost of ppx is higher. Unless demon-
Boston Children’s Hospital/Harvard Medical School, Medicine/ Div. Hematology,
Boston, MA, United States, 2American Thrombosis and Hemostasis Network, strably off-­set by reduction in bleed doses, the net effect may be more
Rochester, NY, United States, 3Newark Beth Israel Med. Ctr., Newark, NJ, United costly therapy for a growing population.
States, 4University of Vermont Medical Center, Burlington, VT, United States, 5Mary
M. Gooley Hemophilia Center, Rochester, NY, United States, 6The Children’s Hospital
of Philadelphia, Philadelphia, PA, United States, 7Childrens Mercy Hospital, Kansas
City, MO, United States, 8Rady Children’s Hospital San Diego, San Diego, CA, United
States, 9Monroe Carell Jr. Children’s Hospital at Vanderbilt, Nashville, TN, United
States, 10St. Jude Children’s Research Hospital, Memphis, TN, United States

Background: Extended half-­life (EHL) FVIII and FIX products are in-
tended to decrease the burden of prophylaxis (ppx). Whether the
properties of these newer agents have led to clinical practice change
remains poorly defined. The ATHNdataset, a U.S. database of 138
ATHN-­affiliated hemophilia treatment center (HTCs), provides an
opportunity to explore longitudinal practice pattern changes.
Aims: To characterize use of FVIII and FIX concentrate types at HTCs,
including rate of switching from standard half-­life (SHL) to EHL products.

Aims: To investigate the relationships between the individual PK and

genes potentially modulating the FVIII survival.
Methods: 85 FVIII single dose (21.4-­51.8 IU/kg) PKs were performed in
33 hemophilia A patients (FVIII:C≤ 2IU/dl) without inhibitors. 20 (60%) of
patients underwent repeated PK (2-­6) with different FVIII concentrates.
The FVIII/time concentrations collected up to 72 hrs were analyzed by
Two-­Compartment Model (TCM) (WinNonlin). Final (K1-­0, K1-­2, K2-­1,
V1) and Secondary Parameters (AlfaHL, BetaHL, Cl, CLD2, Cmax, K1-­
0HL, MRT, V2, Vss) were evaluated. Genes coding for F8, AB0 blood
group, vWF, ASGPR, ITGB3, LDLR, LRP1/2, HSPGs, HPSE receptors were
sequenced in promoter region or/and in exons containing informative
polymorphisms (N=25). Data of each PK parameter were compared by
non-­paired, two-­tails, Student’s t test in patients grouped for genotypes.
Results: The inter-­patients variability of PK outcomes resulted very
large but inter-­occasion variability was definitely smaller: i.e. Beta_HL
range 6.3-­46.0 versus 11.0-­21.5 hrs, respectively. Blood group 0 was
associated with shorter Alpha half-­life than non-­0 (P=0.01). The LDLR
1173 C/T (rs688) displayed the most significant association (TT vs. C-­
carriers, Alpha HL P=0.001). In the ITGB3 gene several in linkage poly-
morphisms showed association with mean values of PK parameters
(i.e. -­425A/C, rs62074393, AA vs. C-­carriers, Beta HL P=0.01). For the
ASGR2 gene the -­95 T/C (rs2289645) polymorphism displayed associ-
ation with several PK parameters (i.e. TT vs. T-­carriers, CLD2 P=0.01).
Conclusions: The study confirms the need for tailoring replacement
therapy in HA, being hard to translate the outcomes of HA population
PK to single patient and suggests receptor gene polymorphisms to be
further analyzed in relation to FVIII PK in HA patients.

PB157 | Real World Data Analysis of US Claims

Database on Coagulation Factor VIII Expenditures
in Patients with Hemophilia A: Standard Half-­life
versus Extended Half-­life Products
D. Spurden1; A. Chhabra2; B. Tortella3; P. Fogarty3; E.
Rubinstein2; S. Harris1; A. Pleil4; J. Alvir2
Pfizer Limited, Tadsworth, United Kingdom, 2Pfizer Inc, New York, United States,
Pfizer Inc, Collegeville, United States, 4Pfizer Inc, San Diego, United States
F I G U R E 1   Longitudinal change in SHL and EHL factor
concentrates use for prophylaxis in U.S. based on ATHNdataset Background: Hemophilia A management requires intravenous factor
VIII (FVIII) infusions to replace the missing coagulation factor. Extended
PB156 | Genetic Polymorphisms Contributing to half-­life (EHL) FVIII products were recently introduced that may require

the Pharmacokinetics of Factor VIII Concentrates fewer infusions compared to standard half-­life (SHL) FVIII products.
Aims: We analyzed US claims data to determine international units
B. Lunghi1; S. Frusconi2; S. Linari3; G. Marchetti4; F.
(IUs) dispensed, expenditure and all-­cause direct expenditure associ-
Bernardi1; G. Castaman5; M. Morfini6
ated with EHLs versus SHLs.
University of Ferrara, Department of Life Science and Biotechnology, Ferrara,
Italy, 2Genetic Diagnostics Unit, Laboratory Department, Florence, Italy, 3Azienda Methods: De-­identified claims data from the Optum® Clinformatics® US
Ospedaliera Universitaria Careggi, Department of Oncology, Center for bleeding claims database were used to retrospectively identify male patients with
disorders, Florence, Italy, 4University od Ferrara, Department of Biomedical and hemophilia A who had used an SHL and/or EHL FVIII replacement product
Specialty Surgical Sciences, Ferrara, Italy, 5Center for Bleeding Disorders, Department
from August 1, 2014, to September 30, 2017, and had data for at least 3
of Oncology, Florence, Italy, 6Italian Association Haemophilia Centers AICE, Milan, Italy
months of dispensation. The key outcome measures were direct expendi-
tures, factor IUs dispensed for the FVIII replacement products and outpa-
Background: The covariates determining the large inter-­patients’
tient (OP) and inpatient (IP) visits. Medians were used to accommodate for
variability of FVIII pharmacokinetics (PK) are poorly known.
skewed distributions. Quarterly data for up to 36 months were analysed.

Results: 314 SHL subjects and 50 EHLs subjects were analyzed. and recovered well within the acceptance criteria (see Table 1). Data
Median age in both groups was similar (SHL=22.50 years; interquar- shows comparable recovery and statistics across sites.
tile range (IQR) 24.00 years; EHL=18.50 years; IQR 19.00 years). Conclusions: N9-­GP showed consistent results and accurate recov-
Median quarterly study period FIX expenditure was $52,575 higher ery using the new FIX-­SFX application on the ACL TOP 50 Series
(1.7 times) for the EHL cohort ($133,168; IQR $107,273) versus the Instrument with SynthAFax in an OSCA.
SHL cohort ($80,593; IQR $99,173). The Median quarterly IU usage
was also higher for EHL (74,742 IU; IQR 57,845) versus SHL users
(60,320 IU; IQR 74,189). Total haemophilia-­related expenditure in- PB159 | Phase 1/2 Trial of Single and Multiple
cluding OP, IP and prescription expenditure was $50,123 higher (1.6 Dose Subcutaneously Administered Factor IX Variant
times) for the EHL cohort ($133,344; IQR $125,673) versus the SHL
CB2679d/ISU304: Pharmacokinetics and Safety
cohort ($83,221; IQR $101,072).
C.W. You1; H. Levy2; M. Lee3; J. Park4; S.-B. Hong4; S. Kim4;
Conclusions: This real-­world data analyses, unadjusted for treatment
H.-J. Shin5; D.Y. Kim5; J.S. Kim6; J.W. Han6; S.-J. Kim6
regimen or disease severity, suggest that both IU dispensation and 1
Eulji University Hospital, Daejeon, Korea, Republic of, 2Catalyst Biosciences,
total FVIII expenditures are significantly higher in hemophilia A pa-
South San Francisco, United States, 3UCLA Fielding School of Public Health,
tients using EHL versus SHL. Further analyses, incorporating essential Biostatistics, Los Angeles, United States, 4ISU Abxis, Sungnam-­si, Korea, Republic
clinical characteristics, should be explored. of, 5Pusan National University Hospital, Busan, Korea, Republic of, 6Yonsei
University College of Medicine, Severance Hospital, Seoul, Korea, Republic of

Background: CB2679d/ISU304 was developed using rational pro-

PB158 | Measuring FIX-­Activity of Nonacog tein design to increase resistance to ATIII inhibition, increase cat-
Beta Pegol (N9-­GP) in a One-­stage Ellagic Acid alytic activity and affinity for FVIIIa resulting in 22-­fold enhanced
Based Clot Assay on the ACL TOP 50 Series potency enabling administration by subcutaneous (SQ) injection for
Instrument: A Two-­site Study use as prophylaxis.
1 1 1 2 1 Aims: Determine pharmacokinetics and safety of SQ CB2679d.
A. Khan ; B. Wilkinson ; K. Shishko ; M. Hansen ; M. Doyle ;
M. Triscott1; M. Ezban2 Methods: Cohort 1 received 75 IU/kg intravenous (IV) BeneFIX fol-
1 lowed by 75 IU/kg IV CB2679d. Cohort 2 and 3 received CB 2679d
Instrumentation Laboratory, Research & Development, Bedford, United States,
Novo Nordisk A/S, Haemophilia Research, Maaloev, Denmark 75 IU/kg IV followed by 75 or 150 IU/kg SQ. Cohort 4 was omitted.
Cohort 5 received 150 IU/kg SQ daily for 6 days.
Background: The Factor IX-­SynthAFax (FIX-­SFX) application on the Safety measures were performed 2 weeks after dosing. FIX antigen
ACL TOP/TOP50 Series Instruments was designed to measure Factor (VisuLizeTM Factor IX Antigen KitAG (Affinity Biologicals)) and activ-
IX clotting activity in plasma samples for aid in diagnosing and moni- ity (one-­stage clotting assay by ACL TOP 700 and Instrumentation
toring hemophilia therapy. This study aimed to demonstrate the re- Laboratories reagents)), anti-­drug antibody and neutralizing antibody
producibility and precision of the above application in accurately were measured at Haematologic Technologies. Half-­life (t1/2) calcula-
measuring the FIX clotting activity of nonacog beta pegol (N9-­GP) tion followed Lee et al, that uses an iterative piecewise fitting al-
spiked plasma samples when compared to a SSC standard. gorithm based on a robust (M-­regression) log-­linear model. Activity
Aims: To demonstrate reproducibility of the FIX-­SFX application to data were adjusted for baseline assuming exponential falloff of IV
recover N9-­GP and SSC spiked samples on the ACL TOP Family 50 administration. Bioavailability was calculated from the AUC0-t activ-
Series at two sites; IL Bedford and Novo Nordisk. ity data.
Methods: Study kits were prepared by IL and distributed to the second Results: IV CB 2679d has 22-­fold greater potency, longer terminal
site. Each site performed a FIX calibration with HemosIL Calibration t1/2 27.0±2.2 versus 21.0±1.1 hrs (P=0.0014) and mean residence
plasma and HemosIL FIX deficient plasma. Two quality control lev- time 35.8±2.5 versus 25.1±1.5 hrs (P=0.0004) compared with
els were tested to verify the calibration. N9-­GP spiked samples were
prepared at three target levels (100, 40 & 5% act.) with FIX deficient
plasma (George King) & Factor diluent. SSC samples were prepared TA B L E   1   Two-­site study data summary
at three target levels (105, 40 and 5% act.) with Factor diluent. All six
Mean (% % Total
samples were tested over five days (N=25 per level per site) on an ACL Samples N Activity) Recovery %CV
TOP 550 CTS analyzer. The mean result from each site was used to
N9-­GP 100% level 50 88.4 88.4 7.2
calculate the % recovery relative to the target. A recovery of ± 30% for
N9-­GP 40% level 50 38.3 95.7 7.1
each sample target level and a total %CV ≤20% was deemed acceptable
N9-­GP 5% level 50 5.7 113.6 10.2
across sites.
SSC 105% level 50 114.4 108.9 6.7
Results: The FIX-­SFX calibration was validated before use. Quality
Controls recovered within the acceptance criteria at both sites. All SSC 40% level 50 41.7 104.3 9.7

sample data (N=50 per sample for N9-­GP & SSC) were comparable SSC 5% level 50 5.9 118.4 12.7

PB160 | A Prospective Observational Study

of Prophylaxis, ACtivity and Effectiveness
(SPACE) in Hemophilia Patients Currently
Treated with Antihemophilic Factor
F I G U R E 1   ISU304-­0 01 trial schema
(Recombinant) in the US
B. Konkle1; D.V. Quon2; L. Raffini3; M. Recht4; V. Radulescu5;
S.L. Carpenter6; A. Dunn7; M. Watt8; J. Booth9
Bloodworks Northwest, Seattle, United States, 2Orthopaedic Hemophilia
Treatment Center, Los Angeles, United States, 3Children’s Hospital of
Philadelphia, Philadelphia, United States, 4Oregon Health & Science University,
Portland, United States, 5University of Kentucky Medical Center, Lexington,
United States, 6Children’s Mercy Hospital, Kansas City, United States,
Nationwide Children’s Hospital and The Ohio State University School of
Medicine, Columbus, United States, 8Shire, Zug, Switzerland, 9Shire, Cambridge,
United States

Background: Maintaining factor activity levels ≥1% of normal during

prophylaxis may not be optimal for all patients (pts) with hemophilia
A (HA). It is unclear if higher factor VIII (FVIII) peaks or troughs are
required to prevent bleeding episodes (BE) in someone with an ac-
tive lifestyle.
F I G U R E 2   FIX activity with six days of subcutaneous dosing Aims: To analyze the relationship between activity level, timing of
factor infusion, and BEs in pts with moderately severe/severe HA
on Antihemophilic Factor (Recombinant) (rFVIII; ADVATE®, Shire,
BeneFIX. SQ bioavailability was 19.8±5.2%. Median [25-­75%] peak
Lexington, MA, US). To integrate consumer information technology
activity 24 hrs after the 6th dose was 15.7% [14.9-­16.6%] and me-
to collect physical activity type/frequency using a pt-­reported eDi-
dian terminal t1/2 63.2 [60.2-­6 4.0] hrs was 2.3-­fold longer than IV.
ary (SPACE eDiary) and an activity tracker (Fitbit®; Fitbit, Inc.).
No subject had a bleed during the study. Adverse events were mild
Methods: SPACE was a 6-­mo, post-­authorization, multicenter, ob-
after 1st through 4th injection except for moderate pain, erythema
servational outcomes study (NCT02190149). Key outcomes were
and redness at SQ injection site in 1 subject after injections 1 and 2.
BEs, rFVIII regimen and dosing, physical activity, functional abil-
Conclusions: SQ PK data supports the aim of achieving 30%
ity, and pt satisfaction. Baseline characteristics were obtained via
activity within 12 days and normal FIX activity levels within 17
case report form and eDiary survey, pt-­reported outcomes includ-
days of initiating daily SQ CB2679d/ISU304 dosing. Complete
ing activity type via eDiary. Activity risk levels were assigned using
safety, bioavailability, PK and activity after daily SQ dosing will
National Hemophilia Foundation (NHF) criteria.
be reported.
Results: 54 pts (aged 11-­58 years) were enrolled (47 prophylaxis, 7 on
demand). Mean annualized bleeding rate on prophylaxis was 8±11.
Pts (N=54) were infused ≤24 h before activity in 35% (1759/4980) of
infusion intervals recorded (Table). More infusions were performed
≤24 h before commencing higher-­risk versus lower-­risk activities. In
SPACE eDiary regression models, prolonging the time between last
infusion and activity start (from ≤12 to 12-­24 h) showed consistent
but statistically nonsignificant increased odds (13-­20%) for activity-­
related bleeds (P≥0.6423).
Conclusions: For pts with HA on prophylaxis, aligning infusion timing
with activity patterns may reduce bleeding risk, although further in-
vestigation in a larger population is needed to confirm the temporal
association. Most pts were satisfied with the eDiary app and Fitbit.
Funding: This study was funded by Shire Development LLC.

TA B L E 1   Time From Most Recent Infusion Prior to Start of Activity Based on NHF Activity Risk Level (eDiary Data)

NHF activity level

All patients (N=54) 1 1.5 2 2.5 3

Number of intervals included in 4980 3761 488 935 330 196

Time between infusion and start of activity, n (%)
 Within 12 h 870 (17.4) 658 (17.5) 107 (21.9) 183 (19.5) 59 (17.9) 39 (19.9)
 Within 24 h 1759 (35.3) 1368 (36.4) 162 (33.2) 331 (35.3) 135 (40.9) 83 (42.3)
 Within 1-­2 d 1382 (27.8) 1047 (27.8) 130 (26.6) 283 (30.3) 92 (27.9) 52 (26.5)
 Within 2-­3 d 802 (16.1) 632 (16.8) 74 (15.2) 124 (13.3) 37 (11.2) 33 (16.8)
 Within 3-­5 d 558 (11.2) 411 (10.9) 57 (11.7) 98 (10.5) 31 (9.4) 19 (9.7)
 Within >5 d 479 (8.6) 303 (8.1) 65 (13.3) 99 (10.5) 35 (10.6) 9 (4.6)

PB161 | Correlation of Factor Levels a 1% increase in basal factor levels was associated with a 0.2
point decrease in the ISTH-­BAT (β=-­0.20, 95%CI: -­0.33-­(-­0.07),
with Bleeding Phenotype in Non-­severe
P=0.003). This association was similar in HA and HB patients (P
for interaction=0.78). Basal factor levels explained only 10%
J. Rejtő1; G. Schuster2; C. Feistritzer3; P. Quehenberger4; S. of the variation in the ISTH-­BAT (R 2=0.10 for both HA and HB
Hofer1; O. Königsbrügge1; E. Grilz1; F. Posch5,6; J. Gebhart1;
C. Ay1; I. Pabinger1
Conclusions: We observed a weak but significant correlation be-
Medical University of Vienna, Department of Medicine I, Vienna, Austria,
2 tween basal FVIII/IX levels and the bleeding severity. This data are
Blutspendezentrale, Linz, Austria, 3Medical University Innsbruck, Department
of Internal Medicine V, Innsbruck, Austria, 4Medical University of Vienna, consistent with the concept that there are other yet-­to-­be-­explored
Department of Laboratory Medicine, Vienna, Austria, 5Medical University determinants of the bleeding phenotype beyond basal factor levels
Vienna, Department of Medicine I, Vienna, Austria, 6Medical University of Graz,
in hemophilia.
Department of Internal Medicine, Graz, Austria

Background: Hemophilia is categorised into severity groups (se-

vere-­, moderate-­and mild hemophilia) based on the basal activ-
ity level of factor VIII (FVIII) or -­IX (FIX). This categorisation is PB162 | The Effect of Inhibitors on Quality
expected to be in accordance with the severity of the bleeding of Life and Pain in Adult Patients with Severe
phenotype. Haemophilia in the PROBE Cohort
Aims: Our study aimed at quantifying the association between the
D. Noone1; C. Chai-Adisaksopha2; R. Curtis3; N. Frick4; M.
severity of the bleeding phenotype and FVIII/ IX basal activity levels
Nichol5; B. O’Mahony1,6; D. Page7; J. Stonebraker8; A. Iorio2;
in 98 non-­severe hemophilia A (HA) and -­B (HB) patients. M.W. Skinner9; PROBE
Methods: Patients with non-­s evere hemophilia, defined as base- 1
Irish Haemophilia Society, Dublin, Ireland, 2McMaster University, Hamilton,
line FVIII/ IX activity of >1 to 40%, were eligible. Data on the Canada, 3Factor VIII Computing, Berkeley, United States, 4National Hemophilia
bleeding history were collected via direct patient consultations Foundation, New York, United States, 5University of Southern California, Sol
Price School of Public Policy, Los Angeles, United States, 6Trinity College, Dublin,
and search of hospital databases. This information was used to
Ireland, 7Canadian Hemophilia Society, Montreal, Canada, 8Poole College of
quantify the bleeding phenotype with the Bleeding Assessment Management, North Carolina State University, Raleigh, United States, 9Institute
Tool of the International Society on Thrombosis and Haemostasis for Policy Advancement Ltd., Washington, United States
Results: Seventy-­
seven (79%) patients with HA and 21 (21%) Background: The development of anti-­factor VIII inhibitors pose a
patients with HB were enrolled (median age 50 years, range 19-­ significant additional burden to people with severe haemophilia A
80). Median levels of FVIII were 13% [range: 1.5%-­39%] and that (PWsHA).
of FIX were 12% [2%-­25%]. The median ISTH-­BAT score was 10 Aims: To describe the effect of inhibitors on the quality of life (QoL)
points [0-­28] overall, 10 [0-­28] points in HA, and 8 [1-­20] points and reporting of pain in PWsHA and inhibitors.
in HB, and did not differ significantly between HA and HB Methods: This analysis utilised data from the PROBE (Patient
(P=0.48). Reported Outcomes, Burdens and Experiences) study. Participants
Lower basal factor levels correlated with a higher ISTH-­
BAT with No Bleeding Disorder (NOBD), PWsHA, and PWsHA with a
score (Spearman’s rho=-­
0.29, P=0.004). In linear regression, current inhibitor were compared. QoL was measured with EQ-­5D

and VAS. Descriptive analyses and logistic regression adjusted for Results: To date, 59 PTPs ≥12 and 24 < 12 years had entered the
age, mobility, presence of other pain, use of pain medication, and study (mean time on study 29.8 months). Over 79% of patients ≥12
reporting of other health problems were used. Standard t-­
tests switched from weekly to 10 (n=12), 14 (n=26) or 21-­day intervals
and chi squared tests were conducted to test for between-­group (n=11). Two patients starting 21-­day dosing switched back to the 14
differences. day interval to reduce their bleeding frequency. 25% of patients < 12
Results: 798 male patients, aged 11-­81 were analysed: 174 (22%) NOBD, were on the extended prophylaxis intervals of once every 10 (n=1) or
582(73%) PWsHA and42 (5%) PSsHA diagnosed with an inhibitor. PWsHA 14 days (n=5). In adults, median (Q1, Q3) AsBR for 7, 10, 14 and 21
and PWsHA with inhibitors did not differ for bleeding rates, presence day regimens was 0.33 (0.00, 2.39), 0 (0.00, 0.68), 0.26 (0.00, 1.54)
or number of target joints, or range of motion. There was a statistically and 0 (0.00, 0.45), respectively. In pediatrics, median AsBR for 7, 10
significant increase of reported life-­threatening bleeds in PWsHA with and 14 day regimens was 0 (0.00, 0.59), 0 (0.00, 3.06) and 0.75 (0.00,
inhibitors over PWsHA (40% vs. 14.5%; P<0.001). 50% of PWsHA with 2.86), respectively. Median number of exposure days (EDs) was 165
inhibitors were on intermittent or regular prophylaxis versus 76% of and 77 (92.8%) subjects have achieved 100 EDs to rIX-­FP; none have
PWsHA. developed inhibitors or antibodies to rIX-­FP.
Main results are reported in the table. Conclusions: Long-­term efficacy and tolerability of prophylaxis with
All EQ5D domains were different (P<0.001) except Anxiety/ rIX-­FP have been demonstrated. rIX-­FP allows extended treatment
Depression between PWsHA and PWsHA with a current inhibitor intervals of 21 days in adults and 10 and 14 days in children for se-
(P=0.317). The strongest interference of acute pain was found for lected patients.
walking ability (P=0.026) and sleep (P=0.056); for chronic pain it was
normal work (P=0.013) and sleep (P=0.090), without differences
among subgroups. PB164 | Burden of Mild Hemophilia A:
Conclusions: PWsHA with inhibitors in the PROBE cohort Systematic Literature Review
have lower QoL, and more chronic and acute pain. The differ-
F. Peyvandi1; A. Lawal2; M.R. Fasulo1; D. Frame3; B. Kim2; M.
ence is not explained by bleeding, target joints or functional
Tavakkoli2; M.C. Lee2; W.Y. Wong2
impairment. 1
University of Milan, Fondazione IRCCS Ca’ Granda Ospedale Maggiore
Policlinico, Angelo Bianchi Bonomi Hemophilia and Thrombosis Center, Milano,
Italy, 2BioMarin Pharmaceutical Inc., Novato, United States, 3Frame Research,
LLC, Brooklyn, United States
PB163 | Efficacy and Safety of Recombinant
Factor IX-­Fusion Protein (rIX-­FP) in Patients with Background: Hemophilia A (HA), an X-­linked, monogenic disorder,
Hemophilia B: Interim Results from an Ongoing is a life-­threatening condition that may require lifelong treatment.
Phase IIIb Extension Study Even mild forms of HA can impact morbidity, quality of life (QOL),
and health care utilization. The overall burden of disease for mild
E. Santagostino1; I. Pabinger2; A. Brainsky3; Y. Li3; W. Seifert4
HA (FVIII activity 5-­4 0%) is not well-­studied, and may vary depend-
Maggiore Hospital Policlinico, Angelo Bianchi Bonomi Hemophilia and
ing on treatment strategies, patient characteristics, and geographic
Thrombosis Center IRCCS Ca’ Granda Foundation, Milan, Italy, 2Universitätsklinik
für Innere Medizin, Medizinische Universität Wien, Vienna, Austria, 3CSL Behring, location of care.
King of Prussia, United States, 4CSL Behring, Marburg, Germany Aims: To systematically assess available evidence of disease burden
and clinical risks in adult patients with mild HA.
Background: rIX-­FP is a fusion protein genetically linking recombinant Methods: A systematic literature review was conducted with
human coagulation Factor IX with recombinant human albumin. It has prospectively determined search criteria. Studies of any de-
an improved pharmacokinetic profile compared with standard FIX prod- sign, published in English 1966-­2017, were eligible. We focused
ucts, allowing a prolonged dosing interval in patients with Hemophilia B. on adults with mild HA or studies with mixed populations of
Aims: As part of the PROLONG-­9FP clinical program, this phase 3b adults and adolescents ≥13 years of age. Studies including pedi-
study evaluates the long-­term safety and efficacy of rIX-­FP over ap- atric patients or without separable outcomes for mild HA were
proximately 3 years. excluded.
Methods: Patients received routine prophylaxis with rIX-­FP; dosing in- Results: Of 1,189 citations (Figure 1), 14 studies met eligibil-
terval could be changed at any 6-­month follow-­up. Treatment intervals ity criteria; mild HA adults (total n=1,740) were a subset of the
could be extended from a weekly dose to every 10 or 14 days with overall study population in all articles. Studies were conducted in
rIX-­FP doses of 50 to 75 IU/kg. Patients ≥18 years old could switch to Europe (57%), N. America (36%), and Japan (7%). Clinical burden
a 21-­day regimen at a dose of 100 IU/kg if they were well controlled of mild HA was reported in a variety of formats, making compari-
on a 14-­day regimen. The primary outcome was number of patients son across studies difficult (Table 1). QOL, cost, and productivity
who developed inhibitors against FIX. Secondary outcomes included loss were reported in only one study each: SF-­36 general health
annualized spontaneous bleeding rate (AsBR) and adverse events. was lower for mild HA versus age-­m atched controls in a Canadian

PB165 | Realworld Health Care Expenditure

and Dispensation of Extended and Standard
Half-­life Recombinant Factor IX Products in
Hemophilia B Patients
D. Spurden1; A. Chhabra2; B. Tortella3; P. Fogarty3; E.
Rubinstein2; S. Harris1; A. Pleil4; J. Alvir2
Pfizer Limited, Tadworth, United Kingdom, 2Pfizer Inc, New York, United States,
Pfizer Inc, Collegeville, United States, 4Pfizer Inc, San Diego, United States

Background: Hemophilia B management requires intravenous fac-

tor IX (FIX) infusions to replace the missing coagulation factor.
Extended half-­life (EHL) FIX products were recently introduced
that require fewer infusions compared to standard half-­life (SHL)
FIX products.
Aims: We analyzed US claims data to determine international units
F I G U R E 1   Study attrition diagram
(IUs) dispensed and expenditures associated with EHLs versus SHLs.
We examined how switching from SHL to EHL impacts pharmacy
cohort (58.1 vs. 70.8, P<0.05). In a US study, mild HA patients expenditure.
missed 6.2 work days per year, of which 4.7 (76%) were due to Methods: Truven Health MarketScan® Research US claims data-
HA. Mild HA health care costs of 793 euro/year were reported in base was queried for FIX IUs dispensed and expenditures. EHL
a Portuguese study. patients were compared with the SHL patients (cross-­sectional
Conclusions: Quantifying outcomes for adult patients with mild HA analysis), starting June 2014 (first EHL claim). We also analysed
remains challenging, with estimates of key QOL and cost data often FIX dispensation and expenditure in patients who switched from
based on small subsets of a single study. While mean annual bleeding SHL (nonacog alfa) to EHL (eftrenonacog alfa) product with at
rates were relatively low, mild HA had measurable impact on QOL least 3 months of data pre and post-­s witch. Descriptive statistics
and productivity and represents an area of therapeutic concern that were calculated with medians used to accommodate for skewed
should be addressed. distributions.
Results: Cross sectional analysis: Quarterly expenditure and IUs dis-
TA B L E   1   Clinical burden of mild HA in published studies
pensed were analyzed for 118 patients over up to 36 months (87
SHL; 31 EHL). Median quarterly study period FIX expenditure was
# mild HA Range in
Outcome # studies ≥age 13 published studies $116763 higher (3.4 times) for the EHL cohort ($165,022) versus the
SHL cohort ($48,259). Median quarterly IU usage was also higher for
Mean annual 3 157 0.41-­4.5
bleeding episodes/year EHL (59,337 IU) versus SHL users (34,088 IU).
episodes Switch analysis: 16 patients switched to EHL FIX and had at least 3
Mean severity 2 33 0.05-­0.07 (as months data pre and post switch. Median quarterly FIX expenditure
score (HSS or proportion of analyzed over up to 3 years pre/post-­switch was $75,064 pre-­switch
Brummel-­ maximum score) and $199,466 post-­switch. Median quarterly metric IUs dispensed
pre-­switch was 62,857 and 68,829 IUs post-­switch.
Mean joint score / 3 89 0.03 (HSS joint);
Conclusions: These real-­world data analyses, unadjusted for treat-
joint pain 5.3 (Colorado);
20% of patients ment regimen or disease severity, suggest that both IU dispensation
reported and total FIX expenditures are higher with EHL versus SHL in US he-
moderate or mophilia B patients. Further analyses, incorporating essential clinical
severe pain in
characteristics, should be explored.
past year
Inhibitor 3 217 4.0%-­7.8% over
development 10 years

HSS = Hemophilia Severity Score (maximum score 3). Brummel-­Ziedins: Bleeding ten-
dency score developed by Brummel-­Ziedins, et al. (2009), incorporating hemarthrosis,
soft tissue hematoma and annual FVIII unit/ kg usage (maximum score 32). HSS joint =
Hemophilia Severity Score, Joint (maximum score=1). Colorado = Colorado joint score
(maximum score=150).

PB166 | Expanding Communications on Background: The Expanding Communication on Hemophilia A

Outcomes (ECHO) study is a prospective, multinational, observa-
Hemophilia A Outcomes (ECHO) Study: Cross-­
tional cohort study investigating patient-­reported outcomes (PROs)
sectional Analysis of Baseline Data
in patients with moderate/severe hemophilia A (PwHA) aged ≥16
C.R.M. Hay1; M. Shima2; M. Makris3; V. Jiménez-Yuste 4; years.
M.W. Skinner5; J. Oldenburg6; K. Fischer7; E. Santagostino8; Aims: To describe cross-­sectional baseline data and to relate PROs
S. von Mackensen9; S. Rauchensteiner10; N. Church11; P.
to clinical data and sociodemographics.
Mathew11; S. Krassova10; C.M. Kessler12; on behalf of the
ECHO Study Investigators Methods: Data collection was planned at baseline and annually for
Manchester University Department of Haematology, Manchester, United ≥2 years. Clinical data were reported from medical records and pa-
Kingdom, 2Nara Medical University, Kashihara, Japan, 3University of Sheffield, tient diaries. PROs were assessed with standardized questionnaires
Department of Infection, Immunity, and Cardiovascular Disease, Sheffield, (electronic or paper) for health-­related quality of life, physical func-
United Kingdom, 4Autónoma University Madrid and La Paz University Hospital,
tioning, adherence, treatment satisfaction, pain, and productivity
Madrid, Spain, 5Institute for Policy Advancement, Ltd, Washington, DC, United
States, 6University Clinic Bonn, Department of Experimental Haematology and impairment. Joint status was assessed by presence of target joints,
Transfusion Medicine, Bonn, Germany, 7Van Creveldkliniek, University Medical joint procedures, Hemophilia Joint Health score, and Pettersson
Center Utrecht, Utrecht, the Netherlands, 8IRCCS Ca’ Granda Foundation,
score. Data analyses were explorative and descriptive.
Maggiore Hospital Policlinico, Milan, Italy, 9University Medical Centre Hamburg-­
Eppendorf, Hamburg, Germany, 10Bayer, Basel, Switzerland, 11Bayer, Whippany, Results: Of 271 PwHA enrolled in ECHO (UK, n=146; Japan, n=76;
NJ, United States, 12Georgetown University Medical Center, Washington, DC, Spain, n=25; US, n=24), 247 were included in this analysis (median
United States
[quartile 1; 3] age, 37 [27; 48] years). Of these, 84.6% had severe

TA B L E   1   Selected PRO Data* at Baseline in ECHO

PRO Possible Σ Total Aged 16-­27 y Aged 28-­36 y Aged 37-­47 y Aged >47 y
PRO Aspect Instrument Range Interpretation (n=247)† (n=66)† (n=57)† (n=61)† (n=63)†

Physical HAL 0-­100 High value indicates 68±25 (n=189) 84±20 (n=39) 76±21 (n=45) 63±23 (n=50) 54±25 (n=55)
functioning good functioning
Work WPAI (activity 0-­100 High value indicates 33±29 (n=184) 20±26 (n=38) 27±23 (n=44) 41±30 (n=49) 38±30 (n=53)
productivity impairment) high impairment
HRQoL EQ-­5D‡,§ 0-­1 High value indicates 0.7±0.3 0.8±0.2 (n=39) 0.8±0.2 0.7±0.3 (n=49) 0.6±0.3 (n=53)
good HRQoL (n=184) (n=43)
HRQoL EQ-­VAS 0-­100 High value indicates 71±21 (n=191) 75±19 (n=39) 75±20 (n=45) 66±19 (n=50) 68±22 (n=57)
good HRQoL
HRQoL SF-­12: PCS 0-­100 High value indicates 44±14 (n=186) 53±14 (n=36) 47±12 (n=45) 42±12 (n=50) 37±13 (n=55)
good HRQoL
Treatment Hemo-­Sat A 0-­100 High value indicates 23±14 (n=185) 20±10 (n=38) 20±13 (n=44) 29±14 (n=49) 22±14 (n=54)
satisfaction dissatisfaction
Treatment Veritas-­PRO 24-­120 High value indicates 44±11 (n=149) 47±13 (n=33) 44±11 (n=39) 44±10 (n=41) 43±11 (n=36)
adherence Veritas-­PRN (both) nonadherence 51±15 (n=30) 50±9 (n=4) 50±8 (n=4) 56±18 (n=8) 49±18 (n=14)
Pain BPI-­SF (total) 0-­10 High value indicates 2.7±2.4 (n=168) 1.8±2.2 (n=31) 2.5±2.3 (n=43) 3.4±2.5 (n=43) 2.7±2.4 (n=51)
more pain

BPI-­SF=Brief Pain Inventory-­Short Form; EQ-­5D=EuroQoL 5-­Dimension questionnaire; EQ-­VAS=EuroQoL visual analog scale; HAL=Haemophilia
Activities List; Hemo-­Sat A=Hemophilia Treatment Satisfaction Questionnaire for adults; HRQoL=health-­related quality of life; PCS=physical compo-
nents summary score of SF-­12; PRO=patient-­reported outcome; SF-­12=Short-­Form Health Survey; Veritas-­PRN=Validated Hemophilia Regimen
Treatment Adherence Scale-­ On Demand; Veritas-­ PRO=Validated Hemophilia Regimen Treatment Adherence Scale-­ Prophylaxis; WPAI=Work
Productivity and Activity Impairment Scale.
Additional PROs examined in ECHO not shown in this table include those assessing well-­being and HRQoL (Haemo-­QoL-­A) as well as patient surveys
for healthcare utilization, treatment consumption, physical activity, and alcohol and smoking status. In total, 8 validated PRO questionnaires and 5
patient surveys were included in ECHO.

Data are mean ± SD.

Mean ± SD EQ-­5D values by hemophilia severity are 0.7±0.3 (n=161) for <1% FVIII (severe hemophilia), 0.7±0.2 (n=11) for 1% to <2% FVIII (moderate
hemophilia), and 0.8±0.2 (n=11) for 2% to 5% FVIII (moderate hemophilia).
Mean ± SD EQ-­5D values by treatment regimen are 0.7±0.2 for on demand (n=34) and 0.7±0.3 for prophylaxis (n=149) at baseline. No information on
length of prophylaxis during a patient’s lifetime was taken into account in this analysis.

hemophilia, 54.7% had target joints, 34.8% had chronic arthropa- during the treatment. Family history of hemophilia was reported in
thy, 9.3% had an inhibitor history, 16.6% were HIV-­positive, and 49.2%, of whom 3% also have family history of inhibitor. Inversion
27.9% were positive for hepatitis C antibodies. Prophylaxis was used of intron 1 and 22 was present in 23/52 (44.2%) and 1 patient
in 80.2% of patients (daily, 6%; every other day, 15%; 3x/wk, 28%; (1.9%), respectively. All the included patients are under exclusive
2x/wk, 18%). PRO data were available from 68% to 77% of patients use of a third-­generation rFVIII of whom 82.1% are on prophylaxis
(Table). PRO scores were generally similar for moderate (15.4%) basis. Time to inhibitor development was 15 ED (95% CI, 9.6-­20.4).
and severe hemophilia and current prophylaxis versus on demand; A total of 61/73 (83.6%) PUPs reached at least 30ED, i.e., 29/61
however, few patients were treated on demand or had moderate (47.5%) with 75ED and 10/61 (16.4%) with 30ED without inhibitor
hemophilia (Table). Increasing age was associated with greater ar- and 22/61 (36.1%) developed inhibitor, of which most are high-­
thropathy, reflecting other covariates on PRO data. titer (n=18; 29%).
Conclusions: Most PRO data were adversely influenced by increas- Conclusions: Our findings confirm that the third generation recom-
ing age, reflecting the disease history apart from general age-­related binant factor VIII induces inhibitor development after a median of 15
developments. Further analyses are necessary to adjust for covari- exposure days in almost 40 percent of all PUPs with HA.
ates and to investigate the influence of different factors, such as he-
mophilia A severity and treatment regimen.

PB168 | The Effect of Unmeasurable

Endogenous Plasma Factor Activity Levels
PB167 | Inhibitor Incidence in Hemophilia on Factor VIII Dosing in Patients with Severe
A under Exclusive Use of a Third-­generation Hemophilia A
Recombinant Factor VIII Concentrate: Results of
P. Chelle1; A. McEneny King1; A. Iorio2; A.N. Edginton1
the HEMFIL Cohort Study 1
University of Waterloo, School of Pharmacy, Kitchener, Canada, 2McMaster
1,2 3 University, Department of Health Evidence, Research Methods and Impact,
L. Lemos Jardim ; D. Gonçalves Chaves ; C.
Brommonschenkel1; M. Portugal Santana3; M. Hermida Hamilton, Canada

Cerqueira4; A. Prezoti5; C. Santos Lorenzato6; V. K Franco7;

J. van derBom2; S. Meireles Rezende1 Background: Severe hemophilia A is a bleeding disorder corresponding
Faculty of Medicine, Universidade Federal de Minas Gerais, Belo Horizonte, to a deficiency of factor VIII (FVIII) and defined as a baseline level (BL)
Brazil, 2Leiden University Medical Center, Leiden, the Netherlands, 3Fundação lower than 0.01 IU.mL−1. To prevent bleeds and ensuing joint dam-
Hemominas, Belo Horizonte, Brazil, 4HEMORIO, Rio de Janeiro, Brazil, 5HEMOES,
Vitorioa, Brazil, 6HEMEPAR, Curitiba, Brazil, 7Hemocentro of Santa Catarina,
age, the prophylactic treatment consists in infusing FVIII regularly
Florianopolis, Brazil to maintain a trough level higher than 0.01 IU.mL−1. However, since
exact measurement of baseline FVIII level is often not possible due to
Background: It has been shown a higher risk for inhibitors with assay limitations (LLOQ = 0.01 IU mL−1), assumptions on BL (usually
the use of recombinant factor VIII (rFVIII) concentrates in com-
parison with plasma-­d erived in patients with hemophilia A (PHA).
However, prospective studies reporting inhibitor incidence under
exclusive use of specific types of rFVIII in an unselected cohort
are still missing.
Aims: To describe the incidence inhibitor in previously untreated
patients (PUPs) with HA under exclusive use of a third-­generation
rFVIII concentrate in the HEMFIL Study, an ongoing cohort.
Methods: PUPs with HA were enrolled consecutively in 5 hemo-
philia treatment centers in Brazil between 05-­
2013 and 31-­
12-­2017 and were followed up to 75 exposure days (ED) or until
inhibitor development. Clinical characteristics were collected using
standardized forms. Time to inhibitor development was calculated
by Kaplan-­Meier survival analyses.
Results: Currently, 73 PHA (FVIII:C< 2%) were included. The me-
dian age at first FVIII infusion was 10 months [IQR, 6.0-­13.8].
Most patients were white (64.3%) and were diagnosed because F I G U R E 1   Dose of FVIII required to maintain 95% of 500 virtual
of bleeding symptoms (83.6%). More than five consecutives ED subjects over various target troughs as a function of the baseline
were prescribed for 13 patients (17.8%) in at least one occasion assumption.

0 IU.mL−1 is assumed) are made during personalized pharmacokinetic

(PK)-­based dose calculations.
Aims: The objective of this work was to assess how assumptions
on BL influenced the determination of a prophylaxis regimen in the
treatment of hemophilia A using a PK-­driven approach.
Methods: Using a validated population pharmacokinetic model for
FVIII, 500 populations of 500 virtual subjects were simulated. Dose
and frequency required to maintain 95% of the population above
each trough level of 0.01, 0.03 and 0.05 IU.mL−1 were calculated as a
function of BL assumptions ranging from 0 to 1 IU.mL−1
Results: Patients with a true BL of 0 IU.mL−1 would require an addi-
tional 40 IU.kg−1 to maintain any target trough as compared to those
with a true BL close to 0.01 IU.mL−1 (Figure 1). For dose individuali-
zation, depending on if the dose is being increased or decreased to
reach a target, the baseline assumption can lead to widely varying
F I G U R E 1   Efmoroctocog alfa half-­life versus age
dose determinations.
Conclusions: Targeting a trough near the LLOQ is a problem unique
to hemophilia that is further complicated by the presence of an un- 12-­17y: 13.1 (11.3-­17.0) hrs; Figure 1). The estimated FVIII activity at
measurable amount of endogenous FVIII. As a result, assumptions 0.5, 72, and 96 hours after 50 IU/kg dose was 84.4 (68.7-­122.0), 2.8
regarding endogenous FVIII levels must be made, which can have a (0.6-­8.0), and 1.5 (0.4-­4.3) IU/dL, respectively (median, range). Five
dramatic effect on dosing regimen design. patients had predicted trough levels < 1 IU/dL at 96 hours. Of these,
one patient’s dose and another patient’s dosing frequency were in-
creased. Three patients continued the initial regimen as troughs < 1

PB169 | Pharmacokinetics of Extended Half-­ IU/dL would likely occur over night. One patient’s PK allowed a dose
reduction to 30 IU/kg twice weekly. The remaining patients had no
life FVIII Products after Recommended Dosing in
dosing changes.
a Pediatric Hemophilia A Population Conclusions: Our results show shorter FVIII half-­life for efmorocto-
T. Jacobs1; P. Barker1; A. McEneny-King2; E. Neufeld3; J. cog alfa in young children. The majority of this small cohort remained
Panetta1; U. Reiss3 on the recommended dosing of 50 IU/kg twice weekly. PK studies
St. Jude Children’s Research Hospital, Pharmaceutical Sciences, Memphis, allowed optimization of dosing in 20% of patients.
United States, 2University of Waterloo, School of Pharmacy, Waterloo, Canada,
St. Jude Children’s Research Hospital, Hematology, Memphis, United States

Background: Prophylactic infusions of factor VIII (FVIII) remain PB171 | Are Current Therapeutic Strategies
standard of care for patients with severe hemophilia A. The intro- with Conventional Products Sufficient to Manage
duction of extended half-­live (EHL) FVIII products may allow for re- the Burden of Haemophilia in Italy?
duction in dose frequency to twice weekly while maintaining trough
R.I. Mazzucchelli1; P. Kritikou2; K.-J. Myren2; J. O’Hara3,4
FVIII levels >1 IU/dL.
Swedish Orphan Biovitrum s.r.l. (Sobi), Parma, Italy, 2Swedish Orphan Biovitrum
Aims: Describe the pharmacokinetics (PK) of EHL FVIII products
AB, Stockholm, Sweden, 3University of Chester, Chester, United Kingdom, 4HCD
after recommended dosing in a pediatric hemophilia A population Economics, Daresbury, United Kingdom
and describe the proportion of patients that required a dose adjust-
ment to achieve a target trough >1 IU/dL. Background: Haemophilia care has improved significantly in the last
Methods: All St. Jude Children’s Research Hospital hemophilia A pa- 2 decades, and recombinant factor replacement therapies (rFVIII &
tients switched to EHL product efmoroctocog alfa with PK studies rFIX) are the mainstay treatment options.
were included. We evaluated the PK using Bayesian estimation and Aims: The aim of the present study was to describe the treatment
a limited sampling schedule developed from published population outcomes for adult persons with haemophilia (PwH) treated with
PK. FVIII half-­life and activity at 0.5, 72, and 96 hours after infusion conventional products in Italy, based on CHESS (Cost of Haemophilia
were estimated. This study was approved by the local institutional in Europe: a Socioeconomic Survey).
review board. Methods: Only PwH that had remained on the same treatment strategy
Results: Fifteen patients (age 1-­17 years) were analyzed. The FVIII (prophylaxis or on-­demand) from start of their treatment were included
half-­life was 11.3 (6.3-­17.0) hours (median, range) and was shorter in the present analysis. All patients were on conventional products; ex-
in younger individuals (1-­5y: 10.0 (6.3-­14.1), 6-­11y: 11.5 (10.2-­14.6), tended half-­life products were not available at the time of the study. Key

TA B L E 1   Presence of target joints, chronic pain, and QoL

Variable Haemophilia Type Primary Prophylaxis Primary On-­Demand

Presence of Target Joints (locations of chronic Haemophilia A 4/15 (27%) 32/60 (53%)
synovitis), n (%)
Presence of Target Joints (locations of chronic Haemophilia B 2/6 (33%) 3/13 (23%)
synovitis), n (%)
Presence of Chronic Pain, n (%) Haemophilia A 12/15 (80%) 39/60 (65%)
Presence of Chronic Pain, n (%) Haemophilia B 2/6 (33%) 5/13 (38%)
Quality of Life, mean (SD) Haemophilia A 0.94 (0.1), n=7 0.85 (0.2), n=21
Quality of Life, mean (SD) Haemophilia B 0.86 (0.1), n=3 1.0 (0.0), n=2

variables were: basic demographic and clinical characteristics, bleeds in Aims: In our specific condition with very recently introduced
the last 12 months, target joints (defined as locations of chronic synovi- prophylaxis, almost exclusively of secondary/tertiary or short
tis), physician-­reported chronic pain and patient-­reported quality of life term type, a high cost demanding therapy, we aimed at finding
(QoL) measured by EQ-­5D. All analyses were descriptive. the optimal solution from the point of view of effectiveness and
Results: CHESS collected data from 280 PwH in Italy. The 186 efficiency.
(66.4%) PwH that did not remain on the same treatment strategy Methods: Our non-­
interventional observational study was per-
were excluded from the analysis. Twenty-­one (7.5%) had always formed on 37 patients with severe form of haemophilia A without
been on prophylaxis (primary prophylaxis; PP), and 73 (26.1%) had inhibitors, all with secondary/tertiary or short term prophylaxis.
always been on-­demand (primary on-­demand; POD). For haemo- Plasma F VIII activity was measured at 0.1 hour and 48 hours after
philia A, 15 and 60 PwH were on PP and POD therapy, respectively. administration of recombinant F VIII in a dosage of 36.7+/-­5.38 IU/
For haemophilia B, 6 and 13 PwH were on PP and POD therapy, kg/24h. Incremental recovery and through level at 48 hours were
respectively. The mean (SD) number of bleeds in the last 12 months evaluated, being compared with the correspondent expected values
were 2.7 (2.3) and 1.3 (1.6) for PP in PwH A and B, respectively; for the type of the administered product.
as opposed to 5.4 (9.5) and 3.8 (6.2) for POD in PwH A and B, re-
spectively. Presence of target joints, chronic pain, and QoL are pre-
sented in the Table.
Conclusions: According to the CHESS results, despite continuation on TA B L E   1   Global situation of FVIII recovery at 48 hours:
the same treatment beyond 12 months, PwH on conventional prod-
ucts in Italy continue to present a notable disease burden in terms of Dosage FVIII Real Estimated recovery
(IU/kg) recovery(IU/dl) (IU/dl) P value
target joints, and chronic pain; especially for those on POD. The data
indicate the need to further optimise haemophilia treatment in Italy. 36.7±5.38 9.1±8.4 9.3±5.1 0.4

TA B L E   2   Distribution of real recovery values (IU/dl) at 48h (% of

PB172 | Practical Considerations in Choosing patients):

a Cost-­effective and-­Efficient Solution for

Dosage FVIII
Prophylaxis in Haemophilia (IU/kg) <1 IU/dl 1.1-­5 IU/dl 5.1-­10 IU/dl >10 IU/dl
M. Serban1; E. Ursu1; A. Traila2; P. Serban2; T. Opait-Dugan2; 36.7±5,38 2.7% 43,2% 21.6% 32.4%
M. Dutescu2; D. Puiulescu3; L. Ritli4; C. Jinca5; E. Boeriu5;
J.M. Patrascu5; S. Cerbu5; S. Arghirescu6
We had to accept once again that „one size doesn’t fit all”, as phar-
Clinical Emergency Children Hospital ‘Louis Turcanu’, European Haemophilia
macokinetic of products can depend on many factors like: age, blood
Treatment Center, Hematology, Timisoara, Romania, 2Rehabilitation Medical Centre
for Children and Adolescents ‘C. Serban’ European Haemophilia Treatment Center, group, vWF level, exercise, joint status and type of replacement
Buzias, Romania, 3Clinical Emergency Children Hospital ‘Louis Turcanu’, Timisoara, product. In our cohort, only in 10.8% cases the through level at 48
Romania, 4University of Medicine and Pharmacy, Oradea, Romania, 5University of
hours matched to the expected values.
Medicine and Pharmacy “Victor Babes”, Timisoara, Romania, 6University of Medicine
and Pharmacy ‘Victor Babes’, Timisoara, Romania The high heterogeneity of 48h recovery values of the same replace-
ment dosage makes obvious that in our situation in which different

Background: Prophylaxis is generally accepted as treatment of trough levels resulted from the same dosage, the individualization

choice in severe forms of haemophilia. But the optimal prophylaxis of prophylaxis in haemophilia care is mandatory, serving not only its

regimen is still under scientific debate. effectiveness, but also its efficiency.

Conclusions: In this challenging situation, it seems that pharmacoki- PB174 | Estimation of Simoctocog Alfa and
netic driven replacement is a promising strategy to improve haemo-
Octocog Alfa Activity using One-­stage and
philia care in a right manner, avoiding too low or two high levels,
Chromogenic Assays in Routine Laboratory
taking into consideration the whole community of persons with
Practice: Results of a Comparative International
Field Study
T. Vicente1; J. Versteden1; C. Borgvall2; J. Bichler1; O.
Walter1; S. Tiefenbacher3; L. Belyanskaya1
PB173 | rVIII-­SingleChain in Surgical 1
Octapharma AG, Lachen, Switzerland, 2Octapharma AB, Stockholm, Sweden,
Prophylaxis: Efficacy and Safety in 43 Surgeries 3
Colorado Coagulation, Laboratory Corporation of America® Holdings,
Englewood, CO, United States
J. Mahlangu1; F. Abdul Karim2; C. Djambas Khayat3; J. Ong4;
S. P’Ng5; J. Oldenburg6; A. Brainsky7; S. Lucas7; I. Pabinger8;
for the Affinity Study Group Background: Precise determination of coagulation factor VIII activ-
University of the Witwatersrand, NHLS and Charlotte Maxeke Hospital, ity (FVIII:C) is essential to ensure efficacy and safety of haemophilia
Johannesburg, South Africa, 2National Blood Centre, Kuala Lampur, Malaysia, A treatment. One-­stage (OS) and/or chromogenic (CHR) assays are
Hotel-­Dieu de France Hospital, Ashrafieh, Lebanon, 4Brokenshire Integrated, commonly used; however, there can be substantial variability in the
Davao City, Philippines, 5Royal Perth Hospital, Perth, Australia, 6University Clinic
Bonn, Institute of Experimental Haematology and Transfusion Medicine, Bonn,
Germany, 7CSL Behring, King of Prussia, United States, 8Medical University Aims: This international field study analysed the performance
Vienna, Clinical Division of Haematology and Haemostaseology, Medical Clinic I, of simoctocog alfa (B-­domain deleted recombinant FVIII [rFVIII]
Vienna, Austria
of human-­
cell line origin) and octocog alfa (full-­
length rFVIII of
hamster-­cell line origin) in FVIII:C assays routinely performed in
Background: rVIII-­SingleChain is a novel recombinant Factor VIII
clinical laboratories.
designed as a B-­domain truncated construct with a covalent bond
Methods: Congenital FVIII-­deficient plasma was spiked with 4 con-
between the heavy and light chain and a higher affinity to von
centrations (1, 5, 30 and 100 IU/dL) of simoctocog alfa or octocog
Willebrand Factor.
alfa. Participating laboratories received blinded samples and deter-
Aims: The surgical sub-­studies of the Affinity program investigate
mined FVIII:C following their standard procedures. The methodo-
the safety and efficacy of rVIII-­SingleChain to control surgical hemo-
logical parameters used by each laboratory were summarised in a
stasis in children and adults with severe Hemophilia A.
Methods: Surgery was defined as a procedure requiring general,
Results: A total of 49 participating laboratories from 9 countries
spinal or regional anesthesia. Dosing of rVIII-­SingleChain was based
provided results (both OS and CHR assays: n = 40; OS assay only:
on the type of surgery, guided by the WFH recommendations. rVIII-­
n = 8; CHR assay only: n = 1). As several laboratories performed
SingleChain was administered as a bolus or a continuous infusion.
more than one assay of a given type, the total number of data
The investigator rated hemostatic efficacy using a 4 point rating
sets was 50 for the OS and 43 for the CHR assay. Mean abso-
scale (poor/none; moderate; good; excellent); treatment success was
lute FVIII:C values were comparable for both products at all con-
defined as a rating of excellent or good.
centrations and for both assays, although ranges appeared to
Results: 43 surgeries were performed on 32 patients aged be-
be slightly larger for simoctocog alfa (Table 1). Mean recoveries
tween 5 and 64y (median, 32). 22 surgeries were orthopedic while
ranged from 97% to 191% for simoctocog alfa and from 93% to
21 surgeries were non-­o rthopedic. 17 surgeries were related to
172% for octocog alfa, with higher recoveries at lower concentra-
hemophilia or its complications, and 1 surgery (appendectomy)
tions. Subgroup analyses by OS and CHR assay reagents showed
was an emergency. Overall, rVIII-­S ingleChain was used as a bolus
minor variations in FVIII:C depending on reagents, but no marked
dose in 35 surgeries, and as a continuous infusion in 8 surgeries.
differences between the two rFVIII products. CHR:OS ratios for
No related AEs or SAEs were observed during the perioperative
simoctocog alfa ranged from 1.05 to 1.19 and from 1.06 to 1.25
period. Hemostatic efficacy was rated as excellent and good in 38
for octocog alfa (Table 1).
(88%) and 5 (12%) surgeries respectively. Of the orthopedic sur-
Conclusions: The accuracy of FVIII:C determination using either OS
geries, 18 (82%) were rated as excellent, and 4 (18%) were rated
or CHR assays routinely performed in clinical laboratories was com-
as good.
parable for simoctocog alfa and octocog alfa.
Conclusions: rVIII-­SingleChain, dosed by bolus or continuous infu-
sion, was well-­tolerated and effective in achieving surgical and peri-
operative hemostasis, with an overall treatment success of 100%.

TA B L E 1   Absolute FVIII:C values as determined by OS and CHR assay and CHR:OS ratios

CHR:OS ratios
at individual
OS assay (n = CHR assay (n = laboratories (n
Target 50) 43) = 43)
tration (IU/ Mean ± SD (IU/ Mean ± SD (IU/
Product dL) dL) Range (IU/dL) dL) Range (IU/dL) Mean ± SD Range

Simoctocog 1 1.8 ± 0.81 0.6-­4.0 1.9 ± 1.02 0.0-­5.6 1.19 ± 0.704 0.33-­3.66
alfa 5 6.5 ± 2.80 4.0-­22.5 6.1 ± 2.14 0.6-­14.2 1.05 ± 0.452 0.09-­2.37
30 32.0 ± 11.14 22.7-­103.0 34.5 ± 3.73 28.2-­47.7 1.13 ± 0.223 0.36-­1.71
100 97.3 ± 15.45 66.2-­171.7 112.4 ± 11.06 96.1-­142.9 1.17 ± 0.189 0.83-­1.59
Octocog alfa 1 1.6 ± 0.68 0.5-­3.1 1.7 ± 0.96 0.0-­5.1 1.25 ± 0.709 0.37-­3.56
5 6.1 ± 1.63 3.7-­11.1 5.9 ± 1.88 0.5-­12.2 1.06 ± 0.428 0.09-­2.44
30 30.9 ± 5.44 22.4-­59.6 34.1 ± 3.62 24.8-­4 4.3 1.12 ± 0.168 0.63-­1.55
100 92.7 ± 10.90 69.0-­141.3 108.5 ± 9.29 95.4-­130.3 1.17 ± 0.134 0.81-­1.46

PB175 | Correlation between the Half-­life of Methods: Male patients (adult/adolescent [≥12 years]; children [0-­
11 years]) with severe haemophilia A (< 1% FVIII) receiving N8-­GP
GlycoPEGylated Recombinant FVIII (N8-­GP) and
at 50 IU/kg underwent pharmacokinetic (PK) evaluations with N8-­
von Willebrand Factor -­Results from Phase 3
GP and their previous standard FVIII (in children only). vWF antigen
Clinical Trials levels (assessed using immunoturbidimetric methodology or ELISA
1 2 3 4
P. Chowdary ; M. Carcao ; W.-H.O. Clausen ; P.A. Holme ; J. in a central laboratory) and N8-­GP half-­life were analysed from all
Møss3; A. Tosetto5; A. Wheeler6; E. Santagostino7 patients.
Royal Free Hospital, The Katherine Dormandy Haemophilia Centre and Results: In total, 42 adults and 25 children with 129 PK profiles
Thrombosis Unit, London, United Kingdom, 2Hospital for Sick Children,
(25 previous product profiles in children; 104 N8-­G P profiles
University of Toronto, Division of Haematology/Oncology and Child Health
Evaluative Sciences, Research Institute, Toronto, Canada, 3 Novo Nordisk in 42 adults/adolescents and 24 children) were evaluated. In all
A/S, Bagsværd, Denmark, 4University of Oslo, Oslo University Hospital, ages, N8-­G P half-­life increased with vWF levels, although this
Department of Haematology, Institute of Clinical Medicine, Oslo, Norway,
trend was less pronounced in children. For children, the relation-
San Bortolo Hospital, Hematology Department, Vicenza, Italy, 6 Vanderbilt
University School of Medicine, Pathology, Microbiology and Immunology, ship between half-­life and vWF levels did not differ between N8-­
Nashville, United States, 7IRCCS Cà Granda Foundation, Maggiore Hospital GP and the patients’ previous products. Higher vWF activity also
Policlinic, Angelo Bianchi Bonomi Hemophilia and Thrombosis Center, correlated with blood type; type O had a lower range of activity
Milan, Italy
versus type A.
Conclusions: The half-­life of N8-­G P increased with vWF levels in
Background: Standard factor VIII (FVIII) products are associated
adults/adolescents, similar to previous observations for stand-
with a relatively short half-­life (~12 hours in adults; ~8-­10 hours in
ard and endogenous FVIII. This trend was less pronounced in
children). The half-­life of standard FVIII (plasma-­derived and recom-
binant FVIII [rFVIII]) products is limited by the tight non-­covalent
association of FVIII in circulation with von Willebrand factor (vWF).
Consequently, the half-­life of such products appears to be driven
by the patients’ endogenous vWF levels. N8-­GP (turoctocog alfa
pegol) is an extended half-­life glycoPEGylated rFVIII being devel-
oped for prophylaxis and the treatment of bleeds in patients with
haemophilia A.
Aims: Determine the relationship between N8-­GP half-­life and vWF
levels in patients who received N8-­GP prophylaxis during the path-
finder™ phase 3 clinical trials, and investigate if this relationship was
similar to that of standard FVIII in children.

PB176 | Chronic Pain in Adults with Severe PB177 | Disease Burden and Remaining Unmet
Haemophilia A and B Need in Patients with Hemophilia A Treated with
J. O’Hara1,2; P. Kritikou3; A. Willemze3; K.-J. Myrén3; D. Primary Prophylaxis
Noone2; K. Khair4,5 J. O’Hara1; F. Castro2; S. O’Hara1; J. Black 2; S. Chaplin1;
L. Ruiz-Casas1; R. Hampton2; C. Sima3
University of Chester, Chester, United Kingdom, 2HCD Economics, Daresbury,
United Kingdom, 3Swedish Orphan Biovitrum AB, Stockholm, Sweden, 4London 1
HCD Economics, Chester, United Kingdom, 2Roche, Basel, Switzerland,
South Bank University, London, United Kingdom, 5Great Ormond Street Hospital, 3
Genentech Inc., RWD Oncology, South San Francisco, United States
London, United Kingdom

Background: The gold standard in severe haemophilia A (HA) is pre-

Background: Chronic pain, primarily as a result of bleeding into a
vention of bleeding using prophylactic treatment initiated at an early
joint or chronic joint damage, is a burden for people with haemo-
age (primary prophylaxis, PPX). Patients (pts) treated with PPX (PPX
philia (PwH), leading to impaired quality of life (QoL).
pts) have improved outcomes, however the remaining disease bur-
Aims: Our aim was to describe how key variables affect the pres-
den has not been reported.
ence and level of severity of chronic pain as assessed by physicians in
Aims: To describe the disease burden in PPX pts with HA. To com-
PwH. A secondary objective was to describe the QoL for PwH with
pare health status, clinical and economic outcomes with those in the
different severities of chronic pain.
general population.
Methods: Data was derived from the “Cost of Haemophilia: A
Methods: Data source was Cost of Haemophilia in Europe: a
Socioeconomic Survey (CHESS)”, on adults with severe haemo-
Socioeconomic Survey (CHESS), a EU5 study of adults with HA. PPX
philia in 5 European countries. Physician-­reported chronic pain
pts have always received prophylaxis, were aged 18-­35 years and
was analysed by presence (yes/no) and by severity level (no pain,
had no history of inhibitors (n=166). Clinical outcomes (presence of
mild, moderate, or severe; based on its interference with occupa-
target joints, chronic pain related to HA), health status (EQ-­5D-­3L)
tion or activities of daily living and the use of analgesic medica-
and economic outcomes (work impairment) were analysed. Several
tion). Presence of target joints (TJs) (defined as locations of chronic
populations were used for comparison: healthy male general popula-
synovitis), and bleeds in the last 12 months were used as variables.
tion from two publicly available data sources, for health status1 and
Patient-­reported QoL, was based on the EQ-­5D. Analyses were
work impairment 2; and comparable non-­PPX patients in CHESS, for
descriptive and results presented as mean (SD) / n (%).
clinical outcomes.
Results: In total, 824 (64%) PwH had chronic pain, with a mean age
Results: 39% of the PPX pts had ≥1 target joint (TJ) and 46% had
of 38.1 (15.2) yr; while 461 (34%) PwH had no pain (mean age of 32.0
some level of chronic pain (8% had at least moderate pain). In com-
[13.0] yr). The number of TJs was 0.5 (0.8) for PwH with no pain,
parison, 62% and 59% of the non-­PPX pts had ≥1 TJ and chronic pain,
increasing to 1.3 (1.3) for PwH with mild pain, 1.6 (1.4) for PwH with
respectively. Health status, as measured by EQ-­5D-­3L index score,
moderate pain, and 3.4 (2.7) for PwH with severe pain. Almost all
was worse in PPX pts compared to the male general population of
(94%) PwH with >2 TJs had chronic pain. Bleeds in the last 12 months
the same age (0.87 vs 0.93), with PPX pts reporting more frequent
ranged from 2.9 (5.7) for PwH with no pain to 10.4 (12.0) for PwH
anxiety/depression (31% vs 14%), self-­care problems (5% vs 1%) and
with severe chronic pain. The EQ-­5D utility scores ranged from 0.94
mobility problems (20% vs 6%). (Table 1) Activity and work impair-
(0.1) for PwH with no pain, to 0.21 (0.4), for PwH with severe chronic
ment analyses (Table 2) show that employment level of PPX pts is
pain, >70% of which reported they had ‘moderate or extreme prob-
lower than that in the general population, with higher level of health-­
lems’ across all domains.
related impairment while working.
Conclusions: Results show chronic pain in adults with severe hae-
mophilia represents a burden for PwH and an unmet medical need. TABLE 1 Health status in PPX pts and healthy male general population
Increased pain severity reported by the physicians correlates with
PPX pats (CHESS) General population,
presence of TJs and bleeds. Chronic pain is associated with de- Outcome n=55 Males (EU5) n=2,570
creased QoL across all domains.
Mean EQ-­5D index 0.87 0.93
EQ-­5D-­3L domain analysis (% pats with a moderate/severe problem)
 Anxiety/Dep 31% 8% (6%-­25%)
  Self-­c are 5% 2% (0.2%-­4%)
 Activities 9% 6% (4%-­15%)
 Pain & Disc 38% 29% (17%-­39%)
 Mobility 20% 5% (3%-­12%)

TABLE 2 Economic outcomes in PPX pts and healthy male general and Inv22-­2), accounts for about 45% of severe HA worldwide. The
populations inversion of intron 1 (Inv1) is the second recurrent F8 disruption
among severe patients, with about 4.5% of prevalence.
US general Aims: Few studies have evaluated the prevalence of the types of
PPX pts, population,
Inv22, whether type 1 or type 2. Therefore, we aimed to report the
18-­35 yo >18 yo
Outcome (N=55) (N=65,389) prevalence of these inversions in a cohort of patients with HA.
Methods: We included 103 severe (FVIII< 1%) HA patients from the
Pts Employed 44% 50.79%
state of Minas Gerais/Brazil, focusing on the genetic frequencies of
Worktime hours missed (absenteeism) 0% 3.5%
Inv1, Inv22-­1 and Inv22-­2. Inv 1 was detected using PCR, and intron
Percent impairment while working, 25.6% 12.9% 22 inversions of F8 were sorted by inverse-­shifting PCR. A combi-
due to health (presenteeism)
nation of different product sizes in 2% agarose gel electrophoresis
Percent overall work impairment due 16.2% 15%
predicted the presence or absence of the mutations.
to health (including missed work and
impairment while working) Results: When all patients are considered, Inv1, Inv22-­1 and Inv22-­2

Percent activity impairment (non-­ 23.1% 22.1%

were present in 5.8% (6/103), 26.2% (27/103) and 5.8% (6/103), respec-
work) due to health tively (Figure 1). A total of 37/103 patients (35.9%) developed inhibitor
(inh+) and 66 (64.1%) did not after 75 exposure days (ED) to FVIII (inh-­).
Conclusions: While PPX has been shown to provide superior efficacy When patients were stratified by inh+ and inh-­groups, Inv1, Inv22-­1
in the HA population compared to other treatment approaches, PPX and Inv22-­2 were present in 13.5% (5/37), 29.7% (11/37), 8.1% (3/37)
pts appear to maintain a degree of joint health impairment, worse and 1.5% (1/66), 24.2% (16/66) and 4.5% (3/66), respectively.
health status, and employment and productivity loss, compared to Conclusions: We conclude that Inv22-­
1 is more prevalent than
the general population. Inv22-­2 in patients with and without inhibitors. Although less preva-
lent, Inv1 accounted for nearly 15% of genetic defects in patients
with inhibitors.

PB178 | Landscape of Intron 1 and Intron 22

Inversions in Severe Hemophilia A Patients and
Inhibitor Development PB179 | The Bone Disease in Haemophilia: The
Role of von Willebrand Factor, Factor VIII and
L.W. Zuccherato1; S. Maria Eloi Santos2; L.L. Jardim3; D.G.
Chaves4; S.M. Rezende3 Thrombin
S. Lancellotti1; M. Sacco2; G. Battafarano3; F. Berruti2; R. De
Universidade Federal de Minas Gerais, Complementary Propaedeutic
Department, Belo Horizonte, Brazil, 2Universidade Federal de Minas Gerais, Cristofaro1,2; A. Del Fattore3
Complementary Propaedeutic Department, BELO HORIZONTE, Brazil, 1
3 Fondazione Policlinico Universitario A. Gemelli, Area Ematologia -­Servizio
Universidade Federal de Minas Gerais, Belo Horizonte, Brazil, 4Fundação
Malattie Emorragiche e Trombotiche, Roma, Italy, 2Università Cattolica del Sacro
Hemominas, Belo Horizonte, Brazil
Cuore, Roma, Italy, 3Bambino Gesù Children’s Hospital, Bone Physiopathology
Group, Multifactorial Disease and Complex Phenotype Research Area, Roma, Italy
Background: Hemophilia A (HA) is an X-­linked disorder due to del-
eterious mutations in the factor VIII (FVIII) gene (F8). A recurrent Background: Reduced bone mineral density (BMD) is a significant
inversion, originated by a homologous recombination event within finding in haemophilia A (HA). However, the pathophysiology is not
F8 intron 22 and its extragenic inversely oriented copies (Inv22-­1 exactly known. BMD seems due to perturbations of the Receptor

F I G U R E 1   Intron 1 and intron 22 inversions (Inv22-­1 and Inv22-­2) of the F8 gene in severe patients with hemophilia A
      87 |
Activator of Nuclear factor-­B (RANK), RANK Ligand (RANKL) and Methods: PTPs with severe hemophilia A (endogenous Factor VIII <
Osteoprotegerin (OPG) pathway. Recent evidence shows that low 1%) and >50 previous exposure days (EDs) to FVIII products were
BMD in HA is due to an increased rate of bone resorption. Moreover, treated either on demand or prophylactically with rVIII-­SingleChain 2
bone cells express protease-­activated receptors (PARs) and FVIII de- or 3 times weekly, or another regimen per the Investigator’s discretion.
ficiency could alter BMD also by decreasing thrombin production. Results: As of 05 Sep 2017, 222 patients were enrolled. The median
Aims: The aim is to identify the specific role played by FVIII, VWF age was 20 years (1-­64); 77 (34.7%) patients were 0-­< 12 years, and
and thrombin on osteoclasts biology. 145 (65.3%) were ≥12 to ≤65 years old. The majority of patients were
Methods: To assess the effects on osteoclastogenesis, blood mon- Caucasian (158, 71.2%); 50 (22.5%) Asian, and 12 (5.4%) were Black or
onuclear cells were stimulated with human macrophage colony African American. A minority of patients (10, 4.5%) were of Hispanic
stimulating factor M-­C SF/RANKL in the presence of the different origin. A target of ≥100 EDs was achieved by 211 (95%) patients; the
coagulation factors (plasma derived VWF/FVIII complex, recombi- total cumulative EDs in the extension study was 65,522 in 494 treat-
nant VWF, recombinant FVIII and thrombin). The activity of cathep- ment years. Of the 211 (95%) patients assigned to prophylaxis, 98
sin K and matrix metalloproteinase (MMP)-­9 was evaluated by real (44.1%) were treated 3 times weekly, and 84 (37.8%) patients were
time expression analysis on mature osteoclasts treated for 48h with treated 2 times weekly. The median annualized spontaneous bleeding
the aforementioned coagulation factors. rate (AsBR) for all prophylaxis regimens, for patients treated 3 times
Results: VWF appears to play a major role, showing by itself ~45% weekly, and 2 times weekly was 0.35 (Q1, Q3: 0.00; 1.09), 0.35 (Q1,Q3:
inhibition of osteoclastogenesis comparable to OPG (the physiologic 0.0; 0.96), and 0.0 (Q1, Q3: 0.0; 1.46) respectively. Of a total of 2273
inhibitor), and even more if is complexed with FVIII (53% inhibi- bleeding events, 2162 were treated with rVIII-­SingleChain. Of these,
tion). Thrombin reduces osteoclast differentiation with variable ef- 2120 were rated for hemostatic efficacy by the investigator: 85.8%
fects (30-­50% inhibition). No statistically significant alterations of were rated as either excellent or good and required 1 or 2 injections to
Cathepsin K and MMP9 expression were revealed in treated mature achieve hemostasis. No PTP developed an inhibitor in the study.
osteoclasts. Conclusions: rVIII-­SingleChain has a favorable safety profile, and is
Conclusions: These findings showed that the osteoclast differentia- effective and well tolerated for on-­demand and prophylaxis treat-
tion is significantly and negatively influenced by VWF/FVIII complex ment in patients who have severe hemophilia A.
and thrombin treatments. By contrast, the resorption activity was not
influenced by the presence of the different factors. These results may
have important implications in the clinical management of hemophilia
PB181 | Standardized Determination of FVIII
patients to better orient the choice of the FVIII concentrate to prevent
Inhibitors with the Improved FVIII INH Kit
bleeding and at the same bone disease in severe HA patients.
L. Rosenmayr; L. Wagner; N.B. Binder
Technoclone Herstellung von Diagnostika und Arzneimitteln GmbH, Vienna,
PB180 | Long Term Safety and Efficacy of rVIII-­ Austria

SingleChain in Previously Treated Patients with

Background: The development of antibodies to Factor VIII:C in the
Severe Hemophilia A: Intermediate Results of an
treatment of patients with haemophilia A is still a problem. The ma-
Extension Study
jority of laboratories continue to perform FVIII inhibitor assays using
J. Mahlangu1; F. Abdul Karim2; O. Stasyshyn3; B. their own in-­house method causing high variability of results.
Korczowski4; A. Brainsky5; S. Lucas5; Y. Li5; I. Pabinger6; for Aims: Aim of the study was to investigate the performance of the FVIII
the AFFINITY Investigators
INH kit, which provides all reagents and instructions and in addition a
University of the Witwatersrand, NHLS and Charlotte Maxeke Hospital,
positive and a negative control, and to correlate the results with those
Johannesburg, South Africa, 2National Blood Centre, Kuala Lumpur, Malaysia,
Institute of Blood Pathology and Transfusion Medicine, National Academy of obtained by the classical Bethesda and by the Nijmegen assay.
Medical Sciences of Ukraine, Lviv, Ukraine, 4Department of Pediatrics, Regional Methods: Haemophilia A patient samples with and without inhibitors
Hospital, University of Rzeszów, Rzeszów, Poland, 5CSL Behring, King of Prussia,
were tested in the classical Bethesda assay, the Nijmegen assay and
United States, 6Clinical Division of Haematology and Haemostaseology, Medical
Clinic I, Medical University Vienna, Vienna, Austria with the FVIII INH assay kit. The FVIII INH kit follows the modified
Nijmegen assay principal and uses buffered normal plasma in a 1+1
Background: rVIII-­SingleChain, a novel B-­domain truncated recom- mixture with patient plasma. The control mix is a 1+1 mixture of imi-
binant Factor VIII comprised of covalently bonded FVIII heavy and dazole buffer with BSA and buffered normal plasma. Patient sample
light chains, was designed to have a high binding affinity to von dilutions are made with imidazole buffer with BSA. After incubation,
Willebrand factor. the residual FVIII activity was determined with the chromogenic assay
Aims: This Phase III multinational extension study investigates the Technochrom FVIII:C and Siron LS for FVIII one stage clotting assay.
long term safety and efficacy of rVIII-­
SingleChain in previously Results: The control values for the positive and negative control
treated patients (PTPs) with severe hemophilia A. were within the confidence range for all methods. No inhibitors were
detected in the Hamophilia A patients without inhibitors.

For Haemophilia A patient samples with inhibitors the results Conclusions: Compared with patients’ previous FVIII prophylaxis,
showed a r = 0.8894 between one stage clotting assay and chromo- Joint ABR was lower with all BAY 94-­9027 prophylaxis regimens, and
genic FVIII assay and a Passing Bablock slope of 1.0949. maintained in the extension study.
The results for the FVIII INH kit showed highest BU/mL, indicating a Study supported by Bayer.
high sensitivity towards FVIII inhibitors. Differences between meth-
ods were sample dependent with SD between 7.8 and 17.8.
Conclusions: The FVIII INH kit can be used with good performance for PB183 | Systematic Review of Efficacy and
determination of FVIII INH in patient samples, assay performance is con-
Factor Consumption of Recombinant Factor
trolled by use of the provided controls. There is a trend towards higher
BUs with FVIII INH kit demonstrating high sensitivity of the assay.
IX Products for Prophylactic Treatment of
Hemophilia B
E. Santagostino1; S. Yan2; T. Matsushita3; L. Alberio 4; J.
PB182 | Decrease in Joint Bleeding Rates
Maggiore Hospital Policlinico, Angelo Bianchi Bonomi Hemophilia and
Compared with Prior Prophylaxis in Patients with Thrombosis Center IRCCS Ca’ Granda Foundation, Milan, Italy, 2CSL Behring,
Hemophilia A Treated with BAY 94-­9027 King of Prussia, United States, 3Nagoya University Hospital, Department
of Transfusion Medicine, Nagoya, Japan, 4CHUV, Department and Central
M.E. Mancuso1; S. Rangarajan2; C. Négrier3; M. Wang4; M. Hematology Laboratory, Lausanne, Switzerland, 5University of Miami Hemophilia
Maas Enriquez5 Treatment Center, Miami, United States
Fondazione IRCCS Ca’ Granda -­Ospedale Maggiore Policlinico, ‘Angelo Bianchi
Bonomi’ Hemophilia and Thrombosis Centre, Milan, Italy, 2University Hospital Background: No direct head-­to-­head study has been conducted to
Southampton NHS Foundation Trust, Southampton, United Kingdom, 3Louis
Pradel University Hospital, University Claude Bernard Lyon 1, Lyon, France,
compare efficacy and safety of recombinant factor IX (rFIX) prod-
Bayer, US Innovation Center, Hematology Research, San Francisco, United ucts in hemophilia B.
States, 5Bayer, Wuppertal, Germany Aims: To systematically review the evidence from Phase III clinical
trials evaluating the use of rFIX products for prophylaxis in hemo-
Background: In hemophilia A, prevention of joint bleeding is a key philia B patients. The outcomes considered were total annualized
aim of prophylactic FVIII treatment; higher FVIII levels protect bleeding rate (ABR), spontaneous ABR (AsBR), joint ABR (AjBR), and
against joint bleeding. BAY 94-­9027, an extended half-­life recom- consumption of rFIX.
binant FVIII, was evaluated in the phase II/III PROTECT VIII study. Methods: A systematic literature search was conducted in both
Aims: To compare patients’ joint annualized bleeding rate (JABR) EMBASE and PubMed. Phase III clinical trials of prophylactic rFIX
during BAY 94-­9027 prophylaxis in the PROTECT VIII study with treatment in previously treated, hemophilia B patients aged ≥12
JABR on a prior FVIII prophylaxis regimen. years, with endogenous FIX levels ≤2% were included. Studies which
Methods: PROTECT VIII included previously treated males 12-­65 years did not meet these criteria, or did not report any of the outcomes
old with severe hemophilia A who had received FVIII in the 12 months mentioned above, were excluded. Relevant data were then ex-
pre study (‘prestudy’). Prophylaxis patients received BAY 94-­9027 twice tracted from these studies.
weekly (2×/wk; 30‒40 IU/kg), or every 5 (E5D; 45‒60 IU/kg) or 7 days Results: The search identified a total of 1,362 articles, with 35 pass-
(E7D; 60 IU/kg) for 26 weeks. Bleeds were documented in electronic ing the title and abstract screen. Following full-­text review, 9 articles
patient diaries. This post-­hoc analysis (data cut Feb 2017) included met the inclusion criteria. Data has been extracted from these ar-
patients who received prestudy prophylaxis and continued in an open-­ ticles and summarized in Tables 1 & 2 (data marked * was obtained
label extension, wherein they could maintain or switch regimen (‘vari- from the clinical study report); information is presented for the
able’ [VAR] patients). Prestudy, main study and extension JABR were weekly infusion regimen where data was available for more than one
analyzed and compared per the intention to treat extension population. prophylaxis schedule. ABR reported in the relevant studies ranged
Results: In total, 82 patients received prestudy prophylaxis and en- from a median of 0-­3.0, AsBR ranged from 0-­1.0, and AjBR ranged
tered the extension: the majority were treated ≥ 3×/week prestudy. from 0-­1.1. The median values reported for rIX-­FP were 0 for ABR,
Median (range) follow up in the main study and extension was 3.9 AsBR and AjBR. Mean values ranged from 1.24-­4.26 for ABR, 0.52-­
(0.8-­4.7) years (n=121, including the 82 analyzed here). Median JABR 2.6 for AsBR, and 0.89-­2.85 for AjBR, with rIX-­FP at the lowest end
was lower in the main study versus prestudy (1.41 vs. 2.00; decrease for each value. Weekly rFIX consumption ranged from 44.5-­105.5
of −0.57). Decreases by group were −1.00 (2x/wk; n=13), −0.81 IU/kg.
(E5D; n=30), −0.05 (E7D; n=20) and −0.58 (VAR; n=19). Low JABR Conclusions: This review identified suitable data for an indirect
was maintained in the extension (median 1.04) with a further de- comparison of rFIX products. Within the limitations of potential
crease in the 2×/wk arm (extension vs. main study −2.67). Reliability differences in patient population and study design, a formal meta-­
is supported by consistent median JABR in on-­demand PROTECT analysis may help to understand how these products compare with
VIII patients (prestudy, 19.50; main study 17.87; extension, 20.49 respect to the parameters of ABR, AsBR, AjBR, and consumption
[n=14]). of rFIX.

TA B L E   1   Details of bleeding rates in rFIX studies identified by the systematic review

No. ABR, AsBR, AjBR,

prophylaxis Age, mean median ABR, mean median AsBR, median AjBR,
Study and regimen rFIX product subjects years (range) (IQR) (SD) (IQR) mean (SD) (IQR) mean (SD)

Santagostino et al. Albutrepenonacog 40 31.6 (SD 15.2) 0 (0, 1.87) 1.24 0 (0, 0) 0.52 0 (0, 1.53) 0.89
2016 (35-­50 IU/kg alfa (rIX-­FP) (1.780)* (1.116)* (1.436)*
Powell et al. 2013 rFIXFc INN 63 Median (range): 3.0 Not 1.0 Not 1.1 Not
(50 IU/kg weekly 28 (12-­71) (1.0-­4.4) reported (0.0-­2.2) reported (0.0-­4.0) reported
Lambert et al. 2007 Nonacog alfa 17 28.32 (12-­61) Not 3.11 Not 0.72 Not Not
(regimen not [for all 34 reported reported reported reported
reported) patients in
Kavakli et al. 2016 Nonacog alfa 25 31.3 2.0 (0, 3.6 (4.6) 1.0 (0, 2.6 (4.1) 0.0 (0, 9.8) 2.1 (3.2)
(100 IU/kg weekly (12.1-­53.7) 13.8) 13.8)
Windyga et al. 2014 BAX326 INN 56 Median (range): 1.99 (0, 4.26 (5.80) 0.0 (0, 1.72 0.0 (0, 2.85
(40-­75 IU/kg 33 (12-­59) [for 23.4) 15.6) (3.26) 21.5) (4.25)
twice-­weekly all 73 patients
prophylaxis) in study]
Collins et al. 2014 Nonacog beta 29 30.0 (SD 15.8) 1.04 Not 0.0 Not Not Not
(40 IU/kg weekly pegol (N9-­GP) (0.00-­ reported (0.0-­ reported reported reported
prophylaxis) 4.00) 0.98)
Young et al. 2016 Nonacog beta 52 32 (SD 14.2) 1.00 Not 0.00 Not 0.00 Not
(40 IU/kg weekly pegol (N9-­GP) (0.00-­ reported (0.00-­ reported (0.00-­ reported
prophylaxis) 2.03) 1.00) 1.97)
Collins et al. 2017 Trenonacog alfa 58 28.8 (SD 14.2) 1.52 3.55 (SD Not Not Not Not
(50-­75 IU/kg (IB1001) (0.00-­ 7.15) reported reported reported reported
twice-­weekly 3.46)
Solano Trujillo et al. BAX326 INN 32 31.9 (14-­55) 0.9 (0, 2.3 (3.7) 0.0 (0, 10) 0.9 (1.95) 0.0 (0, 10) 1.5 (2.66)
2014 (20-­4 0 IU/kg 11.2)

TA B L E   2   Details of consumption in rFIX studies identified by the systematic review

Consumption (IU/kg/week),
Study rFIX product No. prophylaxis subjects mean

Santagostino et al. 2016 (weekly Albutrepenonacog alfa (rIX-­FP) 40 Median: 44.9 (calculated
prophylaxis group) from 194.7 IU/kg/month)
Powell et al. 2013 (weekly prophylaxis rFIXFc INN 63 Median: 45.2
Lambert et al. 2007 Nonacog alfa 17 120.8 (Iorio et al., 2017)
Kavakli et al. 2016 Nonacog alfa 25 95.9
Windyga et al. 2014 BAX326 INN 56 89.1
Collins et al. 2014 (40 IU/kg weekly Nonacog beta pegol (N9-­GP) 29 Not reported
prophylaxis group)
Young et al. 2016 (40 IU/kg weekly Nonacog beta pegol (N9-­GP) 52 44.5
prophylaxis group)
Collins et al. 2017 Trenonacog alfa (IB1001) 58 105.5
Solano Trujillo et al. 2014 BAX326 INN 32 Not reported
90       |

PB184 | Comparing Projected Consumption PB185 | Joint and Functional Assessment

and Cost of Recombinant Factor VIII-­SingleChain Following Secondary and Tertiary Prophylaxis
with Other Recombinant Factor VIII Products in in PWH who were Previously Receiving
the United States Episodic on Demand Factor Replacement
S. Yan1; R. Hutcheson2; L. Stern3; G. Krishnarajah1 Infrequently
CSL Behring, King of Prussia, United States, 2Analytica Laser, Montréal, Canada, A. Dutta; S. Kakati; H. Kaur; S. Kar
Analytica Laser, New York, United States
Assam Medical College and Hospital, Department of Medicine, Dibrugarh, India

Background: rVIII-­SingleChain (SC) is indicated in adults and chil-

Background: Hemophilia is an X-­
linked recessive disease which
dren with hemophilia A (congenital Factor VIII deficiency) for control
results in the deficiency of coagulation factors and affects one in
and prevention of bleeding episodes, routine prophylaxis to prevent
10000 births. The hallmark of hemophiliac is musculoskeletal out-
or reduce the frequency of bleeding episodes, and for perioperative
come. Though Primary prophylaxis can prevent joint damage, most
of the PWH from north east India receive infrequent episodic on
Aims: To compare prophylactic product consumption and cost of SC
demand factor replacement.
versus recombinant factor VIII Fc fusion protein (FC), rurioctocog
Aims: We have studied the role of secondary and tertiary prophy-
alfa pegol (PEG) and octocog alfa (OA), for patients ≥ 12 years of age,
laxis in forty seven patients with hemophilia
from the perspective of US payers.
Methods: Average monthly consumption of product per patient
(IU/kg) was calculated and compared between SC versus FC, PEG
and OA, based on a previously-­d eveloped budget impact model.
Dosing frequency and dosage were based on product prescrib-
ing information (PI), and midpoint value was used where a range
is provided in the PI. Median annualized bleeding rates were
based on PI information and other published data from clinical
trials of the products. Average monthly costs based on current
wholesale acquisition costs (WAC) of the products were also
Results: Average monthly consumption of SC is projected to be
similar or slightly lower than FC and PEG, and lower by > 15% than
OA, for prophylaxis with or without bleeding treatment (Table 1).
F I G U R E 1   HJHS in PWH on infrequent ondemand therapy from
Based on the consumption and WACs, average monthly cost of
Upper Assam region of INDIA
SC per patient is lower than FC, PEG and OA by ~28%, ~24%
and ~11%, respectively, for prophylaxis with or without bleeding

TA B L E 1   Monthly consumption for prophylaxis (IU/per kg), per

patient age ≥ 12

Treatment of
Regular prophylaxis bleeds

SC 379.17 3.85
FC 379.17 5.09
PEG 390.00 5.42
OA 455.00 4.65

Conclusions: While SC showed long lasting hemostatic efficacy in

clinical trials, it may also reduce consumption and cost for US pay-
ers, compared with the extended half-­life products FC, PEG, and the F I G U R E 2   Functional Assesment using FISH scores compared
commonly used standard-­acting OA. between mild moderate and sever PWH on ondemand factor
replacement from Upper Assam region of INDIA

(39 hemophilia A and 8 hemophilia B) attending AMCH who were 1) elaboration of the first version of the instrument, based on
previously receiving episodic treatment. systematic and random observation of 1,800 NC;
Methods: It was a non interventional observational study con- 2) analysis by a multiprofessional outpatient team of hematology
ducted for 3 months, approved by the ethical committee of AMCH. specialists, and later,
Informed consent was signed. Patients were stratified into four
groups, according to age: 3) analysis by 20 nurse-judges with experience in attending to PwH
Group I (1-15 years of age), and 9 national reference hemophilia nurses (RHN), by the Delphi
Group II (15-30 years of age), method. The agreement among nurses was assessed by Fisher′s
Group III (30-45 years of age), Exact test to evaluate the degree of relevance of the characteris-
Group IV (45-60 years of age). tics and the degree of satisfaction about the items of appearance.
The HJHS and FISH was correlated with the age of the pa- The items of at least 70% agreement were considered valid. The
tients and also with the severity of bleeding. One way Analysis study was approved by HEMOPE ethics committee (CEP-
of Variance (ANOVA) was used to determine the differences in HEMOPE n.1.863.411/CAAE 61155316.5.0000.5195) and in-
the mean of the different groups of hemophiliac. SPSS statistical formed consent was obtained from the professionals involved.
software was used and statistical significance was defined as a
p-­v alue < 0.05. Results: The nurses considered all the characteristics “relevant” or
Results: Mean age was 18.02± 10.28 years. 23 (48.9%) were below “extremely relevant” (P>0.05). Regarding the degree of appearance
15 years, 19 (40.4%) were from 16 to 30 years, 4 (8.5%) were from satisfaction, the RHN requested the instrument to be used for other
31 to 45 years and 1(2.1%) was above 45 years. At the time of en- coagulopathies as well, turning the homogeneity test significant
rolment mean total HJHS score for the six main joints and gait was for the characteristic clarity of the statements (P=0.03). The final
42.133 (SD 22.138, median 39) out of which maximum was 124. version was composed of 17 parts, including Treatment, Inhibitor
The most affected joints were knees, followed by elbows and ankles. Status, Independence Degree, Infusion Log and Immune Tolerance.
The mean of total ankle score were significantly lower compared Conclusions: The instrument was considered adequate to the systema-
to total knees score. There was no significant difference between tization of the NC to the PwH, facilitating the identification of patient
HJHS score of mild, moderate and severe PWH. FISH score was high problems. It has been inserted into the service electronic medical record.
in episodic treatment and improved.
Conclusions: There was improvement in HJHS score but it was not
significant in elder patients. Functional assessment also improved
PB187 | Epidemiological Survey and Audit of
after four months of treatment.
Socio-­demographic Characteristics of Patients
with Haemophilia and other Congenital Bleeding
PB186 | Validation of an Instrument for
T. Nwagha1; M. Adediran2; H. Okoye3
Nursing Consultation to the Person with
University of Nigeria Teaching Hospital, Haematology & Immunology, Enugu,
Hemophilia Nigeria, 2Haemophilia Foundation of Nigeria, Kaduna, Nigeria, 3University of
I. Sangi1,2; R. Camelo1; I. Costa1; N. Costa1; F. Lima2; T. Nigeria Teaching Hospital, Haematology & Immunology, Ituku Ozalla, Nigeria

Foundation of Hematology and Hemotherapy of Pernambuco (HEMOPE), Recife, Background: Despite more than 50 years since the first diagnosis of
Brazil, 2Federal University of Pernambuco, Recife, Brazil haemophilia was made in Nigeria, congenital bleeding disorders are
still considered as disease of medical textbooks. There is increas-
Background: Hemophilia is an X-­linked recessive coagulopathy ing evidence of high disease prevalence, disease related morbidity
caused by the deficiency of factors VIII (hemophilia A) or IX and mortality. With WFH′s effort to “bridge the gap” and encourage
(hemophilia B). Brazilian Ministry of Health reinforced promot- individual country government commitment towards these disease
ing health through multiprofessional approach to persons with management Early identification, diagnosis and treatment are key in
hemophilia (PwH). In this context, nursing consultation (NC) to reducing these statistics.
PwH is recommended, offering advantages in treatment and Aims: To study the epidemiology and socio-­demographic variables
adherence. of persons with haemophilia and other congenital bleeding disorders
Aims: This study aimed at constructing and validating an instrument in Nigeria.
for NC to PwH, at the Ambulatory of Coagulopathies of HEMOPE Methods: This was a 12 year retrospective audit of congenital bleed-
(Recife, Brazil). ing data registry of the Haemophilia Foundation of Nigeria (HFN)
Methods: The qualitative cross-­sectional study consisted of between 2005-­2017 Data on age, sex, geographical location of re-
ferring HTC or hospital, type of congenital disorder, clinical severity
type and inhibitor status of persons living with congenital bleeding

disorders were obtained from the HFN registry. Data were analysed assessments met the BMV guidance acceptance criteria with repro-
using the windows SPSS version 20. ducible accuracy, precision, and sensitivity across a reportable range
Results: A total of three hundred and fifty-­one registered with HFN, of 0.01 to 2.5 IU/mL.
350 (99.7%) were males, and only1(0.03%) female. The mean age, SD Conclusions: A method was developed and validated to measure B-­
and range were 15 (11) years, and 4 months-­66years respectively. domain deleted FVIII antigen in human plasma that met the BMV
Most registered persons were from tthe northwest geopolitical guidance’s acceptance criteria for all validation assessments with
zone 40.1% (135/351). Most had haemophilia A, 92.6% (325/351), reproducible recovery across the reportable range.
of severe clinical severity 13.7% (48/351) with only 1 (0.03%) docu-
mented with inhibitors. There was also only 1 documented congeni-
tal platelet disorder (Glanzmann thrombasthenia 0.03%. There was a
PB189 | Exploring Regional Variations in the
significant association between both geopolitical zone, referral HTC
/ hospital and the type of bleeding disorder χ2 292.3, P value 0.000 Cross-­cultural, International Implementation of
and χ2270.4, P value 0.000 respectively. the Patient Reported Outcomes Burdens and
Conclusions: The overall low number of persons registered may be Experience (PROBE) Study
from high disease burden. There is need for government driven pub-
C. Chai-Adisaksopha1; D. Noone2; R. Curtis3; N. Frick4; M.
lic awareness programme as well as its commitment to improving Nichol5; B. O’Mahony2,6; D. Page7; J. Stonebraker8; A. Iorio1;
capacity in diagnosis and treatment. M.W. Skinner9; PROBE
McMaster University, Hamilton, Canada, 2Irish Haemophilia Society, Dublin,
Ireland, 3Factor VIII Computing, Berkeley, United States, 4National Hemophilia
Foundation, New York, United States, 5University of Southern California, Sol
PB188 | Development of a FVIII Antigen Assay Price School of Public Policy, Los Angeles, United States, 6Trinity College, Dublin,
Ireland, 7Canadian Hemophilia Society, Montreal, Canada, 8Poole College of
to Quantify B-­domain Deleted FVIII Antigen in
Management, North Carolina State University, Raleigh, United States, 9Institute
Human Plasma for Policy Advancement Ltd., Washington, United States

M. Robinson; C. Tudan; S. Tiefenbacher

Colorado Coagulation, Laboratory Corporation of America Holdings, Englewood, Background: The Patient Reported Outcomes Burdens and
United States Experience (PROBE) study has developed and validated the multilin-
gual PROBE questionnaire for assessing patient reported outcomes
Background: With the increased use of B-­d omain deleted FVIII in people living with hemophilia (PWH) and participants without
constructs in AAV mediated FVIII gene therapies; there is a bleeding disorders (NOBD). The PROBE questionnaire translations
need to identify FVIII antigen assays that can adequately meas- into local languages account for site specific language and cultural
ure B-­d omain deleted FVIII in human plasma. Evaluations of the peculiarities in the management and life experience of hemophilia.
d omain deleted rFVIII ReFacto AF® in two commonly used
B-­ Aims: To explore the regional variations in the international imple-
FVIII antigen assays in the US, Visualize™ FVIII antigen (Affinity mentation of the PROBE questionnaire and examine the hypoth-
Biologicals™ Inc.) and Asserachrom® VIII antigen (Diagnostica esis that the PROBE score will not be confounded by cross-­cultural
Stago), demonstrated significant recovery issues, with under-­ aspects.
recovery at concentrations ≥ 0.1 IU/mL and over-­recovery at con- Methods: Data were collected from participants in four regions
centrations < 0.1 IU/mL. (Western Pacific, South America, North America and Europe).
Aims: To develop and validate a FVIII antigen assay to measure B-­ Participants were able to select from 23 localized language versions
domain deleted FVIII in human plasma. of the PROBE questionnaire based on their first language. We used
Methods: A FVIII antigen ELISA assay that uses two monoclonal an- analysis of variance methods and multivariable regression to deter-
tibodies for capture and detection with specificity for the A2 domain mine the relative contribution of the variance explained by region
epitope of FVIII was developed and validated to meet the EMA and controlling for hemophilia diagnosis, age group and level of educa-
draft FDA bioanalytical method validation (BMV) guidances. The tion. We also explored interactions between region and the other
assay was calibrated using a ReFacto AF®/Xyntha® reference ma- components.
terial to obtain a calibration range of 0.01 to 1.0 IU/mL. The assay Results: We analyzed 862 questionnaires from 21 countries. Mean
conditions were optimized to minimize von Willebrand factor (vWF) age of participants was 40.03 years (SD 13.89) and 65.20% were
interference while providing consistent and acceptable recovery PWH and 34.80% were NOBD. After adjusting for hemophilia diag-
across the reportable range. Accuracy, precision, dilution linearity nosis, age group and level of education, region contributed 0.44%
and sensitivity were assessed using congenital FVIII deficient plasma to 7.98% of the variance component in sub-­item scores and 0.26%
samples spiked with ReFacto AF®/Xyntha®. in the PROBE score. Years of education contributed 0.34% in the
Results: Titrations of vWF and pre-­incubation with βME were used PROBE score. Age and diagnosis (NOBD, mild, moderate, severe)
to optimize the final assay conditions to reduce steric interfer- contributed 3.42% and 22.42% of the PROBE score. 70.74% of the
ence between the analyte and the assay antibodies. All validation variance was explained by inter-­individual variation.

Conclusions: Variance partitioning for the PROBE score is similar to most critical point, PwHA would recommend a friend with haemo-
that for EQ5D. The results demonstrate that the PROBE question- philia to participate in ECHO (med 95). Five PwH suggested changes
naire is valid to implement for assessing health status among PWH in the survey (e.g. eliminate repetitive questions).
and participants without bleeding disorders across regions. Conclusions: The ECHO PROs selection indicated feasibility.
Additional areas suggested by PWH were not added due to investi-
gator concerns the overall burden would increase and inhibit recruit-

PB190 | Expanding Communications on ment and PRO completion.

Hemophilia A Outcomes (ECHO) Study: Choice

of Patient-­reported Outcomes (PROs) and
Feasibility Testing PB192 | Hemophilia A Subjects with an
Intron-­22 Inversion Mutation Show Robust T-­cell
S. von Mackensen1; C.R.M. Hay2; M. Shima3; M. Makris4; V.
Jiménez-Yuste5; J. Oldenburg6; K. Fischer7; E. Santagostino8; Responses to Epitopes in Both the FVIII Heavy
S. Rauchensteiner9; N. Church10; P. Mathew10; C.M. and Light Chain
Kessler11; M.W. Skinner12; on behalf of the ECHO Study
Investigators D. Gunasekera1; A.F. Karim2; M.V. Ragni3; K.P. Pratt2
1 Uniformed Services University, Medicine, Bethesda, United States, 2Uniformed
University Medical Centre Hamburg-­Eppendorf, Hamburg, Germany,
2 Services University, Medicine MED, Bethesda, United States, 3University of
Manchester University Department of Haematology, Manchester, United
Pittsburgh, Medicine, Pittsburgh, United States
Kingdom, 3Nara Medical University, Kashihara, Japan, 4Department of Infection,
Immunity, and Cardiovascular Disease, University of Sheffield, Sheffield, United
Kingdom, 5Autónoma University Madrid and La Paz University Hospital, Madrid, Background: Almost half of severe hemophilia A (HA) is caused
Spain, 6Department of Experimental Haematology and Transfusion Medicine,
by an inversion mutation in intron 22 of the 26-­exon F8 gene
University Clinic Bonn, Bonn, Germany, 7Van Creveldkliniek, University Medical
Center Utrecht, Utrecht, the Netherlands, 8IRCCS Ca’ Granda Foundation, encoding factor VIII (FVIII). FVIII contains homologous domains
Maggiore Hospital Policlinico, Milan, Italy, 9Bayer, Basel, Switzerland, 10Bayer, A1-­A 2-­B-­A3-­C1-­C 2 and circulates as a heterodimer consisting of
Whippany, NJ, United States, 11Georgetown University Medical Center,
a heavy chain (A1-­A 2-­B) and light chain (A3-­C1-­C 2). The intron-
Washington, DC, United States, 12Institute for Policy Advancement, Ltd,
Washington, DC, United States ­22 inversion precludes the possibility of a full-­length F8 mRNA
being produced; instead, a partial F8 mRNA containing exons 1-­22

Background: The Expanding Communication on Hemophilia A is transcribed. A shorter F8 transcript containing exons 23-­26,

Outcomes (ECHO) study was designed as a 5-­year international, termed F8B, is transcribed from a promoter within intron 22 in

observational registry in adults with moderate/severe hemophilia both non-­h emophilic and hemophilic individuals. F8B encodes a

A (PwHA) exploring associations of different variables on patient-­ C-­terminal C1 domain RNA sequence linked in-­frame to the C2

reported (PRO) and clinical outcomes. In qualitative focus groups, domain sequence. FVIII protein is not detectable in plasma of pa-

10 aspects were identified that were considered important by PwH tients with an intron-­22 inversion.

(n=45), resulting in an inclusion of 9 standardized PRO measures and Aims: To test the hypothesis that putative low-­level intracellular syn-

several ad-­hoc questions concerning socio-­demographic and clinical thesis of partial FVIII proteins encoded by the two partial F8 mRNA

data, health behavior and resource utilization. molecules, if this occurs, confers immune tolerance to all therapeutic

Aims: Testing the feasibility of the complex survey for PwHA con- FVIII regions except a short C1 domain sequence corresponding to

cerning ease and time of completion, number of questions and rel- the inversion break-­point.

evance of aspects. Methods: Cellular immune responses to FVIII were queried using

Methods: A feasibility questionnaire was designed with 10 ques- proteins and synthetic 15-­mer peptides with overlapping sequences

tions using visual analogue scales (VAS) ranging from 0 (difficult/ spanning the FVIII A2 and C2 domains. PBMCs from 20 severe HA

poor) to 100 (good/easy). First enrolled Manchester PwHA (mean subjects with a current, past or no inhibitor history were tested by

age 46.5±12.7, 92.3% severe) were asked to fill in the questionnaire ELISPOT assays to detect cytokine secretion in response to FVIII

after completing PROs. proteins and/or peptides.

Results: PwHA (n=13) needed a median (med) 60 minutes (range 0.5-­ Results: Over half showed robust Th1 and/or Th2 responses to

2.5 hours) to complete the PROs. They thought that the survey did recombinant FVIII-­C2 protein. Incubations with FVIII A2 and C2

not adequately capture some aspects mentioned in the focus groups domain peptides stimulated IFN-­g and/or IL-­5 secretion from intron-

including factor medication (44%) and home & daily life (33%), or ­22 inversion samples, consistent with earlier publications from the

could not be assessed, namely psychological (33%) and economic Conti-­Fine group.

impacts (44%); some (33%) felt that e.g. impact on children should Conclusions: The observed lack of tolerance to multiple FVIII do-

have been assessed. VAS questions revealed that PwHA liked the mains indicates that future strategies to tolerize intron-­22 inversion

survey (med 70), found it easy to complete (med 80), related to their patients should take into account epitopes throughout the thera-

personal situation (med 70), considered the number of questions peutic FVIII protein, rather than focusing on the short C1 domain

inadequate (med 51.5). Although the number of questions was the sequence corresponding to the inversion break-­point.

PB193 | Analysis of HLA-­class-­II (HLAcII) TA B L E 1   Log-­linear Model Results of Dendritic Cell Protein

Processing and Presentation Assays
Peptidomes in Response to Several Therapeutic
Factor (F)VIII Proteins (tFVIIIs) in Healthy Donors
and Hemophilia A (HA) Patients with and without
FVIII Inhibitors (FEIs)
Z. Sauna1; V. Diego2; M. Almeida2; B. Luu2; M. Hofmann3;
J. Hernandez2; R. Rajalingam4; S. Williams-Blangero2;
J. Curran2; J. Powell5; M. Escobar6; J. Blangero2; E. We performed a log-­linear analysis of multi-­way contingency ta-
Maraskovsky7; N. Key8; T. Howard2,9,10 bles (MCTs) of our data. We analyzed a MCT of: 1) a variable con-
Center for Biologics Evaluation and Research, Food and Drug Administration, trasting H with N and I; 2) DRB genotypes; 3) the 4 tFVIIIs, and 4)
Laboratory of Hemostasis, Division of Hematology Research and Review,
tFVIII domain. In a 2nd model, we analyzed a MCT of an inhibitor
Silver Spring, United States, 2University of Texas Rio Grande Valley School
of Medicine, South Texas Diabetes & Obesity Institute, Brownsville, United variable contrasting N and I, and the last 3 variables under the 1st
States, 3CSL Behring, Marburg, Germany, 4University of California San model.
Francisco, Immunogenetics and Transplantation Laboratory, San Francisco,
Results: HA patients I and N have greater IP than baseline, namely
United States, 5CSL Behring, Med Affairs, King of Prussia, United States,
University of Texas Health Sciences Center at Houston, Medicine, Division of H (Table 1A). BDT-­& FL-­rFVIII, respectively, have lesser and greater
Hematology and Oncology, Houston, United States, 7CSL Limited, Research, IP than baseline, pdFVIII. The IP for BDD-­rFVIII was not different
Bio21 Institute, Melbourne, Australia, 8University of North Carolina at Chapel
from pdFVIII. Similarly, patients with inhibitor I have greater IP than
Hill, Medicine, Division of Hematology and Oncology, Chapel Hill, United
States, 9University of Texas Rio Grande Valley School of Medicine, Biomedical those without N (Table 1B). Again, BDT-­& FL-­rFVIII have lesser and
Sciences, Brownsville, United States, 10Haplomics Biotechnology Corp, greater IP than pdFVIII whereas the IPs for BDD-­rFVIII & pdFVIII
Brownsville, United States are not different. The observed higher IPs in affecteds relative to
unaffecteds reflects a significant correlation with the presence of
Background: T cells (TCs) are critical for FEI-­development in HA. the inhibitor risk allele DRB1*15:01. HP and NI have correlations
Dendritic cells (DCs) express HLAcII repertoires and present pep- with DRB1*15:01 of 0.91 and 0.52 (P=0.002 and 0.001).
tides to TCs. We used DC protein presentation assays to study tFVIII Conclusions: DC protein presentation is affected by HLAcII rep-
peptidomes. ertoires but not by HA or inhibitor status. In addition, rFVIIIs have
Aims: Test the hypothesis that protein presentation by DCs is not different IPs relative to pdFVIII with BDT-­and FL-­rFVIII being signifi-
affected by HA. Study factors influencing the immunogenicity po- cantly less and greater than, respectively.
tential (IP) of tFVIIIs as reflected by HLAcII-­bound tFVIII-­derived
peptide density
Methods: DCs from 4 healthy donors (H) and 5 HA patients (P)—2
PB194 | Recombinant Porcine Factor VIII Post-­
with FEIs (I) and 3 without (N)—were incubated with plasma-­derived
(pd)VWF and either pdFVIII or 1 of 3 recombinant (r)FVIII proteins
marketing Safety Studies in the European Union
(rFVIIIs): full-­
length (FL), B-­
domain (BD)-­
deleted (BDD), and BD-­ (EU) and United States (US)
truncated (BDT). DR-­bound peptides were identified by LC-­MS/MS R. Crea1; O. Mauric1; J.-F. Huang2; J. Chen2
and comparison to a proteomics database (Figure 1). 1
Shire, Vienna, Austria, 2Shire, Cambridge, United States

Background: A B-­
deleted, porcine-­
sequence factor VIII
(rpFVIII) with low cross-­reactivity to anti-­human-­F VIII antibodies is
used to treat acquired hemophilia A (AHA).
Aims: To collect data assessing safety, safety-­related factors, utiliza-
tion, and efficacy of rpFVIII in real-­world clinical practice.
Methods: Two multicenter, prospective, noncontrolled, open-­
label, single cohort, noninterventional studies are ongoing in the
EU and US. Patients (pts) >18 years of age with AHA will be en-
rolled once per routine clinical practice and treated with rpFVIII
for bleeding episodes (BEs). Demographic/clinical characteristics,
comorbidities and medical history, characteristics of BEs, treat-
ment indication, rpFVIII treatment details, bleeding control, con-
comitant medications, inhibitor levels, prior rpFVIII and bypassing
agent treatments, and adverse events (AEs)/serious AEs will be
FIGURE 1 Representative Individual HLAcII Peptidomes for Various collected at baseline and follow-­up visits up to 180 d after the last
Therapeutic Factor VIII Proteins (tFVIIIs) dose of rpFVIII.

Results: The EU study began in late 2016 and is ongoing in Austria, PD-­1, on B cells in hemophilic mice after adoptive transfer of regula-
Germany, and Italy, and 4 pts have enrolled. The US study began in tory T cells.
2015 and currently has 16 active sites and 37 pts enrolled. A data Conclusions: Our results, suggest novel therapeutic targets for the
read-­out was carried out in June 2017 for 21 pts. Among evaluable treatment of Hemophilia A patients with inhibitors.
pts, 6 were treated with rpFVIII as first-­line therapy for initial BEs;
4 as first-­line therapy for subsequent BEs; and 6 received rpFVIII as
second-­line or later treatment. Among pts evaluable for efficacy, a PB196 | Phase 2 Trial of Subcutaneously
total of 14 BEs in 11 pts were resolved, while 4 BEs in 3 pts were not
Administered Novel FVIIa Variant, Marzeptacog
resolved. No thromboembolic events were reported.
Conclusions: Preliminary analysis of the US study shows rpFVIII is
Alfa (activated), in Hemophilia A or B with
generally well-­tolerated and effective in AHA. These studies will Inhibitors: Pharmacokinetics, Pharmacodynamics,
provide robust real-­world data on the use of rpFVIII, providing in- Safety and Efficacy
sight into the clinical use of rpFVIII, as well as the optimal regimen to H. Levy1; G. Iosava2; F. Del Greco1; A. Hetherington1; J.
personalize therapy. Hillman1
Funding: These studies were funded by the sponsor, Shire 1
Catalyst Biosciences, South San Francisco, United States, 2Medinvest Institute
International GmbH. of Hematology and Transfusiology, Tbilisi, Georgia

Background: In hemophilia patients with inhibitors, hemostasis

PB195 | Regulatory T Cells in the Modulation can be achieved using bypass agents: activated plasma-­derived
of Hemophilia A prothrombin complex concentrates (aPCC), recombinant Factor
VIIa (FVIIa) or emicizumab that can only treat hemophilia A.
M. Meissner1; J. Gotot1; T. Albert2; J. Oldenburg2; C. Kurts1
Marzeptacog alfa (activated) (MarzAA), a variant recombinant FVIIa
Universitätsklinikum Bonn, Institute for Experimental Immunology, Bonn,
Germany, 2Universitätsklinikum Bonn, Institute for Hematology and Transfusion
with enhanced biological properties enabling subcutaneous (SQ)
Medicine, Bonn, Germany delivery, was developed using a rational protein design approach
and has 7-­fold increased catalytic activity and prolonged duration
Background: Hemophilia A is an X chromosome-­linked bleeding dis- of effect in a human single-­dose intravenous (IV) study that docu-
order caused by a deficiency of blood coagulation factor VIII (FVIII) mented safety. Daily dosing in dogs demonstrated achievement of
that affects approximately 1 in 5,000 males. A major clinical compli- stable trough levels of MarzAA that are believed to be effective for
cation concerning the protein-­replacement therapy with therapeutic prophylaxis.
FVIII is the formation of anti-­F VIII inhibitory antibodies (inhibitors), Aims: Phase 2 study MAA-­201 complies with the Declaration of
which occurs in approximately 30% of all Hemophilia A patients. Helsinki and is approved by recognized medical ethics committees,
Currently accepted clinical protocols for immune tolerance induc- will enroll 12 subjects with inhibitors and an annual bleeding rate
tion (ITI) in Hemophilia A patients do not only eliminate already ex- ≥12 who sign informed consent, to determine if SQ MarzAA can pro-
isting inhibitors, but also induce long-­lasting tolerance towards FVIII vide effective prophylaxis (NTC03407651).
in approximately 66% of all ITI-­receiving Hemophilia A patients. A Methods: 18 mg/kg MarzAA will be infused IV and pharmacokinet-
role of suppressive CD4+CD25+Foxp3+ regulatory T cells has been ics and coagulation parameters measured for 24 hours (Figure). 30
suggested, but the precise mechanisms underlying tolerance in ITI mg/kg MarzAA will be injected SQ and pharmacokinetics and coagu-
is unknown. lation parameters measured for 48 hours. Daily MarzAA SQ therapy
Aims: We here aimed at providing formal and mechanistic evidence will then be injected for 50 days.
for a role of regulatory T cells, by examining the anti-­F VIII immune If no spontaneous bleeding events occur, the subject will proceed to
response in hemophilic mice and healthy control mice. safety follow up 30 days after last dose. Should spontaneous bleed-
Methods: We performed adoptive transfer experiments with regu- ing occur, then dose escalation occurs sequentially to 60, 90 or 120
latory T cells from healthy control mice into hemophilic mice and mg/kg, if needed. PK and bioavailability will be calculated. Subjects
analyzed the humoral and cellular immune response towards FVIII will be monitored for treatment-­
related changes, Quality of Life
after immunizing these mice with recombinant FVIII protein. We es-
pecially focused on the expression of both, activating and inhibitory
molecules on the surface of regulatory T cells and
B cells by flow cytometry.
Results: We demonstrate that regulatory T cells from healthy con-
trol mice, after transfer into hemophilic mice, were able to suppress
B cells in acceptor mice. Furthermore, we show a role of changes in
inhibitory molecules, like Fas and F I G U R E 1   MAA-201 trial schema

measures and tested for inhibitors to MarzAA and FVIIa. ABR will be hemostatic pathway. rhFVIIa (eptacog beta, EB) has an almost
compared to the subjects’ prior ABR. two-­fold greater affinity for EPCR compared to another variant of
Results: Interim results of pharmacokinetics, coagulation changes, human FVIIa, rFVIIa (eptacog alfa, EA)#. Thrombin generation assays
immunogenicity and bleeding rate will be reported. in hemophilia patients w/wo inhibitors show reduced clinical dose
Conclusions: This study design will demonstrate whether daily SQ requirements for EB. As ex vivo studies of thrombin generation do
MarzAA can provide effective prophylaxis in hemophilia patients not predict FVIIa variant efficacy, the PERSEPT1 trial of EB could
with inhibitors. have led to results comparable to those of other variants. Instead,
efficacy at 12 hours was up to 93%, and was sustained through 24
hours without extra or rescue dosing. Thus, the Phase 3 clinical effi-
cacy of EB may validate the clinical relevance of this third hemostatic
PB197 | Potential Clinical Trial Validation of
Differential EPCR-­dependent Activity by Eptacog
Aims: To determine whether clinical evidence supports the func-
Beta, a rhFVIIa Variant, in the Treatment of tional difference in EPCR binding between FVIIa variants.
Bleeding in Hemophilia Patients with Inhibitors Methods: We conducted a literature review for prospectively col-
J. Ducore1; W.A. Alexander2; I. Mitchell3; J. Lawrence 4; J. lected data describing two clinical outcomes of bleeding events
Grandoni4; D. Bonzo 4 treated with the rFVIIa variants EB and EA: rebleeding rate at 24
UC Davis, Sacramento, United States, 2HEMA Biologics, LLC, Louisville, United hours (Table 1), and success in hemarthrosis (Table 2). Interpolated
States, 3Gloval LLC, Broomfield, United States, 4LFB-­USA, Framingham, United rebleeding rates were excluded and hemarthrosis clinical response
was included if reported at 12 and 24 hours. Differential EPCR bind-
ing studies are included in the complete results.
Background: Several laboratories have demonstrated that en-
dothelial protein C receptor (EPCR) binding by pharmaceutical
doses of FVIIa potentially drives a 3rd non-­tissue factor-­dependent

TA B L E 1   Published clinical efficacy and early rebleeding rates

FVIIa Bleeding 12 hour 24 hour Efficacy Rebleeding
Source variant Dosing (mcg/kg) Events (N) Efficacy (%) (%) Rate* (%)

PERSEPT 1 Wang et al 2017 EB 75q3h 252 84.9 96.7 0

PERSEPT 1 Wang et al 2017 EB Initial: 225 213 93.2 99.5 0.2
Subsequent 75q3h
after 9 hour
adept™2 Lentz, et al 2014** EA arm 90q3h (1-­3 doses) 227 93 86.2 13.8
FENOC Astermark, et al EA 90-­120 x2 48 75.6 85.7 4.2
“Megadose” Kenet, et al 2003 EA Initial: 300 Second: 114 -­ 91.4 9.6
200 or 300 @
3-­4  hour

*Early rebleeding -­Prior to 24 hours from first dose, the recurrence of symptoms, need for rescue medication or other objective evidence suggesting
loss of hemostasis after initial clinical response; **Extra doses allowed and administered after initial response.

TA B L E   2   Efficacy in hemarthrosis
Review of efficacy in hemarthrosis was facilitated by the detailed literature review by Treur, et al (2009) where 11 trials were analyzed for
EA efficacy in hemarthrosis

24 hour
Source FVIIa variant Bleeding Events (N) 9 hour Efficacy (%) 12 hour Efficacy (%) Efficacy (%)

PERSEPT 1 Wang, et al EB 396 Pooled efficacy not 88.9 98.2

2017 reported
TREUR Bayes Data Treur, EA 160 / 91 / 121 44 / -­/ -­ -­/ 67 / -­ -­/ -­/ 54
et al 2009
TREUR Bayes Model Treur, EA n/a -­ 66 88
et al 2009
Conclusions: Comparisons from large studies of early rebleed- TA B L E   1   Baseline characteristics of the patients included in the
ing and 12/24 hour efficacy for EA and EB in the setting of BrazIT Study (n=45). * out of 42 patients
joint bleeding yielded consistent differences. This observation Characteristics Results
suggests that the two-­f old difference in affinity for EPCR clini-
Male sex, n (%) 44 (97.8)
cally manifests as an earlier clinical response and lower dose re-
FVIII activity < 2%, n (%) 34 (81.0)*
quirement for rhFVIIa (eptacog beta, EB). Further research into
Age at hemophilia diagnosis in years, median 0.9 (0.6-­1.6)
EPCR binding and clinical efficacy of EB should be conducted to
(IQR -­interquartile range)
more fully characterize the effect of a third rhFVIIa hemostatic
Age at inhibitor diagnosis in years, median (IQR) 3.9 (1.7-­10.7)
Age at ITI (immune tolerance induction) start in 11.8 (4.38-­24.5)
#Grandoni, J et al. Haemophilia 2017, 23(2):300-­308:
years, median (IQR)
Time elapsed between inhibitor diagnosis and 4.9 (1.0-­12.4)
ITI start in years, median (IQR)
PB198 | Immune Tolerance Induction in Pre-­ITI inhibitor titer peak in BU (Bethesda 27.4 (12.7-­57.5)
Patients with Hemophilia A and Inhibitors: unit)/mL, median (IQR)

Preliminary Results of the Brazilian Immune Inhibitor titer immediately before ITI start in 4.2 (2.8-­7.6)
BU/mL, median (IQR)
Tolerance (BrazIT) Study
R. Camelo1; L. Magalhaes1; L. Jardim1; L. Werneck1; A.
Gonçalves2; M.R. Roberti3; D. Tan4; V. Franco5; D. Chaves2;
S. Rezende1
TA B L E   2   Characteristics of patients who completed immune
Federal University of Minas Gerais, Faculty of Medicine, Belo Horizonte, Brazil, tolerance induction (ITI; n=34)
Foundation Hemocentro of Minas Gerais, Belo Horizonte, Brazil, 3Hemocentro
of Goias, Goiania, Brazil, 4Faculty of Medicine of Marilia, Hemocentro of Marilia, P (Mann
Marilia, Brazil, 5Hemocentro of Santa Catarina, Florianopolis, Brazil Successful Failed ITI Whitney
Characteristics ITI (n=29) (n=5) test)
Background: About 20-­30% of patients with HA (PHA) may develop Age at ITI start in 6.4 (2.9-­18.8) 9.5 (8.7-­13.4) 0.78
neutralizing antibodies (inhibitors) against infused factor (F) VIII. years, median (IQR
Immune tolerance induction (ITI) is the treatment of choice to eradi- -­ interquartile
cate inhibitors.
Time elapsed 2.7 (0.5-­8.4) 7.1 (1.0-­10.4) 0.71
Aims: The Brazilian Immune Tolerance (BrazIT) Study aims to inves-
between inhibitor
tigate predictors of response to ITI using a protocol recommended diagnosis and ITI
by the Ministry of Health. start in years,
Methods: Inclusion required confirmed diagnosis of HA with active median (IQR)
inhibitor. Patients were not selected based on “good risk” factors Pre-­ITI inhibitor titer 24.0 140.8 0.92
and all who filled the inclusion criteria and signed informed consent peak in BU (12.0-­52.0) (32.0-­
(Bethesda unit)/mL, 384.0)
were included. The protocol recommended that all PHA started ITI
median (IQR)
with a low-­dose regimen (50IU/kg 3x/week). Upon lack of response
Inhibitor titer 4.0 (2.5-­7.6) 5.6 0.15
(no decline of inhibitor in of at least 20% of peak levels after ITI start immediately before (5.6-­11.2)
within 6 months), high-­dose (HD) regimen (100IU/kg daily) and sub- ITI start in BU/mL,
sequent changing to plasma-­derived (pd) FVIII for those who started median (IQR)
on recombinant FVIII was recommended. Response was based on Inhibitor titer peak 6.0 (2.5-­20.8) 179.2 0.004
inhibitor titer, FVIII half-­life and recovery. during ITI in BU/ (58.4-­
mL, median (IQR) 211.2)
Results: We included 45 patients from 4 Hemophilia Treatment
Plasma-­derived FVIII 20 (69) 4 (80) >0.99
Centers (Table 1) of whom 34 concluded the protocol. A total of
as first type of
29/34 (85%) patients responded to ITI, of whom 13 and 16 were concentrate used, n
total and partial success, respectively; 5/34 (15%) patients failed ITI, (%)
despite changing to HD regimen. A total of 24/34 (70.6%) patients Change to high-­dose 1 (4) 5 (100) 0.0001
started ITI with pdFVIII. Intron 1 and 22 inversions were present in regimen, n (%)
13/18 (72%) and 1/4 (25%) of patients who responded and failed, re- Number of bleeding 9.0 (4.0-­14.0) 20.0 0.012
spectively (p=0.12). Patients who failed treatment had higher inhibi- events during ITI, (19.0-­25.0)
median (IQR)
tor peak during ITI (p=0.004), changed to HD regimen (p=0.0001)
and had more breakthrough bleeding events (p=0.012), compared ITI duration in years, 1.6 (1.0-­2.1) 3.3 (3.1-­3.4) 0.0002
median (IQR)
with those who had success (Table 2).

Conclusions: Brazilian ITI protocol demonstrated 85% success PB200 | Risk Modeling Inhibitor Development
rate despite the long time elapsed between inhibitor diagnosis and
of Haemophilia A Patients in Australia
start of ITI and the older age of patients at start of ITI. Due to
low number, multivariate analysis could not be performed, which
S. Parikh1; H. Tran2,3; S. McRae 4
Australian Haemophilia Centre Directors’ Organisation, Melbourne, Australia,
will be possible with the inclusion of more patients and longer 2
Australian Centre for Blood Diseases, Melbourne, Australia, 3The Alfred
follow-­up. Hospital, Melbourne, Australia, 4Royal Adelaide Hospital, Adelaide, Australia

Background: Development of inhibitors is the most severe and chal-

PB199 | Inhibitor Development and Successful lenging complication associated with factor VIII replacement ther-
Immune Tolerance Induction in a Young Female apy in Haemophilia A (HA) patients. Several studies identifying risk
factors that influence inhibitor development have been published in
Carrier of de novo Severe Hemophilia A
the past however cumulative risk is not clearly evident.
L.W. Zuccherato1; M.D.R.F. Roberti2; L.L. Jardim1; S.M. Aims: The primary aim of this project is to conduct a preliminary
analysis of the risk factors associated with inhibitor development
Universidade Federal de Minas Gerais, Belo Horizonte, Brazil, 2Centro de
among HA patients in Australia and the feasibility of creating a mul-
Hematologia e Hemoterapia de Goiás, Goiania, Brazil
tivariate risk model to enhance clinical decision making.
Methods: Retrospective data was obtained from the Australian
Background: Females are commonly reported as carriers of hemo-
Bleeding Disorders Registry (ABDR) on HA patients (excluding car-
philia A (HA), since only one X chromosome is affected. Skewed ly-
rier patients) from all Haemophilia Treatment Centres (HTCs) with
onization of X chromosome can be associated with low levels of FVIII
appropriate ethics approval. Information was extracted on disease
(< 5%) in carriers, and might require FVIII replacement. Neutralizing
severity, genotype, product usage, treatment regime, treatment rea-
antibodies (inhibitors) against FVIII occur in up to 30% of patients,
son, age at first exposure and exposure days (ED) leading upto first
but is rarely reported in carriers.
inhibitor (titre level ≥ 0.5BU/ml). Mixed modeling (containing both
Aims: Here we report a rare case of a young female presenting very
fixed and random effects of risk factors) plus all other analysis was
low FVIII levels (< 1%) and high-­titer inhibitor development which
performed using Stata v12.1.
successfully disappeared after ITI.
Results: 1644 (68.15%) of the 2412 HA patients used at least one
Methods: A 10-­month-­old female was referred due to bruises and
factor VIII product. 161/1644 HA patients with missing/incom-
hemarthrosis. There was no family history of HA nor other bleed-
plete/historical data were excluded. Data on 1483 HA patients
ing disorder. FVIII activity (FVIII:C) levels were 0.6% and the inhibi-
included 196 Inhibitor (25.87% Severe) and 67 Transient.140/196
tor titer at diagnosis was 0.35 Bethesda Units (BU)/mL. An inverse
(71.42%) developed inhibitors within 50 ED. 922/1483 HA pa-
PCR for intron 22 inversion of F8 showed the carrier status of the
tients were prescribed Prophylaxis of which 440 had that as their
proband, with family members testing negative. Exome sequenc-
first treatment. 3.63% (16/440) developed inhibitor (5 had major
ing of F8 showed no additional deleterious mutations, indicating
surgery/accident/trauma while on prophylaxis, 6 non-­compliant, 3
that the disease is likely caused by a defective skewed epigenetic
paediatric patients and 2 with high risk genetic mutation). Results
X-­inactivation system.
characterised severity, genotype, age at first exposure, ED and
Results: Prophylactic replacement started with plasma-­d erived
prophylaxis as fixed effect variants whereas intensive treatment
(pd) FVIII. After 25 exposure days, a high-­
t iter inhibitor (25.6
therapy and non-­compliance as random effect variants of the mul-
BU/mL) was detected, five weeks after the report of pain and
tivariate risk model.
extensive edema at her left thigh, due to influenza A intramus-
Conclusions: Mixed modeling can be used to predict cumulative risk
cular vaccination, which required five days of FVIII replacement.
of inhibitor development among HA patients.
ITI was initiated with pdFVIII 50 IU/kg 3 times a week. Inhibitor
disappeared after 18 months and partial response was confirmed
by FVIII half-­life of 14.5 hours and FVIII recovery of 58.6%. She
started on prophylactic treatment and is currently maintaining
sustained inhibitor titer below 0.6 BU/mL.
Conclusions: Our report demonstrates that HA carriers with severe
phenotype can develop neutralizing antibodies against FVIII, and
that ITI can be effective. We suggest that HA carriers, particularly
those who require FVIII replacement, should be regularly screened
for FVIII inhibitors.

PB201 | Real-­world Bleeding Frequency in TA B L E 2   ABR and AJBR in Patients on Prophylaxis (n=25)

Inhibitor Patients on Prophylaxis with APCC: No. of Patients (%)

Data Readout of the “FEIBA Global Outcomes Bleeding Rate (Categorized) ABR AJBR
Study (FEIBA GO)”
0 5 (20) 9 (36)
C. Escuriola Ettingshausen1; J. Windyga2; P.A. Holme3; C. 0-­<2 8 (32) 14 (56)
Hermans4; C. Negrier5; A.R. Cid6; S. Rangarajan7; G.R. Chiu8;
0-­<3 11 (44) 18 (72)
V. Cano9; R. Crea10
1 ≥3 14 (56) 7 (28)
HZRM, Haemophilia Centre Rhine-­Main, Frankfurt-­Möerfelden, Germany,
Institute of Hematology and Transfusion Medicine, Warsaw, Poland, 3Oslo
University Hospital and Institute of Clinical Medicine, University of Oslo, Conclusions: This study provides data on long-­term prophylaxis in
Oslo, Norway, 4Saint-­Luc University Hospital, Brussels, Belgium, 5Louis
the real-­world setting by assessing efficacy, HRQoL, and safety of
Pradel Cardiology Hospital, University Claude Bernard, Lyon, France, 6Unidad
Hemostasia y Trombosis, Hospital Universitario y Politécnico La Fe, València, APCC. Preliminary results demonstrate prevention of joint bleeds
Spain, 7University Hospital Southampton NHS Foundation Trust, Hampshire, with FEIBA prophylaxis, which is consistent with that reported in
United Kingdom, 8Shire, Cambridge, United States, 9Shire, Zug, Switzerland,
pts without inhibitors on replacement prophylaxis (Khair K, et al.
Shire, Vienna, Austria
Haemophilia. 2018;24:85-­96).
Funding: This study was funded by the sponsor, Shire International
Background: The “FEIBA GO” study will capture long-­term effi-
cacy, safety, and quality of life (QoL) outcomes in patients (pts)
with hemophilia (hem) and inhibitors treated with FEIBA (activated
prothrombin complex concentrates [APCC]) in routine clinical
practice. PB202 | Addressing Barriers to the Use of
Aims: To describe the hemostatic effectiveness of APCC during Oral Anticoagulant Reversal Agents: A CME
prophylaxis and on demand, including in pts on immune-­tolerance in- Intervention
duction. Secondary objectives include joint functionality outcomes,
J. Spyropoulos
safety, health-­related QoL (HRQoL), activity level, hem-­associated
Medsacape LLC, New York, United States
pain, and healthcare resources used.
Methods: This is a prospective, observational, multicenter, cohort
Background: Oral anticoagulants (OACs) are associated with in-
study in ~55 pts with hem A or B and high-­titer inhibitors treated
creased risk of bleeding. Rapid OAC reversal can be achieved
with APCC before enrollment. Treatment regimens are at the dis-
through use of PCC or plasma. While recommendations exist to help
cretion of attending physicians according to routine clinical practice.
physicians choose the most effective reversal strategy, guidelines
The observation period per pt will be 4 years.
are not always followed, affecting patient outcomes.
Results: A data read-­out was carried out on Oct 23, 2017 on pts with
Aims: To determine if an intervention consisting of continuing medi-
severe hem A (n=48) or B (n=1) and inhibitors from 11 countries. At
cal education (CME) followed by a planned change assessment (PCA)
screening, median age was 18 year (range, 2-­71), and 83% and 10%
could improve performance of physicians related to appropriate use
were white and black/African American, respectively. The mean num-
of reversal agents.
ber of target joints (Table 1) was 1.2 (range, 0-­10), and 32 pts were on
Methods: An online CME activity was developed as a 25-­minute
prophylaxis. For 29 prophylaxis pts for whom first factor VIII inhibi-
roundtable discussion among 4 leading experts. A PCA question-
tor detection was available, median titer was 14.0 BU (range, 1-­150).
naire was administered immediately following the activity in the
Among 25 prophylaxis pts evaluated for efficacy, annualized bleeding
form of intent-­
change questions. Completers also indicated
rates (ABRs) and annualized joint bleeding rates (AJBRs) were reported
potential anticipated barriers that may inhibit implementation of
(Table 2), with median rates of 3.6 (range, 0-­50) and 1.2 (range, 0-­17),
changes. Survey completers were recontacted to complete a follow-
­up assessment approximately 8 weeks later to assess self-­reported
TA B L E 1   Number of Target Joints at Screening in Patients on
actual changes in clinical practice, along with barriers encountered
Prophylaxis (n=32)
to making changes in the real-­world setting.
No. of Results: A total of 750 physicians, completed the PCA, and 711
Sum of Target Joints Patients (%) (95%) indicated intended changes. Among those who intended to
0 21 (65.6) make changes, a total of 2198 changes were planned (Figure 1).
1 3 (9.4) When recontacted for the follow-­up survey, 19 learners completed
2 4 (12.5) the survey. Of those, 16 (84%) reported that they had made 61

>2 4 (12.5) changes, an average of 3.8 changes in practice. The most common
practice changes made included improved ability to prescribe OAC
therapy (62%) and selecting bleeding management strategies based

F I G U R E 1   Summary of planned changes

on the severity of bleeding and patient characteristics (62%). The Methods: Laboratory haemostasis testing included PT, APTT, TT,
most common barriers to implementing changes included financial fibrinogen, lupus anticoagulant and activities of factors VIII, IX, XI,
constraints (22%), difficulty to implement system changes (14%), and and XII using commercially available coagulation methods on BCSXP
lack of availability of PCC (14%). analyzer (Siemens, Germany). The inhibitor modified Bethesda assay
Conclusions: This assessment demonstrated the effectiveness of (Nijmegen) was used to quantify FVIII inhibitor.
CME combined with a PCA intervention and allowed for assessment Results: A 59-­year-­old male patient, 3 yrs ago diagnosed and treated
of quantitative data for explanation of practice changes and barriers for gastric MALT lymphoma with chemotherapy and radiotherapy,
to best practices. came to the Emergency Department (ED) due to large haematoma
on the right hip. The coagulation profile revealed prolonged APTT
of 59s, reduced FVIII activity of 7% and presence of FVIII inhibi-
tor at 4.0 Bethesda Units (BU)/mL with normal results for all other
PB203 | Successful Eradication of Factor FVIII
coagulation tests. Initial IST included prednisolone, but due to in-
Inhibitor by Cyclophosphamide Treatment in a
adequate therapeutic response cyclophosphamide was also intro-
Patient with Acquired Haemophilia A -­A Case duced. Both clinical and laboratory improvements were evident
Report after 2 weeks (APTT=37-­39s). Once discharged, the patient contin-
1 2 1 ued combined therapy with cyclophosphamide and prednisolone.
S. Margetic ; D. Carzavec ; N. Vrkic
Sestre Milosrdnice University Hospital Center, Department of Clinical Chemistry, One month after admittance in ED factor VIII activity rose to 82%
Zagreb, Croatia, 2Sestr