diabetes research and clinical practice 1 2 6 ( 2 0 1 7 ) 2 3 0 2 3 9

Contents available at ScienceDirect

Diabetes Research
and Clinical Practice
journa l home page: www.e lse vier.com/locate/diabres

Albiglutide for the treatment of type 2 diabetes

mellitus: An integrated safety analysis of the
HARMONY phase 3 trials

Bo Ahren a,*, Molly C. Carr b, Karen Murphy b, Christopher Perkins c, Marc Rendell d,
Jason Mallory e, Timothy Wilson f,1, Susan Johnson e,2
a
Department of Clinical Services Division of Medicine, Lund University, Lund, Sweden
b
Pharma Research & Development, GlaxoSmithKline, Collegeville, PA, USA
c
PPD, Inc, Morrisville, NC, USA
d
Department of Medicine, Division of Endocrinology, Creighton University, Omaha, NE, USA
e
Pharma Research & Development, GlaxoSmithKline, Upper Merion, PA, USA
f
Pharma Research & Development, R&D Projects Clinical Platforms, PCPS Qsci Clinical Statistics, Research Triangle Park, Durham, NC, USA

A R T I C L E I N F O A B S T R A C T

Article history: Aims: Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) stimulate the incretin system
Received 19 October 2016 and lower glycaemic parameters in type 2 diabetes mellitus (T2DM). This analysis of clinical
Received in revised form studies of up to 3 years evaluated the safety of albiglutide, a GLP-1 RA, in people with T2DM.
17 January 2017 Methods: Integrated safety analysis included seven phase-3 T2DM studies of albiglutide
Accepted 13 February 2017 compared with placebo and/or active comparators (a dipeptidyl peptidase-4 inhibitor,
Available online 20 February 2017 GLP-1 RA, insulin, sulphonylurea, and thiazolidinedione).
Results: Studies of 32 months (HARMONY 7), 1 year (HARMONY 6), and 3 years (HARMONY
15), reported similar rates of adverse events (AEs) (84.8%, 82.3%), and serious AEs (13.1%,
Keywords:
12.9%) between albiglutide and all comparators, respectively. AEs that did not differ between
Diabetes mellitus
the groups included symptomatic or severe hypoglycaemia as well as nausea (12.0%, 11.3%)
Incretins
and vomiting (5.3%, 4.7%) for albiglutide and all comparators, respectively. According to the
Glucagon-like peptide-1
Medical Dictionary for Regulatory Activities preferred terms, only diarrhoea (13.7%, 9.9%),
Long-term safety
injection-site reaction (9.0%, 2.0%), and peripheral oedema (4.5%, 6.8%) had at least 2% dif-
ference between the albiglutide and all-comparator groups. In a similar integrated analysis,
pancreatitis occurred more often with albiglutide (0.3%, 0.1%). Renal and cardiac function
did not differ between the two groups.
Conclusions: In an integrated analysis of seven phase 3 clinical trials, albiglutide-treated
patients experienced frequencies of AEs (including cardiovascular and renal) similar to
the all-comparators group treated with other T2DM medications or placebo. Albiglutide
treatment was associated with higher rates of diarrhoea and injection-site reactions, but
not increased nausea and vomiting, versus all comparators.
2017 Elsevier B.V. All rights reserved.

* Corresponding author at: Biomedical Centre C11, Lund SE-221 84, Sweden.
E-mail addresses: Bo.Ahren@med.lu.se (B. Ahren), molly.c.carr@gsk.com (M.C. Carr), Karen.J.Murphy@gsk.com (K. Murphy),
christopher.perkins@ppdi.com (C. Perkins), rendell@asndi.com (M. Rendell), jasonmallory@hotmail.com (J. Mallory), Tim.Wilson@
parexel.com (T. Wilson), Susan.l.johnon@pfizer.com (S. Johnson).
1
PAREXEL International, Durham, NC, USA.
2
Pfizer Pharmaceuticals Inc, New York, NY, USA.
http://dx.doi.org/10.1016/j.diabres.2017.02.017
0168-8227/ 2017 Elsevier B.V. All rights reserved.
 diabetes research and clinical practice
1. Introduction
Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) have
been developed as novel glucose-lowering therapies in type
2 diabetes mellitus (T2DM) [1]. Albiglutide is a recombinant
fusion protein consisting of two linked copies of a modified
30-amino acid sequence of human GLP-1 (fragment 736)
genetically fused in series to human albumin [24]. A single
amino acid substitution at position 8 of the GLP-1 sequence
(ala ? gly) confers resistance to proteolysis by dipeptidyl
peptidase-4 (DPP-4) [2]. Fusion with albumin and resistance
to DPP-4 degradation give albiglutide a half-life of approxi-
mately 5 days, allowing for once-weekly dosing [3,5,6].
Albiglutide has been shown to reduce glycaemic parameters
(eg, HbA1c, postprandial glucose excursions, and fasting
plasma glucose [FPG]) in association with enhancement of
glucose-dependent insulin secretion and slowing of gastric
emptying [25,7].
The HARMONY phase 3 program included eight ran-
domised, double-blind or open-label, placebo- and/or active-
controlled trials (Table 1) that evaluated the efficacy and
safety of albiglutide in patients with T2DM who were inade-
quately controlled on their current regimen of diet and exer-
cise, oral antidiabetes drugs (OADs), or basal insulin. The
length of the trials ranged from 32 weeks to 3 years and
included more than 4800 patients (Table 1). The results of
each trial at the primary endpoint and, for 5 of the studies,
at a time point of 3 years have been reported separately
[817]. After 3 years treatment, albiglutide was associated
with reductions in HbA1c and FPG and with modest weight
loss. Given the size and duration of exposure, the program
offers a unique opportunity to evaluate the long-term safety
of albiglutide. We present an integrated analysis of the safety
of albiglutide in the HARMONY phase 3 program, including
data out to 3 years.
2. Methods
2.1. Study design
Seven studies in the HARMONY program (Table 1) were inte-
grated for safety analysis. Five of the phase 3 studies had pre-
planned continuation of randomised treatment out to 3 years.
The integrated studies included patients treated with albiglu-
tide as monotherapy, in combination with one to three OADs,
or with insulin (basal). The pooled all-comparators group in
the integrated analysis includes a range of active comparators
(metformin, sulphonylurea [SU], thiazolidinedione [TZD],
DPP-4 inhibitor, insulin [basal and lispro], or GLP-1 RA [liraglu-
tide]) as well as placebo comparator. HARMONY 8 [15], which
enrolled patients with mild, moderate, and severe renal
impairment, was excluded from this integrated analysis due
to the unique safety characteristics of such patients, although
results of the HARMONY 8 study [15] were included in sepa-
rate integrated analyses of adjudicated pancreatitis and car-
diovascular events, discussed later in this article.
Albiglutide had a starting dose of 30 mg weekly in all the
phase 3 studies. In most of these studies, albiglutide could
1 2 6 ( 2 0 1 7 ) 2 3 0 2 3 9 231
be uptitrated to 50 mg, based on prespecified glycaemic crite-
ria, or was uptitrated by design (Table 1). The HARMONY pro-
gram also permitted the addition of hyperglycaemia rescue
medication for participants who did not meet glycaemic
goals, in line with US Food and Drug Administration (FDA)
guideline recommendations [18]. These studies were con-
ducted in accordance with the International Conference on
Harmonisation Good Clinical Practice guidelines and the Dec-
laration of Helsinki, and all participants provided written
informed consent.
2.2. Key inclusion and exclusion criteria
Detailed inclusion and exclusion criteria for the individual tri-
als have been reported elsewhere [815]. These studies
included men or women 18 years old, with HbA1c 7%10%
(5386 mmol/mol) (or HbA1c 7%10.5% [5391.5 mmol/mol]
for HARMONY 6) with T2DM that was inadequately controlled
with diet and exercise, OADs, or basal insulin. Exclusion crite-
ria included a history of cancer (except non-melanoma skin
cancers) not in remission for 3 years, treated diabetic gastro-
paresis, current symptomatic biliary disease, a history of pan-
creatitis, previous significant gastrointestinal (GI) surgery, or
recent clinically significant cardiovascular disease.
2.3. Statistical analysis
The safety population was defined as patients who received at
least one dose of study medication. Baseline demographics
and on-therapy adverse events (AEs) were tabulated using
descriptive statistics. Safety data were integrated and anal-
ysed irrespective of albiglutide dose (30 mg or 50 mg). Due to
the long half-life of albiglutide, on-therapy AEs included
events that occurred from the first dose of study medication
to within 56 days after the last dose of study medication.
AEs were coded using the Medical Dictionary for Regula-
tory Activities (MedDRA, version 14.0), summarised by Med-
DRA system organ class and preferred term. Select AEs of
interest were summarised using MedDRA query searches.
Hypoglycaemia AEs were characterised using 2013 American
Diabetes Association (ADA) criteria [19] and are summarised
prior to the addition of hyperglycaemia rescue medication
by individual study, due to the differing background antidia-
betic medications and associated risks of hypoglycaemia. Car-
diovascular and pancreatitis events were adjudicated by
respective independent adjudication committees described
elsewhere in detail [14,20,21].
3. Results
3.1. Study participants and treatment exposure
The integrated safety population comprised 4400 T2DM
patients who received at least one dose of study medication
(Table 2). Of the participants in these trials, 5 of which lasted
for 3 years (Table 1), 69.3% treated with albiglutide and 67.5%
treated with a comparator completed their respective
study treatment as planned. The most common reasons for
 232
Table 1 Phase 3 HARMONY program studies [815].

diabetes research and clinical practice

HARMONY Study design Background therapy Study treatments Safety population (n) Primary endpoint
Study (study-treatment
duration, weeks)

1 Double-blind, Pioglitazone metformin Albiglutide 30 mg 150 52 (156)

Reusch et al. [8] Placebo injection 151
2 Double-blind, None Albiglutide 30 mg 101 52 (156)
Nauck et al. [9] Albiglutide 30 mg ? 50 mg 99
Placebo injection 101
3 Double-blind, Metformin Albiglutide 30 mg ? 50 mg 302 104 (156)
Ahren et al. [10] Sitagliptin 100 mg 302
Glimepiride 2 mg ? 4 mg 307
Placebo injection 101
4 Open-label, Metformin sulphonylurea Albiglutide 30 mg ? 50 mg 504 52 (156)
Weissman [11] Insulin glargine 241
5 Double-blind, Metformin + glimepiride Albiglutide 30 mg ? 50 mg 271 52 (156)
Home [12] Pioglitazone 30 mg ? 45 mg 277

1 2 6 ( 2 0 1 7 ) 2 3 0 2 3 9
Placebo 115
6 Open-label Insulin oral antidiabetes drug Albiglutide 30 mg ? 50 mg 285 26 (52)
Rosenstock et al. [13] Insulin lispro 281
7 Open-label, Any combination of metformin Albiglutide 30 mg ? 50 mg 404 32 (32)
Pratley et al. [14] thiazolidinedione Liraglutide 0.6 mg ? 1.2 mg ? 1.8 mg 408
sulphonylurea
8a Double-blind, patients with Metformin thiazolidinedione Albiglutide 30 mg ? 50 mg once weekly 249 26 (52)
renal impairment sulphonylurea Sitagliptin 25 100 mg once daily 246
Leiter et al. [15]
a
Harmony 8 was not included in the integrated safety analysis.
 diabetes research and clinical practice 1 2 6 ( 2 0 1 7 ) 2 3 0 2 3 9 233

discontinuation from treatment in both groups were with-
drawal of consent by the patient and AEs (Table 2). Patient
demographics and baseline characteristics were comparable
between the albiglutide and the all-comparators groups
(Table 3). Approximately 58% of the patients received study
medication for >1 year and approximately 47% for >2 years.
Mean duration of exposure to study medication was similar
between albiglutide (91 weeks) and all comparators
(92 weeks) (Table 2). A similar percentage of patients in both
groups received glycaemic rescue medication (35% vs 36%,
for albiglutide and all comparators, respectively), most com-
monly insulin, SUs, or metformin.
3.2. Overall adverse events
The proportion of patients who experienced an on-therapy AE
or serious AE (SAE) was similar between the treatment groups
(Table 4). Overall, most AEs in either treatment group had a
maximum intensity of mild or moderate, with a similar pro-
portion in both treatment groups experiencing severe AEs.
The most frequently reported on-therapy AEs across the
two groups were in the Infections and Infestations MedDRA
system organ class and the GI disorders system organ class
(Table 5). The most frequently reported AEs by preferred term
(>10% in either the albiglutide or all-comparators group,
respectively) were upper respiratory tract infection (14.0%,
12.7%), diarrhoea (13.7%, 9.9%), and nausea (12.0%, 11.3%).
Diarrhoea (13.7%, 9.9%), injection-site reaction (9.0%, 2.0%),
and peripheral oedema (4.5%, 6.8%) were the only events by
preferred term with at least 2% difference between the
albiglutide and all-comparators groups.
Injection-site reaction was the most frequent AE by pre-
ferred term leading to withdrawal from treatment with
albiglutide (1.6% albiglutide vs 0% all comparators), followed
by nausea (0.6%, 0.5%), vomiting (0.4%, 0.2%), and diarrhoea
(0.3%, 0.4%).
No difference was observed in the total incidence of SAEs
between treatment groups (Table 4). Over the course of these
trials, no single SAE term was reported in >1.0% in either
treatment group (Table 5). More patients in the albiglutide
group reported SAEs of atrial fibrillation (10 vs 3 [0.5% vs
0.1%]), appendicitis (combined terms of appendicitis and per-
forated appendicitis; 6 vs 0 [0.3% vs 0%]), and pneumonia
(combined terms of pneumonia and lobar pneumonia; 15 vs
8 [0.7% vs 0.4%]) than those in the all-comparators group,
respectively.
There was no difference in the incidence of on-therapy
deaths between the treatment groups (22 [1.0%] albiglutide
and 23 [1.0%] all comparators). The MedDRA system organ
class with the highest incidence of deaths included neo-
plasms benign/malignant (7 [0.3%] albiglutide and 10 [0.4%]
all comparators) and cardiac disorders (6 [0.3%] each).
3.3. Adverse events of special interest
3.3.1. Gastrointestinal-related adverse events
GI events in the albiglutide group did not occur appreciably
more than in the all-comparators group. For events typically
associated with GLP-1 RAs such as diarrhoea, nausea, and
vomiting, only diarrhoea tended to be higher with albiglutide.
Most events were of mild to moderate intensity, with low and
similar discontinuation rates (Table 5). Overall, the incidence
of serious GI-related AEs was low (1.3% for albiglutide vs 1.1%
for all comparators).
3.3.2. Injection-site reactions
The potential for injection-site reactions exists for any
injected therapy, including biological agents such as albiglu-
tide. Injection-site reaction-related AEs were identified in
16% of albiglutide-treated patients versus 7% of patients
treated with all comparators. The most common MedDRA-
preferred terms of injection-site-reaction related AEs
were injection-site reactions (191 [9.0%] vs 46 [2.0%]),

Table 2 Patient disposition and exposure.

Albiglutide All comparators

(N = 2116) (N = 2284)

Patient disposition
Safety population (n) 2116 2284
Completed active treatment (%) 69.3 67.5
Discontinued active treatment (%) 30.7 32.5
% of patients on glycaemic rescue medication 35.0 36.1
Most common reasons (>1%) for discontinuing active treatment (%)
Patient withdrew consent from active participation 12.3 13.9
Adverse event 8.3 6.4
Lost to follow-up 3.6 4.4
Noncompliance 2.9 3.2
Protocol violation 1.3 1.4
Termination of study/site by GlaxoSmithKline 0.8 1.4
Duration of exposure
Mean (weeks) (standard deviation) 90.9 (59.28) 91.8 (59.68)
Median (weeks) (minimum, maximum) 74.8 (0, 158) 79.6 (0, 165)
Total person-years 4009.92 4367.00
234 diabetes research and clinical practice 1 2 6 ( 2 0 1 7 ) 2 3 0 2 3 9

Table 3 Demographics and baseline characteristics.

Albiglutide All comparators
(N = 2116) (N = 2284)

Mean age (years) (SD) 54.9 (9.86) 55.2 (9.77)

Mean baseline HbA1c (%) (SD) 8.22 (0.902) 8.21 (0.877)
Mean duration of diabetes (years) (SD) 8.13 (6.214) 7.88 (5.837)
Mean BMI (kg/m2) (SD) 33.02 (5.706) 32.74 (5.602)
Male, n (%) 1079 (51.0) 1197 (52.4)
Race/ethnicity, n (%)
Non-Hispanic White 1018 (48.1) 1085 (47.5)
Hispanic 544 (25.7) 660 (28.9)
Non-Hispanic African American 322 (15.2) 274 (12.0)
Asian 180 (8.5) 220 (9.6)
Other 52 (2.5) 45 (2.0)
Renal impairment by modification of diet in renal disease criteria, n (%)
Normal (eGFR  90 mL/min/1.73 m2) 804 (38.0) 902 (39.5)
Mild (60  eGFR < 90 mL/min/1.73 m2) 1139 (53.8) 1207 (52.8)
Moderate (30  eGFR < 60 mL/min/1.73 m2) 173 (8.2) 174 (7.6)
Severe (eGFR < 30 mL/min/1.73 m2) 0 1 (<0.1)
BMI, body mass index; eGFR, estimated glomerular filtration rate; SD, standard deviation.

Table 4 Overview of on-therapy adverse events.

Description Albiglutide All comparators
(N = 2116) (N = 2284)
n (%) n (%)

Any adverse event 1795 (84.8) 1880 (82.3)

Any serious adverse event 277 (13.1) 294 (12.9)
Any fatal serious adverse event 22 (1.0) 23 (1.0)
Any adverse event by maximum intensity
Mild 651 (30.8) 696 (30.5)
Moderate 827 (39.1) 863 (37.8)
Severe 314 (14.8) 319 (14.0)
Not available 3 (0.1) 2 (0.1)
Any adverse event leading to withdrawal 172 (8.1) 146 (6.4)

injection-site haematoma (55 [2.6%] vs 70 [3.1%]), injection-
site erythema (42 [2.0%] vs 12 [0.5%]), and injection-site rash
(30 [1.4%] vs 1 [<0.1%]) in albiglutide and all-comparators
groups, respectively.
Most injection-site reaction events were of mild intensity,
with no serious events, and resolved within 7 days, regardless
of treatment. Forty-four albiglutide-treated (2.1%) and 1
(<0.1%) all comparators-treated patient were withdrawn from
study treatment due to injection-site reaction-related AEs.
Although most patients who had an injection-site reaction
event were antibody-negative, a higher frequency of at least
1 injection-site reaction was observed among antibody-
positive patients (40%) compared with antibody-negative
patients (14%). There was no apparent correlation between
antibody titre and injection-site reaction event rate or reac-
tion severity. Similar to the overall population, most
injection-site reactions among antibody-positive patients
were mild or moderate, and there was not a higher proportion
leading to discontinuation.
3.3.3. Hypoglycaemia
As described in the Methods section, hypoglycaemia is sum-
marised prior to the addition of rescue medication and by
individual study. Severe hypoglycaemia, requiring the assis-
tance of another person for treatment [22], occurred uncom-
monly in the albiglutide HARMONY studies: 0%1.3% among
patients receiving albiglutide and 0%1.1% among patients
receiving a comparator in the individual HARMONY studies
[814]. Most patients with severe hypoglycaemic events in
clinical studies were receiving concurrent SU or insulin.
When albiglutide was used as monotherapy, the incidence
of symptomatic hypoglycaemia (FPG  3.9 mmol/l) was simi-
lar for albiglutide 30 mg (1%), albiglutide 50 mg (0%), and pla-
cebo (2%) [9]. The rate of symptomatic hypoglycaemia was
higher for albiglutide when used in combination with an SU
(10%19%) or with insulin (19%) compared to combinations
not including an SU or insulin (1%3%). Among patients ran-
domised to the all-comparators group, the incidence of symp-
tomatic hypoglycaemia was 8%33% when used with an SU or
insulin and 2%4% in combinations without these
medications.
3.3.4. Pancreatitis adverse events
Acute pancreatitis has been investigated in association with
GLP-1 RAs [23]. At the conclusion of the phase 3 albiglutide
programme including all 8 studies, a total of 8 patients
 diabetes research and clinical practice 1 2 6 ( 2 0 1 7 ) 2 3 0 2 3 9 235

Table 5 Adverse events, serious adverse events, and gastrointestinal adverse events by system organ class.
System organ class Albiglutide All comparators
(N = 2116) (N = 2284)
n (%) n (%)

Adverse events by system organ class

Any event 1795 (84.8) 1880 (82.3)
Infections and infestations 1064 (50.3) 1132 (49.6)
Gastrointestinal disorders 846 (40.0) 788 (34.5)
Musculoskeletal and connective tissue disorders 614 (29.0) 667 (29.2)
General disorders and administration-site conditions 563 (26.6) 492 (21.5)
Nervous system disorders 484 (22.9) 495 (21.7)
Respiratory, thoracic, and mediastinal disorders 361 (17.1) 395 (17.3)
Injury, poisoning, and procedural complications 348 (16.4) 391 (17.1)
Skin and subcutaneous tissue disorders 324 (15.3) 346 (15.1)
Eye disorders 291 (13.8) 281 (12.3)
Metabolism and nutrition disorders 274 (12.9) 293 (12.8)
Vascular disorders 270 (12.8) 262 (11.5)
Investigations 256 (12.1) 287 (12.6)
Psychiatric disorders 180 (8.5) 185 (8.1)
Cardiac disorders 160 (7.6) 135 (5.9)
Renal and urinary disorders 134 (6.3) 146 (6.4)
Neoplasms benign, malignant and unspecified (including cysts and polyps) 105 (5.0) 124 (5.4)
Blood and lymphatic system disorders 104 (4.9) 143 (6.3)
Ear and labyrinth disorders 94 (4.4) 82 (3.6)
Reproductive system and breast disorders 89 (4.2) 102 (4.5)
Immune system disorders 58 (2.7) 61 (2.7)
Endocrine disorders 39 (1.8) 44 (1.9)
Hepatobiliary disorders 32 (1.5) 47 (2.1)
Congenital, familial, and genetic disorders 4 (0.2) 3 (0.1)
Social circumstances 3 (0.1) 2 (0.1)
Surgical and medical procedures 3 (0.1) 7 (0.3)
Pregnancy, puerperium, and perinatal conditions 0 2 (0.1)
Serious adverse events occurring in 4 patients in either group
Any event 277 (13.1) 294 (12.9)
Chest pain 15 (0.7) 19 (0.8)
Coronary artery disease 12 (0.6) 20 (0.9)
Pneumonia 11 (0.5) 7 (0.3)
Atrial fibrillation 10 (0.5) 3 (0.1)
Acute myocardial infarction 9 (0.4) 13 (0.6)
Myocardial infarction 9 (0.4) 4 (0.2)
Unstable angina 8 (0.4) 8 (0.4)
Transient ischaemic attack 8 (0.4) 5 (0.2)
Noncardiac chest pain 7 (0.3) 4 (0.2)
Osteoarthritis 7 (0.3) 10 (0.4)
Cerebrovascular accident 6 (0.3) 5 (0.2)
Gastroenteritis 6 (0.3) 3 (0.1)
Appendicitis 5 (0.2) 0
Cardiac congestive failure 5 (0.2) 12 (0.5)
Asthma 4 (0.2) 1 (<0.1)
Cellulitis 4 (0.2) 12 (0.5)
Lobar pneumonia 4 (0.2) 1 (<0.1)
Osteomyelitis 4 (0.2) 1 (<0.1)
Nephrolithiasis 3 (0.1) 5 (0.2)
Breast cancer 2 (0.1) 5 (0.2)
Angioedema 1 (<0.1) 4 (0.2)
Acute cholecystitis 1 (<0.1) 7 (0.3)
Hypoglycaemia 1 (<0.1) 6 (0.3)
Back pain 0 7 (0.3)
Peripheral vascular disorder 0 4 (0.2)
(continued on next page)
236 diabetes research and clinical practice
Table 5 (continued)
System organ class
On-therapy gastrointestinal events
Any gastrointestinal adverse event
Gastrointestinal serious adverse event
Gastrointestinal adverse event withdrawal
Gastrointestinal adverse events >5%
Diarrhoea
Nausea
Vomiting
Constipation
(6 albiglutide [0.3%] and 2 all comparators [0.1%, both on the
GLP-1 RA liraglutide]) were adjudicated by the Pancreatitis
Adjudication Committee as having definite or probable pan-
creatitis with at least a possible relationship to study medica-
tion [21].
3.3.5. Cardiovascular events
A cardiovascular meta-analysis performed across 9 studies (8
phase 3 studies [Table 1] and 1 phase 2b study [24]) in the
albiglutide clinical program showed no excess hazard for
adjudicated major adverse cardiovascular event (MACE;
non-fatal myocardial infarction, non-fatal stroke, or cardio-
vascular death) or hospitalization for unstable angina for
albiglutide versus all comparators (hazard ratio [HR] 1.0; 95%
confidence interval [CI] 0.68, 1.49; p = 0.983) [20]. Hypertension
was the most commonly reported cardiovascular event in
both groups (7.9% albiglutide vs 8.3% all comparators). Atrial
fibrillation, chest pain, coronary artery disease, and periph-
eral oedema were other events reported in >1.0% of patients
in either group (but each in <2% of patients). There were more
albiglutide-treated patients than all comparator treated
patients who experienced at least 1 on-therapy AE of atrial
fibrillation or atrial flutter (1.4% vs 0.6%) [20].
Over 3 years, change from baseline in heart rate at any
visit with albiglutide was approximately 12 beats higher
than with all comparators. There were no meaningful
changes in blood pressure with albiglutide (range of mean
change [mmHg] systolic: 0.7 to 1.9; diastolic 0.7 to 0.1) or
all comparators (range of mean change systolic: 0.6 to 2.3;
diastolic 1.0 to 0.3).
3.3.6. Liver events
Elevations of gamma-glutamyltransferase (GGT) and alanine
aminotransferase (ALT) were the most common on-therapy
liver AEs, with a higher incidence of GGT elevations in the
albiglutide versus all-comparators group (1.2% vs 0.7%). How-
ever, the proportion of patients with elevations 3x upper
limit of normal was similar in the albiglutide and all-
comparators groups. ALT elevations were 0.7% in both groups.
3.3.7. Thyroid cancer events
In the integrated phase 3 population, few patients were diag-
nosed with thyroid cancer. Two patients were diagnosed with
medullary thyroid cancer (1 on albiglutide and 1 in the all-
comparator group on placebo); both had an elevated
1 2 6 ( 2 0 1 7 ) 2 3 0 2 3 9
Albiglutide All comparators
(N = 2116) (N = 2284)
n (%) n (%)
846 (40.0) 788 (34.5)
27 (1.3) 25 (1.1)
40 (1.9) 40 (1.8)
290 (13.7) 226 (9.9)
254 (12.0) 258 (11.3)
112 (5.3) 108 (4.7)
106 (5.0) 94 (4.1)
calcitonin level at baseline, indicating a pre-existing condi-
tion. Three cases were diagnosed as follicular (1 on albiglu-
tide, 1 on sitagliptin) and/or papillary (1 on sitagliptin)
thyroid cancer. In routine annual monitoring of calcitonin, a
comparable number of patients had a result 50 pg/mL
(0.4% in each group) or 100 pg/mL (0.2% in each group).
3.3.8. Systemic allergic reactions
There were no notable differences in the incidence of hyper-
sensitivity events (17% vs 15.2%) or discontinuation (0.5% vs
0.4%) associated with these potential events between albiglu-
tide and all-comparators groups, respectively. The incidence
of SAEs identified as potential hypersensitivity events were
reported in 0.4% of patients in each group, including events
of asthma, angioedema, and one AE of anaphylaxis in an
albiglutide-treated patient.
Overall, 5.9% of albiglutide-treated patients tested positive
for antialbiglutide antibodies. None were neutralizing, except
in one patient whose antialbiglutide antibodies existed before
treatment initiation. AEs were generally similar between
patients who were antibody-positive and antibody-negative,
although a higher percentage of patients who were
antibody-positive had a systemic allergic reaction compared
with patients who were antibody-negative (2.5% vs 1.6%).
3.3.9. Renal adverse events
Over 60% of patients across the integrated phase 3 studies
had mild or moderate renal impairment (estimated glomeru-
lar filtration rate [eGFR]  30 and <90 mL/min/1.73 m2 at base-
line) (Table 3). AEs in the renal organ class occurred in a
similar proportion of patients in the albiglutide and all-
comparators groups (6.3% vs 6.4%) An equal number of these
T2DM patients experienced AEs of renal failure and impair-
ment (1.0% vs 0.9%), respectively.
When evaluated by eGFR subgroup, there was a higher
event rate for GI AEs in patients treated with albiglutide
who had moderate renal impairment (48.0% albiglutide vs
33.9% all comparators) compared to those treated with
albiglutide who had either mild renal impairment (40.6%
albiglutide vs 35.5% all comparators) or normal renal function
(37.3% albiglutide vs 33.4% all comparators). There was a sim-
ilar trend of increase in nausea AEs as renal function declined
in the albiglutide and all-comparators groups (moderate:
16.8% vs 16.1%; mild: 12.5% vs 11.8%; and normal: 10.3% vs
9.6% with albiglutide vs all comparators, respectively). An
 diabetes research and clinical practice
evaluation of the safety profile of albiglutide in a dedicated
trial specifically in patients with mild or moderate renal
impairment has been published previously [15].
4. Discussion
This report presents a large integrated safety analysis of the
GLP-1 RA albiglutide, from seven studies encompassing 4400
patients with T2DM followed for up to 3 years. The analysis
includes patients across the spectrum of T2DM disease pro-
gression, which mimics real-world treatment paradigms.
Comparison of the albiglutide 30-mg arm with the 50-mg
arm suggested no significant changes in safety profile due
to dose escalation of albiglutide [9]; therefore safety data were
integrated and analysed irrespective of dose (30 mg or 50 mg).
Overall, patients treated with albiglutide had a similar
incidence of AEs when compared with the pooled all-
comparators group. The most common AEs with albiglutide
were GI and injection-site reactions, which were generally
mild and transient. These events were also the most common
reason for withdrawal from albiglutide treatment, although
the withdrawal rate for GI-related events was similar between
albiglutide and comparators.
Treatment with GLP-1 RAs is known to be associated with
GI AEs such as diarrhoea, nausea, and vomiting [2527]. The
present finding that only diarrhoea tended to be higher with
albiglutide compared to the comparators, but not nausea
and vomiting, suggests that albiglutide has a lower risk for
GI AEs than other GLP-1 RAs. This may be explained by the
larger size of the molecule, which reduces the ability of
albiglutide to enter the brain through the blood-brain barrier
[28]. Albiglutide may not have direct contact with areas of the
brain thought to mediate nausea-related AEs, such as the area
postrema [14,29]. More detailed analysis of GI AEs with
albiglutide treatment has recently been published [30].
Pancreatic safety is another potential area of concern with
regard to GLP-1 RAs. The FDA and the European Medicines
Agency (EMA) have evaluated pancreatic safety with
incretin-based drugs because there is an ongoing discussion
whether treatment is associated with increased risk for pan-
creatitis [23]. Previous analyses with other GLP-1 RAs have
demonstrated no significant increased risk for pancreatitis,
although in several studies a non-significant trend has been
observed [31]. Overall, the albiglutide pancreatitis event rate
was comparable to rates reported from clinical trials of other
approved GLP-1 RAs and within the range reported from
observational studies of patients with T2DM [21,32,33]. How-
ever, each of the few adjudicated cases of definite or probable
acute pancreatitis at least possibly related to study treatment
in the albiglutide HARMONY phase 3 trials occurred in a
patient receiving a GLP-1 RA [21]. Ongoing studies will further
assess the pancreatic safety of GLP-1 RAs.
The FDA and the EMA require that before being adopted
into clinical practice, all new drug treatments for diabetes
mellitus must demonstrate no substantial increase in cardio-
vascular risk. Albiglutide studies showed no excess hazard of
MACE with albiglutide versus all comparators, which is simi-
lar to other GLP-1 RAs [34]. Treatment with GLP-1 RAs
increases heart rate, and a small exenatide study suggests
1 2 6 ( 2 0 1 7 ) 2 3 0 2 3 9 237
involvement of the sympathetic nervous system in this phe-
nomenon [35]. Reflex tachycardia could explain the increase
seen in the combined analysis of albiglutide studies of atrial
fibrillation or atrial flutter with albiglutide compared with
the all-comparators group, albeit at very low frequencies in
both groups. Overall, the meta-analysis suggested cardiovas-
cular safety with albiglutide [20]. A conclusion awaits results
of the ongoing albiglutide dedicated cardiovascular outcomes
study [ClinicalTrials.gov NCT02465515].
The HARMONY studies also showed that the risk for hypo-
glycaemia is very low with albiglutide, which can be
explained by the glucose dependency of its effects [1]. Fur-
thermore, no new safety signal was found after long-term
use of albiglutide in patients with T2DM and mild to moderate
renal impairment.
Limitations of these analyses include the fact that, even
considering the large database, the number of the patients
experiencing some AEs of special interest were small. In addi-
tion, the integrated analysis of trials did not evaluate differ-
ences in AE frequencies between albiglutide and the
separate classes of comparator treatments (eg, incretin-
targeted and nonincretin-targeted). We also did not evaluate
differences in the 3-year results of the many subgroups of
patients characterized by, for example, age, duration of dia-
betes, or ethnicity, although ethnic differences in responsive-
ness to GLP-1 RAs are suggested by comparison of studies of
predominantly Asian populations versus non-Asian popula-
tions [36].
In summary, in this integrated analysis of seven phase 3
clinical studies of up to 3 years duration, the safety and toler-
ability profile of the once-weekly GLP-1 RA albiglutide was
compared with that of the pooled series of alternative treat-
ments, including T2DM medications and placebo. The two
profiles were generally similar. The groups did not differ in
cardiovascular or renal safety. Hypoglycaemia was low with
albiglutide except when it was coadministered with an SU
or insulin. Albiglutide had an increased risk of diarrhoea
and injection-site-related AEs compared with the group of
all comparators, but the frequencies of nausea and vomiting
did not differ. The data presented here suggest that a large-
size GLP-1 RA, such as albiglutide, is associated with a low
risk for GI AEs. These observations are useful in the context
of the individualised and patient-centred approach for clini-
cians to select treatments for their patients that has been rec-
ommended by the ADA and the European Association for the
Study of Diabetes [37].
Funding sources
Funding for this work and the related studies was provided by
GlaxoSmithKline: ClinicalTrials.gov NCT00849056;
NCT00849017; NCT00838903; NCT00838916; NCT00839527;
NCT00976391; NCT01128894; NCT01098539.
Conflict of interest
BA has received honoraria for lecturing and/or consulting
from GlaxoSmithKline (GSK), Merck, Sharp and Dohme
(MSD), Novartis, Novo Nordisk, and Sanofi, all of which are
238 diabetes research and clinical practice
companies producing therapies based on GLP-1 or DPP-4.
MCC, KM, and JM are employees of and hold stock in GSK.
CP is an employee of PPD Inc, Morrisville, NC, USA. MR has
received research support from GSK. TW is an employee of
PAREXEL International, Durham, NC, USA, and was an
employee of GSK during the conduct of all trials and develop-
ment of this manuscript. He holds stock in GSK. SJ is an
employee of Pfizer Pharmaceuticals Inc, New York, NY, USA,
and was an employee of GSK during the conduct of all trials
and development of this manuscript.
Authorship contributions
All authors meet the criteria set forth by the International
Committee of Medical Journal Editors. BA analysed and inter-
preted the data; MCC and KM analysed and interpreted the
data and wrote sections of the manuscript; CP contributed
to the acquisition of data, and analysed and interpreted the
data; MR analysed and interpreted the data; JMM, THW, and
SJ contributed to the concept and design of the study, con-
tributed to the acquisition of data, analysed and interpreted
the data; JMM wrote the initial draft of the manuscript. Each
author commented critically on the content, approved the
manuscript for submission and assume responsibility for
the direction and content.
Acknowledgments
The authors thank Douglas L. Wicks, MPH, CMPP of
GlaxoSmithKline, Collegeville, PA, USA, for the management
of the manuscript development. Editorial support was pro-
vided by William Ho, PhD, and Joelle Suchy, PhD, (assistance
with the production of draft outline, first draft, referencing,
assembling tables and figures, and collating author com-
ments), of MediTech Media, Hamilton, NJ, USA, and Elizabeth
Rosenberg, PhD, of AOI Communications, L.P. Exton, PA, USA,
(assistance with manuscript revisions, fact checking and
preparing the final draft for submission) and was funded by
GSK.
R E F E R E N C E S
[1] Nauck MA, Meier JJ. The incretin effect in healthy individuals
and those with type 2 diabetes: physiology, pathophysiology,
and response to therapeutic interventions. Lancet Diabetes
Endocrinol 2016;4:52536. http://dx.doi.org/10.1016/S2213-
8587(15)00482-9.
[2] Bush MA, Matthews JE, De Boever EH, Dobbins RL, Hodge RJ,
Walker SE, et al. Safety, tolerability, pharmacodynamics and
pharmacokinetics of albiglutide, a long-acting glucagon-like
peptide-1 mimetic, in healthy subjects. Diabetes Obes Metab
2009;11:498505. http://dx.doi.org/10.1111/j.1463-
1326.2008.00992.x.
[3] Rosenstock J, Reusch J, Bush M, Yang F, Stewart M, Albiglutide
Study Group. Potential of albiglutide, a long-acting GLP-1
receptor agonist, in type 2 diabetes: a randomized controlled
trial exploring weekly, biweekly, and monthly dosing.
Diabetes Care 2009;32:18806. http://dx.doi.org/10.2337/dc09-
0366.
1 2 6 ( 2 0 1 7 ) 2 3 0 2 3 9
[4] Young MA, Wald JA, Matthews JE, Scott R, Hodge RJ, Zhi H,
et al. Clinical pharmacology of albiglutide, a GLP-1 receptor
agonist. Postgrad Med 2014;126:8497. http://dx.doi.org/
10.3810/pgm.2014.11.2836.
[5] Matthews JE, Stewart MW, De Boever EH, Dobbins RL, Hodge
RJ, Walker SE, et al. Pharmacodynamics, pharmacokinetics,
safety, and tolerability of albiglutide, a long-acting glucagon-
like peptide-1 mimetic, in patients with type 2 diabetes. J Clin
Endocrinol Metab 2008;93:48107. http://dx.doi.org/10.1210/
jc.2008-1518.
[6] Ahren B. Islet G protein-coupled receptors as potential targets
for treatment of type 2 diabetes. Nat Rev Drug Discov
2009;8:36985. http://dx.doi.org/10.1038/nrd2782.
[7] Rosenstock J, Stewart MW. Albiglutide. Drugs Fut
2010;35:70112. http://dx.doi.org/10.1358/
dof.2010.035.09.1516042.
[8] Reusch J, Stewart MW, Perkins CM, Cirkel DT, Ye J, Perry CR,
et al. Efficacy and safety of once-weekly glucagon-like
peptide 1 receptor agonist albiglutide (HARMONY 1 trial): 52-
week primary endpoint results from a randomized, double-
blind, placebo-controlled trial in patients with type 2 diabetes
mellitus not controlled on pioglitazone, with or without
metformin. Diabetes Obes Metab 2014;16:125764. http://dx.
doi.org/10.1111/dom.12382.
[9] Nauck MA, Stewart MW, Perkins C, Jones-Leone A, Yang F,
Perry C, et al. Efficacy and safety of once-weekly GLP-1
receptor agonist albiglutide (HARMONY 2): 52 week primary
endpoint results from a randomised, placebo-controlled trial
in patients with type 2 diabetes mellitus inadequately
controlled with diet and exercise. Diabetologia
2016;59:26674. http://dx.doi.org/10.1007/s00125-015-3795-1.
[10] Ahren B, Johnson SL, Stewart M, Cirkel DT, Yang F, Perry C,
et al. HARMONY 3: 104-week randomized, double-blind,
placebo- and active-controlled trial assessing the efficacy
and safety of albiglutide compared with placebo, sitagliptin,
and glimepiride in patients with type 2 diabetes taking
metformin. Diabetes Care 2014;37:21418. http://dx.doi.org/
10.2337/dc14-0024.
[11] Weissman PN, Carr MC, Ye J, Cirkel DT, Stewart M, Perry C,
et al. HARMONY 4: randomised clinical trial comparing once-
weekly albiglutide and insulin glargine in patients with type
2 diabetes inadequately controlled with metformin with or
without sulfonylurea. Diabetologia 2014;57:247584. http://
dx.doi.org/10.1007/s00125-014-3360-3.
[12] Home PD, Shamanna P, Stewart M, Yang F, Miller M, Perry C,
et al. Efficacy and tolerability of albiglutide versus placebo or
pioglitazone over 1 year in people with type 2 diabetes
currently taking metformin and glimepiride: HARMONY 5.
Diabetes Obes Metab 2015;17:17987. http://dx.doi.org/
10.1111/dom.12414.
[13] Rosenstock J, Fonseca VA, Gross JL, Ratner RE, Ahren B, Chow
FC, et al. Advancing basal insulin replacement in type 2
diabetes inadequately controlled with insulin glargine plus
oral agents: a comparison of adding albiglutide, a weekly
GLP-1 receptor agonist, versus thrice-daily prandial insulin
lispro. Diabetes Care 2014;37:231725. http://dx.doi.org/
10.2337/dc14-0001.
[14] Pratley RE, Nauck MA, Barnett AH, Feinglos MN, Ovalle F,
Harman-Boehm I, et al. Once-weekly albiglutide versus once-
daily liraglutide in patients with type 2 diabetes inadequately
controlled on oral drugs (HARMONY 7): a randomised, open-
label, multicentre, non-inferiority phase 3 study. Lancet
Diabetes Endocrinol 2014;2:28997. http://dx.doi.org/10.1016/
S2213-8587(13)70214-6.
[15] Leiter LA, Carr MC, Stewart M, Jones-Leone A, Scott R, Yang F,
et al. Efficacy and safety of the once-weekly GLP-1 receptor
agonist albiglutide versus sitagliptin in patients with type 2
diabetes and renal impairment: a randomized phase III study.
 diabetes research and clinical practice
Diabetes Care 2014;37:272330. http://dx.doi.org/10.2337/
dc13-2855.
[16] Tanzeum [package insert] Research Triangle Park, NC, USA: [27]
GlaxoSmithKline LLC; 2015. http://
www.diabetesincontrol.com/wp-content/uploads/2014/09/
www.gsksource.com_gskprm_htdocs_documents_
TANZEUM-PI-MG-IFU-COMBINED.PDF.
[17] Eperzan [Summary of Product Characteristics] United
Kingdom: GlaxoSmithKline LLC; 2014. http://www.ema. [28]
europa.eu/docs/en_GB/document_library/EPAR_-_Product_
Information/human/002735/WC500165117.pdf.
[18] Guidance for Industry Diabetes mellitus: Developing drugs
and therapeutic biologics for treatment and prevention.
Rockville, MD, USA: U.S. FDA Center for Drug Evaluation and [29]
Research; 2008. http://www.fda.gov/downloads/Drugs/.../
Guidances/ucm071624.pdf. Accessed June 22, 2016.
[19] Seaquist ER, Anderson J, Childs B, Cryer P, Dagogo-Jack S, Fish
L, et al. Hypoglycemia and diabetes: a report of a workgroup [30]
of the American Diabetes Association and the Endocrine
Society. Diabetes Care 2013;36:138495. http://dx.doi.org/
10.2337/dc12-2480.
[20] Fisher M, Petrie MC, Ambery PD, Donaldson J, Ye J, McMurray [31]
JJ. Cardiovascular safety of albiglutide in the Harmony
programme: a meta-analysis. Lancet Diabetes Endocrinol
2015;3:697703. http://dx.doi.org/10.1016/S2213-8587(15) [32]
00233-8.
[21] Johnson SL, Anderson M, Enns R, et al. Incidence of
adjudicated acute pancreatitis across the albiglutide phase III [33]
clinical program. Poster THR-645 presented at: 97th Annual
Meeting and Expo of the Endocrine Society; March 58, 2015;
San Diego, CA.
[22] American Diabetes Association Workgroup on
Hypoglycemia. Defining and reporting hypoglycemia in [34]
diabetes: a report from the American Diabetes Association
Workgroup on Hypoglycemia. Diabetes Care 2005;28:12459.
http://dx.doi.org/10.2337/diacare.28.5.1245.
[23] Egan AG, Blind E, Dunder K, de Graeff PA, Hummer BT, [35]
Bourcier T, et al. Pancreatic safety of incretin-based drugs
FDA and EMA assessment. N Engl J Med 2014;370:7947.
http://dx.doi.org/10.1056/NEJMp1314078.
[24] Seino Y, Inagaki N, Miyahara H, Okuda I, Bush M, Ye J, et al. A
randomized dose-finding study demonstrating the efficacy [36]
and tolerability of albiglutide in Japanese patients with type 2
diabetes mellitus. Curr Med Res Opin 2014;30:1095106.
http://dx.doi.org/10.1185/03007995.2014.896327.
[25] Aroda VR, Ratner R. The safety and tolerability of GLP-1
receptor agonists in the treatment of type 2 diabetes: a [37]
review. Diabetes Metab Res Rev 2011;27:52842. http://dx.doi.
org/10.1002/dmrr.1202.
[26] Ahmann A, Rodbard HW, Rosenstock J, Lahtela JT, de Loredo
L, Tornoe K, et al. Efficacy and safety of liraglutide versus
placebo added to basal insulin analogues (with or without
metformin) in patients with type 2 diabetes: a randomized,
1 2 6 ( 2 0 1 7 ) 2 3 0 2 3 9 239
placebo-controlled trial. Diabetes Obes Metab
2015;17:105664. http://dx.doi.org/10.1111/dom.12539.
Jendle J, Grunberger G, Blevins T, Giorgino F, Hietpas RT,
Botros FT. Efficacy and safety of dulaglutide in the treatment
of type 2 diabetes: a comprehensive review of the dulaglutide
clinical data focusing on the AWARD phase 3 clinical trial
program. Diabetes Metab Res Rev 2016. http://dx.doi.org/
10.1002/dmrr.2810.
Hottenstein CS, Szapacs ME, Maier CC. Investigation of
blood-brain barrier penetration of albiglutide in mice
[abstract]. Diabetes Care 2016;65(supple1):A221360. http://
app.core-apps.com/tristar_ada16/abstract/
92b0eac0f984437a5bebc08c57516262.
Kenward H, Pelligand L, Savary-Bataille K, Elliott J. Nausea:
current knowledge of mechanisms, measurement and
clinical impact. Vet J 2015;203:3643. http://dx.doi.org/
10.1016/j.tvjl.2014.10.007.
Leiter LA, Mallory JM, Wilson TH, Reinhardt RR.
Gastrointestinal safety across the albiglutide development
programme. Diabetes Obes Metab 2016;18:9305. http://dx.
doi.org/10.1111/dom.12679.
Meier JJ, Nauck MA. Risk of pancreatitis in patients treated
with incretin-based therapies. Diabetologia 2014;57:13204.
http://dx.doi.org/10.1007/s00125-014-3231-y.
Anderson SL, Trujillo JM. Association of pancreatitis with
glucagon-like peptide-1 agonist use. Ann Pharmacother
2010;44:9049. http://dx.doi.org/10.1345/aph.1M676.
Noel RA, Braun DK, Patterson RE, Bloomgren GL. Increased
risk of acute pancreatitis and biliary disease observed in
patients with type 2 diabetes: a retrospective cohort study.
Diabetes Care 2009;32:8348. http://dx.doi.org/10.2337/dc08-
1755.
Tahrani AA, Barnett AH, Bailey CJ. Pharmacology and
therapeutic implications of current drugs for type 2 diabetes
mellitus. Nat Rev Endocrinol 2016;12:56692. http://dx.doi.
org/10.1038/nrendo.2016.86.
Smits MM, Muskiet MH, Tonneijck L, Hoekstra T, Kramer MH,
Diamant M, et al. Exenatide acutely increases heart rate in
parallel with augmented sympathetic nervous system
activation in healthy overweight males. Br J Clin Pharmacol
2016;81:61320. http://dx.doi.org/10.1111/bcp.12843.
Kim YG, Hahn S, Oh TJ, Park KS, Cho YM. Differences in the
HbA1c-lowering efficacy of glucagon-like peptide-1
analogues between Asians and non-Asians: a systematic
review and meta-analysis. Diabetes Obes Metab
2014;16:9009. http://dx.doi.org/10.1111/dom.12293.
Inzucchi SE, Bergenstal RM, Buse JB, Diamant M, Ferrannini E,
Nauck M, et al. Management of hyperglycemia in type 2
diabetes, 2015: a patient-centered approach: update to a
position statement of the American Diabetes Association and
the European Association for the Study of Diabetes. Diabetes
Care 2015;38:1409. http://dx.doi.org/10.2337/dc14-2441.