European Journal of Heart Failure (2012) 14, 992–999

doi:10.1093/eurjhf/hfs092

Pentraxin-3 in chronic heart failure: the

CORONA and GISSI-HF trials
Roberto Latini 1*, Lars Gullestad 2,3, Serge Masson 1, Ståle Haugset Nymo 4,
Thor Ueland 4, Ivan Cuccovillo 5, Mari Vårdal 6, Barbara Bottazzi 5,
Alberto Mantovani 5,7, Donata Lucci8, Nobuhito Masuda 9, Yukio Sudo 9,
John Wikstrand 10, Gianni Tognoni 11, Pål Aukrust3,12, and Luigi Tavazzi13, on behalf of
the Investigators of the Controlled Rosuvastatin Multinational Trial in Heart Failure
(CORONA) and GISSI-Heart Failure (GISSI-HF) trials
1
Department of Cardiovascular Research, Istituto di Ricerche Farmacologiche Mario Negri, via Giuseppe La Masa 19, 20156 Milan, Italy; 2Department of Cardiology, Oslo University
Hospital, Rikshospitalet, Oslo, Norway; 3Faculty of Medicine, University of Oslo, Oslo, Norway; 4Research Institute for Internal Medicine, Rikshospitalet, Oslo, Norway; 5Istituto
Clinico Humanitas, Rozzano, Italy; 6Department of Biostatistics, Oslo University Hospital Rikshospitalet, Oslo, Norway; 7Department of Translational Medicine, Università degli Studi
di Milano, Milan, Italy; 8ANMCO Research Center, Florence, Italy; 9Perseus Proteomics, Tokyo, Japan; 10Wallenberg Laboratory for Cardiovascular Research, Sahlgrenska Academy,
Gothenburg University, Gothenburg, Sweden; 11Consorzio Mario Negri Sud, Santa Maria Imbaro, Italy; 12Section of Clinical Immunology and Infectious Diseases, Oslo University
Hospital Rikshospitalet, Oslo, Norway; and 13GVM Hospitals of Care and Research, Ettore Sansavini Health Science Foundation, Cotignola, Italy

Received 11 April 2012; revised 16 May 2012; accepted 17 May 2012; online publish-ahead-of-print 27 June 2012

Aims Pentraxin-3 (PTX3) is a component of the humoral arm of innate immunity which can regulate inflammatory pro-
cesses. Since the role of inflammation in the progression of chronic heart failure (HF) is debated, we investigated
the prognostic value of PTX3 and the effect of a statin in two large populations of patients with HF.
.....................................................................................................................................................................................
Methods Plasma levels of PTX3 were measured at randomization and after 3 months in 1457 patients enrolled in the Con-
and results trolled Rosuvastatin Multinational Trial in HF (CORONA) and 1233 patients enrolled in the GISSI-Heart Failure
trial (GISSI-HF). The relationships between baseline PTX3 levels or their changes over time and mortality were eval-
uated with multivariable Cox proportional hazard models including clinical factors, high sensitivity C-reactive protein
(hsCRP), and N-terminal pro brain natriuretic peptide (NT-proBNP). PTX3 concentration [median (Q1 –Q3) ¼ 5.34
(3.55–7.64) ng/mL, n ¼ 2690] was higher in females, in older patients, and those with lower body mass index. Base-
line elevated PTX3 was associated with a higher risk of all-cause mortality [759 events, hazard ratio (HR) for 1 SD
increase 1.20, 95% confidence interval (CI) 1.12–1.30, P , 0.0001], cardiovascular mortality (587 events, HR 1.27,
95% CI 1.17–1.38, P , 0.0001), or hospitalization for worsening HF (720 events, HR 1.21, 95% CI 1.12–1.30,
P , 0.0001), and marginally improved discrimination. Three-month changes in PTX3 were associated with fatal
events after adjustment for hsCRP or NT-proBNP. Rosuvastatin lowered hsCRP levels but significantly raised PTX3.
.....................................................................................................................................................................................
Conclusion In two independent clinical trials that enrolled patients with chronic HF, PTX3 was consistently associated with
outcomes. The opposite effects of a statin on hsCRP and PTX3 call for further investigation.
.....................................................................................................................................................................................
Trial NCT00336336 (GISSI-HF), NCT00206310 (CORONA).
registration
-----------------------------------------------------------------------------------------------------------------------------------------------------------
Keywords Heart failure † Immune system † Prognosis

controversial research. While it is accepted that expression and

Introduction release of circulating inflammatory molecules are related to more
The role of inflammation in the progression and prognostic stratifi- severe HF and worse outcomes,1 – 6 it is still debated whether inflam-
cation of heart failure (HF) is a topic of active and sometimes mation is a cause or just a consequence of the disease.7 – 9

* Corresponding author. Tel: +39 239014454, Fax: +39 0233200049, Email: roberto.latini@marionegri.it
Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2012. For permissions please email: journals.permissions@oup.com.
Pentraxin-3 in chronic HF
C-reactive protein (CRP) is the most established marker of in-
flammation and provides independent prognostic information in
various cardiovascular disorders.10 Post-hoc analysis of the Con-
trolled Rosuvastatin Multinational Trial in HF trial (CORONA)
showed that CRP decreased significantly in 3 months with rosuvas-
tatin,11 and the drug improved outcomes in patients with higher
CRP at baseline.12 We hypothesized that another pentraxin
(pentraxin-3 or PTX3) which, unlike CRP, is produced at the site
of inflammation, including the inflamed vascular endothelium,13
might add information on the role of inflammation in chronic HF.
Pentraxin-3 is related to outcomes in acute myocardial infarc-
tion independently of other accepted risk markers such as tropo-
nin, natriuretic peptides, and CRP.14 In older adults from the
community, PTX3 is associated with cardiovascular disease and all-
cause mortality.15 Its prognostic role in chronic HF has been
reported in studies of no more than 200 patients.16 – 18 Since the
standard for validation of biomarkers should include an adequate
sample size, testing in large independent trials and assessment of
incremental value, PTX3 was assayed in 1233 patients from the
GISSI-Heart Failure trial (GISSI-HF)19,20 and 1457 patients from
CORONA.11 Since in both trials patients were randomly assigned
to rosuvastatin or placebo, this seemed an ideal setting also for
exploring the effects of a statin on PTX3 in chronic HF.
Methods
GISSI-HF and CORONA were two randomized, double-blind,
placebo-controlled, multicentre trials that enrolled 6975 and 5011
patients with chronic HF, respectively.11,19,20 Criteria for eligibility
were different: in GISSI-HF, patients had clinical evidence of stable
HF, irrespective of the aetiology and level of left ventricular ejection
function (LVEF); in CORONA, patients were ≥60 years old, had is-
chaemic aetiology, and an LVEF ≤ 40%. Eligibility criteria have been
published.21,22 All patients in CORONA and only the eligible patients
in GISSI-HF were equally allocated to 10 mg of rosuvastatin or match-
ing placebo, once daily. All patients gave written informed consent for
blood sampling. The study populations are representative of white
Caucasians in Italy for GISSI-HF, and in North-Eastern Europe for
CORONA.
Blood sampling and biochemical analyses
Venous blood samples were drawn on EDTA after overnight fasting, at
randomization, and at the 3-month follow-up visit, from 1233 repre-
sentative patients from GISSI-HF and 1457 from CORONA (see Sup-
plementary material online for more details).
Statistical analysis
The analyses are presented for the combined populations (2690
patients) and in some instances as separate data from the 1233
GISSI-HF patients and the 1457 CORONA patients. Concentrations
of PTX3 and two other circulating biomarkers [N-terminal pro brain
natriuretic protein (NT-proBNP) and high sensitivity CRP (hsCRP)]
were expressed as median and 25% and 75% quartiles (Q1– Q3).
See the Supplementary material online for more details on analyses.
Results
Table 1 reports the demographic and clinical characteristics of
patients in the pooled population and by tertiles of baseline
993
PTX3 concentration. Patients in the upper tertile were older,
with lower body mass index (BMI), and more frequently with HF
of ischaemic aetiology, severe HF symptoms, atrial fibrillation,
and diabetes. They had lower serum triglycerides levels and
higher plasma concentrations of hsCRP and NT-proBNP. Major
differences between the two trial populations were age (older
patients in CORONA), higher prevalence of ischaemic aetiology
of HF, and more advanced New York Heart Association
(NYHA) class in CORONA (Supplementary material online,
Table S1). The characteristics of the patients in the subgroups
studied for biomarkers were not different from those of the re-
spective populations of the main trials (data not shown).
Correlates of high pentraxin-3
At univariate analysis, the baseline PTX3 concentration was higher
in females than males, in older patients, in those with lower BMI,
and those with reduced estimated glomerular filtration rate
(eGFR) (Figure 1). There was a modest but significant relationship
between baseline concentrations of PTX3 and hsCRP (Spearman
r ¼ 0.22, P , 0.0001) or PTX3 and NT-proBNP (r ¼ 0.35, P ,
0.0001).
The variables associated with high PTX3 were advanced NYHA
class, age, BMI, ischaemic aetiology of HF, and LVEF in GISSI-HF,
but only NYHA class and BMI in CORONA.
Prognostic value of baseline pentraxin-3
concentrations
Median follow-up was 33 (27 –38) months in CORONA and 47
(37 –55) months in GISSI-HF. The regression splines show a con-
tinuous, monotonous increase in the hazard ratio (HR) for all-
cause and cardiovascular mortality with increasing levels of PTX3
in the pooled population (Figure 2).
In the pooled population, baseline PTX3 concentration was in-
dependently associated with the risk of all-cause mortality or car-
diovascular mortality in multivariable Cox models (Table 2).
Addition of baseline hsCRP to the model slightly weakened the as-
sociation between baseline PTX3 concentration and outcome, and
it was no longer significant when NT-proBNP was entered instead
of hsCRP (Table 2). The binary variable entered in the multivariable
Cox models (CORONA vs. GISSI-HF) and a term of interaction
between PTX3 and the two trials were never significant, suggesting
no major differences in the prognostic value of PTX3 in the two
trials. Baseline PTX3 was associated with hospitalization for wor-
sening HF only in models that did not include NT-proBNP
(Table 2).
In the pooled population, log-transformed concentrations of
baseline PTX3 significantly improved discrimination for all-cause
mortality [c-index, P ¼ 0.0005; net reclassification index (NRI) ¼
0.19, P ¼ 0.02] and for cardiovascular mortality (c-index, P ¼
0.0003; NRI ¼ 0.20, P ¼ 0.02) compared with a multivariable
model containing clinical risk factors. The discriminative value of
baseline PTX3 was not maintained when baseline concentrations
of NT-proBNP and hsCRP were introduced into the models (all-
cause mortality: c-index, P ¼ 0.03; NRI ¼ –0.003, P ¼ 0.97; car-
diovascular mortality: c-index, P ¼ 0.81; NRI ¼ 0.019, P ¼ 0.82).
994 R. Latini et al.

Table 1 Baseline characteristics of the pooled population by tertiles of baseline pentraxin-3 concentration

Variable All T1 (0.44– 4.09 T2 (4.11–6.66 T3 (6.67–139.83 P-value

ng/mL) ng/mL) ng/mL)
...............................................................................................................................................................................
No. patients (%) 2690 (100%) 898 (33.38) 896 (33.31) 896 (33.31)
Age (years) 69.3 + 9.3 66.3 + 10.3 70.6 + 8.4 70.9 + 8.3 ,0.001
Females 21.7% 19.4% 21.7% 24.0% 0.02
BMI (kg/m2) 27.1 + 4.5 27.4 + 4.4 27.1 + 4.5 26.7 + 4.7 ,0.001
HR (b.p.m.) 71.1 + 12.1 69.3 + 11.9 70.6 + 11.3 73.4 + 12.7 ,0.001
SBP/DBP (mmHg) 127/77 127/77 127/77 127/77 0.23/0.35
Ischaemic aetiology 77.5% 67.4% 82.7% 82.4% ,0.001
NYHA classes III– IV 48.7% 35.2% 51.0% 60.0% ,0.001
LVEF (%) 32.4 + 8.1 32.1 + 7.6 32.5 + 7.8 32.7 + 8.9 0.84
Laboratory values
Serum creatinine (mg/dL) 1.25 + 0.37 1.21 + 0.37 1.26 + 0.37 1.28 + 0.37 ,0.001
eGFR (mL/min/1.73 m2) 62.6 + 19.7 66.5 + 21.5 61.4 + 18.5 59.8 + 18.3 ,0.001
HDL-cholesterol (mg/dL) 48.0 + 13.5 48.0 + 12.9 48.2 + 13.2 47.8 + 14.4 0.28
LDL-cholesterol (mg/dL) 130 + 39 131 + 41 132 + 37 126 + 39 0.04
Triglycerides (mg/dL) 164 + 115 175 + 133 166 + 109 153 + 99 ,0.001
Co-morbidities
Atrial fibrillation/flutter 17.7% 12.4% 19.1% 21.8% ,0.001
Diabetes 26.1% 22.4% 26.8% 29.0% 0.001
Pharmacological therapy for HF
Beta-blockers 72.4% 73.6% 74.1% 69.4% 0.05
ACEIs 80.9% 81.6% 81.3% 79.9% 0.36
ARBs 13.6% 14.9% 14.7% 11.2% 0.02
Aldosterone receptor antagonists 39.6% 38.9% 38.6% 41.3% 0.29
Digitalis 31.2% 24.8% 31.3% 37.6% ,0.001
Biomarkers at baseline
PTX3 (ng/mL) 5.34 (3.55– 7.64) 2.94 (2.26–3.55) 5.34 (4.69–5.98) 9.02 (7.64– 11.37) NA
hsCRP (mg/L) 3.00 (1.30– 6.70) 2.33 (1.02–4.73) 2.88 (1.30–6.72) 4.20 (1.85– 9.20) ,0.001
NT-proBNP (pg/mL) 1050 (428–2394) 651 (296– 1363) 1097 (456– 2241) 1778 (786– 3758) ,0.001

All-cause mortality (per 100 person-years) 9.55 (8.89– 10.25) 6.35 (5.51–7.32) 9.86 (8.71–11.14) 13.23 (11.85– 14.78) ,0.001
Cardiovascular mortality (per 100 person-years) 7.39 (6.82– 8.01) 4.39 (3.70–5.21) 7.95 (6.93–9.12) 10.55 (9.32–11.94) ,0.001

Data are presented as mean + standard deviation or median (Q1 –Q3) for non-normally distributed continuous variables and as frequency for categorical variables.
ACEI, ACE inhibitor; ARB, angiotensin II type 1 receptor blocker; BMI, body mass index; eGFR, estimated glomerular filtration rate (simplified Modification of Diet in Renal Disease
equation); HF, heart failure; HR, heart rate; hsCRP, high sensitivity C-reactive protein; LVEF, left ventricular ejection fraction; NA, not applicable (stratification variable);
NT-proBNP, N-terminal pro brain natriuretic peptide; NYHA, New York Heart Association functional class; PTX3, pentraxin-3; SBD/DBP, systolic/diastolic arterial blood
pressure.

There was no significant interaction between median baseline
PTX3 levels and the randomized treatments (rosuvastatin vs.
placebo) on the fatal outcomes in each trial (P for interaction for
all-cause mortality, CORONA P ¼ 0.53, GISSI-HF P ¼ 0.51; for
cardiovascular mortality, CORONA P ¼ 0.89, GISSI-HF P ¼ 0.50)
or in the pooled population.
Effects of rosuvastatin on pentraxin-3
After 3 months of treatment, PTX3 in the rosuvastatin arm
increased significantly vs. placebo in the pooled population
(Figure 3) and in individual trial populations. In contrast, the
hsCRP concentration dropped more after 3 months in the rosu-
vastatin arm than in the placebo arm (Supplementary material
online, Table S2).
Prognostic values of 3-month changes in
pentraxin-3
Demographic and clinical characteristics of patients grouped by
tertiles of changes in baseline PTX3 concentration are shown in
the Supplementary material online, Table S3. The prognostic
value of changes in PTX3 concentrations over 3 months was
tested in nested multivariable Cox models adjusted for clinical
risk factors, baseline PTX3 concentration, trial, and combinations
of the baseline concentrations of the other two circulating biomar-
kers hsCRP and/or NT-proBNP. In the pooled population, both
baseline and 3-month increases in PTX3 concentrations predicted
fatal events (Table 3). In models containing baseline PTX3 and
hsCRP, and changes in PTX3, these three variables predicted out-
comes. After addition of NT-proBNP, changes in PTX3, but not
Pentraxin-3 in chronic HF
Figure 1: Baseline plasma pentraxin-3 (PTX3) concentrations
in relation to selected clinical characteristics. Box plots showing
the baseline median PTX3 concentration in clinical subgroups
based on sex, age, body mass index (BMI), and estimated glom-
erular filtration rate (eGFR). Groups were compared with the
Kruskall – Wallis or Mann– Whitney tests. The number of patients
in each category is indicated below the bars.
baseline PTX3, remained significantly associated with outcomes
(Table 3). Moreover, changes in PTX3 were still significantly asso-
ciated with all-cause mortality [HR for 1 SD increase 1.18, 95%
confidence interval (CI) 1.12– 1.25, P , 0.0001], cardiovascular
mortality (HR 1.16, 95% CI 1.09–1.25, P , 0.0001), or hospitaliza-
tion for worsening HF (HR 1.16, 95% CI 1.11–1.21, P , 0.0001)
when baseline concentrations of NT-proBNP and hsCRP were
entered in the Cox models.
Similar analyses were repeated separately for each trial popula-
tion and consistently showed that increases in PTX3 concentra-
tions were associated with fatal outcomes, even after adjustment
for baseline concentrations of hsCRP and/or NT-proBNP (Supple-
mentary material online, Table S4). Changes in PTX3 concentra-
tions did not improve discrimination for all-cause mortality
(c-index, P ¼ 0.52; NRI ¼ 0.015, P ¼ 0.87) or cardiovascular mor-
tality (c-index, P ¼ 0.18; NRI ¼ –0.028, P ¼ 0.76).
There was no significant interaction between changes in PTX3
levels and the effect of the randomized treatments (rosuvastatin
or placebo) on the fatal outcomes in the pooled population
(adjusted P for interaction for all-cause mortality ¼ 0.33; for car-
diovascular mortality, P ¼ 0.53).
Discussion
The main results of the present study can be summarized as
follows: (i) PTX3, a component of the humoral arm of the
innate immune system, was measured in plasma samples from
2690 patients with chronic HF enrolled in two clinical trials,
CORONA and GISSI-HF, that differed slightly for selection criteria;
(ii) common associates of elevated PTX3 are age, advanced NYHA
class, and lower BMI; (iii) changes in PTX3 concentration over
time, but not baseline PTX3, can predict fatal outcomes, independ-
ently of hsCRP and of the benchmark prognostic marker in HF,
NT-proBNP; (iv) rosuvastatin (10 mg/day for 3 months)
995
Figure 2: Association between baseline concentrations of
pentraxin-3 (PTX3) and mortality. The association between base-
line PTX3 concentration and all-cause mortality (A) or cardiovas-
cular mortality (B) in the pooled population was assessed by
regression splines. The solid lines indicate estimated hazard
ratios (with 95% confidence intervals as dashed lines) and the
vertical lines indicate the tertile cut-off points.
significantly lowered hsCRP but raised PTX3 compared with
placebo; and (v) the study was adequately powered and the popu-
lation representative of different European healthcare systems.
Pentraxin-3 is rapidly induced in various cell subsets such as per-
ipheral blood leucocytes, myeloid dendritic cells, and vascular
endothelium under the stimulation of inflammatory cytokines.13
PTX3 shares some similarities with CRP, but differs in terms of
structural domain, gene organization, cellular and tissue sources,
inducing stimuli, and ligands recognized. PTX3 is stored in poly-
morphonuclear cells and is released on activation to attenuate
their recruitment at the site of inflammation.23 Thus, in contrast
to CRP that is primary synthesized in the liver and thereby reflects
systemic inflammation, PTX3 is, at least in part, synthesized at the
site of inflammation. The majority of early published studies on
PTX3 as a marker of outcomes in healthy individuals24 and in
patients with chronic HF16 – 18 have been carried out in cohorts
of 37 –196 Asian subjects, thus not allowing firm conclusions to
be drawn. Here we estimated to have at least 80% power (at
996 R. Latini et al.

Table 2 Prognostic value of baseline pentraxin-3 in the pooled population

Outcome Model Biomarker Wald x2 HR (95% CI) P-value

...............................................................................................................................................................................
All-cause mortality (759 events, 28.3%) 1. Clinical risk factors + PTX3 PTX3 24.2 1.20 (1.12–1.30) ,0.0001
2. Clinical risk factors + PTX3 + hsCRP PTX3 11.8 1.14 (1.06–1.23) 0.001
hsCRP 39.5 1.22 (1.14–1.29) ,0.0001
3. Clinical risk factors + PTX3 + NT-proBNP PTX3 0.2 1.02 (0.94–1.11) 0.23
NT-proBNP 158 1.71 (1.57–1.86) ,0.0001
Cardiovascular mortality (587 events, 21.9%) 1. Clinical risk factors + PTX3 PTX3 31.5 1.27 (1.17–1.38) ,0.0001
2. Clinical risk factors + PTX3 + hsCRP PTX3 20.5 1.22 (1.12–1.32) ,0.0001
hsCRP 19.4 1.17 (1.09–1.25) ,0.0001
3. Clinical risk factors + PTX3 + NT-proBNP PTX3 1.2 1.05 (0.96–1.16) 0.28
NT-proBNP 163 1.88 (1.71–2.07) ,0.0001
Hospitalization for worsening HF 1. Clinical risk factors + PTX3 PTX3 25.1 1.21 (1.12–1.30) ,0.0001
(720 events, 26.9%) 2. Clinical risk factors + PTX3 + hsCRP PTX3 19.3 1.19 (1.10–1.28) ,0.0001
hsCRP 5.4 1.08 (1.01–1.15) 0.02
3. Clinical risk factors + PTX3 + NT-proBNP PTX3 0.5 1.03 (0.95–1.12) 0.48
NT-proBNP 176 1.74 (1.60–1.89) ,0.0001

Cox proportional hazard models for fatal outcomes with baseline PTX3 modelled as a continuous log-transformed variable. Baseline concentrations of hsCRP and NT-proBNP
were entered as log-transformed data. Clinical variables entered into the first model (PTX3 only) and significant (P , 0.05) for all-cause mortality and for cardiovascular mortality:
age, sex, BMI, NYHA, LVEF, eGFR. Variables entered into the first model but not significant: ischaemic aetiology of HF, study trial (CORONA vs. GISSI-HF). Clinical variables
entered into the first model (PTX3 only) for hospitalization for worsening of HF and significant (P , 0.05): age, sex, NYHA, eGFR, ischaemic aetiology of HF. Variables entered
into the first model but not significant: BMI, study trial (CORONA vs. GISSI-HF).
BMI, body mass index; CI, confidence interval; eGFR, estimated glomerular filtration rate; HF, heart failure; HR, hazard ratio; hsCRP, high sensitivity C-reactive protein; LVEF, left
ventricular ejection fraction; NT-proBNP, N-terminal pro brain natriuretic peptide; NYHA, New York Heart Association functional class; PTX3, pentraxin-3.

our population, in agreement with observations made in white and

Figure 3: Three-month changes in pentraxin-3 (PTX3) and
high sensitivity C-reactive protein (hsCRP) concentrations in
the pooled population. Three-month absolute changes in PTX3
or hsCRP concentrations, shown as median values, were com-
pared in the placebo and rosuvastatin arms of the pooled popu-
lation with the Mann-Whitney test. The number of patients in
each group is shown with the bars.
a¼ 0.05) to detect effect sizes of 1.18 and 1.21 (expressed as HR)
for the association between above vs. below the median PTX3
concentration, and all-cause mortality or cardiovascular mortality,
respectively.
Our data suggest that circulating PTX3 levels are also high in Cau-
casians with chronic HF. In healthy subjects, PTX3 is higher in women,
rises with age, and is inversely related to BMI.24 In our study, the same
anthropometric factors independently influenced PTX3, as well as
NYHA class. In adjusted analyses, renal function did not seem to in-
fluence PTX3 levels or its predictive value (data not shown) much in
Chinese Americans without cardiovascular disease in the Multi-
Ethnic Study of Atherosclerosis.25 In this respect, PTX3 differs
from established markers such as natriuretic peptides and troponins,
which are highly dependent on renal function.
The prognostic value of repeated measurements of PTX3 has
been assessed here for the first time in chronic HF. A single meas-
urement of PTX3 is associated with worse outcome, in patients
with chronic HF16 – 18 or in the general population.26 We found
that baseline concentrations of PTX3 and 3-month changes in
PTX3 were independently associated with fatal outcomes in
2690 patients with chronic and stable HF. On the other hand,
after addition of NT-proBNP to the prognostic model, changes
in PTX3 but not baseline PTX3 levels remained associated with
outcome; this is an unexpected finding which we cannot explain.
Several points merit comment in relation to this issue: (i) the as-
sociation of changes in PTX3 with mortality was stronger overall
than that of baseline PTX3 levels; (ii) the prognostic value of
changes in PTX3 was independent of hsCRP and of the natriuretic
peptides, the strongest marker of outcome in chronic HF;
(iii) results were similar for the two trials; (iv) NT-proBNP
remained by far the strongest prognostic biomarker in our
patients; and (v) baseline concentrations of PTX3, but not their
changes over time, modestly improved the prognostic discrimin-
ation beyond clinical risk factors.
In the CORONA trial, rosuvastatin (10 mg/day for 3 months)
reduced plasma hsCRP levels by 20 –30% compared with placebo,
and patients with higher baseline levels seemed to benefit more
than those with lower levels.27 In GISSI-HF, the same regimen of
rosuvastatin consistently reduced hsCRP,19 in accordance with its
anti-inflammatory properties. The present analyses showed an
Pentraxin-3 in chronic HF 997

Table 3 Prognostic value of 3-month changes in pentraxin-3 in the pooled population

Outcome Model Biomarker Wald HR (95% CI) P-value

x2
...............................................................................................................................................................................
All-cause mortality
(629 events, 26.2%) 1. Clinical risk factors + PTX3 DPTX3 (%) 27.3 1.17 (1.10–1.24) ,0.0001
Baseline PTX3 13.6 1.09 (1.04–1.14) ,0.0001

2. Clinical risk factors + PTX3 + hsCRP DPTX3 (%) 29.8 1.17 (1.11–1.24) ,0.0001
Baseline PTX3 4.7 1.06 (1.01–1.11) 0.03
hsCRP 37.7 1.24 (1.15–1.32) ,0.0001

3. Clinical risk factors + PTX3 + NT-proBNP DPTX3 (%) 32.7 1.18 (1.11–1.25) ,0.0001
Baseline PTX3 0.1 1.01 (0.95–1.08) 0.14
NT-proBNP 145 1.73 (1.59–1.90) ,0.0001

4. Clinical risk DPTX3 (%) 34.4 1.18 (1.12–1.25) ,0.0001

factors + PTX3 + hsCRP + NT-proBNP Baseline PTX3 0.0 0.99 (0.93–1.07) 0.87
hsCRP 14.2 1.15 (1.07–1.24) ,0.0001
NT-proBNP 127 1.68 (1.54–1.84) ,0.0001
Cardiovascular mortality
(483 events, 20.1%) 1. Clinical risk factors + PTX3 DPTX3 (%) 13.4 1.15 (1.07–1.24) ,0.0001
Baseline PTX3 19.9 1.11 (1.06–1.15) ,0.0001

2. Clinical risk factors + PTX3 + hsCRP DPTX3 (%) 14.3 1.15 (1.07–1.24) ,0.0001
Baseline PTX3 10.0 1.08 (1.03–1.13) 0.002
hsCRP 23.4 1.21 (1.12–1.31) ,0.0001

3. Clinical risk factors + PTX3 + NT-proBNP DPTX3 (%) 18.9 1.16 (1.09–1.25) ,0.0001
Baseline PTX3 0.9 1.03 (0.97–1.10) 0.34
NT-proBNP 153.0 1.92 (1.73–2.13) ,0.0001

4. Clinical risk DPTX3 (%) 19.7 1.16 (1.09–1.25) ,0.0001

factors + PTX3 + hsCRP + NT-proBNP Baseline PTX3 0.3 1.02 (0.95–1.09) 0.57
hsCRP 7.2 1.12 (1.03–1.22) 0.007
NT-proBNP 138.0 1.87 (1.69–2.08) ,0.0001

Hospitalization for worsening HF (650 1. Clinical risk factors + PTX3 DPTX3 (%) 47.4 1.18 (1.12–1.23) ,0.0001
events, 27.2%) Baseline PTX3 35.3 1.27 (1.18–1.38) ,0.0001

2. Clinical risk factors + PTX3 + hsCRP DPTX3 (%) 46.9 1.18 (1.12–1.23) ,0.0001
Baseline PTX3 29.5 1.25 (1.15–1.36) ,0.0001
hsCRP 3.1 1.06 (0.99–1.14) 0.08

3. Clinical risk factors + PTX3 + NT-proBNP DPTX3 (%) 45.2 1.16 (1.11–1.21) ,0.0001
Baseline PTX3 3.5 1.09 (1.00–1.19) 0.06
NT-proBNP 147 1.70 (1.56–1.85) ,0.0001

4. Clinical risk DPTX3 (%) 44.9 1.16 (1.11–1.21) ,0.0001

factors + PTX3 + hsCRP + NT-proBNP Baseline PTX3 3.3 1.09 (0.99–1.19) 0.07
hsCRP 0.06 1.01 (0.94–1.08) 0.80
NT-proBNP 142 1.69 (1.55–1.84) ,0.0001

Cox proportional hazard models for outcomes with 3-month changes (DPTX3) and baseline PTX3 modelled as continuous variables. Baseline concentrations of hsCRP and
NT-proBNP were entered as log-transformed data. Clinical variables entered into the first model (PTX3 only) and significant (P , 0.05): age, sex, BMI (not for hospitalization for
worsening HF), NYHA, LVEF, eGFR; variables tested in the models but not significant: ischaemic aetiology of HF (significant only for hospitalization for worsening HF), study trial
(CORONA vs. GISSI-HF).
BMI, body mass index; CI, confidence interval; eGFR, estimated glomerular filtration rate; HF, heart failure; HR, hazard ratio; hsCRP, high sensitivity C-reactive protein; LVEF, left
ventricular ejection fraction; NT-proBNP, N-terminal pro brain natriuretic peptide; NYHA, New York Heart Association functional class; PTX3, pentraxin-3.
998
opposite, consistent effect on PTX3. This was unexpected since
statins are known to suppress the PTX3 expression of human endo-
thelial cells and fibroblast-like synoviocytes.28,29 Previous evidence
on the effect of statins on PTX3 levels in man is, however, inconclu-
sive, and flawed by the very limited number of subjects. In an uncon-
trolled study, pitavastatin (2 mg daily for 6 months) had no effect on
plasma PTX3 and hsCRP levels in 35 patients with hypercholesterol-
aemia, but showed a trend toward a small decrease in a subgroup of
12 patients with high baseline PTX3.30 A direct toxic effect of rosu-
vastatin on skeletal muscle and ensuing release of PTX3 from vascu-
lar cells is unlikely since the drug did not cause any excess
muscle-related symptoms or elevation in the levels of creatine
kinase in the CORONA and GISSI-HF trials.11,19 On the other
hand, statins may alter the innate immunity behaviour of stimulated
phagocytic cells and enhance the production of macrophage and
neutrophil extracellular traps that contain antimicrobial proteins
and PTX3.31,32 This very speculative hypothesis, based on clinical
and epidemiological observations of a lower susceptibility to
severe bacterial infection in patients receiving statin therapy, needs
to be validated in appropriate experimental models. These findings
suggest that CRP and PTX3 may reflect overlapping, but also differ-
ent, inflammatory pathways (e.g. systemic vs. local inflammation).
Study limitations
The present study has several limitations besides the obvious se-
lection of patients according to trial inclusion criteria. We have
investigated only short-term variations in PTX3 concentration,
while it may be more clinically relevant to have insights on
changes over longer periods in stable ambulatory patients. We
also could not investigate the mechanisms responsible for PTX3
elevation in patients receiving rosuvastatin.
To our knowledge, this is the first demonstration of the prog-
nostic value of an inflammatory marker in chronic HF, independent
of and in addition to a natriuretic peptide. PTX3 may represent a
marker of cardiovascular pathology reflecting the vascular involve-
ment by inflammatory processes. The present findings support the
need for a comprehensive, unbiased assessment of potential roles
played by inflammatory pathways and innate immunity, in particular
in chronic HF. The opposite effects of rosuvastatin on PTX3 and
hsCRP activation and outcomes call for further investigation.
Supplementary material
Supplementary material is available at European Journal of Heart
Failure online.
Acknowledgements
Reagents for measuring PTX3 in the CORONA and GISSI-HF trials
were kindly donated by Perseus Proteomics, Japan. Reagents for
measuring NT-proBNP in the GISSI-HF trial were kindly donated
by Roche Diagnostics, Rotkreuz, Switzerland.
Funding
The CORONA trial was funded by AstraZeneca. The GISSI-HF trial
was funded by Società Prodotti Antibiotici (SPA, Italy), Pfizer, Sigma
Tau, and AstraZeneca. This study was supported in part by funding
R. Latini et al.
under the European Commission (HEALTH-F4-2008-202156 TOLER-
AGE), the European Research Council – Advanced Grant (project
HIIS), and grants from Fondazione CARIPLO (Ricerca Scientifica e
Trasferimento Tecnologico, project 2009-2582) and Regione Lombar-
dia (Ricerca Indipendente in Ambito Oncologico, Cardiocerebrovasco-
lare ed Intensivistico, and ‘Progetto Ricerca Indipendente 2010:
Caratterizzazione della pentraxina lunga PTX3 come biomarcatore in
patologia cardiovascolare’).
Conflicts of interest: A.M. and B.B. are inventors of patents on
pentraxin-3 and obtain royalties on related reagents. R.L. and S.M.
have received institutional grants and honoraria from Roche Diagnos-
tics. L.G. has received consulting or advisory board fees from AstraZe-
neca. J.W. is a former medical advisor at AstraZeneca, Mölndal,
Sweden. The remaining authors have no conflict of interest to report.
References
1. Wrigley BJ, Lip GY, Shantsila E. The role of monocytes and inflammation in the
pathophysiology of heart failure. Eur J Heart Fail 2011;13:1161 –1171.
2. Vasan RS, Sullivan LM, Roubenoff R, Dinarello CA, Harris T, Benjamin EJ,
Sawyer DB, Levy D, Wilson PW, D’Agostino RB; Framingham Heart Study. In-
flammatory markers and risk of heart failure in elderly subjects without prior
myocardial infarction: the Framingham Heart Study. Circulation 2003;107:
1486 –1491.
3. Anand IS, Latini R, Florea VG, Kuskowski MA, Rector T, Masson S, Signorini S,
Mocarelli P, Hester A, Glazer R, Cohn JN; Val-HeFT Investigators. C-reactive
protein in heart failure: prognostic value and the effect of valsartan. Circulation
2005;112:1428 –1434.
4. Ueland T, Aukrust P, Yndestad A, Otterdal K, Frøland SS, Dickstein K, Kjekshus J,
Gullestad L, Damås JK. Soluble CD40 ligand in acute and chronic heart failure. Eur
Heart J 2005;26:1101 –1107.
5. Radauceanu A, Ducki C, Virion JM, Rossignol P, Mallat Z, McMurray J, Van
Veldhuisen DJ, Tavazzi L, Mann DL, Capiaumont-Vin J, Li M, Hanriot D,
Zannad F. Extracellular matrix turnover and inflammatory markers independently
predict functional status and outcome in chronic heart failure. J Card Fail 2008;14:
467 –474.
6. Christensen HM, Frystyk J, Faber J, Schou M, Flyvbjerg A, Hildebrandt P,
Raymond I, Klausen TW, Kistorp C. a-Defensins and outcome in patients with
chronic heart failure. Eur J Heart Fail 2012;14:387 –394.
7. Heymans S, Hirsch E, Anker SD, Aukrust P, Balligand JL, Cohen-Tervaert JW,
Drexler H, Filippatos G, Felix SB, Gullestad L, Hilfiker-Kleiner D, Janssens S,
Latini R, Neubauer G, Paulus WJ, Pieske B, Ponikowski P, Schroen B,
Schultheiss HP, Tschöpe C, Van Bilsen M, Zannad F, McMurray J, Shah AM. Inflam-
mation as a therapeutic target in heart failure? A scientific statement from the
Translational Research Committee of the Heart Failure Association of the Euro-
pean Society of Cardiology. Eur J Heart Fail 2009;11:119 –129.
8. Anand SS, Yusuf S. C-reactive protein is a bystander of cardiovascular disease. Eur
Heart J 2010;31:2092 –2096.
9. Schulz R, Heusch G. C-reactive protein: just a biomarker of inflammation or a
pathophysiological player in myocardial function and morphology? Hypertension
2011;57:151 –153.
10. Araújo JP, Lourenço P, Azevedo A, Friões F, Rocha-Gonçalves F, Ferreira A,
Bettencourt P. Prognostic value of high-sensitivity C-reactive protein in heart
failure: a systematic review. J Card Fail 2009;15:256 –266.
11. Kjekshus J, Apetrei E, Barrios V, Böhm M, Cleland JG, Cornel JH, Dunselman P,
Fonseca C, Goudev A, Grande P, Gullestad L, Hjalmarson A, Hradec J,
Jánosi A, Kamenský G, Komajda M, Korewicki J, Kuusi T, Mach F, Mareev V,
McMurray JJ, Ranjith N, Schaufelberger M, Vanhaecke J, van Veldhuisen DJ,
Waagstein F, Wedel H, Wikstrand J, CORONA Group. CORONA Group. Rosu-
vastatin in older patients with systolic heart failure. N Engl J Med 2007;357:
2248 –2261.
12. McMurray JJ, Kjekshus J, Gullestad L, Dunselman P, Hjalmarson A, Wedel H,
Lindberg M, Waagstein F, Grande P, Hradec J, Kamenský G, Korewicki J,
Kuusi T, Mach F, Ranjith N, Wikstrand J, CORONA Study Group. Effects of
statin therapy according to plasma high-sensitivity C-reactive protein concentra-
tion in the Controlled Rosuvastatin Multinational Trial in Heart Failure
(CORONA): a retrospective analysis. Circulation 2009;120:2188 – 2196.
13. Bottazzi B, Doni A, Garlanda C, Mantovani A. An integrated view of humoral
innate immunity: pentraxins as a paradigm. Annu Rev Immunol 2010;28:157 – 183.
14. Latini R, Maggioni AP, Peri G, Gonzini L, Lucci D, Mocarelli P, Vago L, Pasqualini F,
Signorini S, Soldateschi D, Tarli L, Schweiger C, Fresco C, Cecere R, Tognoni G,
Mantovani A; Lipid Assessment Trial Italian Network (LATIN) Investigators.
Pentraxin-3 in chronic HF
Prognostic significance of the long pentraxin PTX3 in acute myocardial infarction.
Circulation 2004;110:2349 –2354.
15. Jenny NS, Arnold AM, Kuller LH, Tracy RP, Psaty BM. Associations of pentraxin 3
with cardiovascular disease and all-cause death: the Cardiovascular Health Study.
Arterioscler Thromb Vasc Biol 2009;29:594–599.
16. Suzuki S, Takeishi Y, Niizeki T, Koyama Y, Kitahara T, Sasaki T, Sagara M, Kubota I.
Pentraxin 3, a new marker for vascular inflammation, predicts adverse clinical out-
comes in patients with heart failure. Am Heart J 2008;155:75 –81.
17. Kotooka N, Inoue T, Aoki S, Anan M, Komoda H, Node K. Prognostic value of
pentraxin 3 in patients with chronic heart failure. Int J Cardiol 2008;130:19– 22.
18. Matsubara J, Sugiyama S, Nozaki T, Sugamura K, Konishi M, Ohba K, Matsuzawa Y,
Akiyama E, Yamamoto E, Sakamoto K, Nagayoshi Y, Kaikita K, Sumida H,
Kim-Mitsuyama S, Ogawa H. Pentraxin 3 is a new inflammatory marker correlated
with left ventricular diastolic dysfunction and heart failure with normal ejection
fraction. J Am Coll Cardiol 2011;57:861 –869.
19. GISSI-HF InvestigatorsTavazzi L, Maggioni AP, Marchioli R, Barlera S, Franzosi MG,
Latini R, Lucci D, Nicolosi GL, Porcu M, Tognoni G. Effect of rosuvastatin in
patients with chronic heart failure (the GISSI-HF trial): a randomised, double-
blind, placebo-controlled trial. Lancet 2008;372:1231 –1239.
20. GISSI-HF Investigators, Tavazzi L, Maggioni AP, Marchioli R, Barlera S, Franzosi MG,
Latini R, Lucci D, Nicolosi GL, Porcu M, Tognoni G. Effect of n-3 polyunsaturated
fatty acids in patients with chronic heart failure (the GISSI-HF trial): a randomised,
double-blind, placebo-controlled trial. Lancet 2008;372:1223 –1230.
21. Tavazzi L, Tognoni G, Franzosi MG, Latini R, Maggioni AP, Marchioli R, Nicolosi GL,
Porcu M, GISSI-HF Investigators. Rationale and design of the GISSI heart failure trial:
a large trial to assess the effects of n-3 polyunsaturated fatty acids and rosuvastatin in
symptomatic congestive heart failure. Eur J Heart Fail 2004;6:635 –641.
22. Kjekshus J, Dunselman P, Blideskog M, Eskilson C, Hjalmarson A, McMurray JV,
Waagstein F, Wedel H, Wessman P, Wikstrand J; CORONA Study Group. A
statin in the treatment of heart failure? Controlled rosuvastatin multinational
study in heart failure (CORONA): study design and baseline characteristics. Eur
J Heart Fail 2005;7:1059 –1069.
23. Deban L, Russo RC, Sironi M, Moalli F, Scanziani M, Zambelli V, Cuccovillo I,
Bastone A, Gobbi M, Valentino S, Doni A, Garlanda C, Danese S, Salvatori G,
Sassano M, Evangelista V, Rossi B, Zenaro E, Constantin G, Laudanna C,
999
Bottazzi B, Mantovani A. Regulation of leukocyte recruitment by the long pen-
traxin PTX3. Nat Immunol 2010;11:328–334.
24. Yamasaki K, Kurimura M, Kasai T, Sagara M, Kodama T, Inoue K. Determination of
physiological plasma pentraxin 3 (PTX3) levels in healthy populations. Clin Chem
Lab Med 2009;47:471 –477.
25. Dubin R, Shlipak M, Li Y, Ix J, de Boer IH, Jenny N, Peralta CA. Racial differences in
the association of pentraxin-3 with kidney dysfunction: the Multi-Ethnic Study of
Atherosclerosis. Nephrol Dial Transplant 2010;26:1903 –1908.
26. Jenny NS, Arnold AM, Kuller LH, Tracy RP, Psaty BM. Associations of pentraxin 3
with cardiovascular disease and all-cause death: the Cardiovascular Health Study.
Arterioscler Thromb Vasc Biol 2009;29:594 – 599.
27. McMurray JJ, Kjekshus J, Gullestad L, Dunselman P, Hjalmarson A, Wedel H,
Lindberg M, Waagstein F, Grande P, Hradec J, Kamenský G, Korewicki J,
Kuusi T, Mach F, Ranjith N, Wikstrand J; CORONA Study Group. Effects of
statin therapy according to plasma high-sensitivity C-reactive protein concentra-
tion in the Controlled Rosuvastatin Multinational Trial in Heart Failure
(CORONA): a retrospective analysis. Circulation 2009;120:2188 –2196.
28. Morikawa S, Takabe W, Mataki C, Wada Y, Izumi A, Saito Y, Hamakubo T,
Kodama T. Global analysis of RNA expression profile in human vascular cells
treated with statins. J Atheroscler Thromb 2004;11:62–72.
29. Yokota K, Miyoshi F, Sato K, Asanuma Y, Akiyama Y, Mimura T. Geranylgeranyl-
pyrophosphate regulates secretion of pentraxin 3 and monocyte chemoattractant
protein-1 from rheumatoid fibroblast-like synoviocytes in distinct manners. Clin
Exp Rheumatol 2011;29:43–49.
30. Ohbayashi H, Miyazawa C, Miyamoto K, Sagara M, Yamashita T, Onda R. Pitavas-
tatin improves plasma pentraxin 3 and arterial stiffness in atherosclerotic patients
with hypercholesterolemia. J Atheroscler Thromb 2009;16:490 –500.
31. Chow OA, von Köckritz-Blickwede M, Bright AT, Hensler ME, Zinkernagel AS,
Cogen AL, Gallo RL, Monestier M, Wang Y, Glass CK, Nizet V. Statins
enhance formation of phagocyte extracellular traps. Cell Host Microbe 2010;8:
445 –454.
32. Jaillon S, Peri G, Delneste Y, Frémaux I, Doni A, Moalli F, Garlanda C, Romani L,
Gascan H, Bellocchio S, Bozza S, Cassatella MA, Jeannin P, Mantovani A. The
humoral pattern recognition receptor PTX3 is stored in neutrophil granules
and localizes in extracellular traps. J Exp Med 2007;204:793–804.