Abstract
Background:
Galectin-3 (Gal-3) is involved in collagen deposition and inflammation and is a prognostic biomarker in heart failure (HF).
Methods and Results:
Gal-3 and other markers of fibrosis or cardiac stress were measured serially in 413 patients with mild HF randomized to the mineralocorticoid receptor antagonist canrenone or placebo to evaluate treatment effect and association with clinical outcome. Gal-3 increased slightly over 6 months in both arms of the study and was associated with clinical endpoints.
Conclusions:
Although Gal-3 showed prognostic value, the effect of canrenone on clinical outcomes was unaffected by baseline concentrations of biomarkers of fibrosis or cardiac stress.
Galectin-3 (Gal-3), a protein of the β-galactoside-binding lectin family, is involved in collagen deposition and inflammation,1
and has been shown to be a prognostic biomarker in heart failure (HF).2
Because the role of circulating Gal-3 levels in the individualizing of therapy and management of HF remains to be elucidated,3,4
we investigated the effect of the mineralocorticoid receptor antagonist (MRA) canrenone on Gal-3 levels and their interaction on clinical outcomes in patients with mild, chronic HF with reduced ejection fraction (EF) enrolled in a clinical trial.
Methods
AREA IN-CHF was a multicenter, randomized, double-blind, placebo-controlled trial comparing canrenone with placebo in 467 patients with mild, chronic HF over 12 months of follow-up.5
The investigation conformed with the principles of the Declaration of Helsinki, the protocol was approved by the ethics committees of all participating centers and informed consent was obtained from all enrolled subjects. The prespecified primary endpoint was a change in echocardiographic left ventricular (LV) end-diastolic volume over 12 months. Secondary endpoints included total deaths, hospitalization for cardiac causes, and the combination of cardiac death and hospitalization for cardiac causes. Blood samples were collected at randomization and after 6 months and sent to a central laboratory for measurement of B-type natriuretic peptide (BNP), procollagen type III aminoterminal peptide (PIIINP), and aldosterone (at baseline). Gal-3 at baseline (n=413) and 6 months (n=339) was measured with a chemiluminescent microparticle immunoassay (Abbott Diagnostics).
The effects of time (baseline and 6 months) and study treatment on biomarker concentrations were assessed with a 2-way ANOVA analysis. Logistic regression analyses were done to test the biomarker’s associations with clinical events at 12-month follow-up, considering Gal-3 as continuous, with the only exception of the non-linear relation with hospitalization for cardiac causes. The multivariable models were adjusted for treatment arm, age, hospitalization for HF in the previous year, diabetes, chronic obstructive pulmonary disease (COPD), creatinine, prescription of diuretics and of angiotensin-converting enzyme inhibitor (ACEi)/angiotensin-receptor blocker (ARB), with a backward selection forcing the single marker in the model. Finally, an interaction between Gal-3, BNP, PIIINP, aldosterone, and study treatment on clinical outcomes was assessed by Breslow-Day test for the homogeneity of the odds ratios (ORs). A 2-sided P<0.05 was considered statistically significant.
Results
Patients with baseline Gal-3 concentration above the median (10.8 ng/mL) were older, more frequently diabetic, with a prior hospitalization for HF, with peripheral artery disease, and on diuretics or RAAS inhibitors (Table). Gal-3 weakly correlated with BNP (Spearman r=0.11, P=0.03), aldosterone (r=0.12, P=0.02) and PIIINP (r=0.11, P=0.03). Higher baseline Gal-3 levels were independently associated with diabetes mellitus (P=0.009), older age (P=0.01), concomitant therapy with ACEi or ARB (P=0.02), and lower LVEF (P=0.04).
Table.
Demographic and Clinical Characteristics at Randomization According to Galectin-3 Median Concentration (10.8 ng/mL)
| Variable |
All |
Gal-3 ≤median |
Gal-3 >median |
P value |
| No. (%) |
413 |
208 (50.4) |
205 (49.6) |
|
| Demographic and clinical data |
| Age (years) |
62.6±9.3 |
60.2±9.7 |
65.0±8.3 |
<0.0001 |
| Age >70 years (No. (%)) |
93 (22.5) |
32 (15.4) |
61 (29.8) |
0.0005 |
| Females (No. (%)) |
68 (16.5) |
34 (16.4) |
34 (16.6) |
0.95 |
| BMI (kg/m2) |
26.9±3.6 |
26.8±3.3 |
27.0±3.8 |
0.830 |
| SBP (mmHg) |
128.4±16.9 |
128.0±15.9 |
128.8±17.9 |
0.77 |
| DBP (mmHg) |
78.6±8.8 |
79.0±8.0 |
78.2±9.5 |
0.33 |
| HR (beats/min) |
66.7±11.3 |
65.5±11.0 |
67.9±11.5 |
0.03 |
| Clinical history (n (%)) |
| Hypertension |
195 (47.2) |
94 (45.2) |
101 (49.3) |
0.41 |
| Diabetes mellitus |
84 (20.4) |
30 (14.4) |
54 (26.3) |
0.003 |
| Myocardial infarction |
202 (48.9) |
108 (51.9) |
94 (45.9) |
0.20 |
| Ischemic etiology |
210 (50.9) |
109 (52.4) |
101 (49.3) |
0.52 |
| Diagnosis of HF >12 months |
313 (75.8) |
164 (78.9) |
149 (72.7) |
0.144 |
| Prior hospitalization for HF |
196 (47.5) |
81 (38.9) |
115 (56.1) |
0.0005 |
| Prior stroke |
10 (2.4) |
3 (1.4) |
7 (3.4) |
0.22 |
| COPD |
43 (10.4) |
14 (6.7) |
29 (14.2) |
0.014 |
| Prior revascularization |
143 (34.6) |
69 (33.2) |
74 (36.1) |
0.53 |
| Implantable cardiac defibrillator |
22 (5.3) |
9 (4.3) |
13 (6.3) |
0.36 |
| Peripheral vascular disease |
12 (2.9) |
1 (0.5) |
11 (5.4) |
0.003 |
| Paroxysmal AF |
56 (13.6) |
23 (11.1) |
33 (16.1) |
0.13 |
| Left bundle branch block |
122 (29.5) |
54 (26.0) |
68 (33.2) |
0.11 |
| Echocardiographic data |
| LVEF (%) |
39.5±8.7 |
40.4±8.7 |
38.5±8.6 |
0.03 |
| LVEDV (mL) |
161.7±60.7 |
158.0±52.5 |
165.4±68.1 |
0.67 |
| LV mass (g) |
270±75 |
267±70 |
273±80 |
0.44 |
| Relative wall thickness |
0.33±0.06 |
0.33±0.05 |
0.33±0.06 |
0.40 |
| Left atrial diameter (cm) |
4.2±0.7 |
4.1±0.6 |
4.3±0.8 |
0.02 |
| E/A |
1.21±0.94 |
1.10±0.74 |
1.33±1.11 |
0.65 |
| Concomitant pharmacological therapy (n (%)) |
| Diuretics |
288 (69.7) |
128 (61.5) |
160 (78.1) |
0.0003 |
| β-blockers |
325 (78.7) |
164 (78.9) |
161 (78.5) |
0.86 |
| ACEi or ARBs |
400 (96.9) |
195 (93.8) |
205 (100.0) |
0.0005 |
| Digitalis |
104 (25.2) |
52 (25.0) |
52 (25.4) |
0.95 |
| Nitrates |
112 (27.1) |
50 (24.0) |
62 (30.2) |
0.16 |
| Statins |
183 (44.3) |
100 (48.1) |
83 (40.5) |
0.13 |
| Antiplatelets |
236 (57.1) |
124 (59.6) |
112 (54.6) |
0.28 |
| Laboratory data |
| Serum creatinine (mg/dL) |
1.06±0.26 |
1.02±0.22 |
1.11±0.28 |
0.0004 |
| Serum potassium (mEq/L) |
4.3±0.4 |
4.3±0.4 |
4.4±0.4 |
0.39 |
| WBC (mm−3) |
6,779±1,707 |
6,763±1,749 |
6,796±1,666 |
0.89 |
| Circulating biomarkers |
| Galectin-3 (ng/mL) |
10.8 [9.0–13.3] |
9.1 [7.4–9.9] |
13.3 [12.0–15.8] |
– |
| PIIINP (μg/L) |
5.4 [4.0–7.2] |
5.1 [3.8–6.8] |
5.8 [4.4–7.4] |
0.02 |
| Aldosterone (pg/mL) |
117 [74–178] |
113 [72–171] |
125 [75–187] |
0.17 |
| BNP (ng/L) |
88 [36–187] |
67 [30–138] |
112 [52–222] |
<0.0001 |
ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin-receptor blocker; BMI, body mass index; BNP, B-type natriuretic peptide; COPD, chronic obstructive pulmonary disease; DBP, diastolic blood pressure; HF, heart failure; HR, heart rate; LVEDV, left ventricular end-diastolic volume; PIIINP, procollagen type III aminoterminal peptide; SBP, systolic blood pressure; WBC, white blood cells.
A trend towards a larger beneficial effect of canrenone was apparent in patients with above median concentrations of Gal-3, BNP, and aldosterone, but not of PIIINP (Figure). However, no interactions between the randomized treatments and the 4 biomarkers were observed for the combined endpoint.
Circulating Gal-3 concentration tended to increase slightly over 6 months, both in the canrenone (absolute difference=0.1 [−1.4 to 3.4] ng/mL, median [Q1–Q3]) and in placebo groups (absolute difference=0.6 [−1.6 to 3.2] ng/mL, between-treatment difference: P=0.86). Canrenone significantly decreased BNP levels over 6 months vs. placebo (P<0.0001).
Gal-3 was associated with all-cause death (14 events (3.6%), adjusted OR for 1 unit increment=1.07 [1.01–1.13], P=0.01) and hospital admission for cardiac cause (36 events (9.1%), OR for Gal-3 ≥12 vs. Gal-3 <12=2.17 [1.04–4.50], P=0.04), but not with the composite endpoint of cardiac death or hospital admission for cardiac cause (42 events (10.7%), OR for 1 unit increment=1.04 [0.99–1.09], P=0.13). BNP showed by far the strongest associations with the 3 endpoints (P≤0.0001).
Discussion
Higher Gal-3 level was weakly but independently related with worse fatal and non-fatal outcomes. Even if the sample size of AREA IN-CHF was inadequate for a proper evaluation of Gal-3 as a prognostic factor, the consistency of this limited data set was still supported by the clear prediction of 12-month outcomes by BNP and the decrease in BNP by canrenone associated with its beneficial effect on outcomes, both echocardiographic and clinical.5,6
The effect of canrenone on clinical outcomes at 12-month follow-up was unaffected by baseline concentrations of the 4 biomarkers. Nonetheless, its clinical benefit (lower incidence of cardiac death/hospitalization for cardiac causes) tended to be higher in patients with Gal-3 levels over the median concentration, although the interaction was not statistically significant (Figure). This is in agreement with other studies on MRAs.3,7
In contrast, in Val-HeFT valsartan significantly reduced hospitalizations for HF in patients with lower than the median levels of Gal-3 at entry,8
while in CORONA rosuvastatin reduced the primary endpoint of cardiovascular death, non-fatal myocardial infarction, and non-fatal stroke in patients with Gal-3 <19.0 ng/mL.9
Low Gal-3 may reflect reversible as opposed to irreversible myocardial fibrosis, as suggested by the findings from the PARAMOUNT trial in HF with preserved EF where LCZ696 reduced left atrial volume more effectively in patients with sST2 and Gal-3 levels below their respective medians.10
However, there are no clinical trials prospectively selecting patients with HF based on Gal-3 concentrations.
Conclusions
Gal-3 in mild HF was associated with the severity of HF and clinical outcomes. However, BNP but not Gal-3 levels over 6 months were affected by canrenone. The effect of canrenone on clinical outcomes was unaffected by baseline concentrations of biomarkers of fibrosis or cardiac stress. However, a non-statistically significant trend for a larger beneficial effect of canrenone was apparent in patients with above median concentrations of Gal-3, BNP, and aldosterone.
Acknowledgments
We thank Giuseppe Fraterrigo and Rossana Spatola for their skillful assistance with performing Gal-3 assays.
Conflicts of Interest
The funding source (Therabel GiEnne Pharma SpA) had no role in the trial design, conduct, data collection, analyses and data interpretation.
M.V. is employed by Therabel GiEnne Pharma SpA. L.G., D.L., and A.P.M. are employed at Heart Care Foundation, ANMCO Research Center, an independent research institution that received unrestricted funding from Therabel GiEnne Pharma SpA to conduct the study. The echocardiography (G.F.M.) and the biomarker core laboratories (S.M., E.B.N., and R.L.) received a research grant from Therabel GiEnne Pharma SpA. F.C., A.B., P.G.C., D.D.C., and A.D.L. have no conflicts of interest to declare.
References
- 1.
Yang RY, Rabinovich GA, Liu FT. Galectins: Structure, function and therapeutic potential. Expert Rev Mol Med 2008; 10: e17.
- 2.
Chen A, Hou W, Zhang Y, Chen Y, He B. Prognostic value of serum galectin-3 in patients with heart failure: A meta-analysis. Int J Cardiol 2015; 182: 168–170.
- 3.
Koukoui F, Desmoulin F, Galinier M, Barutaut M, Caubère C, Evaristi MF, et al. The prognostic value of plasma galectin-3 in chronic heart failure patients is maintained when treated with mineralocorticoid receptor antagonists. PLoS One 2015; 10: e0119160.
- 4.
Edelmann F, Holzendorf V, Wachter R, Nolte K, Schmidt AG, Kraigher-Krainer E, et al. Galectin-3 in patients with heart failure with preserved ejection fraction: Results from the Aldo-DHF trial. Eur J Heart Fail 2015; 17: 214–223.
- 5.
Boccanelli A, Mureddu GF, Cacciatore G, Clemenza F, Di Lenarda A, Gavazzi A, et al. Anti-remodelling effect of canrenone in patients with mild chronic heart failure (AREA IN-CHF study): Final results. Eur J Heart Fail 2009; 11: 68–76.
- 6.
Gupta DK, Wang TJ. Natriuretic peptides and cardiometabolic health. Circ J 2015; 79: 1647–1655.
- 7.
Gandhi PU, Motiwala SR, Belcher AM, Gaggin HK, Weiner RB, Baggish AL, et al. Galectin-3 and mineralocorticoid receptor antagonist use in patients with chronic heart failure due to left ventricular systolic dysfunction. Am Heart J 2015; 169: 404–411.e3.
- 8.
Anand IS, Rector TS, Kuskowski M, Adourian A, Muntendam P, Cohn JN. Baseline and serial measurements of galectin-3 in patients with heart failure: Relationship to prognosis and effect of treatment with valsartan in the Val-HeFT. Eur J Heart Fail 2013; 15: 511–518.
- 9.
Gullestad L, Ueland T, Kjekshus J, Nymo SH, Hulthe J, Muntendam P, et al. Galectin-3 predicts response to statin therapy in the Controlled Rosuvastatin Multinational Trial in Heart Failure (CORONA). Eur Heart J 2012; 33: 2290–2296.
- 10.
Zile MR, Jhund PS, Baicu CF, Claggett BL, Pieske B, Voors AA, et al; Prospective Comparison of ARNI With ARB on Management of Heart Failure With Preserved Ejection Fraction (PARAMOUNT) Investigators. Plasma biomarkers reflecting profibrotic processes in heart failure with a preserved ejection fraction: Data from the Prospective Comparison of ARNI With ARB on Management of Heart Failure With Preserved Ejection Fraction Study. Circ Heart Fail 2016; 9: e002551.
