Heterogeneity Of Early B-type Natriuretic Peptide Change In HFrEF Patients Treated With Sacubitril/Valsartan

Abstract

<h3>Introduction</h3> Through inhibition of neprilysin, sacubitril/valsartan (Sac/Val), an angiotensin receptor neprilysin inhibitor (ARNI), may cause initial increase in B-type natriuretic peptide (BNP) concentrations. The frequency, variability, and predictors of BNP rise during treatment of heart failure with reduced ejection fraction (HFrEF) from Sac/Val are not well-described. <h3>Hypothesis</h3> Changes in BNP after initiation of Sac/Val in HFrEF are variable and correlate with changes in NT-proBNP and ucGMP. <h3>Methods</h3> We classified BNP changes from baseline (pre Sac/Val) to Weeks 4 and12 among 367 patients treated with Sac/Val (regardless of dose) in an individual patient data analysis pooled from the EVALUATE-HF (n=221) and PROVE-HF (n=146) studies. Clinical predictors of BNP changes to Week 12 were analyzed in linear regression models and correlations with changes in NT-proBNP and ucGMP were examined using Spearman's rank correlation coefficient (rho). <h3>Results</h3> Across the two trials, median [IQR] concentration of BNP at Baseline, Week 4 and Week 12 was 145 [55, 329], 136 [50, 338] and 135 [51, 299] ng/L, respectively. There was no significant change from baseline to Week 4 (0% [-30%, +41%], p=0.36) or Week 12 (+1% [-36%, +50%], p=0.97). In adjusted analyses, predictors of BNP rise were similar to those of NT-proBNP rise and included atrial fibrillation, worse renal function, lower BMI, and lower concentrations of BNP/NT-proBNP at baseline (<b>Figure</b>). Change in BNP was strongly and linearly associated with change in NT-proBNP (rho=0.81, p<0.001), and the median [IQR] NT-proBNP change/BNP change was 66% [51%, 86%] (<b>Figure</b>). In contrast, change in BNP was only weakly associated with change in ucGMP (rho=0.19, p<0.001), and increases in ucGMP were observed regardless of whether BNP was decreased (+11% [-34%, +115%]), unchanged (+34% [-15%, +205%]), or increased (+57% [-12%, +14%]). <h3>Conclusions</h3> In this pooled analysis, following initiation of Sac/Val in HFrEF there was no significant overall increase in BNP concentrations. Early changes in BNP correlated closely with change in NT-proBNP but weakly with changes in ucGMP. Patients treated with Sac/Val demonstrate increase in ucGMP regardless of the BNP trajectory. These data argue against use of BNP rise as a biomarker of Sac/Val efficacy in HFrEF