Changes in NT-proBNP, Safety, and Tolerability in HFpEF Patients With a WHF Event (HFpEF Decompensation) Who Have Been Stabilized and Initiated at the Time of or Within 30 Days Post-decompensation (PARAGLIDE-HF)

This study used a randomized, double-blind, double-dummy, active-controlled, parallel group design conducted across 100 centers in the US and Canada. The study duration was a maximum of 20 months (minimum follow up was 8 weeks).

Randomized patients were deemed hemodynamically stabilized and needed to meet all inclusion and none of the exclusion criteria. Patients were randomized 1:1 to LCZ696 or valsartan. Initial dose at randomization was determined based on the patient's previous dose of or lack of ACEi/angiotensin receptor blocker (ARB) immediately prior to current worsening heart failure (WHF) event (heart failure with preserved ejection fraction \[HFpEF\]) decompensation, or at the time of post-decompensation randomization.

LCZ696 dose or valsartan dose levels may have been increased to the targeted desired dose of 97/103 mg \[200 mg\] BID or valsartan 160 mg BID on an every 2-week basis or earlier based on clinical need and investigator judgment. Every effort was made to titrate to and maintain patients on the target dose level, as tolerated by the patient.

To maintain the blinding, patients were required to take their assigned active treatment tablet along with placebo matching the opposite treatment BID.

The protocol had 4 amendments. Protocol Version 00 (Original Protocol) included a double-blind phase through Week 8 followed by an open-label phase during Weeks 8 to 12. Protocol Amendment 01 omitted the open-label phase and followed patients for a maximum of 20 months in a double-blinded treatment phase. Throughout all protocol versions, the primary endpoint remained the time-averaged proportional change in N-terminal pro-brain natriuretic peptide (NT-proBNP) from Baseline to Weeks 4 and 8. The most recent protocol amendment (Amendment 04) reduced the sample size to approximately 450 patients (from 800) with 85% power for the primary endpoint, deemphasizing the statistical power for key secondary clinical endpoints; however, clinical events were still assessed as secondary endpoints.

No efficacy analyses include "OPEN LABEL' data. After Protocol Amendment 01, the open-label option was removed from the study, only the 233 patients randomized in the Double-blind Phase Sacubitril+ Valsartan (LCZ696) and the 233 patients randomized in the Double-blind Phase Valsartan arms were included in the efficacy analysis.