Factors Associated With Noncompletion During the Run-In Period Before Randomization and Influence on the Estimated Benefit of LCZ696 in the PARADIGM-HF Trial

Background-The 8442 patients randomized in the Prospective Comparison of Angiotensin Receptor-Neprilysin Inhibitor With an Angiotensin-Converting Enzyme Inhibitor to Determine Impact on Global Mortality and Morbidity in Heart Failure (PARADIGM-HF) trial, in which sacubitril/valsartan (LCZ696) reduced both death and HF hospitalization more than enalapril, were a subset of 10 521 patients entering sequential, single-blind run-in periods (enalapril 10 mg twice daily for 2 weeks followed by LCZ696 200 mg twice daily for 4 to 6 weeks) to ensure short-term tolerability of the 2 study medications. We identified the predictors of run-in noncompletion and estimated the implications of noncompletion for the overall study result. Methods and Results-Patient factors associated with run-in noncompletion were defined in multivariable logistic regression models. The effectiveness of LCZ696 in a broader cohort approximating the full run-in population was estimated by weighting randomized patients according to the inverse probability of run-in completion; 2079 (19.8%) subjects discontinued the study during the run-in period, including 1102 (10.5%) during the enalapril phase and 977 (9.3%) during the LCZ696 phase. In multivariable models, lower systolic blood pressure, lower estimated glomerular filtration rate, higher N-terminal pro-B-type natriuretic peptide, and ischemic cause of heart failure were associated with higher risk for run-in noncompletion. Repeat analysis of the effect of randomized treatment giving greater weight to randomized patients resembling those who did not complete the run-in did not alter the hazard ratio favoring LCZ696 over enalapril for the primary end point of cardiovascular death or heart failure hospitalization, or the additional key end points of cardiovascular death and all-cause mortality. Conclusions-Patients with lower blood pressure, lower glomerular filtration rate, and more severe heart failure were at higher risk for noncompletion during the run-in period of PARADIGM-HF. Weighted analysis of key study outcomes accounting for the effect of run-in noncompletion did not alter the benefit of LCZ696 over enalapril.