Abstract 4015: Validity of the Combined Efficacy plus Safety Composite Endpoint (Net Clinical Benefit) in TRITON-TIMI 38

Abstract

BACKGROUND: Composite endpoints (CE) are frequently used in clinical trials for parsimonious summarization of treatment effects. The construction of the CE is generally based on the premise that each component endpoint is interchangeable. However, for this assumption to be valid, each component should be of equal or comparable clinical importance, occur with similar frequency, and be equally responsive to treatment intervention. OBJECTIVE: To examine the validity of the combined efficacy plus safety “quadruple” CE of Net Clinical Benefit utilized in TRITON. METHODS: A hierarchical (excluding double counting) CE analysis was performed using all-cause death, nonfatal MI, nonfatal stroke and TIMI Major or TIMI Major plus Minor bleeding. A formal assessment of heterogeneity was also conducted using the Cochran’s Q and I 2 tests. RESULTS: Figure (top panel) shows a wide variability in the contribution of each component to the CE (death 23%, MI 55%, stroke 7%, major bleeding 15%). There was significant heterogeneity of treatment effect across the components (P=0.0006) with treatment differences primarily attributable to nonfatal MI and bleeding with little or no impact on death or nonfatal stroke. Similar observations are seen with TIMI major plus minor bleeding (bottom panel). CONCLUSIONS: The benefit of prasugrel is driven by nonfatal MI, the most prevalent but arguably not the most important component of the CE. A large gradient in clinical importance, prevalence, and treatment effect across components challenges the validity of “net clinical benefit” CE in TRITON. Furthermore, lack of information about the type of MI prevented precludes an informed benefit-risk assessment.