A critical evaluation of the EFS endpoint in AML: Does induction treatment failure timing have a profound impact on study design and results?

Abstract

Despite emerging novel therapies, treating acute myeloid leukemia (AML) remains challenging. Complexities persist in designing pivotal clinical trials and establishing acceptable endpoints for AML. Recent FDA guidance for drug and biological products development for AML outlines considerations for trial design. The guidance defines overall survival (OS) and event-free survival (EFS) as endpoints representing clinical benefit for AML therapies without curative intent. We highlight the EFS definition, particularly the assignment of day 1 as the event date for patients with induction treatment failures (ITFs), as recommended in the guidance. Through a comprehensive simulation study, our results show that the guidance EFS definition performs adequately with high complete remission (CR) rates but may pose challenges for low CR rates. When the experimental arm CR rate is 5% or less over the control, the use of the ITF events at day 1 for EFS definition leads to a critical power decrease, hampering the ability to predict survival benefit for a moderate OS duration. We further expand upon the EFS definition with the event date at ITF period end. Our goal is to inform investigators and regulatory agencies about the implications and limitations of various EFS definitions for future pivotal trials in AML.