Patient-Reported Outcome Time-to-Event Estimands in Lung Cancer Trials: What Is the Question?

• Activity Number: 577 - Estimands: Some Perspectives from an FDA Working Group
• Type: Topic Contributed
• Date/Time: Thursday, August 6, 2020 : 3:00 PM to 4:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Mallorie Fiero* and Jessica Roydhouse and Jonathon Vallejo and Jiaxi Zhou and Ting-Yu Chen and Bellinda King-Kallimanis and Paul Kluetz and Rajeshwari Sridhara
• Companies: U.S. Food and Drug Administration and Menzies Institute for Medical Research, University of Tasmania and FDA and FDA and FDA and FDA and FDA and FDA
• Keywords: Estimands; Patient-reported outcomes; Time-to-event analysis; Oncology

Abstract

A time-to-event analysis, such as time-to-deterioration (TTD), is the most common type of inferential patient-reported outcome (PRO) analysis in lung cancer trials. However, varying definitions point to different scientific questions of TTD, which makes interpretation of patient experience difficult. A TTD analysis compares time from randomization to deterioration of a symptom between two treatment arms. Deterioration is defined using a threshold that identifies the point in which a patient experiences a clinically meaningful worsening of a symptom score from baseline. In order to better understand what TTD questions were being asked in trials, we reviewed TTD estimands in lung cancer trials submitted to FDA from 2008-2019 based on the ICH E9(R1) addendum. Alignment of the estimand with the PRO objective should be explicitly pre-specified to facilitate transparency in interpreting patient experience in oncology trials. Further work is needed on TTD to better understand the utility of TTD in FDA’s risk-benefit assessment.