A0601
Title: Resilience measures for the surrogate paradox
Authors: Layla Parast - University of Texas at Austin (United States) [presenting]
Abstract: Surrogate markers are often used in clinical trials to evaluate treatment effects when primary outcomes are costly, invasive, or take a long time to observe. However, reliance on surrogates can lead to the "surrogate paradox, where a treatment appears beneficial based on the surrogate but is actually harmful with respect to the primary outcome. Formal measures are proposed to assess resilience against the surrogate paradox. The setting assumes an existing study in which the surrogate marker and primary outcome have been measured (Study A) and a new study (Study B) in which only the surrogate is measured. Rather than assuming transportability of the conditional mean functions across studies, a class of functions is considered for Study B that deviates from those in Study A. Using these, the distribution of potential treatment effects is estimated on the unmeasured primary outcome and defines resilience measures, including a resilience probability, resilience bound, and resilience set. The approach complements traditional surrogate validation methods by quantifying the plausibility of the surrogate paradox under controlled deviations from what is known from Study A. The performance of the proposed measures is investigated via a simulation study and application to two distinct HIV clinical trials.
