F014 Handling censored covariates in Huntington disease studies

<h3></h3> Cognitive dysfunction is a symptom of Huntington disease (HD) and can serve as an early marker for evaluating treatments intended at delaying motor skill decline. Understanding cognitive dysfunction as a function of age at clinical diagnosis is a prerequisite for their use in clinical trials, but this remains challenging because patients drop out or the study ends, resulting in a censored value for age at clinical diagnosis. While right-censored time- to-event outcomes such as age at clinical diagnosis have been studied for decades, handling time-to-event covariates, also known as censored covariates, is now of growing interest. So far, the literature has assumed that the covariate censoring is noninformative, i.e., event time is independent of censoring time, and has yet to establish methods when covariate censoring is in fact informative. Here, we present five methods capable of handling both informative and noninformative covariate censoring, namely a complete case, inverse probability weighting (IPW), augmented complete case, augmented IPW, and maximum likelihood estimators. We establish the robustness properties of these estimators in terms of when these estimators remain consistent despite incorrect distributional assumptions. All methods are applied to ENROLL-HD, a non-interventional observational study of HD, and compare the effect estimates and variances of each estimator.