A Comparison of Estimand and Estimation Strategies for Clinical Trials in Early Parkinson's Disease

TL;DR

This paper evaluates different estimand and estimation strategies for clinical trials in early Parkinson's disease, focusing on handling intercurrent events and comparing estimators through simulation based on real cohort data.

Contribution

It introduces and compares multiple imputation-based estimators tailored for specific estimand strategies in Parkinson's disease trials, addressing intercurrent events.

Findings

Estimators based on missing-at-random assumptions show varying bias and MSE.

Inclusion of time-varying ICE indicators improves estimator performance.

Simulation results guide optimal estimand and estimator choices for PD trials.

Abstract

Parkinson's disease (PD) is a chronic, degenerative neurological disorder. PD cannot be prevented, slowed or cured as of today but highly effective symptomatic treatments are available. We consider relevant estimands and treatment effect estimators for randomized trials of a novel treatment which aims to slow down disease progression versus placebo in early, untreated PD. A commonly used endpoint in PD trials is the MDS-Unified Parkinson's Disease Rating Scale (MDS-UPDRS), which is longitudinally assessed at scheduled visits. The most important intercurrent events (ICEs) which affect the interpretation of the MDS-UPDRS are study treatment discontinuations and initiations of symptomatic treatment. Different estimand strategies are discussed and hypothetical or treatment policy strategies, respectively, for different types of ICEs seem most appropriate in this context. Several estimators based on multiple imputation which target these estimands are proposed and compared in terms of bias, mean-squared error, and power in a simulation study. The investigated estimators include methods based on a missing-at-random (MAR) assumption, with and without the inclusion of time-varying ICE-indicators, as well as reference-based imputation methods. Simulation parameters are motivated by data analyses of a cohort study from the Parkinson's Progression Markers Initiative (PPMI).