Estimating treatment duration effects via clone-censor-weight: a breast cancer case study

2026-05-19

AI Summary

This study addresses bias in estimating treatment duration effects from observational survival data arising from time-varying confounding and immortal time bias. Building on the clone–censor–weight (CCW) framework to emulate a target trial, it clarifies core identifying assumptions and distinguishes artificial from natural censoring. Under scenarios involving baseline and time-varying confounders, the authors systematically compare the performance of inverse probability of censoring weighting (IPCW), G-formula, and doubly robust estimators. The approach is applied to a breast cancer cohort to evaluate the comparative effectiveness of two versus five years of adjuvant tamoxifen therapy. Results reveal substantial estimation uncertainty, underscoring the necessity of sensitivity analyses and offering methodological guidance for causal inference in complex observational studies.

Abstract

In this work, we study the estimation of treatment duration effects in observational survival data, where treatment and covariate histories evolve over time and longer observed durations are only attainable among individuals who remain event-free and under follow-up, leading to immortal time bias under naive analyses. The cloning-censoring-weighting (CCW) framework provides a practical approach to emulate target trials of treatment duration strategies, but several methodological aspects remain insufficiently understood. We focus on static treatment duration strategies under two settings of increasing complexity: baseline confounding only, and confounding with time-varying covariates. We formalize the assumptions underlying CCW, with particular emphasis on treatment admissibility, relaxed intervention rules, and the distinction between artificial and natural censoring. We then compare several estimation approaches after cloning and censoring, including inverse probability of censoring weighting (IPCW), the G-formula, and doubly robust estimators, through simulation studies assessing robustness, variability, and sensitivity to censoring model misspecification. Finally, we apply the framework to a Breast Cancer cohort to emulate a target trial comparing 2 versus 5 years of adjuvant tamoxifen in early stage breast cancer. Due to the small number of events and limited support for the 2-year strategy, estimates are associated with substantial uncertainty. These findings highlight both the practical relevance and the limitations of CCW, and underscore the importance of sensitivity analyses in complex longitudinal observational settings.

Problem

• treatment duration effects
• immortal time bias
• observational survival data
• time-varying confounding
• target trial emulation

Innovation

• cloning-censoring-weighting
• treatment duration effect
• immortal time bias
• doubly robust estimation
• observational survival data