Department of Statistics and Actuarial Science, The University of Hong Kong, Hong Kong.
Department of Statistics and Actuarial Science, The University of Hong Kong, Hong Kong.
Contemp Clin Trials. 2021 May;104:106353. doi: 10.1016/j.cct.2021.106353. Epub 2021 Mar 9.
In oncology clinical trials, the primary endpoint is often time to an event of clinical interest, e.g., time to disease progression or time to death. As a result, progression-free survival (PFS: the time from initiation of treatment till disease progression or death whichever occurs first) and overall survival (OS: the time from initiation of treatment till death) are the focus of statistical analysis in comparison of two treatment arms. It is often argued that PFS may serve as a surrogate endpoint for OS, while the validity of such surrogacy is still under debates in different types of cancer. In practice, one may observe a significant difference in PFS but no significant difference in OS; or vice versa. We provide a concordance index (C-index) to measure the degree of concordance between PFS and OS, and elaborate on the PFS vs OS discrepancies using the C-index using simulation studies and real trial analysis.
在肿瘤学临床试验中,主要终点通常是临床关注事件的时间,例如疾病进展时间或死亡时间。因此,无进展生存期(PFS:从治疗开始到疾病进展或死亡的时间,以先发生者为准)和总生存期(OS:从治疗开始到死亡的时间)是比较两种治疗方案的统计分析的重点。人们常认为 PFS 可以作为 OS 的替代终点,但这种替代的有效性在不同类型的癌症中仍存在争议。在实践中,人们可能会观察到 PFS 有显著差异但 OS 无显著差异;或者反之亦然。我们提供了一个一致性指数(C-index)来衡量 PFS 和 OS 之间的一致性程度,并通过模拟研究和真实试验分析使用 C-index 详细说明 PFS 与 OS 的差异。