Department of Statistics and Actuarial Science, University of Waterloo, Waterloo, ON N2L 3G1, Canada.
Stat Med. 2018 May 30;37(12):1947-1959. doi: 10.1002/sim.7641. Epub 2018 Mar 26.
Therapeutic advances in cancer mean that it is now impractical to performed phase III randomized trials evaluating experimental treatments on the basis of overall survival. As a result, the composite endpoint of progression-free survival has been routinely adopted in recent years as it is viewed as enabling a more timely and cost-effective approach to assessing the clinical benefit of novel interventions. This article considers design of cancer trials directed at the evaluation of treatment effects on progression-free survival. In particular, we derive sample size criteria based on an illness-death model that considers cancer progression and death jointly while accounting for the fact that progression is assessed only intermittently. An alternative approach to design is also considered in which the sample size is derived based on a misspecified Cox model, which uses the documented time of progression as the progression time rather than dealing with the interval censoring. Simulation studies show the validity of the proposed methods.
癌症治疗的进步意味着，现在已经不切实际地在基于总生存期的基础上进行评估实验性治疗的 III 期随机试验。因此，近年来，无进展生存期的复合终点已被常规采用，因为它被认为能够更及时、更具成本效益地评估新干预措施的临床获益。本文考虑了针对无进展生存期评估治疗效果的癌症试验设计。特别是，我们根据疾病死亡模型推导了样本量标准，该模型联合考虑了癌症进展和死亡，同时考虑了仅间歇性评估进展的事实。还考虑了另一种设计方法，其中根据指定不当的 Cox 模型推导样本量，该模型将记录的进展时间用作进展时间，而不是处理区间 censoring。模拟研究表明了所提出方法的有效性。
