Chen Li, Burkard Mark, Wu Jianrong, Kolesar Jill M, Wang Chi
Division of Cancer Biostatistics, Department of Internal Medicine, University of Kentucky, Lexington, Kentucky.
Markey Cancer Center, University of Kentucky, Lexington, Kentucky.
Stat Med. 2023 Feb 10;42(3):388-406. doi: 10.1002/sim.9622. Epub 2022 Dec 27.
With the rapid development of new anti-cancer agents which are cytostatic, new endpoints are needed to better measure treatment efficacy in phase II trials. For this purpose, Von Hoff (1998) proposed the growth modulation index (GMI), that is, the ratio between times to progression or progression-free survival times in two successive treatment lines. An essential task in studies using GMI as an endpoint is to estimate the distribution of GMI. Traditional methods for survival data have been used for estimating the GMI distribution because censoring is common for GMI data. However, we point out that the independent censoring assumption required by traditional survival methods is always violated for GMI, which may lead to severely biased results. In this paper, we construct both nonparametric and parametric estimators for the distribution of GMI, accounting for the dependent censoring of GMI. Extensive simulation studies show that our nonparametric estimators perform well in practical situations and outperform existing estimators, and our parametric estimators perform better than our nonparametric estimators and existing estimators when the parametric model is correctly specified. A phase II clinical trial using GMI as the primary endpoint is provided for illustration.
随着具有细胞生长抑制作用的新型抗癌药物的迅速发展，在II期试验中需要新的终点指标来更好地衡量治疗效果。为此，冯·霍夫（1998年）提出了生长调节指数（GMI），即两个连续治疗阶段的疾病进展时间或无进展生存时间之比。在以GMI作为终点指标的研究中，一项重要任务是估计GMI的分布。由于GMI数据中删失情况常见，传统的生存数据方法已被用于估计GMI分布。然而，我们指出传统生存方法所需的独立删失假设对于GMI总是不成立的，这可能导致严重有偏的结果。在本文中，我们构建了GMI分布的非参数和参数估计量，同时考虑了GMI的相依删失。大量的模拟研究表明，我们的非参数估计量在实际情况下表现良好且优于现有估计量，并且当参数模型正确设定时，我们的参数估计量比非参数估计量和现有估计量表现更好。本文提供了一项以GMI作为主要终点指标的II期临床试验作为例证。
