用g方法解决ALTA-1L试验中的治疗切换问题:探索模型规范的影响
Amani Al Tawil1,2, Sean McGrath3, Robin Ristl4
1Institute for Medical Information Processing, Biometry, and Epidemiology (IBE), Faculty of Medicine, Ludwig-Maximilians-Universität München, Marchioninistr. 15, 81377, Munich, Germany. altawil@ibe.med.uni-muenchen.de.
在临床试验中切换治疗可能会掩盖真正的治疗效果. 这项研究发现,布里加替尼可能比克里佐替尼提供更大的整体生存益处,估计有10-45%的改善,与初步分析相反.
科学领域:
- 在瘤学瘤学.
- 生物统计学 生物统计学
- 临床试验 临床试验
背景情况:
- 随机临床试验中的治疗转换使因果推断复杂化.
- 治疗意图 (ITT) 分析可能不会准确地反映治疗效应,当切换发生时.
- 假设的治疗策略,不包括切换,对决策有兴趣.
研究的目的:
- 用g方法重新分析ALTA-1L试验数据,以估计假设治疗策略的因果关系.
- 为了比较"总是布里加替尼"与"总是克里佐替尼"在没有治疗转换的情况下的整体存活率 (OS) 的有效性.
- 评估治疗转换对估计治疗效果的影响.
主要方法:
- 使用定向非循环图来建模临床环境并识别混因素.
- 使用审查权重的反向概率 (IPCW) 和因果推理的参数g公式.
- 在重新分析ALTA-1L试验数据 (NCT02737501) 中,根据基线和时间变化的共变量进行调整.
主要成果:
- 根据IPCW方法,累计危险比率 (cHR) 在0.38-0.73之间,风险比率 (RR) 在0.52-0.79之间.
- 参数g-公式的结果是cHRs在0.61-0.72之间,RRs在0.71-0.79.
- 这两种方法都表明,ITT分析低估了布里加替尼的OS益处.
结论:
- ITT分析 (cHR:0.82) 可能低估了布里加替尼的益处10-45% (IPCW) 或10-20% (g-公式).
- 敏感性分析至关重要,因为单个分析可能会产生异常结果.
- 在治疗转换的情况下,G方法提供了更准确的治疗效应估计.
更多相关视频
10:50Using the FishSim Animation Toolchain to Investigate Fish Behavior: A Case Study on Mate-Choice Copying In Sailfin Mollies
Published on: November 8, 2018
05:15The Spatial Memory Game: Testing the Relationship Between Spatial Language, Object Knowledge, and Spatial Cognition
Published on: February 19, 2018
相关概念视频
Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
Comparing the Survival Analysis of Two or More Groups
Mechanistic Models: Compartment Models in Individual and Population Analysis
Strategies for Assessing and Addressing Confounding
Confounding can be addressed at both the design phase of a study and through analytical methods after data...
Crossover Experiments
Crossover designs are performed even with smaller sample sizes since the samples can act as their controls. These are better than simple randomized trials since patients are exposed to all the treatments.
Assumptions of Survival Analysis
