文献速递流行病学与生物统计临床试验与 GCP

预测概率简化计算:一种快速准确的临床试验决策方法

Predictive Probabilities Made Simple: A Fast and Accurate Method for Clinical Trial Decision-Making.
PubMed 文献 · Stat Med · 2025-01-01

提出一种利用p值或后验概率近似计算贝叶斯预测概率的方法,以减轻临床试验期中监测与试验模拟的计算负担。通过五项模拟研究,在二分类、时间事件、有序终点及历史借用和纵向建模等主要分析策略下,将该近似法与完整预测概率比较。结果显示该快速近似法在所有五项应用中表现良好,可显著减少试验模拟计算量,从而模拟更多虚拟试验以更精确估计试验操作特征。

要点

  • 贝叶斯预测概率常用于临床试验期中监测的疗效与无效停止规则。
  • 新近似法利用p值或后验概率计算预测概率,显著降低计算负担。
  • 五项模拟研究表明该近似法与标准蒙特卡洛插补法高度一致,在所有应用中表现良好。

核心结论

  • 该快速预测概率近似法在所有五项应用中表现良好,显著减少试验模拟的计算负担,允许模拟更多虚拟试验以更精确估计试验操作特征。

研究设计与样本

  • 方法学研究,提出利用p值或后验概率近似计算预测概率,并通过五项模拟研究比较该近似法与完整预测概率。

主要发现

  • 该近似法与标准蒙特卡洛插补法计算预测概率高度一致。
  • 五项模拟研究涵盖二分类、时间事件、有序终点、历史借用和纵向建模等主要分析策略,近似法在所有应用中表现良好。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

Bayesian predictive probabilities are commonly used for interim monitoring of clinical trials through efficacy and futility stopping rules. Despite their usefulness, the calculation of predictive probabilities, particularly in pre-experiment trial simulation, can be a significant challenge. We introduce an approximation for computing predictive probabilities using either a p $$ p $$ -value or a posterior probability that significantly reduces this burden. We show the approximation has a high degree of concordance with standard Monte Carlo imputation methods for computing predictive probabilities, and present five simulation studies comparing the approximation to the full predictive probability for a range of primary analysis strategies: Dichotomous, time-to-event, and ordinal endpoints, as well as historical borrowing and longitudinal modeling. We find that this faster method of predictive probability approximation works well in all five applications, thus significantly reducing the computational burden of trial simulation, allowing more virtual trials to be simulated to achieve greater precision in estimating trial operating characteristics.

来源:Stat Med,查看原文。版权归原作者所有。

引用

  1. Marion J, Lorenzi E, Allen-Savietta C, et al. Predictive Probabilities Made Simple: A Fast and Accurate Method for Clinical Trial Decision-Making. Stat Med. 2025. PMID: 40459275. doi:10.1002/sim.70120

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。