Rationale Estimating the causal effect of an intervention at individual level, also called individual treatment effect (ITE), may help in identifying response prior to the intervention. Objectives We aimed to develop machine learning (ML) models which estimate ITE of an intervention using data from randomised controlled trials and illustrate this approach with prediction of ITE on annual chronic obstructive pulmonary disease (COPD) exacerbation rates. Methods We used data from 8151 patients with COPD of the Study to Understand Mortality and MorbidITy in COPD (SUMMIT) trial ([NCT01313676][1]) to address the ITE of fluticasone furoate/vilanterol (FF/VI) versus control (placebo) on exacerbation rate and developed a novel metric, Q-score, for assessing the power of causal inference models. We then validated the methodology on 5990 subjects from the InforMing the PAthway of COPD Treatment (IMPACT) trial ([NCT02164513][2]) to estimate the ITE of FF/umeclidinium/VI (FF/UMEC/VI) versus UMEC/VI on exacerbation rate. We used Causal Forest as causal inference model. Results In SUMMIT, Causal Forest was optimised on the training set (n=5705) and tested on 2446 subjects (Q-score 0.61). In IMPACT, Causal Forest was optimised on 4193 subjects in the training set and tested on 1797 individuals (Q-score 0.21). In both trials, the quantiles of patients with the strongest ITE consistently demonstrated the largest reductions in observed exacerbations rates (0.54 and 0.53, p<0.001). Poor lung function and blood eosinophils, respectively, were the strongest predictors of ITE. Conclusions This study shows that ML models for causal inference can be used to identify individual response to different COPD treatments and highlight treatment traits. Such models could become clinically useful tools for individual treatment decisions in COPD. Data are available upon reasonable request. The anonymised individual participant data set of SUMMIT (Sponsor ID HZC113782) and IMPACT (Sponsor ID CTT116855) were provided upon request by clinicalstudydatarequest.com with the approval of GlaxoSmithKline Research & Development Ltd. Protocols, reporting and analysis plans and clinical study reports are available for both studies. [1]: /lookup/external-ref?link_type=CLINTRIALGOV&access_num=NCT01313676&atom=%2Fthoraxjnl%2F78%2F10%2F983.atom [2]: /lookup/external-ref?link_type=CLINTRIALGOV&access_num=NCT02164513&atom=%2Fthoraxjnl%2F78%2F10%2F983.atom

中文翻译：

---

通过随机对照试验中的因果机器学习估计个体治疗对 COPD 恶化的影响

基本原理 估计个体水平干预的因果效应，也称为个体治疗效应 (ITE)，可能有助于在干预之前确定反应。目标 我们的目标是开发机器学习 (ML) 模型，该模型使用随机对照试验的数据来估计干预的 ITE，并通过预测 ITE 对年度慢性阻塞性肺疾病 (COPD) 恶化率来说明这种方法。方法 我们使用了来自 8151 名 COPD 患者的数据来了解 COPD 中的死亡率和发病率 (SUMMIT) 试验 ([NCT01313676][1])，以确定糠酸氟替卡松/维兰特罗 (FF/VI) 与对照（安慰剂）的 ITE并开发了一种新的指标 Q-score，用于评估因果推理模型的功效。然后，我们对来自 Informing the Pathway of COPD Treatment (IMPACT) 试验 ([NCT02164513][2]) 的 5990 名受试者验证了该方法，以估计 FF/umeclidinium/VI (FF/UMEC/VI) 与 UMEC/VI 的 ITE恶化率。我们使用因果森林作为因果推理模型。结果 在 SUMMIT 中，因果森林在训练集 (n=5705) 上进行了优化，并在 2446 名受试者上进行了测试（Q 得分 0.61）。在 IMPACT 中，因果森林在训练集中的 4193 名受试者上进行了优化，并在 1797 名个体上进行了测试（Q 得分 0.21）。在这两项试验中，具有最强 ITE 的患者的分位数始终表现出观察到的恶化率的最大降低（0.54 和 0.53，p<0.001）。肺功能差和血液嗜酸性粒细胞分别是 ITE 的最强预测因素。结论 这项研究表明，用于因果推理的 ML 模型可用于识别个体对不同 COPD 治疗的反应并突出治疗特征。这些模型可能成为临床上有用的慢性阻塞性肺病个体治疗决策工具。数据可根据合理要求提供。SUMMIT（赞助商ID HZC113782）和IMPACT（赞助商ID CTT116855）的匿名个人参与者数据集是根据clinicalstudydatarequest.com的要求并经葛兰素史克研究开发有限公司批准提供的。协议、报告和分析计划以及临床研究报告均可供使用对于这两项研究。[1]: /lookup/external-ref?link_type=CLINTRIALGOV&access_num=NCT01313676&atom=%2Fthoraxjnl%2F78%2F10%2F983.atom [2]: /lookup/external-ref?link_type=CLINTRIALGOV&access_num=NCT02164513&atom=%2Fthoraxjnl%2F7 8%2F10 %2F983.原子