抗癌疗法长期疗效评估中的统计学考量
Statistical considerations in long-term efficacy evaluation of anti-cancer therapies.
作者信息
Li Ruobing, Zhang Jingyi, Wang Jingzhao, Wang Jun
机构信息
Office of Biostatistics and Clinical Pharmacology, Center for Drug Evaluation, National Medical Products Administration, Beijing, China.
Research Center of Biostatistics and Computational Pharmacy, China Pharmaceutical University, Nanjing, China.
出版信息
Front Pharmacol. 2023 Oct 3;14:1265953. doi: 10.3389/fphar.2023.1265953. eCollection 2023.
Anti-cancer therapy has been a significant focus of research. Developing and marketing various types and mechanisms of anti-cancer therapies benefit a variety of patients significantly. The long-term benefit to patients in evaluating the risk-benefit ratio of anti-cancer therapy has become a significant concern. This paper discusses the evaluation of long-term efficacy within the estimand framework and summarizes the various strategies for addressing potential intercurrent events. Non-proportional hazards of survival data may arise with novel anti-cancer therapies, leading to potential bias in conventional evaluation methods. This paper reviews statistical methods for addressing this issue, including novel endpoints, hypothesis testing, and efficacy estimation methods. We also discuss the influences of treatment switching. Although advanced methods have been developed to address the non-proportional hazard, they still have limitations that require continued collaborative efforts to resolve issues.
抗癌治疗一直是研究的重点。开发和推广各种类型及作用机制的抗癌疗法，使众多患者显著受益。评估抗癌治疗的风险效益比给患者带来的长期益处，已成为一个重大关注点。本文讨论了在估计量框架内对长期疗效的评估，并总结了应对潜在并发事件的各种策略。新型抗癌疗法可能会出现生存数据的非比例风险，从而导致传统评估方法存在潜在偏差。本文回顾了用于解决这一问题的统计方法，包括新型终点、假设检验和疗效估计方法。我们还讨论了治疗转换的影响。尽管已经开发出先进方法来解决非比例风险问题，但它们仍存在局限性，需要持续的合作努力来解决这些问题。
Zotero 插件