药物非劣效临床试验剖析

Dissection of non-inferiority clinical trials to establish effectiveness

非劣效试验虽然历经数十年的实践,但至今仍有争议,尤其是国际相关指南在我国被误解和误用,需要重新认识。一般来说,非劣效检验只能回答受试药是否优于安慰剂,确认是否有效,难以获得与阳性对照药的比较性结论,即非劣效试验不产生真正的非劣效结论,除非另设疗效差异没有临床意义的界值。其次,非劣效界值是疗效差异的可信区间限定值,属于统计学界值,但其大小的选择又归结为临床可接受的样本量多少。因此非劣效界值不能等同于与阳性对照药的疗效差异,更不是受试药与阳性对照药相比在临床上可接受的最大疗效损失。另外,非劣效试验在我国用于仿制药评价具有历史原因,由于不能产生比较性和可互换性结论,理应摒弃,应回归到临床终点的生物等效试验。本文从基本的分析原理出发,以国际指南为依据,采用实例阐明以上关键问题,试图提出一些解决方案,为我国非劣效试验的正确开展提供参考。

Although non-inferiority trials have been practiced for decades,they continue to be a subject of ongoing debate,especially due to the misunderstanding and misuse of international guidelines,necessitating a reevaluation.Generally,current non-inferiority testing methods can only determine whether a test drug is superior to a placebo,confirming its efficacy,but they struggle to provide comparative conclusions with positive control drugs.That is to say,non-inferiority trials do not produce genuine non-inferiority conclusions unless a clinical threshold for efficacy difference is additionally set.Furthermore,the non-inferiority margin represents a statistical limit in the confidence interval of the efficacy difference.However,the choice of its size boils down to clinical judgment,correlating to the clinically acceptable sample size.Therefore,the non-inferiority margin should be neither equated with an efficacy difference,nor considered the maximum clinically acceptable loss of efficacy when comparing the test drug with a positive control drug.Moreover,the use of non-inferiority trials for evaluating generic drugs in China has historical reasons.Since these trials cannot produce comparative conclusions,they should be discarded in favor of bioequivalence trials with clinical endpoints.This article,starting from basic principles and relying on international guidelines,uses examples to clarify the above key issues and attempts to propose some solutions,providing a reference for the proper conduct of non-inferiority trials in China.