高变异药物生物等效性评价确切样本量计算

发布时间：2018-11-21 16:03

【摘要】：目的探讨高变异药物生物等效性评价(BE)的确切样本量计算方法。方法基于常规2×2交叉设计生物等效性评价双单侧检验原理和统计分布理论,阐明其确切样本量估计的方法。分析美国食品药品监督管理局(FDA)和欧洲药品管理局(EMA)对高变异药物BE研究指南所涉及的方法学内涵,推导出高变异药物生物等效性评价两种重复交叉设计下确切样本量计算的方法。借助SAS编程计算出两种重复交叉设计下不同参数设定的确切样本量列表,并与已发表文献的公式近似法结果进行比较,分析其差别规律。结果因为EMA指南利用参比制剂个体内变异调整生物等效限值的标准比FDA指南更加严格,故其在同样的参数设定下所需的样本量更大。与文献公式近似法的结果对比,当几何均数的比值(GMR)为1时,确切法的样本量均大于近似法,而对于GMR不为1的情形,确切法普遍小于或等于近似法。本文还给出EMA指南要求下确切样本量计算的SAS程序代码,可容易实现不同参数设定下的确切样本量计算。结论本文给出的FDA和EMA指南高变异药物BE研究生物等效性评价的确切样本量计算方法具有严谨的统计理论基础,提供的SAS程序代码简明实用,希望能为开展高变异药物BE研究的样本量确定提供方法学支持。
[Abstract]:Objective to study the exact sample size calculation method for bioequivalence evaluation of highly variable drugs (BE). Methods based on the principle of two-sided test and statistical distribution theory for evaluating the bioequivalence of conventional 2 脳 2 cross-design, the method of accurate sample size estimation was expounded. To analyze the methodological implications of the Food and Drug Administration (FDA) and European Drug Administration (EMA) guidelines for the study of highly variable drugs (BE). A method for calculating the exact sample size under two repeated cross designs for bioequivalence evaluation of highly variable drugs was derived. By means of SAS programming, the exact sample size list of different parameter setting under two repeated cross design is calculated, and compared with the results of the formula approximation method published in the literature, the difference law is analyzed. The result is that the EMA guidelines are more stringent than the FDA guidelines in adjusting bioequivalence limits using reference agents' variations in vivo, so it requires a larger sample size under the same parameters. Compared with the results of the approximate method in literature, when the ratio of geometric mean (GMR) is 1, the sample size of the exact method is larger than that of the approximate method, but for the case of GMR not equal to 1, the exact method is generally less than or equal to the approximate method. In this paper, the SAS program code for exact sample size calculation under EMA guidelines is also given, which can easily realize the exact sample size calculation under different parameter settings. Conclusion the exact sample size calculation method of the FDA and EMA guidelines for the study of bioequivalence evaluation of high variant drug BE is based on rigorous statistical theory and provides a concise and practical SAS program code. It is hoped to provide methodological support for the determination of sample size for the study of high-variant drug BE.
【作者单位】：南方医科大学南京临床医学院南京医科大学金陵临床医学院南京军区南京总医院医学统计学教研室;中国药科大学药学院;
【基金】：国家自然科学基金资助项目(81473066)
【分类号】：R96

【相似文献】