Bootstrap方法根据预试验数据指导生物等效性试验样本量的案例分析
|
摘要
在仿制药生物等效性评价中,通过预试验可以为正式试验的样本量选择提供重要依据。除了几何均值比值和变异系数(CV),个体间一致性趋势也是重要信息。在具体工作实践中,常碰到一些和整体趋势不太一致的个体数据,纳入后查表所得到的样本例数远超预期,简单剔除又担心在正式试验中这些异常个体是正常情况,这给申办方样本例数决策带来了一定的困扰。本文提出了使用Bootstrap自助法重抽样模拟不同样本例数下的正式试验效能,预测其结果,并在效能-样本例数曲线中寻找拐点,以及随样本例数增加效能增加速度减缓的曲线片段,从而找到合适的样本例数使获益风险比最大化,科学指导生物等效性试验。
In bioequivalence study,conducting a pilot trial is useful to determine sample size. Besides geometric mean ratios( GMR) and coefficient of variation( CV),it's also significant to do a consistent tendency research between individuals. However,in some cases there are outliers that can not be deleted directly and will be leading to a huge sample size.Here,we propose the simulation of power in different sample size using bootstrap. By studying the simulation curve,we can find how much power we will get when increasing the sample size. Therefore,we can find the best sample size to conduct the bioequivalence study properly.
In bioequivalence study,conducting a pilot trial is useful to determine sample size. Besides geometric mean ratios( GMR) and coefficient of variation( CV),it's also significant to do a consistent tendency research between individuals. However,in some cases there are outliers that can not be deleted directly and will be leading to a huge sample size.Here,we propose the simulation of power in different sample size using bootstrap. By studying the simulation curve,we can find how much power we will get when increasing the sample size. Therefore,we can find the best sample size to conduct the bioequivalence study properly.
引文
[1]DILETTI E,HAUSCHKE D,STEINIJANS V W.Sample size determination for bioequivalence assessment by means of confidence intervals[J].Int J Clin Pharmacol Ther Toxicol,1991,30(Suppl.1)(1):S51.
[2]LIU J P.Use of the repeated cross-over designs in assessing bioequivalence[J].Statistics Med,2010,14(9):1067-1078.
[3]EFRON B.Bootstrap Methods:Another look at the Jackknife[J].Annals Statistics,1979,7(1):1-26.
[4]谢益辉,朱钰.Bootstrap方法的历史发展和前沿研究[J].统计与信息论坛,2008,23(2):90-96.
[5]PIGEOT I,HAUSCHKE D,SHAO J.The bootstrap in bioequivalence studies[J].J Biopharm Stat,2011,21(6):1126-1139.
[2]LIU J P.Use of the repeated cross-over designs in assessing bioequivalence[J].Statistics Med,2010,14(9):1067-1078.
[3]EFRON B.Bootstrap Methods:Another look at the Jackknife[J].Annals Statistics,1979,7(1):1-26.
[4]谢益辉,朱钰.Bootstrap方法的历史发展和前沿研究[J].统计与信息论坛,2008,23(2):90-96.
[5]PIGEOT I,HAUSCHKE D,SHAO J.The bootstrap in bioequivalence studies[J].J Biopharm Stat,2011,21(6):1126-1139.