以簡易方法決定對等性假說樣本數之研究

Abstract

對等性假設檢定(Equivalence hypothesis)在評估學名藥與專利藥間的差異扮演著很重要的角色。常使用雙單尾檢定(Two one-sided tests procedure, TOST)來檢定對等性假設上的兩種處理的對等性，而當兩處理之間的差異不為0時，其檢定力函數(Power function)並非對稱型分配，故沒有確切公式可算出樣本數，而一般近似的樣本數公式，其計算出來的檢定力(power)常產生不足或是過多的現象，雖然後來有幾何平均的方式來推估所需樣本數。但由於是採用迭代方法，故計算方式較為複雜，為了改善此點，本論文採取較簡單的迭代方法，也採用多重主要端點樣本數方法的概念來推導出所需的樣本數。並將本論文及其他方法所算出的樣本數進行比較。並以實例說明生物對等性的對數尺度及臨床相等性的原始尺度之應用。

The equivalence hypothesis plays an important role on assessing the differences between the generic drugs and original patented drugs. The two-one-sided test ( TOST) procedure is usually applied to test the equivalence hypothesis based on average of two treatments. When the difference in population between the two treatments is not 0 but still within the equivalence limits, the power function is not symmetric. There is no exact formula to calculate the sample sizes. Hence the approximate formulas are proposed to determine the sample size for the equivalence hypothesis. Consequently the sample sizes may provide either insufficient power or unnecessarily high power. Geometric approach to the sample size determination for the equivalence hypothesis has been proposed. However , it requires a complicated iterative procedure. In this thesis, we suggest an iterative method in conjunction of the ideal of multiple co-primary endpoints sample size method for determination of the sample size for equivalence hypothesis. A numerical study was conducted to compare the performance of our proposed method with the current methods. Numerical examples are used to illustrate the applications to bioequivalence on the logarithmic scale and to clinical equivalence on the original scale.