最大概似估計式與最小變方不偏估計式在評估平均生体相等性之比較

Comparison of MLE and MVUE for Evaluation of Average Bioequivalence

劉仁沛(Jen-Pei Liu) ； 許富閔(Fu-Min Xu)

《中國統計學報》 43卷3期 (2005/09) Pp. 239-258

https://doi.org/10.29973/JCSA.200509.0001

摘要

原廠藥研發需要花費巨大財力與長期投資，所以原廠藥藥價是非常昂貴。待原廠藥專利期滿後，其它藥廠則可仿造相同藥品，這種仿照藥稱爲學名藥，因學名藥不須長期研發投資，所以學名藥價格較原廠藥爲便宜。因此可節省醫療費用的支出。爲確保學名藥的療效和安全性與原廠藥相同，藥政單位要求廠商必須証明學名藥與原廠藥具有平均生体相等性後才核准上市。最大概似估計式已被採用在評估平均生体相等性，但最大概似估計式會高估平均生体相等性。所以我們建議採用最小變方不偏估計式評估平均生体相等性。我們並執行一個模擬研究經，以經驗型Ⅰ誤差與經驗檢定力來比較兩種估計式之優劣。我們亦以實際資料說明最小變方不偏估計式在評估生体相等性上的應用。

關鍵字

平均生体相等性 ； 模擬研究 ； 經驗型誤 ； 經驗檢定力 ； 信賴區間包含率

並列摘要

The research and development of an innovative drug product takes 10-12 years and 800 millions to 1 billion US dollars. Therefore, it is a costly time-consuming, and highly risky endeavor. One way to reduce the drug cost is to introduce generic drugs after the patent of the innovative drugs expires. Currently, most regulatory agencies iii the world only require evidence of average bioequivalence from in vivo bioequivalence trials to approve the generic drugs. Currently, most of heath regulatory agencies in the world such as the U.S. Food and Drug Administration and Taiwan Department of Heath use maximum likelihood estimator (MLE) for evaluation of average bioequivalence. Since MLE is not an unbiased estimator for the relative bioavailability, therefore, we propose the minimum variance unbiased estimator (MVUE) to assess the average bioequivalence. We performed a simulation study to compare the bias, mean square error, empirical size, empirical power and probability coverage between MLE and MVUE on the various combinations of parameters and sample size under 2×2 crossover design and higher-order crossover design. The simulation results showed that in addition to unbiasedness, the size and power of the MVUE is comparable to those of MLE. A numerical example illustrates the procedure of the MVUE for evaluation of average bioequivalence.

並列關鍵字

Average bioequivalence ； simulation study ； empirical size ； empirical power ； probability coverage

參考文獻

Bradu, D.,Mundlak, Y(1970).Estimation in lognormal linear models.Journal of the American Statistical Association.65,198-211.

Chow, S.C,Liu, J.P.(2000).Design and Analysis of Bioavailability and Bioequivalence Studies.New York:Marcel Dekker.

Liu, J.P.,Weng, C.S(1992).Estimation of direct formulation effect under lognormal distribution in bioavailability/bioequivalence studies.Statistics in Medicine.11,881-896.

O`Brien, P.C.(1984).Procedure for comparing samples with multiple endpoints.Biometrics.40,1079-1087.

U.S. FDA(2003).Guidance for industry on bioavailability and bioequivalence studies for orally administered drug products-general considerations.Rockville, Maryland:Center for Drug Evaluation and Research, Food and Drug Administration.