Exact sample size calculation for bioequivalence evaluation for highly variable drugs
10.13699/j.cnki.1001-6821.2017.12.022
- VernacularTitle:高变异药物生物等效性评价确切样本量计算
- Author:
Tian-Tian LIU
1
;
Meng-Jie LU
;
Chong-Chong ZHOU
;
Wei-Hua ZHONG
;
Jin YANG
;
Yu-Xiu LIU
Author Information
1. 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
- Keywords:
highly variable drug;
bioequivalence;
sample size
- From:
The Chinese Journal of Clinical Pharmacology
2017;33(12):1152-1157
- CountryChina
- Language:Chinese
-
Abstract:
Objective To explore the exact sample size calculation method of bioequivalence evaluation (BE) for highly variable drugs.Methods Based on the two one-sided tests of bioequivalence evaluation for conventional 2 × 2 cross-over design and the statistical distribution theories to elucidate the exact sample size estimation method.Under the Food and Drug Administration (FDA) guidance and the European Medicines Agency (EMA) guideline on the investigation of bioequivalence for highly variable drugs,by using the reference-scaled average bioequivalence(RSABE) criteria to define the relationship between biological equivalent limits and various within-subject variability of reference product,the sample size required for replicated 3-and 4-period cross-over design is deduced exactly respectively.Using SAS progranaming to calculate the exact listings of sample size for different parameter settings of two types of replicated cross-over design,and compared with the formula approximation results published.Results Because of stricter RSABE criteria for EMA guideline than for FDA guidance,the required sample size is larger under the same parameter settings for EMA guideline.Compared with the results of formula approximation method,when the geometric mean ratio(GMR) is 1.0,the exact sample sizes are greater.When the GMR is increased or decreased from 1.0 the required sample sizes rise,and the exact sample size is generally less than or equal to the approximation result.The SAS program codes are provided to easily calculate the exact sample size under various parameter settings.Conclusion The methods of the sample size calculation proposed under FDA guidance and EMA guideline are exact with rigorous statistical theory.The SAS program codes are concise and practical,hope to provide methodological support for the sample size calculation of bioequivalence evaluation for highly variable drugs.
