溶出曲线的常见评价方法及优劣性探讨  

Comparison of commonly used evaluation methods of dissolution curves and their advantages and disadvantages

作　　者：马莉 祝传勇 何小琳 段湘宁 高志晨 吕昭云 谭德讲[1] MA Li;ZHU Chuan-yong;HE Xiao-lin;DUAN Xiang-ning;GAO Zhi-chen;Lü Zhao-yun;TAN De-jiang(National Institutes for Food and Drug Control,Beijing 102629,China;Beijing Union Pharmaceutical Factory,Beijing 102629,China)

机构地区：[1]中国食品药品检定研究院,北京102629 [2]北京协和药厂,北京102629

出　　处：《中国新药杂志》2021年第22期2091-2097,共7页Chinese Journal of New Drugs

基　　金：中国食品药品检定研究院药品质量安全和能力建设资助项目;北京市科技计划课题资助项目(Z151100000115013)。

摘　　要：目的:对溶出曲线常见评价方法的性能进行比较分析,并探讨当溶出数据呈较高变异时,无法使用f_(2)因子的情况下,哪种替代方法较为合适。方法:以双环醇片作为实例,收集20批溶出数据,使用不同评价方式进行比较,包括非模型依赖方式、模型依赖差异性比较方式和等效性比较方式(建立模型参数等效区间)。结果:在溶出数据变异较大(>10%)时,非模型依赖法中的f_(2)-bootstrap法的结果与多变量统计距相比,更接近f_(2)因子法的结果;模型依赖法比较中的t检验法和多变量统计矩法的结果,与f_(2)因子法有较大差异;首次利用模型参数获得的各参数容忍区间;可作为模型参数的等效区间供今后溶出一致性的评估。结论:当选择非模型依赖法时,当原f_(2)因子法不适用时,f_(2)-bootstrap法可作为f;因子法的替代方法;新提出模型参数的等效区间能够弥补t检验的不足,更好地从质量等效层面上评价溶出曲线。Objective: To compare and analyze the performance of commonly used evaluation methods for dissolution curves, and explore more suitable alternative method when f;factor cannot be used due to highly variable dissolution data. Methods: Taking bicyclol tablets as example, 20 batches of dissolution data were collected and compared using different evaluation methods, including model-independent methods, model-dependent difference test, and model-dependent equivalence test of the model parameters. Results: Among the several model-independent methods, the results of the f_(2)-bootstrap method were closer to the f_(2) method than the multivariate statistic distance(MSD) method when the dissolution data had greater variation(>10%). The results of t-test and MSD for comparing the parameters of model-dependent method were significantly different with the f_(2) method. The tolerance intervals of each parameter obtained by using model parameters for the first time could be used as the equivalent interval of model parameters for evaluation of dissolution consistency in the future. Conclusion: When the f;method is not applicable, the f_(2)-bootstrap method can be used as an alternative method. The newly proposed equivalence interval of the model parameters can make up for the deficiency of t-test, and better evaluate the dissolution curves from the view of quality consistency.

关 键 词：溶出曲线 f_(2)因子 f_(2)-bootstrap 多变量统计矩 模型依赖法 等效区间 

分 类 号：R917[医药卫生—药物分析学]