Kharbach M, Cherrah Y, Vander Heyden Y, Bouklouze A
Pharmaceutical and toxicological analysis research team, laboratory of pharmacology and toxicology, faculty of medicine and pharmacy, university Mohammed V. Souissi, avenue Med Belarbi El Alaoui, BP 6203, 10000 Rabat, Morocco; Department of analytical chemistry and pharmaceutical technology, CePhaR, Vrije Universiteit Brussel (VUB), Laarbeeklaan 103, 1090 Brussels, Belgium.
Pharmaceutical and toxicological analysis research team, laboratory of pharmacology and toxicology, faculty of medicine and pharmacy, university Mohammed V. Souissi, avenue Med Belarbi El Alaoui, BP 6203, 10000 Rabat, Morocco.
Ann Pharm Fr. 2017 Nov;75(6):446-454. doi: 10.1016/j.pharma.2017.07.003. Epub 2017 Aug 7.
According to the Food and Drug Administration and the European Good Manufacturing Practices (GMP) guidelines, Annual Product Review (APR) is a mandatory requirement in GMP. It consists of evaluating a large collection of qualitative or quantitative data in order to verify the consistency of an existing process. According to the Code of Federal Regulation Part 11 (21 CFR 211.180), all finished products should be reviewed annually for the quality standards to determine the need of any change in specification or manufacturing of drug products. Conventional Statistical Process Control (SPC) evaluates the pharmaceutical production process by examining only the effect of a single factor at the time using a Shewhart's chart. It neglects to take into account the interaction between the variables. In order to overcome this issue, Multivariate Statistical Process Control (MSPC) can be used. Our case study concerns an APR assessment, where 164 historical batches containing six active ingredients, manufactured in Morocco, were collected during one year. Each batch has been checked by assaying the six active ingredients by High Performance Liquid Chromatography according to European Pharmacopoeia monographs. The data matrix was evaluated both by SPC and MSPC. The SPC indicated that all batches are under control, while the MSPC, based on Principal Component Analysis (PCA), for the data being either autoscaled or robust scaled, showed four and seven batches, respectively, out of the Hotelling T 95% ellipse. Also, an improvement of the capability of the process is observed without the most extreme batches. The MSPC can be used for monitoring subtle changes in the manufacturing process during an APR assessment.
根据美国食品药品监督管理局和欧洲药品生产质量管理规范（GMP）指南，年度产品回顾（APR）是GMP中的一项强制性要求。它包括评估大量定性或定量数据，以验证现有工艺的一致性。根据联邦法规第11部分（21 CFR 211.180），所有成品应每年进行质量标准审查，以确定药品规格或生产中是否需要任何变更。传统的统计过程控制（SPC）通过使用休哈特控制图一次仅检查一个因素的影响来评估药品生产过程。它忽略了变量之间的相互作用。为了克服这个问题，可以使用多元统计过程控制（MSPC）。我们的案例研究涉及一次APR评估，在一年中收集了164个在摩洛哥生产的含有六种活性成分的历史批次。根据欧洲药典专论，通过高效液相色谱法对六种活性成分进行测定，对每个批次进行了检查。对数据矩阵进行了SPC和MSPC评估。SPC表明所有批次均处于受控状态，而基于主成分分析（PCA）的MSPC，对于自动缩放或稳健缩放的数据，分别显示有4个和7个批次超出了霍特林T 95%椭圆。此外，在去除最极端批次后，观察到工艺能力有所提高。MSPC可用于在APR评估期间监测生产过程中的细微变化。
