Design of Experiments (DoE) applied to Pharmaceutical and Analytical Quality by Design (QbD)

According to Quality by Design (QbD) concept, quality should be built into product/method during pharmaceutical/analytical development. Usually, there are many input factors that may affect quality of product and methods. Recently, Design of Experiments (DoE) have been widely used to understand the effects of multidimensional and interactions of input factors on the output responses of pharmaceutical products and analytical methods. This paper provides theoretical and practical considerations for implementation of Design of Experiments (DoE) in pharmaceutical and/or analytical Quality by Design (QbD). This review illustrates the principles and applications of the most common screening designs, such as two-level full factorial, fractionate factorial, and Plackett-Burman designs; and optimization designs, such as three-level full factorial, central composite designs (CCD), and Box-Behnken designs. In addition, the main aspects related to multiple regression model adjustment were discussed, including the analysis of variance (ANOVA), regression significance, residuals analysis, determination coefficients (R2, R2-adj, and R2-pred), and lack-of-fit of regression model. Therefore, DoE was presented in detail since it is the main component of pharmaceutical and analytical QbD.

根据质量源于设计（Quality by Design, QbD）的理念，在药物研发或分析方法开发过程中，应将质量属性内嵌于产品与方法的全流程中。通常存在诸多输入因素，可能对产品及分析方法的质量产生影响。近年来，实验设计（Design of Experiments, DoE）已被广泛应用于解析多维度输入因素及其交互作用对药物产品与分析方法输出响应的影响。本文针对实验设计（DoE）在药物及分析领域质量源于设计（QbD）中的落地实施，提供了理论与实践层面的考量依据。本综述阐述了两类主流实验设计的原理与应用：其一为常用的筛选设计，包括二水平全因子设计、部分因子设计以及普拉凯特-伯曼设计；其二为优化设计，涵盖三水平全因子设计、中心复合设计（Central Composite Design, CCD）与Box-Behnken设计。此外，本文还讨论了与多元回归模型拟合相关的核心要点，包括方差分析（ANOVA）、回归显著性检验、残差分析、决定系数（R²、R²-adj及R²-pred）以及回归模型失拟检验。鉴于实验设计（DoE）是药物及分析领域质量源于设计（QbD）的核心组成部分，本文对其进行了详细阐释。