Department of Chemical and Biochemical Engineering, Rutgers University, Piscataway, NJ, USA.
Department of Chemical and Biomolecular Engineering, University of Delaware, Newark, DE, USA.
Int J Pharm. 2021 Jun 1;602:120643. doi: 10.1016/j.ijpharm.2021.120643. Epub 2021 Apr 24.
To modernize drug manufacturing, the pharmaceutical industry has been moving towards implementing emerging technologies to enhance manufacturing robustness and process reliability for production of regulation compliant drug products. Although different science and risk based technologies, like Quality-by-Design, have been used to illustrate their potential, there still exist some underlying obstacles. Specifically, for the production of oral solid drug products, an in-depth process understanding, and predictive modeling of powder mixing in continuous powder blenders is one such major obstacle and originates from the current limitations of the experimental and modeling approaches. Though first principle based discrete element modeling (DEM) approach can address the above issues, it can get very computationally intensive which limits its applications for predictive modeling. In the proposed work, we aim to address this limitation using a multi-zonal compartment modeling approach, which is constructed from DEM. The approach provides a computationally efficient and mechanistically informed hybrid model. The application of the proposed approach is first demonstrated for a periodic section of the blender, followed by its extension for the entire continuous powder blender and the obtained model predictions are validated. The proposed approach provides an overall assessment of powder mixing along axial and radial directions, which is an important requirement for the quantification of blend uniformity. Given the low computational cost, the developed model can further be integrated within the predictive flowsheet model of the manufacturing line.
为了实现药品制造现代化，制药行业一直在努力采用新兴技术，以提高生产符合法规要求的药品的制造稳健性和工艺可靠性。虽然已经使用了不同的基于科学和风险的技术，如质量源于设计（Quality-by-Design），来说明它们的潜力，但仍然存在一些潜在的障碍。具体来说，对于口服固体制剂的生产，深入的工艺理解和连续粉末混合器中粉末混合的预测建模是一个主要障碍，其源于当前实验和建模方法的局限性。虽然基于第一原理的离散元建模（DEM）方法可以解决上述问题，但它的计算量非常大，限制了其在预测建模中的应用。在拟议的工作中，我们旨在使用多区域隔室建模方法来解决这个限制，该方法是由 DEM 构建的。该方法提供了一种计算效率高且具有机理启示的混合模型。该方法的应用首先在混合器的一个周期性部分进行演示，然后扩展到整个连续粉末混合器，并验证了得到的模型预测。所提出的方法提供了沿轴向和径向的粉末混合的全面评估，这是量化混合均匀度的重要要求。由于计算成本低，所开发的模型可以进一步集成到制造线的预测流程图模型中。
