Amasya Gulin, Badilli Ulya, Aksu Buket, Tarimci Nilufer
Department of Pharmaceutical Technology, Faculty of Pharmacy, Ankara University, Ankara, Turkey.
Department of Pharmaceutical Technology, Faculty of Pharmacy, Ankara University, Ankara, Turkey.
Eur J Pharm Sci. 2016 Mar 10;84:92-102. doi: 10.1016/j.ejps.2016.01.003. Epub 2016 Jan 16.
With Quality by Design (QbD), a systematic approach involving design and development of all production processes to achieve the final product with a predetermined quality, you work within a design space that determines the critical formulation and process parameters. Verification of the quality of the final product is no longer necessary. In the current study, the QbD approach was used in the preparation of lipid nanoparticle formulations to improve skin penetration of 5-Fluorouracil, a widely-used compound for treating non-melanoma skin cancer. 5-Fluorouracil-loaded lipid nanoparticles were prepared by the W/O/W double emulsion - solvent evaporation method. Artificial neural network software was used to evaluate the data obtained from the lipid nanoparticle formulations, to establish the design space, and to optimize the formulations. Two different artificial neural network models were developed. The limit values of the design space of the inputs and outputs obtained by both models were found to be within the knowledge space. The optimal formulations recommended by the models were prepared and the critical quality attributes belonging to those formulations were assigned. The experimental results remained within the design space limit values. Consequently, optimal formulations with the critical quality attributes determined to achieve the Quality Target Product Profile were successfully obtained within the design space by following the QbD steps.
采用质量源于设计（QbD），这是一种涉及所有生产过程的设计与开发以实现具有预定质量的最终产品的系统方法，你在一个确定关键制剂和工艺参数的设计空间内开展工作。不再需要对最终产品的质量进行验证。在当前研究中，QbD方法被用于脂质纳米粒制剂的制备，以提高5-氟尿嘧啶（一种广泛用于治疗非黑素瘤皮肤癌的化合物）的皮肤渗透性。通过W/O/W双乳液-溶剂蒸发法制备了载5-氟尿嘧啶的脂质纳米粒。使用人工神经网络软件评估从脂质纳米粒制剂获得的数据，以建立设计空间并优化制剂。开发了两种不同的人工神经网络模型。发现这两种模型获得的输入和输出的设计空间极限值均在知识空间内。制备了模型推荐的最佳制剂，并确定了这些制剂的关键质量属性。实验结果保持在设计空间极限值内。因此，通过遵循QbD步骤，在设计空间内成功获得了具有确定的关键质量属性以实现质量目标产品概况的最佳制剂。
