KRKA, d.d., Novo mesto, Šmarješka cesta 6, 8501 Novo mesto, Slovenia; University of Ljubljana, Faculty of Pharmacy, Aškerčeva 7, 1000 Ljubljana, Slovenia.
KRKA, d.d., Novo mesto, Šmarješka cesta 6, 8501 Novo mesto, Slovenia; University of Ljubljana, Faculty of Pharmacy, Aškerčeva 7, 1000 Ljubljana, Slovenia.
Int J Pharm. 2019 Apr 5;560:261-272. doi: 10.1016/j.ijpharm.2019.01.069. Epub 2019 Feb 8.
Process analytical technology (PAT) has become an important factor in design, analysis and control of complex technological processes. In the present study, pellet coating process was monitored using four different PAT approaches, i.e. near-infrared (NIR) spectroscopy, Raman spectroscopy, in-line image analysis, and spatial filtering technique (SFT). Robustness and accuracy of a novel in-line image analyser was studied during the active ingredient coating process on the pilot scale in the first part of the study. In the second part, multivariate model for in-line monitoring of critical quality attributes, such as moisture content and coating thickness, was calibrated with off-line NIR measurements of laboratory scale samples. The quality of the model was evaluated during the pilot scale batches. In the last part of the study, applicability of two different Raman process analysers for real time moisture content and coating quantity determination was investigated extensively. In addition, Raman spectral data was correlated with the in-line SFT measurements. The results and their interpretations present novel aspects of multiple process analysing techniques, which could help improve pellet coating process monitoring and control. Moreover, presented multivariate model calibration approaches could significantly reduce time, costs, and effort needed to implement PAT into the pharmaceutical industry.
过程分析技术（PAT）已成为设计、分析和控制复杂技术过程的重要因素。在本研究中，使用四种不同的 PAT 方法监测了颗粒包衣过程，即近红外（NIR）光谱、拉曼光谱、在线图像分析和空间滤波技术（SFT）。在研究的第一部分，在中试规模上对新型在线图像分析仪在活性成分包衣过程中的稳健性和准确性进行了研究。在第二部分，使用离线近红外测量的实验室规模样品对在线监测关键质量属性（如水分含量和涂层厚度）的多元模型进行了校准。在中试规模批次中评估了模型的质量。在研究的最后一部分，广泛研究了两种不同的拉曼过程分析仪实时测定水分含量和涂层量的适用性。此外，还对拉曼光谱数据与在线 SFT 测量进行了相关性分析。研究结果及其解释呈现了多种过程分析技术的新方面，这有助于改善颗粒包衣过程的监测和控制。此外，所提出的多元模型校准方法可以显著减少将 PAT 引入制药行业所需的时间、成本和精力。
