Novel Process Analytic Technology for Continuous Bioprocessses

摘要 这项建议旨在演示如何实时测量蛋白质产品质量 外来物质污染的属性和快速评估可用于 制造商在连续蛋白质生产中基于风险的决策。而质量是由 设计一直是保持产品质量的成功方法，它的采用是一种 难以实时测量产品质量的副产品。我们将在实验中 展示两个用于在线蛋白质质量评估的新平台(1)微/纳米流体 用于评估高和低分子量蛋白质杂质、蛋白质结构和 结合，糖基化和(2)扫描源拉曼(SS-拉曼)-使用新的方法 可调谐激光器和低成本的固定波长探测器，而不是传统的昂贵的 光谱仪或干涉仪-测量所有点的工艺参数和产品质量 这一过程。我们之前已经演示了使用微流控和纳流控检测来 离线测量蛋白质质量属性，构建自动采样系统进行饲喂 生物反应器上清液进行微/纳流控分析，并用拉曼光谱进行鉴定 并以药物中发现的浓度量化溶解在缓冲溶液中的治疗性蛋白质 该产品的体积等于或小于一剂。这项工作将是第一次 这些技术用于实时蛋白质质量评估的演示。此外，这一点 这项工作将开发快速检测外来病原体的新方法，并展示其 用于CHO细胞培养。我们将演示使用拉曼光谱来获得新陈代谢 蛋白质的指纹(细胞内代谢产物)和足迹(细胞外代谢产物) 在受到各种病原体攻击时产生CHO细胞，并对这些指纹和 脚印是正常的还是异常的，表明存在潜在的感染。我们会 还评估快速下一代测序和生物信息学方法的适用性 目的：快速检测CHO细胞连续培养过程中的病毒污染。最后，这项提议将 研究将这些新技术集成到一个连续的单克隆中的影响 抗体制造工艺。这项工作完成后，将改进(A)产品和工艺 通过演示使用实时蛋白质属性测量(B)过程获得的知识 通过在制造和生产过程中实现关键质量属性的直接测量进行控制 待测试的反馈或前馈控制策略，以及(C)通过开发 快速外源试剂检测，以更快地确保生物药物的无菌，这可以 提高患者安全，保障药品供应，降低经营风险。

Abstract This proposal aims to demonstrate how real-time measurement of protein product quality attributes and rapid assessment of adventitious agent contamination can be used for manufacturers’ risk-based decision making in continuous protein production. While Quality by Design has been a successful approach for maintaining product quality, its adoption is a byproduct of difficulties in measuring product quality in real time. We will experimentally demonstrate two novel platforms for in-line protein quality assessment (1) a micro/nanofluidic platform for assessing high- and low-molecular weight protein impurities, protein structure and binding, and glycosylation and (2) swept source Raman (SS-Raman) – using a novel approach of a tunable laser and low-cost, fixed-wavelength detector instead of a traditional, expensive spectrometer or interferometer – to measure process parameters and product quality at all points of the process. We have previously demonstrated the use of micro- and nano-fluidic assays to measure protein quality attributes off-line, constructed an automated sampling system to feed bioreactor supernatant to the micro/nanofluidic assays, and used Raman spectroscopy to identify and quantify therapeutic proteins dissolved in buffer solutions at concentrations found in drug product and at volumes equivalent to or less than one dose. This work will be the first demonstration of these technologies for real-time protein quality assessment. Additionally, this work will develop novel approaches for rapid adventitious agent detection and demonstrate their use in CHO cell culture. We will demonstrate the use of Raman spectroscopy to obtain metabolic fingerprints (intracellular metabolites) and footprints (extracellular metabolites) of protein producing CHO cells when challenged by various pathogens and classify these fingerprints and footprints as either normal or anomalous, indicating the presence of a potential infection. We will also evaluate rapid Next Generation Sequencing and bioinformatic approaches for their suitability to rapidly detect virus contaminations of continuous CHO cell culture. Finally, this proposal will investigate the impact of integrating these novel technologies into a continuous monoclonal antibody manufacturing process. At its completion this work will improve (a) product and process knowledge by demonstrating the use of real-time protein attribute measurements (b) process control by enabling direct measurement of critical quality attributes during manufacturing and feedback or feedforward control strategies to be tested, and (c) safety through development of rapid adventitious agent assays to more rapidly assure sterility of biopharmaceuticals which can improve patient safety, assure supply of medicines, and reduce business risk.