【文献分享】机器学习与代谢建模结合辅助实时PAT系统提升细胞培养过程控制
分析方法处方与工艺


基本信息

- 标题:Machine learning and metabolic modelling assisted implementation of a novel process analytical technology in cell and gene therapy manufacturing 

- 标题:机器学习和代谢建模帮助在细胞和基因治疗生产中实施一种新的过程分析技术

- 作者:Dikicioglu, Duygu 等

- 发表日期:2023年1月16日

- 期刊:Scientific Reports

- 链接: https://www.nature.com/articles/s41598-023-27998-2

摘要(abstract)

本文描述了一种新颖的基于折射率的过程分析技术(PAT)系统——Ranger系统,它被用于实时监控HEK293T细胞培养在慢病毒载体(LVV)生产过程中的代谢活性。PAT系统能够迅速识别出生物反应器pH值与培养代谢活性之间的关系,并利用这一关系制定了一个pH操作策略,与未经优化的生物过程相比,在最少数量的生物反应器实验中实现了代谢活性的1.8倍增长。通过实施PAT技术获得的培养代谢活性的提高,并未与LVV产量的增加相关联。我们采用了代谢建模策略来阐明这些生物过程水平事件与HEK293T细胞代谢之间的关系。建模显示,在低pH(pH 6.40)环境下培养HEK293T细胞直接影响了细胞内pH的维持和细胞内氧气的可用性。我们提供的证据表明,提高的代谢活性是对低pH压力的应对,以维持有利的细胞内环境,而不是表明HEK293T细胞培养处于更优活性状态从而导致LVV产量提高的标志。预测策略被用来构建数据模型,这些模型识别出新型的PAT系统不仅与过程pH有直接关系,还与氧气可用性有关;这两个参数之间的相互作用和相互依赖性直接影响了在生物过程水平观察到的响应。我们呈现的数据表明,使用这种新型基于折射率的PAT系统进行的过程控制和干预,有助于调整和快速优化生产环境,实现适应性过程控制,从而提高过程性能和稳健性。

Process analytical technology (PAT) has demonstrated huge potential to enable the development of improved biopharmaceutical manufacturing processes by ensuring the reliable provision of quality products. However, the complexities associated with the manufacture of advanced therapy medicinal products have resulted in a slow adoption of PAT tools into industrial bioprocessing operations, particularly in the manufacture of cell and gene therapy products. Here we describe the applicability of a novel refractometry-based PAT system (Ranger system), which was used to monitor the metabolic activity of HEK293T cell cultures during lentiviral vector (LVV) production processes in real time. The PAT system was able to rapidly identify a relationship between bioreactor pH and culture metabolic activity and this was used to devise a pH operating strategy that resulted in a 1.8-fold increase in metabolic activity compared to an unoptimised bioprocess in a minimal number of bioreactor experiments; this was achieved using both pre-programmed and autonomous pH control strategies. The increased metabolic activity of the cultures, achieved via the implementation of the PAT technology, was not associated with increased LVV production. We employed a metabolic modelling strategy to elucidate the relationship between these bioprocess level events and HEK293T cell metabolism. The modelling showed that culturing of HEK293T cells in a low pH (pH 6.40) environment directly impacted the intracellular maintenance of pH and the intracellular availability of oxygen. We provide evidence that the elevated metabolic activity was a response to cope with the stress associated with low pH to maintain the favourable intracellular conditions, rather than being indicative of a superior active state of the HEK293T cell culture resulting in enhanced LVV production. Forecasting strategies were used to construct data models which identified that the novel PAT system not only had a direct relationship with process pH but also with oxygen availability; the interaction and interdependencies between these two parameters had a direct effect on the responses observed at the bioprocess level. We present data which indicate that process control and intervention using this novel refractometry-based PAT system has the potential to facilitate the fine tuning and rapid optimisation of the production environment and enable adaptive process control for enhanced process performance and robustness.


文献阅读主要内容分享

概述 

本文介绍了一种新型的过程分析技术(PAT),即基于折射率的Ranger系统,用于实时监测HEK293T细胞培养在生产慢病毒载体(LVV)过程中的代谢活性。研究发现,通过调整生物反应器的pH值,可以显著提高细胞的代谢活性,但这种增加的活性并未转化为LVV产量的增加。文章还运用了代谢建模策略,探讨了生物过程水平事件与HEK293T细胞代谢之间的关系,并利用机器学习方法构建数据模型,预测了PAT系统与过程pH和氧气可用性之间的直接关系。

整体评估

该研究在细胞和基因治疗制造领域展示了PAT技术的新应用,特别是在监测和控制细胞代谢活性方面。研究的原创性在于结合了机器学习和代谢建模,以深入理解复杂生物过程中的代谢动态。该研究对提高生物制药过程的效率和稳健性具有潜在影响。

实验方法

研究使用了基于折射率的PAT系统(Ranger系统)来监测细胞培养的代谢活性,并通过调整生物反应器的pH值来优化代谢活性。此外,研究还采用了代谢建模和机器学习技术来分析数据,构建预测模型。

结论 

研究发现,通过调整pH值可以显著提高细胞的代谢活性,但这并不会导致LVV产量的增加。代谢建模揭示了低pH环境下细胞内部维持pH和氧气可用性的挑战,以及细胞如何响应这些压力以维持有利的细胞内环境。

优势

- 创新地将PAT技术应用于细胞和基因治疗的制造过程。

- 结合机器学习和代谢建模,为理解复杂的生物过程提供了新的视角。

- 通过实验和建模揭示了细胞代谢活性与生产效率之间的关系。

弱点

- 研究主要集中在pH值对细胞代谢活性的影响,未涉及其他可能影响因素。

- 尽管代谢活性提高,但未观察到LVV产量的增加,这可能表明存在其他限制产量的因素。

问题和建议

- 未来的研究是否可以探索除了pH值之外的其他因素如何影响细胞的代谢活性和LVV产量?

- 是否可以进一步研究在低pH环境下提高LVV产量的策略?

整体评价

根据上述标准,本文提供了PAT在细胞和基因治疗制造中的新应用,并展示了结合实验和计算方法来优化生物过程的潜力。尽管存在一些局限性,但研究的原创性和对领域的贡献使其成为一篇有价值的学术论文。