(69d) Towards Scalable and Cost-Effective Plasmid DNA Manufacturing

Advanced Therapy Medicinal Products (ATMPs) are a novel class of therapeutics that have demonstrated promising results in the cure of life-threatening diseases, such as cancer and neurodegenerative disorders. ATMPs have attracted global interest that translates into 27 approved products, 1,454 ATMP manufacturers worldwide, and 2,220 active clinical trials, as of 2022 [1, 2]. The adoption of scalable and cost-effective manufacturing processes is key for meeting the unprecedented demand for these products. Plasmid DNA (pDNA) is a critical raw material used as carrier of the gene of interest [3]. Currently, pDNA supply is limited and costly and the demand is expected to increase rapidly over the coming years, mainly due to its key role in the manufacturing of cell and gene therapies and mRNA and viral vector vaccines [4]. Scaling up the pDNA manufacturing process to meet current and future demand is one of the main challenges of the ATMP sector [4].

In this work, we employ Process Systems Engineering (PSE) methodologies as an approach to assist systematic and proactive decision-making throughout the stages of process design and optimization. We propose an integrated computational framework to aid wider adoption of ATMPs by ensuring efficient timelines, cost-effectiveness, and scalability in the production of pDNA. Specifically, we identify key process and economic parameters for the production of pDNA and develop techno-economic models for different scales, phase I clinical trials to commercialization [6]. Through uncertainty and global sensitivity analysis, we quantify key process and economic uncertainties in the manufacturing process (Figure 1a) and apportion them to key process parameters (Figure 1b) [5]. We thus highlight cost drivers and limitations that steer decision-making, as well as the importance of economies of scale in the production of pDNA. Finally, we perform Bayesian optimization to identify optimal sets of process parameters that lead to reduced costs, in an attempt to decrease the cost of goods (COGs), improved lead times, increased productivity, and reduced the environmental footprints of the process [7]. Optimized manufacturing recipes identified with the proposed framework can achieve up to a 170% increase in batch size and a 34.7% decrease in the operating cost per batch (Figure 1a).

Figure 1. (a) Batch size variability for the production of pDNA in a commercial scale. The black bar in the center is the interquartile range, the white dot inside the black bar is the median value, and the black lines stretched from the bar are the lower/upper adjacent values. Bayesian optimization is used to maximize the batch size. (b) Global sensitivity analysis of the nine critical input parameters on the batch size. First-order effects (𝑆𝑖) are displayed in the diagonal and second-order effects (𝑆𝑖𝑗) are presented in the upper and lower triangular.

References

[1] Hunt TD (2023) CELL & GENE: state of the industry briefing. https://alliancerm.org/arm-event/sotibriefing/. Accessed 28 Feb 2023

[2] The TransMission: Current FDA Approved Gene and Cell Therapies. https://www.mirusbio.com/blog/fda-approved-gene-cell-therapies. Accessed 27 Feb 2023

[3] M. Gosse, C. Jones, D. Jesus, S. D’Costa, 2022, Regulatory & supply chain implications for plasmids as critical starting materials in the manufacture of viral vector gene therapy products. Cell and Gene Therapy Insights, 8(2), 279–286.

[4] J. Ohlson, 2020, Plasmid manufacture is the bottleneck of the genetic medicine revolution, Drug Discovery Today, 25(11), 1891–1893.

[5] S. Kucherenko, 2013, SobolHDMR: A general-purpose modeling software. Methods in Molecular Biology, 1073, 191–224.

[6] R. Ferreira, D. Petrides, 2021, Plasmid DNA (pDNA) Large Scale Manufacturing – Process Modeling and Techno-Economic Assessment (TEA) using SuperPro Designer, doi: 10.13140/RG.2.2.12780.28800.

[7] M. Balandat, B. Karrer, D. R. Jiang, S. Daulton, B. Letham, A. Gordon Wilson, E. Bakshy, 2020, BoTorch: A Framework for Efficient Monte-Carlo Bayesian Optimization, Advances in Neural Information Processing Systems, 33.