(588cq) An End-to-End, Plug-and-Play Digital Twin for Continuous Manufacturing of mRNA-Based Therapeutics

There is increasing interest in mRNA-based therapeutics due to their adaptability for various disease targets and relatively rapid production timelines. However, scaling these therapies to meet global demand poses significant challenges. Traditional batch processes are often time-consuming, expensive, and difficult to scale, driving the industry toward continuous manufacturing. Continuous processes can increase throughput and reduce costs but demand robust real-time monitoring and control strategies. Digital twins—virtual models that replicate real-world processes—are gaining attention as a means to address these complexities, as they allow researchers to test and refine process parameters in a controlled virtual environment before implementing costly physical changes.^[1,2]

In a typical mRNA manufacturing pipeline, multiple unit operations are involved including an upstream in vitro transcription reaction synthesizes the mRNA strand from nucleotide triphosphates, enzymes, and template DNA.^[2,3] The downstream phase often involves chromatography and tangential flow filtration (TFF) chromatography and tangential flow filtration (TFF), which remove impurities and concentrate the mRNA product.^[4] The formulation stage then encapsulates the purified mRNA into lipid nanoparticles (LNPs), which enhance delivery and stability.^[1] Finally, the product may be subjected to lyophilization for improved shelf life and easier distribution.^[5] While prior studies have successfully developed digital twins of individual unit operations^[2,3,5]—often focusing on a single chromatography system or the kinetics of the in vitro transcription reaction—there remains a notable gap when it comes to integrating the entire pipeline under one unified framework. Moreover, many existing solutions use proprietary software, limiting collaboration and making it difficult to adapt them to new processes or scales.

In this work, we develop an end-to-end, plug-and-play digital twin specifically tailored for continuous mRNA manufacturing. Built on validated mechanistic simulations, our open-source platform integrates the entire production pipeline—upstream (IVT process), downstream purification (including chromatography and tangential flow filtration), LNP formulation, and lyophilization for final product stabilization. By representing each unit operation through modular sub-models, the software provides a holistic view of how changes in one unit operation can affect product quality and yield in subsequent unit operations. A key feature of this platform is its user-friendly interface, which makes it easy to configure process parameters and switch between unit operations. Additionally, an integrated SQL database automatically records simulation runs, parameter sets, and performance outcomes, enabling users to track the evolution of their process designs over time. The plug-and-play architecture makes it straightforward to adapt the software to other biopharmaceutical processes by adding or replacing specific unit operations. Users can rapidly incorporate different purification methods, new reaction conditions, or alternative formulations without rebuilding the entire framework from scratch. This presentation will also describe case studies demonstrating the use of the digital twin to evaluate alternative process configurations and optimize yield in mRNA-based therapeutics. The proposed end-to-end digital twin software provides a foundation for future integration of real-time data transfer and process control.

Keywords: Digital twin, biopharmaceutical manufacturing, mechanistic model, mRNA vaccine

Acknowledgments: This research was supported by the U.S. Food and Drug Administration under the FDA BAA-22-00123 program, Award Number 75F40122C00200.

References

• Shahab, M. A., Destro, F., & Braatz, R. D. (2025). Digital twins in biopharmaceutical manufacturing: Review and perspective on human-machine collaborative Intelligence. arXiv preprint arXiv:2504.00286.
• Hengelbrock, A., Schmidt, A., & Strube, J. (2024). Digital twin fundamentals of mRNA in vitro transcription in variable scale toward autonomous operation. ACS Omega, 9(7), 8204-8220.
• Stover, N. M., Ganko, K., & Braatz, R. D. (2024). Mechanistic modeling of in vitro transcription incorporating effects of magnesium pyrophosphate crystallization. Biotechnology and Bioengineering, 121(9), 2636-2647.
• Bhoyar, S., Kumar, V., Foster, M., Xu, X., Traylor, S. J., Guo, J., & Lenhoff, A. M. (2024). Predictive mechanistic modeling of loading and elution in protein A chromatography. Journal of Chromatography A, 1713, 464558.
• Srisuma, P., Barbastathis, G., & Braatz, R. D. (2024). Mechanistic modeling and analysis of thermal radiation in conventional, microwave-assisted, and hybrid freeze drying for biopharmaceutical manufacturing. International Journal of Heat and Mass Transfer, 221, 125023.