The rapid evolution of messenger RNA (mRNA) technology has opened unprecedented avenues in modern pharmacology, extending far beyond prophylactic vaccines into the realm of acute therapeutic intervention. In the highly volatile environment of the intensive care unit (ICU), acute organ failure—such as acute kidney injury, hepatic failure, and acute respiratory distress syndrome—carries a persistently high mortality rate. Transient mRNA therapeutics offer a paradigm-shifting approach by temporarily reprogramming cellular machinery to produce therapeutic proteins, thereby providing acute organ support without the risks of permanent genomic alteration. However, deploying these potent biological agents in critically ill patients requires precise dynamic dosing and rigorous real-time monitoring to prevent adverse hyper-responses. This paper proposes a novel translational medicine framework that integrates hypothetical transient mRNA therapeutic regimens with advanced artificial intelligence, specifically reinforcement learning and time-series foundation models, to optimize individualized dosing strategies. By synthesizing current biological capabilities with state-of-the-art computational critical care techniques, we present an end-to-end conceptual pipeline for the adaptive administration of mRNA therapies. Through a rigorously structured discussion on methodological design, evaluation planning, and ethical deployment, this work aims to bridge the gap between molecular innovation and algorithmic clinical decision support, ultimately paving the way toward a new era of precision critical care medicine.

Historically, therapeutic interventions for acute organ support have been primarily supportive, relying on mechanical devices such as continuous renal replacement therapy or extracorporeal membrane oxygenation, alongside systemic pharmacological agents.

Shirali, Ali, Schubert, Alexander, & Alaa, Ahmed (2023). Pruning the Way to Reliable Policies: A Multi-Objective Deep Q-Learning Approach to Critical Care. IEEE Journal of Biomedical and Health Informatics, vol. 28, no. 10, pp. 6268-6279, Oct. 2024. https://doi.org/10.1109/JBHI.2024.3415115 https://doi.org/10.1109/JBHI.2024.3415115

Sheikhalishahi, Seyedmostafa, Balaraman, Vevake, & Osmani, Venet (2019). Benchmarking machine learning models on multi-centre eICU critical care dataset. https://doi.org/10.1371/journal.pone.0235424 https://doi.org/10.1371/journal.pone.0235424

Spathis, Dimitris, & Hyland, Stephanie L. (2022). Looking for Out-of-Distribution Environments in Multi-center Critical Care Data. https://arxiv.org/pdf/2205.13398v2 https://arxiv.org/pdf/2205.13398v2

Bansal, Aryaman, & Sharma, Divya (2025). Benchmarking Offline Multi-Objective Reinforcement Learning in Critical Care. https://arxiv.org/pdf/2512.08012v1 https://arxiv.org/pdf/2512.08012v1

Jagd, Katja Naasunnguaq, DeVries, Rachael, & Winther, Ole (2025). Towards Self-Supervised Foundation Models for Critical Care Time Series. https://arxiv.org/pdf/2509.19885v1 https://arxiv.org/pdf/2509.19885v1

Shi, Tongyue, Ma, Jun, Yu, Zihan, Xu, Haowei, Xiong, Minqi, Xiao, Meirong, Li, Yilin, Zhao, Huiying, & Kong, Guilan (2024). Stochastic Parrots or ICU Experts? Large Language Models in Critical Care Medicine: A Scoping Review. https://arxiv.org/pdf/2407.19256v1 https://arxiv.org/pdf/2407.19256v1

Alwakeel, Mahmoud, Nagori, Aditya, Wong, An-Kwok Ian, Chaisson, Neal, Krishnamoorthy, Vijay, & Kamaleswaran, Rishikesan (2025). Performance of Large Language Models in Answering Critical Care Medicine Questions. https://arxiv.org/pdf/2509.19344v1 https://arxiv.org/pdf/2509.19344v1

Yadav, Shashank, & Subbian, Vignesh (2025). Failure Modes of Time Series Interpretability Algorithms for Critical Care Applications and Potential Solutions. https://arxiv.org/pdf/2506.19035v1 https://arxiv.org/pdf/2506.19035v1

Burger, Manuel, Sergeev, Fedor, Londschien, Malte, Chopard, Daphné, Yèche, Hugo, Gerdes, Eike, Leshetkina, Polina, Morgenroth, Alexander, Babür, Zeynep, Bogojeska, Jasmina, Faltys, Martin, Kuznetsova, Rita, & Rätsch, Gunnar (2024). Towards Foundation Models for Critical Care Time Series. https://arxiv.org/pdf/2411.16346v1 https://arxiv.org/pdf/2411.16346v1

Liu, Chia-Fu, & Mostafavi, Ali (2023). Equitable Optimization of Patient Re-allocation and Temporary Facility Placement to Maximize Critical Care System Resilience in Disasters. https://arxiv.org/pdf/2306.07545v2 https://arxiv.org/pdf/2306.07545v2