(T-056) A novel mechanistic mathematical model of liver transplant immune dynamics predicts important drivers of patient outcome

Julia Bruner – University of Florida; Kyle Adams – University of Florida; Skylar Grey – University of Florida; Mahya Aghaee – University of Florida; Sergio Duarte – University of Florida; Ali Zarrinpar – University of Florida; Helen Moore – University of Florida

Enter your core abstract text here using the suggested layout. Abstracts are text only: No figures/tables. Do not paste author or affiliation information in the body of your abstract. Max character count not including spaces for abstract body is 2800.:

Objectives:
Despite substantial improvement in short-term outcomes over the past 3 decades, long-term outcomes after liver transplantation have not significantly improved. With the introduction of contemporary immunosuppressive agents, causes of morbidity and mortality have largely shifted from early rejection-related complications to longer-term immunosuppression-related toxicity. Though immunosuppression is still indispensable, a personalized, optimized management strategy is desirable. We used mechanistic mathematical modeling to study this setting.

Methods:
To construct our model, we extensively searched the literature, reviewing key mechanisms. For robustness of the model and parameters, we only included dynamics of the most-essential components: dynamics between the donor liver (which we refer to as the allograft), antigen-presenting cells (APCs), helper T cells (Th), cytotoxic T cells (Tc), regulatory T cells (Treg), and interleukin-2 (IL-2). We used ordinary differential equations to represent these dynamics mathematically. Parameterization of the model included careful calculations based on data from the literature for 29 of 41 parameters; the remaining parameters were estimated. Using Sobol sensitivity analysis, we identified key pathways affecting the health of the allograft. To evaluate model performance, we will fit the most-influential parameters to data obtained from liver transplant recipients.

Results:
We found that the most-important factors in allograft injury were six parameters involved in dynamics between Tc, IL-2, and the allograft itself. Specifically, these parameters relate to death of the allograft cells, immune attack of the allograft by Tc, and IL-2 mediated proliferation of Tc. [1]

Conclusions:
Our results suggest that therapeutic and diagnostic strategies targeting these components are likely to be most impactful in treating and monitoring rejection. This has implications for the development and combination of therapies, and monitoring patients to refine the control of the immune response to the allograft. In the future, we will fit the model to data and validate with holdout data, before applying it to the clinical setting to personalize immunosuppression.

Citations: [1] Julia Bruner, Kyle Adams, Skylar Grey, Mahya Aghaee, Sergio Duarte, Ali Zarrinpar, Helen Moore. Understanding Immune Dynamics in Liver Transplant Through Mathematical Modeling, 2024, https://doi.org/10.48550/arXiv.2411.17789.

Keywords: liver transplantation, mechanistic mathematical modeling, global sensitivity analysis