A stochastic agent-based model for naive CD8^± T cell recirculation dynamics in mice

Understanding the dynamics of T cell recirculation is vital for predicting immune responses and providing mechanistic insights into T cell migration. In this work we present a stochastic agent-based mathematical model for naive CD8⁺ T cell dynamics and recirculation patterns in mice. This work includes the ability to interrogate the earliest stages of T cell trafficking using a dataset with uniquely dense temporal sampling. Measurements beginning at 10 min post-transfer enabled characterization of the rapid redistribution phase that is often missed in studies with coarser sampling intervals. The model structure includes the bloodstream, lymph nodes, spleen, and other parts of the body where naive T cells recirculate. The model is governed by migration parameters designed according to the physiological nature of the system components. A parameter estimation technique was used to fit the model to experimental data while maintaining a reasonable level of stochasticity. The results show that the simulated naive CD8⁺ T cell counts in the different tissues matched the experimental data up to 47 h post T cell transfer. The model incorporates lymph node size heterogeneity and distinct splenic red and white pulp compartments to better capture naive CD8⁺ T cell recirculation dynamics. The robustness of the model was demonstrated by simulating the effect of FTY720 drug on T cell counts in the bloodstream at two different drug doses and over the span of one week, which is much longer than the original training period. This work offers a novel tool for quantifying T cell recirculation parameters and enables predictive capabilities for T cell dynamics.