Victoria Kulesh 1,2,3, Kirill Peskov 1,2,3, Gabriel Helmlinger 4, Gennady Bocharov 3,5,6
1 Research Center of Model-Informed Drug Development, Sechenov First Moscow State Medical University (Moscow, Russia), 2 Modeling & Simulation Decisions FZ-LLC (Dubai, United Arab Emirates), 3 Marchuk Institute of Numerical Mathematics RAS (INM RAS) (Moscow, Russia), 4 Quantitative Medicines Consulting (Boston, United States), 5 Institute for Computer Science and Mathematical Modelling, Sechenov First Moscow State Medical University (Moscow, Russia), 6 Moscow Center of Fundamental and Applied Mathematics at INM RAS (Moscow, Russia)
Introduction
CD4+ T lymphocytes represent a central component of the human adaptive immune system, coordinating and regulating immune responses throughout life. Immune cell homeostasis reflects the diversity of CD4+ T-lymphocyte subpopulations across functional phenotypes (e.g., Th1, Th2) and maturation stages (e.g., naïve, effector, memory). In addition, T-cell development and trafficking – through homing and tissue distribution – are key determinants of CD4+ T-cell homeostasis. Immune aging encompasses thymic involution, a marked decline in T-cell numbers (particularly within the naïve pool), the accumulation of more differentiated subsets, and systemic inflammation. Together, these processes contribute to immunosenescence and can substantially impair the ability of the organism to eliminate infectious agents, malignant cells and control autoimmune disorders [1]. To elucidate the mechanisms underlying age-dependent changes in immune homeostasis and to identify targetable homeostatic processes, a quantitative approach based on mathematical modelling is required.
Objective
To develop a mechanistic physiologically-based mathematical model describing the CD4+ T-lymphocyte homeostasis across the human lifespan, incorporating maturation, differentiation, migration and age effects on distinct cell subpopulations and integrating multi-level information and data on cellular kinetics and age-dependent dynamics.
Methods
A stepwise modelling strategy was developed to describe the CD4+ T-lymphocyte kinetics and tissue distribution and to explore age-dependent and regulatory patterns in homeostatic processes. Model development was driven by and based on an extensive quantitative dataset of CD4+ T-cell kinetic characteristics and concentrations in blood and tissues across narrowly defined age ranges, leveraging generalized estimates from our previously conducted systematic review and meta-analysis [2]. The model was formulated as a multi-compartment system of ordinary differential equations (ODEs), covering four thymocyte- and six CD4+ T-lymphocyte subpopulations across five physiological compartments (thymus, blood, lymphoid tissue, gastrointestinal tract, and lung). A range of candidate functional age dependencies was explored to identify regulatory mechanisms driving age-related changes in homeostasis and to reproduce age-dependent concentration profiles. In parallel, reciprocal cellular feedback functions were assessed as an alternative to explicit age-dependent parameterization. Model evaluation included (1) validation against total CD4+ T-cell data, (2) simulations of homeostatic perturbation following thymectomy, and (3) global sensitivity analysis using partial rank correlation coefficients (PRCC) at representative ages (0, 1, 20, 50, and 80 years) [3].
Results
The CD4+ T-lymphocyte homeostasis model comprised 24 ODEs describing immune cell counts; 26 parameters were fixed based on prior experimental information and 19 parameters were estimated using data derived from meta-analysis of cell concentrations for the first years of life. Age effects were captured using 13 empirical hyperbolic functions of age. The model identified reduced thymic output, together with age-related shifts in naïve and activated cell proliferation, central- and effector-memory differentiation, recent thymic emigrant (RTE) survival, and blood–tissue trafficking, as key determinants shaping CD4+ T-cell homeostasis across the human lifespan.
Across compartments and subpopulations, three distinct phases of age dynamics emerged: 0–18 years, 18–50 years, and >50 years. During the first phase, less differentiated subsets (RTE and naïve) exhibited a sharp increase followed by a decline, with a maximum at ~4 years of age. The second phase was characterized by a gradual expansion of cell numbers into adulthood (~40 years of age), whereas the third phase showed an additional rise in cell counts around ~64 years of age for most subpopulations, followed by a subsequent decline.
Global sensitivity analysis supported a stage-specific control structure: thymocyte and naïve homeostasis predominantly drove an early CD4+ T cell differentiation, whereas clonal expansion dominated the maintenance of memory and effector subsets, with the overall influence of differentiation and expansion processes decreasing with age. Simulations of a complete thymectomy condition indicated that increased naïve cells proliferation and reduced RTE cell death are important compensatory mechanisms following thymic loss; however, they were insufficient to restore long-term CD4+ T-cell counts, i.e. decades after thymectomy.
Conclusion
By integrating heterogeneous clinical and experimental observations in a multiscale mechanistic framework, the proposed physiologically-based mathematical model enables quantitative prediction of age-dependent CD4+ T-cell dynamics and assessment of physiological changes in response to perturbations affecting immune homeostasis. Moreover, the model can serve as a core module for a drug-disease modeling platform to support drug and vaccine developments, in line with model-informed drug discovery and development (MIDD) principles.
References:
1 – Montecino-Rodriguez E, Berent-Maoz B, Dorshkind K. Causes, consequences, and reversal of immune system aging. J Clin Invest. 2013;123(3):958-965. doi:10.1172/JCI64096
2 – Kulesh V, Peskov K, Helmlinger G, Bocharov G. Systematic review and quantitative meta-analysis of age-dependent human T-lymphocyte homeostasis. Front. Immunol. 2025;16:1475871. doi:10.3389/fimmu.2025.1475871
3 – Marino S, Hogue IB, Ray CJ, and Kirschner DE. A methodology for performing global uncertainty and sensitivity analysis in systems biology. J Theor Biol. (2008) 254:178–96. doi: 10.1016/j.jtbi.2008.04.011
Reference: PAGE 34 (2026) Abstr 11898 [www.page-meeting.org/?abstract=11898]
Poster: Drug/Disease Modelling - Other Topics