Dissertations and Theses (Open Access)
Author ORCID Identifier
0000-0002-5264-521X
Date of Graduation
8-2026
Document Type
Dissertation (PhD)
Program Affiliation
Quantitative Sciences
Degree Name
Doctor of Philosophy (PhD)
Advisor/Committee Chair
Ken Chen
Committee Member
Stefano Casarin
Committee Member
May Daher
Committee Member
Eleonora Dondossola
Committee Member
Ziyi Li
Committee Member
Peng Wei
Abstract
Adoptive cell therapies (ACT) have shown promising progress in combating cancer, but clinical responses remain heterogeneous. Key determinants, including the functional states of infused cells, tumor–immune interactions, and dosing strategies, are difficult to resolve experimentally in the dynamic and heterogeneous tumor microenvironment, limiting optimization of ACT products and treatment regimens. To address the gap, we developed ABMACT (Agent-Based Model for Adoptive Cell Therapies), a computational framework that reconstruct key cellular functions, interactions, and molecular signatures to recapitulate cell population dynamics associated with differential treatment responses. ABMACT encodes autonomous “virtual cells” using literature-informed rules calibrated with experimental data, enabling simulation of an evolving tumor–immune ecosystem across diverse experimental settings.
ABMACT recapitulated tumor regression and cell population dynamics in lymphoma and glioblastoma xenograft models treated with NK cell products, enabling inference of the cellular properties and kinetics underlying differential tumor control. Using cell line–calibrated agents, it reconstructed differential responses in lymphoma patient cohorts treated with optimal versus suboptimal cord blood NK cells. Statistical modeling of paired scRNA-seq and tumor radiance data identified functional signatures of NK cytotoxicity, while co-culture simulations revealed distinct NK phenotypic trajectories. In silico perturbations identified key determinants of treatment efficacy, including effector-to-target ratios, cytotoxic killing probability, and serial killing capacity. The framework also enabled virtual dose optimization, predicting that early high-dose, high-efficacy NK treatment outperformed fractionated dosing and could rescue ineffective therapies.
Together, ABMACT captured cell population dynamics in experimental observations and provided mechanistic insights into cellular and molecular factors determining effective tumor control. The framework complements experimental models, enabling rapid hypothesis generation, dosing optimization, and rational design of next-generation NK cell therapies.
Recommended Citation
Wang, Yujia, "Agent-Based Modeling of Cellular Dynamics in Adoptive Cell Therapies" (2026). Dissertations and Theses (Open Access). 1549.
https://digitalcommons.library.tmc.edu/utgsbs_dissertations/1549
Keywords
Agent-based model, mechanistic modeling, single-cell, adoptive cell therapy, CAR-NK cell therapy