A Minimal Model Framework for Robust CAR-T Cell and Oncolytic Virus Combination Therapy

Area of Science:

• Oncology
• Immunotherapy
• Mathematical Modeling

Background:

• Glioblastoma is a highly lethal brain cancer.
• Combination therapy with CAR-T cells and oncolytic viruses shows potential but lacks mechanistic understanding.
• Synergistic mechanisms in glioblastoma combination immunotherapy require elucidation.

Purpose of the Study:

• To develop and validate mathematical models for predicting glioblastoma combination immunotherapy outcomes.
• To analyze the interaction between IL-13Rα2-targeting CAR-T cells and the oncolytic virus C134.
• To assess the utility of quasi-steady-state (QSS) approximations in simplifying complex immunotherapy models.

Main Methods:

• Development of a minimal mathematical model framework for glioblastoma immunotherapy.
• Application of timescale separation and quasi-steady-state (QSS) approximations to reduce model complexity.
• Comparison of full and QSS models using patient-derived glioblastoma data and Akaike Information Criterion (AIC).

Main Results:

• The QSS model, with 9 parameters, achieved comparable fits to the full 11-parameter model.
• QSS models were generally favored by AIC, indicating improved parsimony and predictive power.
• CAR-T cell exhaustion dynamics did not significantly improve model fits within a 72-hour window.

Conclusions:

• Simplified QSS formulations effectively capture viral dynamics in combination immunotherapy.
• The QSS model provides a practical and accurate framework for optimizing glioblastoma combination immunotherapies.
• Further investigation into CAR-T cell exhaustion may be warranted beyond the initial 72-hour timeframe.