In the development of cell-based cancer therapies, quantitative mathematical models of cellular interactions are instrumental in understanding treatment efficacy. Efforts to validate and interpret mathematical models of cancer cell growth and death hinge first on proposing a precise mathematical model, then analyzing experimental data in the context of the chosen model. In this work, we present the first application of the sparse identification of non-linear dynamics (SINDy) algorithm to a real biological system in order discover cell-cell interaction dynamics in in vitro experimental data, using chimeric antigen receptor (CAR) T-cells and patient-derived glioblastoma cells. By combining the techniques of latent variable analysis and SINDy, we infer key aspects of the interaction dynamics of CAR T-cell populations and cancer. Importantly, we show how the model terms can be interpreted biologically in relation to different CAR T-cell functional responses, single or double CAR T-cell-cancer cell binding models, and density-dependent growth dynamics in either of the CAR T-cell or cancer cell populations. We show how this data-driven model-discovery based approach provides unique insight into CAR T-cell dynamics when compared to an established model-first approach. These results demonstrate the potential for SINDy to improve the implementation and efficacy of CAR T-cell therapy in the clinic through an improved understanding of CAR T-cell dynamics.

In the development of cell-based cancer therapies, quantitative mathematical models of cellular interactions are instrumental in understanding treatment efficacy. Efforts to validate and interpret mathematical models of cancer cell growth and death hinge first on proposing a precise mathematical model, then analyzing experimental data in the context of the chosen model. In this work, we implement the sparse identification of non-linear dynamics (SINDy) algorithm to discover cell-cell interaction dynamics in in vitro experimental data, derived from chimeric antigen receptor (CAR) T-cells and patient-derived glioblastoma cells. By combining the techniques of latent variable analysis and SINDy, we infer key aspects of the interaction dynamics of cancer and CAR T-cell populations. Importantly, we show how the model terms can be interpreted biologically in relation to different CAR T-cell functional responses, single or double CAR T-cell-tumor cell binding models, and Allee effects in either of the CAR T-cell or cancer cell populations. As an application of data-driven model discovery, this work serves to identify biological interactions, and has the potential to improve the implementation and efficacy of CAR T-cell therapy in the clinic through an improved understanding of CAR T-cell dynamics.

No abstract

The combination of chimeric antigen receptor (CAR) T cell therapy, which delivers large numbers of tumor reactive T cells, and oncolytic viral therapy, causing activation of host immune responses, is an attractive approach for improving outcomes for patients with glioblastoma (GBM). Here we present data from two independent phase I clinical trials evaluating IL13Rα2-targeted CAR therapy (NCT02208362) and C134 oncolytic viral (OV) therapy (NCT03657576) for the treatment of recurrent GBM (rGBM), along with preclinical studies supporting the utility of combining these two therapies. For NCT02208362, locoregional delivery of IL13Rα2-targeted CAR T cells were evaluated in heavily pretreated patients with rGBM. Interrogating biomarkers of clinical response revealed that levels of intratumoral T cells prior to treatment were positively associated with overall survival; furthermore, two patients who achieved a complete response had the highest levels of intratumoral CD3+ T cells pre-therapy. These findings suggest that therapeutic strategies which increase endogenous immune infiltrates could augment CAR T cell mediated responses. For NCT03657576, intratumoral delivery of C134, a herpes simplex virus (HSV-1) that has been genetically engineered to safely replicate and kill glioma tumor cells, is also being evaluated for treatment of rGBM. We report findings from a patient treated intratumorally with 1 × 106 pfu of C134. At 6-7 weeks post treatment this patient had MRI changes that suggested possible recurrence or pseudoprogression, and therefore underwent resection with biopsy assessment. Evaluation of virus-treated areas showed increased immune infiltrates as compared to untreated tumor sites, suggesting that C134 activated host immune responses. These clinical findings provide the rationale for evaluating a combination therapy of C134 OV and IL13Rα2-CAR T cells to potentially reshape the tumor microenvironment (TME) and enhance CAR therapy. In orthotopic GBM models in nude mice, we show that co-treatment with the two agents gave no adverse reaction, and more notably pre-treatment with C134 re-shaped the TME by increasing immune cell infiltrates and enhanced the efficacy of sub-therapeutic doses of CAR T cell therapy delivered either intraventricularly or intratumorally. Ongoing preclinical studies aim to provide detailed phenotypic analysis, as well as a mechanistic understanding of this combination approach to support the potential benefit of a soon to be opened combination trial evaluating C134 and IL13Rα2-CAR T cells. In this clinical trial in patients with IL13Rα2+ rGBM and anaplastic astrocytoma, increasing doses of intratumorally administered C134 will be followed by dual intracranial intratumoral and intraventricular administration of IL13Rα2-targeted CAR T cell therapy. Citation Format: Christine E. Brown, Agata Xella, Jonathan C. Hibbard, Vanessa Salvary, Brenda Aguilar, Jamie Wagner, Bruce Dezube, Knut Niss, Lynn Bayless, James Edinger, Jianmei Leavenworth, Stephen J. Forman, Behnam Badie, James M. Markert, Kevin A. Cassady. Oncolytic viral reshaping of the tumor microenvironment to promote CAR T cell therapy for glioblastoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr CT541A.