Data driven model discovery and interpretation for CAR T-cell killing using sparse identification and latent variables.

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.

Potency monitoring of CAR T cells.

The effector potency of chimeric antigen receptor (CAR) T cell therapeutic products is essential to their clinical antitumor responses, and potency monitoring is a critical quality control method for CAR T cell therapy platforms. While many in vitro assays enable high-throughput assessment of CAR T cell cytotoxicity, it has been challenging for these assays to reflect the in vivo therapeutic effect due to their nature as short-term methods that fail to recapitulate the high tumor burden environment. Here, we describe two in vitro co-culture methods to evaluate CAR T cell recursive killing potential at high tumor cell loads. In these assays, long-term cytotoxic function and proliferative capacity of CAR T cells are examined in vitro over 7days. Further, these assays are coupled with profiling CAR T cell expansion, cytokine production and phenotypes. These methods provide a facile approach to assess CAR T cell potency and to elucidate the functional variations across different CAR T cell products.

Characterization of an RNA binding protein interactome reveals a context-specific post-transcriptional landscape of MYC-amplified medulloblastoma.

Pediatric medulloblastoma (MB) is the most common solid malignant brain neoplasm, with Group 3 (G3) MB representing the most aggressive subgroup. MYC amplification is an independent poor prognostic factor in G3 MB, however, therapeutic targeting of the MYC pathway remains limited and alternative therapies for G3 MB are urgently needed. Here we show that the RNA-binding protein, Musashi-1 (MSI1) is an essential mediator of G3 MB in both MYC-overexpressing mouse models and patient-derived xenografts. MSI1 inhibition abrogates tumor initiation and significantly prolongs survival in both models. We identify binding targets of MSI1 in normal neural and G3 MB stem cells and then cross referenced these data with unbiased large-scale screens at the transcriptomic, translatomic and proteomic levels to systematically dissect its functional role. Comparative integrative multi-omic analyses of these large datasets reveal cancer-selective MSI1-bound targets sharing multiple MYC associated pathways, providing a valuable resource for context-specific therapeutic targeting of G3 MB.