Abstract 849: Evaluation of a transcriptomics-based precision oncology platform for patient-therapy alignment in treatment resistant malignancies

Background: Predicting in vivo response to antineoplastics remains an elusive challenge. We evaluated two novel transcriptomic-based, tumor-agnostic systems biology tools, OncoTarget and OncoTreat. Both methodologies predict tumor response to a diverse repertoire of clinically relevant oncology drugs, based on tumor Master Regulator (MR) analysis and de novo, experimentally-assessed drug mechanism-of-action in cognate cell lines. Specifically, OncoTarget identifies high-affinity inhibitors of individual MR proteins, while OncoTreat identifies drugs that invert the transcriptional activity of hyper-connected MR modules. Methods: We enrolled patients with several distinct, poor prognosis malignancies that had progressed on multiple standard therapies, and developed low-passage, patient-derived xenograft (PDX) models. We assessed in vivo tumor response to 35 predicted patient-drug pairings in the first seven successfully engrafted models, including three basal-like breast cancers, a pancreatic ductal carcinoma, a KIT/PDGFR wildtype gastrointestinal stromal tumor, a colon cancer, and a recurrent atypical meningioma. Results: Both OncoTarget and OncoTreat produced highly significant 30-day disease control rates (68% and 91%, respectively) and markedly delayed time to treatment failure versus vehicle control by Kaplan-Meier analysis (p < 10−4, log-rank test). Predicted drugs significantly outperformed randomly selected antineoplastic drugs that were not predicted by either methodology. Importantly, of 18 OncoTreat-predicted drugs, 15 showed MR-module inversion in vivo, demonstrating conservation of drug effect on patient tumor MR modules between carefully selected cognate cell lines and PDXs. Further, extensive benchmarking of OncoTarget and OncoTreat on TCGA tumor cohorts and prospectively profiled tumors demonstrates that candidate drugs can be identified for the vast majority of patients. Conclusions: Complementary precision cancer medicine paradigms are needed to broaden the clinical benefit realized through genetic profiling and immunotherapy. We present extensive preclinical validation of two transcriptomic-based approaches. Our results suggest OncoTarget and OncoTreat may substantively complement existing PCM approaches. Importantly, as RNA-based prediction tools are relatively affordable and allow initial predictions within two weeks of receipt of specimen, OncoTarget and OncoTreat are readily scalable for the design of basket and umbrella clinical trials that would enroll diverse patient populations. Citation Format: Prabhjot S. Mundi, Filemon S. Dela Cruz, Michael V. Ortiz, Adina Grunn, Daniel Diolaiti, Mariano J. Alvarez, Andrew L. Kung, Andrea Califano. Evaluation of a transcriptomics-based precision oncology platform for patient-therapy alignment in treatment resistant malignancies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 849.