Abstract 6214: OncodynamiX as an artificial intelligence (AI) based platform for precision medicine

Abstract Introduction: In spite of the recent advances in cancer research, cancer incidence and mortality rate still remain high. Although genomic based approaches are used more extensively for identifying the right treatment approach, only a few patients benefit from this because a) only a limited number of mutations have been classified as driver mutations that are of ‘significance’ and b) in case of multiple mutations, the approach to identify right drug or drug combinations is not well understood. In this regard, there is a need for approaches that use artificial intelligence where exhaustive omics data is effectively processed, mined and utilized to device personalized therapy. Therefore, we used OncodynamiX an AI driven Cancer Biology platform to identify the right drug (s) for patients that had limited or no treatment options. We present here 2 case studies, one patient with stage 4 uterine serous adenocarcinoma and another one with endometrial adenocarcinoma. Methods: OncodynamiX platform has over 15 million data points, including mutation, CNA, mRNA and protein data for 1500 cell lines as well as potency, efficacy, target, phenotype and network information on more than 250 drugs. Based NGS data from the patients, OncodynamiX platform identified direct or indirect targetable alterations and then picked drugs or compounds shown to be active in the altered pathways. Based on this, a drug-gene alteration matrix was created which formed the basis for identification of appropriate drug (s). Results: The first patient with uterine serous adenocarcinoma had the following mutations: PIK3CA (Gain of function, GoF), TP53 (Loss of Function, LoF), PPP2R1A (LoF), APC (Conservation of function) ; Copy number amplification or high expression was reported for the following genes: BRD4, MYC, NOTCH3, CCNE1 and MUC16. Based on the drug-gene alteration matrix and all available drug data, WEE1 inhibitor was suggested to be the best therapy option for this patient. Accordingly, the patient was treated with WEE1 inhibitor in a then ongoing clinical trial. The second patient with endometrial adenocarcinoma had FBXW7, BRCA1, and PIK3R1 LoF, PIK3CA GoF and TP53 switch of function mutations, amplifications in CCNE1, FGFR1, NSD3, ZNF217 and ZNF70033 and loss of PTEN and FAS. For this patient, Pazopanib and Evrolimus combination was recommended and the patient was treated accordingly. Both patients responded well for the treatment as measured by RECIST response and continued with the treatment for over a year. Conclusion: The studies presented above highlight the promise of this AI-based approach in personalized medicine. In multiple cases we have used OncodynamiX approach where clear clinical benefit was demonstrated. To further validate this approach, we are initiating a clinical trial with 250 patients. Such approaches should be adopted more widely for significant improvement in life expectancy as well as in quality of life of patients. Citation Format: Dhanalakshmi Sivanandhan, Priyanka R. Bhargav, Sumanth M. Vasista, Sirisha Narayanbhatla, Suman Kamath, Oguru Sailaja, Dipanjan Chakraborty, Sundaresh Babu. OncodynamiX as an artificial intelligence (AI) based platform for precision medicine abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts) ; 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84 (6Suppl): Abstract nr 6214.