Abstract 7115: Mutational signatures in PDXs for improved understanding of drug response and companion biomarkers identification

Abstract Genomic DNA editing is a continuous process that occurs during the entire cell life span. The type and frequency of these modifications can be related to the physiological or pathological activity of intrinsic mechanisms such as DNA surveillance and repair or to extrinsic events that may induce an alteration of DNA sequence by exposure to agents that directly or indirectly induce accumulations of DNA alterations. In the past few years, large-scale analyses have revealed mutational signatures across human cancer types. These signatures can be used as markers of defective internal processes, such as DNA repair deficiency, or external exposures, such as carcinogens, like tobacco, or genotoxic therapies such as radiation and chemotherapy. Our TumorGraft platform, a collection of 1500 patient-derived xenograft generated from more than 50 different types of cancer, is one of the most comprehensive preclinical oncology platforms worldwide, and aims to recapitulate the variety of the patient population and tumor biology complexity. This platform is currently used to evaluate the efficacy of new drugs, and all our models are very well characterized at the molecular level, including whole transcriptome, proteomics and phospho-proteomics, quantification, and genomic variation calls. This allows accurate selection of models with the molecular characteristics of the target patient population, as well as to identify biomarkers predictive of treatment efficacy. This could eventually lead to the development of companion biomarkers in the clinical setting to improve the identification of patients that will benefit from the treatment and those that should be spared. To maximize the chances of identifying relevant molecular traits associated with tumor response, we have been utilizing the Pharmaco-Pheno-Multiomic (PPMO) integration workflow, a machine learning approach which combines phenotypic and therapeutic response profiles with multiple omics datatypes to generate complex biomarker profiles. To provide additional insights on the molecular characteristics of our tumors, we decided to perform the mutational signatures profiling of our models by using whole exome sequencing data. The results obtained were crossed with patient’s clinical history and showed good correlation between the mutational signatures identified and the treatments received by the patient, in particular for exposure to platinum salts. The addition of PDXs mutational signatures strongly improves our knowledge on these models and confers an important parameter for model selection. Moreover, the integration of mutational signature profiles in the PPMO workflow would make a significant contribution for companion biomarkers identification. Citation Format: Haia Khoury, Stefano Cairo, Mara Gilardi, Michael Ritchie, Gilad Silberberg. Mutational signatures in PDXs for improved understanding of drug response and companion biomarkers identification 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 7115.