Abstract 3647: Validation of a computational reasoning model for precision oncology with large-scale patient-derived xenograft data

Abstract Comprehensive tumor molecular testing is key in precision oncology. However, interpretation of complex molecular profiles in clinical practice can be challenging and subjective. The Digital Drug Assignment (DDA) system is a computational reasoning model that scores treatment options based on full tumor genomic data. This approach was predictive of relative benefit of the agents as used in the SHIVA01 trial (DOI: 10.1038/s41698-021-00191-2). Here, we analyzed patient-derived xenograft (PDX) data from Gao et al. (DOI: 10.1038/nm.3954). Molecular profiles of 179 solid tumors were processed by DDA, the resulting personalized drug scores were evaluated with treatment outcome data in PDX models for 1151 molecularly targeted agent (MTA) monotherapies, using 23 MTAs, targeting key oncogenic signaling pathways. In the PDX avatar system, multiple treatments can be tested for the same tumor, and outcomes can be measured in clinical metrics, enabling direct evaluation of drug ranking. To this end, cohorts were formed based on DDA ranking positions for each MTA-tumor pair. DDA ranking position was predictive of relative treatment benefit (Table, upper panel). Because the difference between neighboring ranks is probably minor, and to increase group sample sizes, larger cohorts were formed by combining neighboring rank groups. Here, a clear pattern of gradual decrease of treatment benefit with ranking distance from the top cohort was demonstrated (Table, lower panel). In this study, utilizing the unique advantages of the PDX system, we could demonstrate the predictive power of DDA treatment ranking. Digital drug assignment can serve as a practical and economic in silico avatar system for treatment planning and can potentially address challenges with complex molecular profiles in routine clinical settings. Citation Format: Robert Doczi, Akos Takacs, Anna Dirner, Dora Lakatos, Maria Kocsis-Steinbach, Barbara Vodicska, Dora Tihanyi, Eniko Kispeter, Reka Szalkai-Denes, William T. Beck, Petak Istvan. Validation of a computational reasoning model for precision oncology with large-scale patient-derived xenograft data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3647.