Abstract PO1-15-02: Comprehensive characterization of genetic interactions in breast cancer reveals therapeutic vulnerabilities

Abstract Background: Genome-informed and genome-targeted precision treatment for breast cancer have achieved remarkable progress in improving clinical outcomes for patients with specific genetic alterations. However, treatment efficacy is compromised by the current practice of basing treatment decision-making solely on single driver alterations, without considering the role of genetic interactions. Consequently, it is of great necessity to conduct systematic investigations to determine the clinical relevance of genetic interactions. Methods: We established a large-scale multi-omics cohort (N=873) and a real-world clinical sequencing cohort (N=4,421) representing the Asian breast cancer population. Detailed treatment records were collected. We then investigated the prognostic and predictive effects of genetic interactions based on multivariate Cox proportional hazards model and logistic regression model. To validate our findings, we utilized patient-derived organoids and tumor fragments to confirm the associations observed between genetic interactions and drug response. Results: Through integrated analysis of genomics, transcriptomics, proteomics, and metabolomics, we constructed a network comprising 54 co-occurring events and 38 mutually exclusive events, elucidating their association with dysregulated biological processes. External validations were performed in TCGA-BRCA, MSK-IMPACT, METABRIC, and AACR-GENIE datasets, respectively. Furthermore, we systematically revealed the prognostic effects of genetic interactions across distinct clinical subtypes. In triple-negative breast cancer, we found that the co-occurrence of PIK3CAmut-FOXA1mut was associated with unfavorable distant metastasis-free survival while TP53mut-MYBamp and TP53mut-CCNE1amp were associated with decreased overall survival. Additionally, we characterized the genetic interactions that impact the clinical outcomes of patients undergoing specific treatments in the neoadjuvant, adjuvant, and advanced settings. Notably, we identified associations such as TP53mut-AURKAamp with tamoxifen resistance, ERBB2amp-PAK1amp with resistance to trastuzumab-pertuzumab combinations, germline BRCA1mut-MYCamp with sensitivity to PARP1 inhibitors, and TP53mut-MYBamp with immunotherapy resistance. Conclusion: Overall, the consideration of genetic interactions may enhance our understanding of the heterogeneity in treatment response and complement ongoing efforts in precision oncology. Our study suggests that decision-making regarding genome-informed and genome-targeted treatment should extend beyond the scope of single driver alterations. Citation Format: Caijin Lin, Xi Jin, Ding Ma, Chao Chen, Xin Hu, Yi-Zhou Jiang, Zhi-Ming Shao. Comprehensive characterization of genetic interactions in breast cancer reveals therapeutic vulnerabilities [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO1-15-02.