Determining patterns and mechanisms of drug resistance is fundamentally required for improving clinical outcome of cancer treatments. The ability to study multiple samples from different metastatic sites of the same patient is a clinically challenging task, which has become possible with the advent of “rapid” autopsy procedures (<6 hours from death) conducted on the patients deceased from cancer. We have obtained and analyzed whole-exome, whole genome and transcriptome sequencing data from advanced breast cancer samples derived from multiple cancer lesions of the same patient post-treatment through the Massachusetts General Hospital Rapid Autopsy program. We have integrated genomic and transcriptomic data through advanced clonal reconstruction methods (PhylogicNDT) to investigate how the patients developed resistance to multiple lines of anti-cancer therapy. Additionally, we have used previously collected cell-free DNA samples to establish both a spatial and temporal picture of cancer drug resistance and progression. We have examined clonal heterogeneity and resistance to targeted therapies in 10 metastatic ER+ breast cancer patients over 120 distinct metastatic sites. The majority of patients in this cohort were treated with CDK4/6 cell cycle inhibitors in addition to hormone therapy. We discovered that multiple metastatic sites often share similar clonal structure, but overall progeny of several sibling clones spreads throughout the body, creating “families” of more closely related lesions. These distinct lesion populations often develop independent resistance mechanisms to the many lines of treatment that the patents have received. At least 8 out of 10 patients showed multiple known and potentially novel resistance mechanisms in separate branches of the phylogenetic tree, often converging on distinct mutations in the same resistance genes. Notably we have identified ESR1, KRAS and chromatin modifier recurrent and convergent mutations across the cohort. Temporal analysis using cfDNA taken over the course of treatment allows us to identify clones that expanded in response to a given treatment. Using RNA expression and pathway analysis, we identify unique transcriptional programs, differentially expressed genes and ER-signaling changes between distinct clones within one patient, as well as compare genetically similar clones across patents.

Citation Format: Ignaty Leshchiner, Elizabeth E. Martin, Christopher T. Chen, Elizaveta Leshchiner, Thomas Zhang, Christopher Pinto, Filippo Utro, Kahn Rhrissorrakrai, Chaya Levovitz, Brian P. Danysh, Kara Slowik, Raquel A. Jacobs, Laxmi Parida, Gad Getz, Dejan Juric. Convergence of resistance patterns in breast cancer after multiple lines of treatment through analysis of rapid autopsy samples [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 37.