Abstract 1241: Deciphering heterogeneity and tumor evolution in bladder cancer using multi-omics and liquid biopsies from rapid autopsies

Abstract Introduction: Metastatic bladder cancer (mBLCA) exhibits significant heterogeneity at the genomic, transcriptomic, molecular, and epigenetic levels, contributing to challenges in diagnosis, prognosis, and treatment. While 75% of mBLCA cases are urothelial carcinomas, 25% show divergent differentiation with variant histological features. This work characterizes intra-patient and inter-patient heterogeneity through comprehensive genomic and transcriptomic analyses of a rapid autopsy cohort of 20 patients with multiple tumors, matched normal tissue, and cell-free DNA (cfDNA). Our findings provide critical insights into tumor heterogeneity, clonal evolution, metastatic seeding, and the tumor microenvironment. Approach: Using bioinformatics tools, we reconstructed clonal evolution trees from whole-genome sequencing (WGS) analysis of tumors and inferred migration patterns to determine metastatic seeding. COSMIC mutational signatures were examined to correlate histological subtypes with clinical history. Single-nucleus RNA sequencing (snRNA-seq) revealed cell types and transcriptional heterogeneity using the human protein atlas and known marker annotations. To investigate alterations and clonality in tumors and their matched cfDNA, we used custom methodologies to evaluate mutations, CNAs, SVs, and rearrangements. Results: The clonal evolution trees and bulk RNA-seq highlight cisplatin resistance in the plasmacytoid urothelial carcinoma subtype, driven by upregulation of the DNA damage response pathway. Significant mutational heterogeneity (∼20-30% subclonal) emerges in later stages of tumor evolution, potentially contributing to therapy resistance. Additionally, a substantial proportion (60-70%) of structural variants arise later in tumor progression, particularly in driver genes. snRNA-seq revealed distinct clusters of epithelial cells, macrophages, fibroblasts, and other cell types. cfDNA analysis captured over 80% of subclonal deleterious tumor mutations in BLCA driver genes. Nucleosome profiling of cfDNA helps distinguish mBLCA from healthy samples and identify key transcription factors to differentiate between variants. Conclusion: This comprehensive study of mBLCA, leveraging a rapid autopsy dataset, uncovers critical insights into variant histologies across multiple omics layers. Integrating cfDNA analysis captures intra-patient and inter-patient tumor heterogeneity, providing a holistic view of clonal dynamics. By elucidating the patterns of clonal evolution, this work advances our understanding of tumor progression and establishes a foundation for precision therapies targeting adaptive and evolving clones. Citation Format: Pushpa Itagi, Samantha Schuster, Sonali Arora, Patricia Galipeau, Thomas Persse, Funda Vakar-Lopez, John K. Lee, Petros Grivas, Robert Montgomery, Jonathan Wright, Hung-Ming Lam, Andrew Hsieh, Gavin Ha. Deciphering heterogeneity and tumor evolution in bladder cancer using multi-omics and liquid biopsies from rapid autopsies [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 1241.