Abstract 2004: Interrogating the genomic co-evolution of the immune system and cancer along the time and treatment continuum.

Abstract Clonal hematopoiesis (CH) is exemplified by the expansion of mutant hematopoietic stem and progenitor cells that may give rise to altered immune repertoires. Driven largely by age and chronic inflammation, CH is historically linked to myeloid malignancies. Although emerging evidence points to a modest prevalence of CH within solid tumors, the impact of these naturally occurring alterations on anti-tumor immunity and therapeutic responses remain less defined. Here, we employed somatic mutation calling algorithms to detect single nucleotide variants (SNVs) in both epithelial and immune compartments of tumors. Our preliminary analyses of single-cell RNA-sequencing (scRNA-seq) of 32 primary, therapy-naïve prostate tumors encompassing 11 patients identified mutations representative of both CH and prostate cancer. Within the tumor-epithelial compartment, we identified recurrent FOXA1 class 1C mutations in 27% (3/11), alongside NCOR1 mutations in 18% (2/11), IDH1 in 9% (1/11), and SPOP in 9% (1/11) of patients. Of note, these SNVs were detected across multiple samples for the same patient. Interestingly, we also detected germline HOXB13 G84E variants, associated with increased risk for prostate cancer, in 2 cases at a high clonality within epithelial cells (>75%). CH alterations were identified in 27% (3/11) of patients based on the detection of exonic, nonsynonymous SNVs in myeloid cells. Specifically, these variants affected SF3B2 (variant allele frequency; VAF = 0.30), UBE2A (VAF = 0.30), and KMT2C (VAF = 0.43) that are predicted deleterious across multiple classifier algorithms. Patients with detectable CH alterations displayed immunologically hotter tumor microenvironments, with CH+ patients having significantly higher proportions of tumor-infiltrating monocytes (16.22% vs 7.13%, Fisher’s exact test p<0.0001), macrophages (6.49% vs 2.88%, p<0.0001), CD8+ T cells (10.35% vs 3.33%, p<0.0001), CD4+ T cells (2.81% vs 0.74%, p<0.0001), and B cells (1.81% vs 0.59%, p<0.0001), and lower epithelial fractions (50.94% vs 69.81%, p<0.0001). In summary, we leveraged de novo mutational algorithms to detect CH alterations in prostate cancer at the single-cell resolution. Further work will integrate our mutational signatures with transcriptomics to elucidate the impact of both cancer- and immune-intrinsic alterations on anti-tumor immunity. Additionally, we will incorporate mitochondrial DNA-based lineage tracing algorithms on matched peripheral blood, primary, and metastatic tumor samples across diverse solid tumor types to deconvolute the spatiotemporal regulation of naturally occurring CH alterations across cancer progression. We envision our findings will illuminate novel drivers of immune function and regulation within solid tumors that may be exploited for future immunomodulatory therapies. Citation Format: John Roy Lozada, Christine Luo, Devin Schmeck, Jeffrey S. Miller, Akash Patnaik, Emmanuel S. Antonarakis, Frank Cichocki, Justin Hwang. Interrogating the genomic co-evolution of the immune system and cancer along the time and treatment continuum [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 2004.