Abstract 7351: Delineating the prostate cancer tumor microenvironment through scRNA-seq profiling at different timepoints

Itzel Valencia,Hubert Pakula,Mohamed Omar,Massimo Loda,Luigi Marchionni

DOI: https://doi.org/10.1158/1538-7445.am2024-7351

IF: 11.2

2024-03-22

Cancer Research

Abstract:Abstract Distant metastasis remains the leading cause of death in patients with prostate cancer (PCa), however, the underlying mechanisms of this process remain unclear. Using samples from genetically-engineered mouse models (GEMMs) of PCa, we have identified eight different stromal cell states linked to epithelial mutations that may impact PCa progression to metastatic stages. To further investigate the unique intercellular interactions in the tumor microenvironment (TME) involved in mediating disease progression, we use single cell RNA sequencing (scRNA-seq) to profile primary PCa samples from different GEMMs including TRAMP, Hi-MYC, and FVBN, at different timepoints. Analysis of the scRNA-seq data revealed a significant enrichment of three clusters, designated as stromal c6 (s6), epithelial c7 (e7), and epithelial c10 (e10), exclusively in TRAMP, each expressing proliferative markers such as Mki67 and Top2a. To investigate the intercellular interactions among these clusters, we conducted ligand-receptor (LR) interaction analysis using CellChat. Notably, s6 is expressing the genes encoding ligands Bmp4, Fn1, and Igf1, which interact with the receptors Bmpr1b/Bmpr2, Itga3/Itgb1, and Igf1r, respectively expressed in e7. Subsequently, the transcriptomic profiles of these cells were used to build and validate a robust gene signature that can predict metastasis and survival in patients with PCa using a large cohort of 1239 samples, 930 of which were used for training while the remaining 309 samples were used for testing the model’s performance. In conclusion, these results underscore the role of the TME mediating PCa progression and highlight the potential of leveraging the TME to extract predictive and prognostic features which can be later employed to build predictive and prognostic models for patients with PCa. Citation Format: Itzel Valencia, Hubert Pakula, Mohamed Omar, Massimo Loda, Luigi Marchionni. Delineating the prostate cancer tumor microenvironment through scRNA-seq profiling at different timepoints [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7351.

oncology