Abstract 3197: Transcriptional characterization of immunological features observed in localized and metastatic prostate cancer

Abstract

Abstract Objectives: There is a paucity of data describing the immunologic subtypes of prostate cancer, and the features that contribute to them, in either the localized or metastatic disease setting. The objective of our effort was to utilize publicly available datasets in both disease settings to better understand the prevalence of immunologic subtypes within each disease setting and to explore biological mechanisms that may regulate them. Methods: Publicly available datasets utilized were The Cancer Genome Atlas (TCGA), for localized disease, and Stand Up To Cancer (SU2C), for metastatic disease. Evaluating a pan-cancer unsupervised clustering analysis of over 5000 tumors in TCGA with 20 biologically distinct immune related transcriptional signatures we identified sample classes with distinct profiles. The expression level of signatures within each sample class was taken as a composite score to identify the immunological characteristics of individual tumors. This characterization was then applied to the SU2C dataset to interrogate 444 mCRPC tumors. Results: The results of the pan-cancer analysis identified 4 individual sample classes; hot, cold, and 2 altered phenotypes. Of the two altered phenotypes identified, one represented only prostate cancer samples, and was the focus of our investigation in this study. The TCGA dataset showed, 60% cold, 12% hot, and 28% altered. A transcriptional signature representing Macrophage Colony Stimulating Factor 1 (CSF-1) response showed positive correlation with effector cells(CD4 & CD8), while it showed a negative correlation with suppressive cell(Tregs & MDSCs) highlighting the potential for positive immune modulation through this pathway. The SU2C dataset showed, 83% cold, 6% hot, and 11% altered. Cancer testis antigens (CTAs) were expressed at a higher level in SU2C compared to TCGA. We also observed positive correlation between the expression of CTAs and immune suppressive mechanisms and a negative correlation with inflammation. These observations were unique to the metastatic disease setting suggesting a role of CTA expression and increased immune suppressive environment in mCRPC. Conclusions: Through characterization of publicly available datasets, TCGA & SU2C, we were able to define frequencies and characteristics of hot, cold, and altered immune subtypes within different disease settings in prostate cancer. Further validation of unique subtypes and mechanisms identified in this study in larger cohorts would be required to understand the true biological impact of these observations. Citation Format: Justin Lucas, Clemente Aguilar, Michael Gormley, Shibu Thomas, Natalie Hutnick, John Loffredo. Transcriptional characterization of immunological features observed in localized and metastatic prostate cancer [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 3197.