Immune Subtyping of Prostate Adenocarcinoma Reveals Tumor Microenvironment Heterogeneity and Enables a Multigene Recurrence‐Risk Model

Background Prostate adenocarcinoma (PRAD) exhibits marked heterogeneity in its immune microenvironment, which influences tumor progression and therapeutic response. However, a comprehensive delineation of immune infiltration patterns and their prognostic relevance in PRAD remains incomplete. Methods We analyzed TCGA‐PRAD transcriptomes using ssGSEA, CIBERSORT, ESTIMATE, and complementary deconvolution frameworks (xCell, TIMER, EPIC, and MCPcounter). Immune subtypes were derived by unsupervised clustering of immune‐related gene expression. Differential immune characteristics, clinical associations, and putative immune‐evasion features were examined. Prognostic immune genes were identified via univariate Cox and LASSO‐Cox regression using biochemical recurrence–free survival (BCRFS) as the primary endpoint (PFI/OS exploratory). A multigene risk score was constructed and evaluated by Kaplan–Meier analysis, time‐dependent ROC, calibration, and a nomogram integrating clinical covariates. Results Two immune subtypes—ImmunityH and ImmunityL—were identified with distinct tumor microenvironment (TME) features. ImmunityH displayed higher stromal/immune scores, enhanced antigen presentation and interferon signaling, upregulated HLA expression, enrichment of CD8 + T cells, activated memory CD4 + T cells, and M1 macrophages, whereas ImmunityL was characterized by increased M2 macrophages and resting immune populations. A multigene signature stratified patients into high‐ and low‐risk groups with significantly different BCRFS. The risk score correlated negatively with antitumor immune cells (e. g. , memory B cells and CD8 + T cells) and positively with immunosuppressive components (e. g. , cancer‐associated fibroblasts). Incorporation of the risk score with PSA, Gleason/ISUP grade, and pathologic T/N stage in a nomogram improved individualized recurrence‐risk prediction. Conclusion This study delineates the immune heterogeneity of PRAD and proposes a robust immune‐related prognostic model that supports risk stratification and may inform immunotherapeutic decision‐making.