Prostate cancer (PCa) is an extensive heterogeneous disease with a complex cellular ecosystem in the tumor microenvironment (TME). However, the manner in which heterogeneity is shaped by tumors and stromal cells, or vice versa, remains poorly understood. In this study, single‐cell RNA sequencing, spatial transcriptomics, and bulk ATAC‐sequence are integrated from a series of patients with PCa and healthy controls. A stemness subset of club cells marked with SOX9^highAR^low expression is identified, which is markedly enriched after neoadjuvant androgen‐deprivation therapy (ADT). Furthermore, a subset of CD8⁺CXCR6⁺ T cells that function as effector T cells is markedly reduced in patients with malignant PCa. For spatial transcriptome analysis, machine learning and computational intelligence are comprehensively utilized to identify the cellular diversity of prostate cancer cells and cell‐cell communication in situ. Macrophage and neutrophil state transitions along the trajectory of cancer progression are also examined. Finally, the immunosuppressive microenvironment in advanced PCa is found to be associated with the infiltration of regulatory T cells (Tregs), potentially induced by an FAP⁺ fibroblast subset. In summary, the cellular heterogeneity is delineated in the stage‐specific PCa microenvironment at single‐cell resolution, uncovering their reciprocal crosstalk with disease progression, which can be helpful in promoting PCa diagnosis and therapy.

Purpose: Tumor heterogeneity and burden, which impact treatment outcome in prostate cancer, are rarely evaluated using next-generation imaging. Experimental Design: The trial prospectively included 37 patients who had an early PSA progression (≤2 ng/mL) during castration and high-risk (PSA doubling time ≤10 months) nonmetastatic disease by conventional imaging. All patients underwent both ⁶⁸Ga-PSMA and ¹⁸F-FDG PET/CT. Lesions were classified into PSMA+FDG± lesions and PSMA-FDG+ lesions. The primary endpoint was the prevalence of PSMA-FDG+ disease. Tumor burden, predictors for positive imaging, and suitability for oligometastases-directed therapy (OMDT) were also evaluated. Results: All patients were treated with RP and the median duration of castration was 23 months. The median PSA at imaging was 0.57 ng/mL. Overall, 114 lesions were detected in 29 of the 37 patients. A high prevalence (73%) of N+/M+ disease was observed. Of the 114 lesions, 81 were PSMA+FDG± and 33 were PSMA-FDG+. Per patient level, 9 men (24%; 95% confidence interval: 10%–39%) showed at least one new PSMA-FDG+ lesions. A short PSA doubling time (P = 0.009, OR = 8.000) was associated with PSMA+FDG± disease, while a high Gleason grade group (P = 0.022, OR = 13.091) with PSMA-FDG+ disease. Nineteen patients (51%) with 51 lesions, including 10 PSMA-FDG+ lesions, could be enrolled for OMDT. Among different disease stages, PSMA-FDG+ disease was rarely detected in the hormone-sensitive cohort, but frequently found in the castration-resistant cohort. Conclusions: Using ⁶⁸Ga-PSMA and ¹⁸F-FDG PET, we observed a high prevalence of N+/M+ disease and a significant proportion of PSMA-FDG+ disease in patients with an early PSA progression during castration (ChiCTR1900022634).