Abstract 1294: Deciphering single-cell heterogeneity and cellular ecosystem dynamics during prostate cancer progression.

Abstract Prostate cancer (PC) progresses from benign epithelium through pre-malignant lesions, localized tumors, metastatic dissemination, and castration-resistant stages, with some cases exhibiting phenotype plasticity under therapeutic pressure. However, high-resolution insights into how cellular phenotypes and ecological interactions shift across these successive stages remain incomplete. Here, we present the Prostate Cancer Cell Atlas (PCCAT) by integrating ∼710,000 single cells from 197 human samples covering a spectrum of tumor stages. By further incorporating bulk transcriptomic cohorts, spatial transcriptomics, and experimental validations, this comprehensive analysis dissects the multicellular landscape of PC and identifies key epithelial, stromal, and immune programs associated with disease progression. We highlight a dynamic and heterogeneous continuum of malignant and non-malignant epithelial states, and uncover several cell states strongly linked to aggressive progression and poor prognosis, including lineage plasticity-like malignant cells, neuroendocrine tumor cells, matrix cancer-associated fibroblasts (mCAFs), and SPP1-expressing macrophages (SPP1+ Mph). Furthermore, we identify shared immune-suppressive states enriched in advanced disease, including proliferative T cells, activated regulatory T cells, exhausted CD8+ T cells, and SPP1+ Mph, highlighting convergent microenvironmental mechanisms of immune evasion. Among these, SPP1+ Mph emerge as a key immunosuppressive population, progressively accumulate along PC progression, exhibiting an M2-like immunosuppressive phenotype, associating with bone metastasis, and spatially colocalizing with mCAFs at the tumor-stroma interface in NEPC. Functionally, SPP1 blockade restrains bone metastatic progression and alleviates cancer-induced bone pain by reprogramming tumor-associated macrophages and reshaping CAF and CD8+ T-cell phenotypes. Lastly, by integrating PCCAT with spatial transcriptomics, we identify distinct spatial cellular neighborhood (CNs) that transition from epithelial-dominated to stroma-enriched ecosystems along PC evolution. Among tumor-associated CNs, CN1 reflects neuroendocrine and lineage-plasticity features, whereas CN4 captures aggressive luminal malignant programs. Among microenvironmental CNs, mCAF-enriched CN7 expands with tumor progression and predicts adverse prognosis, while iCAF-enriched CN8 diminishes and associates with favorable outcome. Notably, a CN7-CN8-derived ecosystem index robustly predicts biochemical recurrence across multiple patient cohorts. Overall, our study provides a high-resolution reference of PC cellular ecosystems and reveals the dynamic cell states, spatial niches, and immunoregulatory interactions that shape disease progression, offering mechanistic insights and therapeutic opportunities. Citation Format: Faming Zhao, Han Zeng, Jianming Zeng, Canping Chen, Xiaofan Zhao, Tingting Zhang, Kunlun Wang, Gulsu Sener, Jingui Liu, George V. Thomas, Rosalie C. Sears, Joshi J. Alumkal, Amy Moran, Gordon B. Mills, Ece S. Eksi, Ji Zheng, Peter S. Nelson, Zheng Xia. Deciphering single-cell heterogeneity and cellular ecosystem dynamics during prostate cancer progression 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 1294.