Search

n/a

Fatal neuroendocrine differentiation (NED) of castration-resistant prostate cancer is a recurrent mechanism of resistance to androgen deprivation therapies (ADT) and antiandrogen receptor pathway inhibitors (ARPI) in patients. The design of effective therapies for neuroendocrine prostate cancer (NEPC) is complicated by limited knowledge of the molecular mechanisms governing NED. The paucity of acquired genomic alterations and the deregulation of epigenetic and transcription factors suggest a potential contribution from the microenvironment. In this context, whether ADT/ARPI induces stromal cells to release NED-promoting molecules and the underlying molecular networks are unestablished. Here, we utilized transgenic and transplantable mouse models and coculture experiments to unveil a novel tumor-stroma cross-talk that is able to induce NED under the pressure of androgen deprivation. Castration induced upregulation of GRP78 in tumor cells, which triggers miR29-b–mediated downregulation of the matricellular protein SPARC in the nearby stroma. SPARC downregulation enabled stromal cells to release IL6, a known inducer of NED. A drug that targets GRP78 blocked NED in castrated mice. A public, human NEPC gene expression dataset showed that Hspa5 (encoding for GRP78) positively correlates with hallmarks of NED. Finally, prostate cancer specimens from patients developing local NED after ADT showed GRP78 upregulation in tumor cells and SPARC downregulation in the stroma. These results point to GRP78 as a potential therapeutic target and to SPARC downregulation in stromal cells as a potential early biomarker of tumors undergoing NED. Significance: Tumor–stroma cross-talk promotes neuroendocrine differentiation in prostate cancer in response to hormone therapy via a GRP78/SPARC/IL6 axis, providing potential therapeutic targets and biomarkers for neuroendocrine prostate cancer.

Immunotherapy, including the use of checkpoint inhibitors, is a potent therapeutic approach for some cancers, but has limited success with prostate tumors, in which immune suppression is instigated by the tumor. The immunosuppressive capacity of mast cells, which promote adenocarcinoma development in the prostate, prompted our investigation on whether mast cells promote tolerance to SV40 Large-T antigen, the transforming oncogene in transgenic adenocarcinoma of the mouse prostate (TRAMP) mice. The incidence of adenocarcinoma was reduced in the offspring of a cross between TRAMP mice and mast cell–deficient KitWsh mice. TRAMP mice are tolerant to the SV40 Large T antigen, which is otherwise immunogenic in normal syngeneic B6 mice. Genetic ablation of mast cells in TRAMP mice restored their ability to mount a tumor-specific cytotoxic T-cell response. In KitWsh-TRAMP mice, the restored T-cell immunity correlated with the reduced activity of polymorphonuclear myeloid-derived suppressor cells (PMN-MDSC), along with their reduced expression of Arg1, Nos2, and Stat3. Having found that CD40L-expressing mast cells can interact in vivo with CD40-expressing PMN-MDSC, we then determined that only KitWsh-TRAMP mice reconstituted with mast cells expressing CD40L could restore PMN-MDSCs suppressive functions, T-cell unresponsiveness and adenocarcinoma development. Thus, mast cells have an immunoregulatory effect on PMN-MDSCs activity through CD40L-CD40 interaction, favoring immunosuppression and tumor onset. In prostate cancer patients, in silico analyses correlated poor clinical outcomes with high expression of genes related to mast cells and PMN-MDSCs. Cancer Immunol Res; 6(5); 552–65. ©2018 AACR.