Abstract Advanced prostate cancer (PCa) in patients still progresses to metastatic relapse after curative treatment such as primary tumor removal. This is likely due to the early disseminated PCa cells that stayed dormant initially and got reactivated to proliferate into overt metastases. However, we don’t know when and to which organs PCa cells disseminate, whether and how primary tumor removal affects the dissemination and dormancy. To answer these, we injected C4-2B cells subcutaneously into immunodeficient NSG mice and randomized the mice for either having the tumors removed at week 2 or keeping the tumors till the endpoints. We then euthanized the mice and semi-quantified the disseminated C4-2B cells in various mouse organs via human-specific genomic PCR at weeks 2/6/8. We serendipitously found PCa metastatic recurrence at a distant site in one mouse at 5 weeks post removal of the C4-2B PCa subcutaneous tumor, suggesting that the mice without observed recurrence may have dormant disseminated tumor cells (DTC). We further observed that tumor removal limited DTC in the bones, specifically, in the bone cortex. In clinic, metastatic PCa tumors were found in patient bone marrows and bone marrow DTC presence was associated with poor prognosis. Therefore, we hypothesized that cells in the bone cortex induce and maintain DTC dormancy. We first explored the dormancy-inducing effects of various types of cells in bone. We co-cultured C4-2B cells in direct contact with osteoblasts MC3T3-E1, macrophages Raw264.7, mesenchymal progenitors OP-9, fibroblasts NIH3T3, or endothelial cells HUVEC. The dormancy phenotypes such as proliferation and marker gene changes of the co-cultured C4-2B cells were compared to C4-2B cells cultured alone. We found that only MC3T3-E1 cells, but not other cells, inhibited C4-2B proliferation, increased the expression of dormancy markers such as NR2F1, and decreased the expressions of proliferation markers such as Ki67 and Cyclin D1, suggesting that osteoblasts induced C4-2B dormancy. Direct physical contacts were necessary as dormancy phenotypes were not observed in C4-2B cells treated with the conditioned media of MC3T3-E1 cells cultured alone, nor in C4-2B cells in transwell co-culture with MC3T3-E1 cells. We then performed RNA-Seq in C4-2B/MC3T3-E1 co-culture to profile the dormancy signature, which was loaded into a novel artificial intelligence (AI) platform to screen for drugs mimicking this signature. One candidate from the AI pipeline, PF-562271, a focal adhesion kinase (FAK) inhibitor, was validated with dormancy-mimicking effect for C4-2B cells in vitro. Further studies are focusing on the in vivo effects and the mechanisms of action of PF-562271. Our study profiled the C4-2B PCa cell disseminations post primary tumor removal. We defined a unique osteoblast-induced dormancy signature and revealed physical contact requirement in dormancy. We also recapitulated osteoblast-induced PCa cell dormancy in vitro using an FAK inhibitor, which could potentially inhibit PCa bone metastatic progression and recurrence clinically. Citation Format: Ruihua Liu, Shang Su, Ke Liu, Jing Xing, Mary Stangis, Diego P. Jacho, Eda D. Yildirim-Ayan, Cara M. Gatto-Weis, Haiquan Yu, Bin Chen, Xiaohong Li. Prostate cancer (PCa) cell dormancy in bone depends on physical contacts between PCa cells and osteoblasts and can be induced via FAK inhibition [abstract]. In: Proceedings of the AACR Special Conference: Advances in Prostate Cancer Research; 2023 Mar 15-18; Denver, Colorado. Philadelphia (PA): AACR; Cancer Res 2023;83(11 Suppl):Abstract nr B038.
Read full abstract