Overcoming natural killer cell dysfunction in the tumor microenvironment of advanced prostate cancer

Abstract

Natural killer (NK) cells are associated with good prognosis in patients with metastatic castration-resistant prostate cancer (mCRPC). However, NK cell function is hindered in the immunosuppressive and hypoxic tumor microenvironment (TME) of mCRPC. To improve NK cell-mediated responses against mCRPC, I developed a novel Tri-specific Killer Engager (TriKE) that engages with the NK cell activating receptor CD16, binds to prostate-specific membrane antigen (PSMA) that is highly and specifically expressed on mCRPC, and has an interleukin (IL)-15 moiety that is essential for NK cell survival, proliferation, and priming. My central hypothesis is that hypoxia and MDSC-induced suppression of NK cell in the TME of mCRPC can be overcome by CD16+IL-15 co-stimulation with PSMA TriKE treatment.PSMA TriKE was produced using the mammalian expression system and tested using healthy donor and prostate cancer patient-derived NK cells. Flow cytometry-based functional and dye dilution proliferation assays were used to compare activation and proliferation of healthy donor and prostate cancer patient-derived NK cells treated with PSMA TriKE or IL-15. NK cell cytolytic capacity against PSMA+ and PSMA- prostate cancer lines and patient-derived xenograft (PDX) were examined using live cell imaging assays. In various assays, hypoxic (1% oxygen) culture condition and cytokine-induced myeloid-derived suppressor cells (MDSC) were incorporated to better examine PSMA TriKE function in the physiological setting of mCRPC. In vivo testing of the PSMA TriKE was performed using C4-2 xenograft model and expanded NK cells in NSG mice. The PSMA TriKE significantly enhanced expansion of peripheral blood NK cells derived from healthy donors and prostate cancer patients. Additionally, the PSMA TriKE induced significantly higher NK cell degranulation and intracellular interferon (IFN)-γ and tumor necrosis factor (TNF)-α buildup, when compared to IL-15 treatment, after incubation with C4-2 cells. Responses of NK cells derived from prostate cancer patients were equivalent to healthy controls. Although NK cell activation was not observed against PSMA knockout (KO) C4-2, indicating specificity of PSMA TriKE treatment, bystander killing of PSMA-KO C4-2 cells was achieved with PSMA TriKE treatment when wildtype C4-2 cells were cocultured. This suggests a potential PSMA TriKE benefit in controlling tumor antigen escape, through natural cytotoxicity once primed. In settings that mimic the TME of mCRPC, NK cells treated with the PSMA TriKE during prolonged exposure to hypoxia showed retention of cytotoxicity against C4-2 cells while IL-15 treated NK cells showed greatly impaired cytotoxicity. Similarly, MDSC suppressed NK cell cytotoxicity in the presence of IL-15, but PSMA TriKE treatment abrogated MDSC-induced suppression. Finally, PSMA TriKE enhanced tumor control and improved survival of mice as compared to IL-15 and no treatment groups in vivo. To summarize, PSMA TriKE demonstrates potential in overcoming suppression on NK cells in the TME of mCRPC and is a promising candidate for advanced prostate cancer therapy.