Lewis Y antigen as a novel target for CAR (chimeric antigen receptor) T-cell therapy in patients with neuroendocrine prostate cancer.

Abstract

217 Background: Neuroendocrine prostate cancers (NEPC) currently have no effective therapies. While CAR (Chimeric Antigen Receptor) T cell therapies targeting PSMA and PSCA receptors are currently in phase 1/2 studies in metastatic castrate-resistant prostate cancer (mCRPC), antigen targets for NEPC remain limited. Lewis Y antigen (LeY) is an oncofetal antigen expressed only in adult tumour cells and is considered a target for CAR T cell therapy in haematological, lung, ovarian and colon cancer. We previously found that LeY is a promising target for CAR T cell therapy in prostate cancer and that preconditioning LeY-directed CAR T cells with carboplatin further enhances tumour responses in mCRPC. We investigated if LeY can be used as a novel CAR T cell target for NEPC. Methods: The expression of the Lewis Y antigen was assessed in our patient-derived prostate cancer xenograft collection using immunohistochemistry (IHC). To assess the efficacy of LeY-CAR T cells, we used specimens established from a patient (Patient 508) with NEPC who had incurable metastatic disease and failed standard treatments. We generated CAR T cells from his peripheral blood mononuclear cells (PBMCs) to test their specificity and efficacy in vitro against established cell lines. Results: In our MURAL PDX collection of 59 prostate tumours, 81% (9/11) NEPCs expressed LeY antigen on IHC. Next, we successfully produced anti-LeY CAR T cells from the patient’s PBMCs using retroviral transduction. Flow cytometry verified efficient transduction of the CAR constructs. Previous studies found that a 1:1 ratio of CD4 + :CD8 + T cells improved outcomes following CAR T cell therapy 5,6 , so we further phenotyped patient T cells by flow cytometry. The CD4 + :CD8 + T cell ratio was nearly 1:1 (51.7% and 41.4% of total T cells, respectively). Next, we used an immunobead assay to investigate CAR-T specificity in releasing proinflammatory cytokines. Significant cytokine production was observed after a 4-hour co-culture with LeY+ human ovarian cancer cell line, OVCAR3, vs co-culture with control LeY- MDA-MB435 cells (p=0.0004), demonstrating the specificity of the CAR T cells.Lastly, we showed that patient CAR T cells effectively killed LeY+ cells (OVCAR3), but not control LeY- cells (MDA-MB435; p<0.0001). Tumour cell death was measured by the amount of released chromium following 16-hour co-culture. CAR T cells killed LeY+ target cells more effectively than their non-transduced equivalents at a range of effector-to-target ratios, with up to 15-fold increase in cell death. Conclusions: Our preclinical studies show that most NEPC tumors express the LeY antigen and that patient-derived CAR T cells specifically kill those cells. These findings identified a potential CAR T treatment target for NEPC for a phase 1/2 study.