Natural Killer (NK) cells are a promising source for Chimeric Antigen Receptor (CAR) cell-based therapiesdue to their MHC independency for activation, intrinsic ability against tumor cells, rapid cytokine production, and low potential to cause graft-versus-host-disease. Conventional CARs only recognize cell surface antigens, sparing numerous intracellular tumor-associated antigens. MAGE-A4 intracellular antigen is highly expressed in different cancers and has low expression in normal tissues. Following its expression within cells, MAGE-A4 is presented on the cell surface in association with HLA-A*02, conferring a cancer-specific antigen profile. This project uses a TCR-like CAR, and an innovative human GITR signaling domain, developed by our collaborator, Prof. Shiku Hiroshi at Mie University. This CAR construct demonstrates the ability to recognize the intracellular tumor-associated antigen MAGE-A4 when presented by HLA-A*02 molecules. This anti-MAGE-A4 CAR is presently undergoing evaluation in a multi-institutional phase 1 study.

The primary aim of this study is to create an off-the-shelf CAR-NK cell product, using the anti-MAGE-A4 CAR, designed to target the HLA-A*02 complex along with a specific epitope of the MAGE-A4 antigen and to and to evaluate the therapeutic efficiency of these cells in the treatment of MAGE-A4+ solid tumors including melanoma.

LentiX-HEK293T cells were transfected with MAGE-A4 CAR DNA and helper plasmids to produce LV viral particles. K562 cell line were used for viral titration. NK-92 cell line were transduced using lentivirus particles with MOIs 10, 5, and 3. After 48 hours, the cells were evaluated by flow cytometry for CAR expression. Subsequently, anti-MAGE-A4 CAR NK-92 positive cells were enriched using magnetic beads. Meanwhile, target cell lines A375-Luc (MAGE-A4+), and HCT116-Luc (MAGE-A4-) were evaluated for MAGE-A4 expression by flow cytometry. To measure the specificity and cytotoxicity of anti-MAGE-A4 CAR NK-92, these cells were co-cultured with target cell lines expressing luciferase at different effector to target ratios. Cytotoxicity-mediated killing was assessed using bioluminescence-imaging systems (IVIS), quantifying the decrease of luminescence from baseline.

We conducted NK cell transduction using MOIs of 10, 5, and 3. Our observations revealed transduction efficiencies of 52%, 49.1%, and 48.7% respectively. Next, CAR positive cells were selected and the percentage of anti-MAGE-A4 CAR-NK-92 cells increased to 99.4% and these cells were used for co-culture with target cell lines. The results of MAGE-A4 expression in A375 cell line was 68.6%, and 2.54% in HCT116. The cytotoxicity of anti-MAGE-A4 CAR-NK-92 effector cells against A375 target cell line is around 80% in the ratio of 1:1 and 5:1. MAGE-A4-CAR-NK cells demonstrated enhanced effectiveness in killing A375 cells when compared to NK92-WT cells (p < 0.05), underscoring the specificity of the CAR vector.

Previous Studies done by Dr. Shiku team have already demonstrated the effectiveness of TCR-like CAR-T cells against MAGE-A4+ tumors. Our study is the first to demonstrate the effectiveness of this anti-MAGE-A4 CAR incorporating GITR as a costimulatory domain in NK cells.

This study has the potential to create a new off-the-shelf advanced cell therapy approach against MAGE-A4+ malignancies.