CD19 CAR-NK CONSTRUCTION WITH A NEW HINGE AND GITRL CO-EXPRESSION FOR TREG ACTIVITY INHIBITION

• Published in: Hematology, Transfusion and Cell Therapy Vol. 45; p. S548
• Main Authors: DF Schmidt, S Ebrahimabadi, MC Tirapelle, DT Covas, V Picanço-Castro
• Format: Journal Article
• Language: English
• Published: Elsevier 01-10-2023
• ISSN: 2531-1379
• DOI: 10.1016/j.htct.2023.09.1007

Introduction: Chimeric Antigen Receptor T (CAR-T) cell therapy has emerged as a promising approach in which autologous T cells are engineered to recognize a specific target. Although the clinical responses are encouraging, nearly 60% of large B cell lymphoma patients have disease progression after anti-CD19 CAR-T cell treatment, and neurotoxicity remains a challenge. NK cells are an attractive resource, given their innate capacity to eliminate tumor cells, and the prospect to use allogeneic sources, unlike T cells. This allows the development of well-characterized “off-the-shelf” CAR-NK cells, which also present a reduced risk of cytokine release syndrome and neurotoxicity. Furthermore, several studies indicate that Tregs play a central role in the tumor microenvironment of CD19+ cancers. Thus, targeting these immunosuppressive cells is a strategy to improve CAR-based therapies. In this study, we designed a novel anti-CD19 CAR (CAR19 Ig) featuring an Immunoglobulin (Ig)-like hinge, providing flexibility to access the target. We also constructed a fourth-generation CAR co-expressing GITRL (CAR19 Ig-GITRL) ‒ a modulator that blocks Tregs inhibitory activity. We hypothesize that this innovative approach could boost the antitumor response encompassing the use of “off-the-shelf” CAR-NK cells. Objectives: This study aims to evaluate the cytotoxic potential of CAR19 Ig in eliminating CD19+ cells, while also modifying this CAR into a CAR co-expressing GITRL, and assess its potential for Treg blockade. Methods: Lentiviral vectors with CAR19 Ig or CAR19 Ig-GITRL were constructed. NK-92 cells were transduced with lentivirus and the CAR-positive population was enriched by sorting. Then, the in vitro potential of CAR-NK92 cells against the RAJI (CD19+) cell line was evaluated by coculture at an Effector-to-Target (E:T) ratio of 2:1 for 5h. Cytotoxicity after coculture was evaluated by flow cytometry. Sorted cells were cryopreserved in a solution of 90% FBS and 10% DMSO. Results: NK-92 cell transduction resulted in 18% CAR expression in CAR19 IgGITRL-NK-92 and 13% of CAR in CAR19 Ig-NK-92 cells. At first, the sorting was performed with CAR19 Ig-NK-92, leading to over 90% CAR expression. We evaluated the CAR19 Ig-NK-92 cells in vitro cytotoxic potential and coculture of the cells demonstrated 22% cytotoxicity, while untransduced NK-92 cells did not eliminate target tumor cells in the same conditions. Additionally, we conducted an assessment of CAR-NK expression, potency, and proliferative capacity after freezing, which remained unaffected by the cryopreservation procedure. To carry on evaluating CAR19 Ig-GITRL-NK-92 cells, they were also sorted resulting in about 90% CAR expression. Conclusion/discussion: Usually, lentiviral transduction of NK cells is challenging, which leads to low transduction efficiency. Hence, our findings show a remarkable achievement, attaining a CAR-positive cell population above 90% via sorting. Besides, our study establishes the feasibility of freezing these cells while maintaining stable CAR expression post-thawing, a valuable advantage for “off-to-shelf” therapy. Moreover, our new CAR, featuring an IgG hinge, exhibited increased cytotoxicity against target cells, underscoring the effectiveness of the CAR construct to recognize CD19 and activate NK cells. Subsequently, the next phases of this project will evaluate the cytotoxic potential of CAR19 Ig-GITRL-NK-92, along with exploring its impact on Tregs.