Co-Transduced CD19/BCMA Dual-Targeting CAR-T Cells for the Treatment of Non-Hodgkin Lymphoma

CAR-T cells targeting CD19 have accomplished impressive responses in relapsed/refractory non-hodgkin lymphoma (NHL) patients, but disease recurrence is a concerning hurdle to overcome. This complication is mainly due to a selective pressure on tumor cells expressing CD19, which leads to loss of expression or downregulation to levels that cannot be detected by CAR-T cells. Simultaneous targeting of more than one antigen has been proposed as a compelling strategy to prevent tumor escape. CD19 is expressed along B-cell differentiation, and BCMA is expressed not only on plasmatic cells but also on mature B-cells. Thus, most NHL subtypes express both CD19 and BCMA. Here, we hypothesized that dual CD19 and BCMA targeting could reduce tumor antigen escape and enhance the antitumor efficacy of CAR-T cells.

CAR constructs targeting CD19 (ARI0001) or BCMA (ARI0002h) were engineered in our institution and both have demonstrated safety and efficacy for treating patients with B-cell lymphoid neoplasms and multiple myeloma, respectively. Both CARs contain the 41BB co-stimulatory domain in tandem with the CD3z. We tested the antitumoral efficacy of three different dual-targeting CD19/BCMA strategies: 1) a bicistronic vector encoding both CAR constructs separated by a T2A cleavage peptide (ARI0001-T2A-ARI0002h); 2) co-transduced T cells with ARI0001 and ARI0002h lentiviral vectors to obtain a product with three CAR populations comprising both single-CARs and a double positive population; 3) pooled ARI0001 and ARI0002.

Conventional functional assays such as cytokine secretion and specific lysis after short-term co-culture of CAR-T cells with NHL cell lines co-expressing CD19 and BCMA at constitutive levels revealed that all strategies were active but, did not discriminate between single- and dual-CARs groups. In order to mimic a potential relapse due to CD19 downregulation, we developed a CD19low density cell line of Burkitt lymphoma through its expression under a weak promoter. After 7-days of co-culture of CAR-T cells with RAMOS-CD19Low, only the co-transduction product maintained tumor elimination. Next, we analyzed the capacity of single versus dual CAR-T cells to kill malignant B cells and release cytokines in co-culture with NHL Patient Derived Lymphoma Spheroids. We tested samples from patients with 1) follicular lymphoma, 2) diffuse large B cell lymphoma (DLBCL), and 3) DLBCL patients who relapsed or had refractory disease post-CD19 CAR-T cell treatment. Under these more challenging conditions, differences among effector groups were maximized. ARI0001 and bicistronic CAR-T cells displayed similar effector functions in terms of killing capacity and cytokine production, especially when CD19 was expressed at high densities. ARI0002h presented antitumor efficacy but not as high as ARI0001. Co-transduction and pool products showed increased cytokine production and killing, being capable of B-cell elimination in all cases. A high disease burden of Burkitt lymphoma in vivo model revealed that co-transduction product significantly increased mice survival compared to ARI0001. This increase in survival was even more remarkable when the model was CD19low. No significant differences were observed between the co-transduction and pool groups.

In summary, we propose a relatively simple preparation of a dual-CAR-T cells targeting CD19 and BCMA, which is particularly effective against NHL cell lines expressing low CD19 densities. Co-transduction is advantageous with respect to the pool since it only requires one manufacturing process, decreasing the cost of the therapy and increasing patients' access to these novel therapies. Our studies provide a strong rational to test CD19/BCMA dual-CAR-T cells in clinical trials for the treatment of NHL patients.

Juan:Gyala Therapeutics S.L.: Research Funding. Urbano-Ispizua:Gilead: Consultancy; Celgene: Consultancy; Miltenyi: Consultancy. Guedan:Novartis: Patents & Royalties; Tmunity Therapeutics: Patents & Royalties; Gilead: Research Funding; Spark Therapeutics: Research Funding.