Discovery of JNJ-89853413, a First-in-Class CD33xVδ2 T-Cell Engager for the Treatment of Acute Myeloid Leukemia

Targeted immunotherapy in acute myeloid leukemia (AML) remains a substantial clinical challenge due to the heterogenous nature of the AML cancer cells and the lack of AML-specific antigens. While several immunotherapies targeting CD33, including CAR-T and NK cell therapies, are being evaluated clinically, challenges have been experienced with immunotherapies targeting CD33 due to lack of tolerability. In addition, on-target off-tumor toxicities due to target expression on normal myeloid cells represent a major concern. Vδ2 T-cells represent a promising subset of T-cells to explore for novel and innovative cancer immunotherapy. They represent ~5% of the T-cell population in the peripheral blood of healthy donors and are prevalent in a broad set of cancers. Vδ2 T-cells are less affected by inhibition by immune check points on cancer cells and do not activate inhibitory Tregs, which may limit activity of CD3 bispecific T-cell engagers. In addition, differential expression of phosphoantigen-activated butyrophilins contributes to greater activity of Vδ2 T-cells against cancer cells over normal cells. Targeting CD33 (a well validated AML target that is expressed on ~90 % AML cells) with Vδ2 T-cell engagement could potentially provide advantages with regards to safety and efficacy compared to the existing therapies. We developed JNJ-89853413, a novel first-in-class human bispecific antibody comprising a CD33 arm paired with a high-affinity Vδ2 binder. JNJ-89853413 induced potent and selective T-cell mediated cytotoxicity on CD33⁺ cancer cells with specific activation of Vδ2 T-cells. Furthermore, JNJ-89853413 elicited dose-dependent cytotoxicity of primary AML blasts. In the presence of healthy donor PBMCs, JNJ-89853413 showed preferential cytotoxicity of co-cultured cancer cells with no cytotoxicity of healthy CD33-expressing monocytes or natural killer (NK) cells. JNJ-89853413 showed no impact on the colony forming potential of CD34⁺ hematopoietic stem and progenitors cells obtained from healthy donors in colony forming unit assays. In addition, JNJ-89853413 inhibited tumor growth of disseminated AML in in vivo regression xenograft models with expanded enriched Vδ2 T cells as effectors. Taken together, JNJ-89853413 represents a novel potential best- and first-in-class opportunity for AML immunotherapy that is progressing to clinical trials with a potential for cure and improved tolerability compared to existing treatments.