Adapter CAR T cells to counteract T-cell exhaustion and enable flexible targeting in AML

Although the landscape for treating acute myeloid leukemia (AML) patients has changed substantially in recent years, the majority of patients will eventually relapse and succumb to their disease. Allogeneic stem cell transplantation provides the best anti-AML treatment strategy, but is only suitable in a minority of patients. In contrast to B-cell neoplasias, chimeric antigen receptor (CAR) T-cell therapy in AML has encountered challenges in target antigen heterogeneity, safety, and T-cell dysfunction. We established a Fab-based adapter CAR (AdCAR T-cell; 4dCAR T-cell) platform with flexibility of targeting and control of AdCAR T-cell activation. Utilizing AML cell lines and a long-term culture assay for primary AML cells, we demonstrate AML-specific cytotoxicity using anti-CD33, anti-CD123, and anti-CLL1 adapter molecules in vitro and in vivo. Notably, we show for the first time the feasibility of sequential application of adapter molecules of different specificity in primary AML co-cultures. Importantly, using the AML platform, we demonstrate that chronic T-cell stimulation and antigen heterogeneity are well-known causes of resistance and that the AdCAR platform might offer effective strategies to ameliorate these via mitigating death receptor signaling and tonic CAR T-cell signaling. Building on the previously described Adapter CAR-T cell platform we developed a fragment antigen-binding (Fab)-based approach that enables flexibility of targeting and control of AdCAR T-cell activation. Using an AdCAR directed against a biotinylated adapter molecule in the context of a specific linker, we show AdCARs are highly functional against multiple AML cell lines in vitro and in vivo, and against primary AML (pAML) cells, using anti-CD33, anti-CD123, and anti-CLL1 adapter molecules (AMs). Because many AML-associated target antigens are internalized, we studied receptor-mediated endocytosis of various AM formats and its impact on half-life and cytotoxicity. We provide the first evidence that serial use of AMs against different target antigens is feasible and highly potent in eliminating pAML cells in ex vivo long-term co-cultures.

D. Nixdorf, M. Sponheimer, D. Berghammer, F. Engert, U. Bader, N. Philipp, M. Kazerani, T. Straub, L. Rohrbacher, L. Wange, S. Dapa, D. Atar, C. M. Seitz, K. Brandstetter, A. Linder, M. von Bergvelt, H. Leonhardt, J. Mittelstädt, A. Kaiser, V. Bücklein, M. Sukbwele