Background: Targeted immunotherapies such as CAR-T cells are limited to several well-known lineage markers for which the on-target off-tumor toxicities are clinically tolerated. CART cells are or have been developed individually for different hematologic malignancies based on the relevant lineage antigens (e.g. CD19 for B cell, BCMA for myeloma, CD7 for T cells, CD33 for AML). By targeting a pan-hematologic antigen, a single "drug" could be used for all hematologic indications, thereby accelerating clinical research. CD45 is expressed on most hematologic malignancies and is, therefore, an attractive target for pan-hematologic cancer immunotherapy. However, targeting CD45 with CART cells is limited by (i) expression of CD45 on hematopoietic stem cells (HSC) and their progeny (ii) T cell (including CART cell) fratricide, and (iii) the central function of CD45 in immune cell signaling. To render CD45-specific immunotherapy a potentially viable pan-hematologic CART option we built upon prior work in which we had deleted the pan-myeloid antigen CD33 in HSC. In the current work, however, we sought to modify the relevant CD45 epitope in both T-cells and HSCs in such a way that it is unrecognized by the anti-CD45 CAR-T cells while maintaining its surface expression and intracellular phosphatase function (Figure 1A). Methods: After mapping the amino acid residues on the extracellular domain of human CD45 that are required for binding by three different anti-CD45 scFv's, we devised a CRISPR base editing (BE) approach to install a single mutation at the identified epitope. This allowed us to generate a CD45 molecule that is "invisible" to CAR-T cells made from these anti-CD45 scFv's, thus rendering them resistant to fratricide or suicide. Controls were CD45WT (non-edited) or CD45KO (cells in which CD45 expression was deleted by Cas9-based indel formation). We tested the efficacy of the various CART against leukemia specimens in vitro and in vivo. We then installed the anti-CD45 resistant mutation in primary human CD34+ HSC and assessed the functional consequences of CD45edited hematopoiesis. Results: T-cells transduced with anti-CD45 CARs (CART-45) show substantially impaired ex vivo expansion compared to CD19-directed CAR-T cells due to fratricide, as expected. However, CART-45 cells in which the targeted epitope was altered through adenine base editing (CART-45edited) were resistant to fratricide, enabling the expansion of CD45-directed CAR-T cells comparable to that of CD45KO or CAR19 control T cells. Unlike CD45KO CAR-T cells, CD45editedcells retained CD45 phosphatase expression which is crucial for CAR-T cell function in xenograft tumor models. In mice engrafted with a CD19+ B-ALL cell line, CD45edited CART-19 function was equivalent to CD45WT CART-19, and superior to CD45KO CART-19 (Mock vs CD45KO, p = <0.001); Mock vs CD45edited, p=0.97. These results indicate that CD45 is needed for the proper function of CART cells. We then tested CART-45 in a patient-derived xenograft (PDX) model of human AML and found that CART-45 expanded in the peripheral blood and effectively eliminated tumor cells with no detectable blasts in peripheral blood or bone marrow 4 weeks post CART-45 injection. To protect the hematopoietic system from CD45-directed on-target/off-tumor toxicity, we base-edited the targeted CD45 epitope in human HSCs, achieving the desired Tyr->Cys conversion in 62% of human HSCs (median, 51%-91% range). CD45edited HSC maintained equivalent colony formation in vitro when compared to CD45WT HSC. Crucially, CD45edited HSC transplanted into immunodeficient mice led to equivalent hematopoietic engraftment and differentiation when compared with CD45WT cells, whereas CD45KO hematopoiesis was lost over time (Figure 1B). Conclusions: Here we developed a novel platform that combines pan-hematologic anti-CD45 CAR-T cells with an engineered hematopoietic stem cell transplant endowed with selective resistance to CD45-specific immunotherapy while maintaining CD45 expression and function. The CD45 phosphatase domain is necessary and sufficient to maintain hematopoiesis and T cell function in humanized mice. This platform creates a cancer-specific antigen in all residual host hematopoietic cells, paving the way to profound myeloablation. This platform could be used to treat most hematopoietic malignancies and perhaps other diseases requiring profound hematopoietic ablation. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal
Read full abstract