Th127 - Engineered Single Amino Acid Substitutions Protect Therapeutic Cells From CD123 Targeted Immunotherapy in vitro

Therapeutic depletion of diseased cells using monoclonal antibodies, T cell engagers or chimeric antigen receptor (CAR) cells is very effective. Whereas B cell targeted immunotherapies are highly specific, the co-expression of myeloid markers such as CD33 or CD123 on healthy hematopoietic stem and progenitor cells (HSPCs) bears the risk for unintentional myelotoxicity. In order to enable therapeutic targeting of such proteins it was proposed to transplant engineered HSCs in which the target was removed. However, since removing the protein abolishes its function, this approach is limited to redundant proteins and bears a certain risk for antigen-negative tumor relapse. Here, we show feasibility for shielding the therapeutic cells from targeted therapy while preserving the function of the targeted receptor. Structure-guided computer modeling identified 28 single amino acid (aa) substitutions in CD123 designed to interfere with CSL362 (talacotuzumab) binding but preserve CD123 function. Using flow cytometry of cells expressing the variants we identified variants for which the single aa substitution resulted in complete non-binding while CD123 remained normally expressed. Cells expressing non-binding variants were shielded from antibody dependent cellular cytotoxicity (ADCC), T cell engager or CAR T-mediated killing. Protein characterization of select variants demonstrated normal biophysical properties of most variants. Further functional investigation is ongoing for variants with normal binding of Interleukin-3. Thus, our data demonstrates the feasibility to engineer proteins harboring single aa substitutions that shield therapeutic cells from all tested cell depleting modalities while preserving their function. We propose that this concept could be broadly applicable to targeted immunotherapies.