Abstract 5011: Cancer immunotherapy targeting CD47: Wild type SIRPαFc is the ideal CD47-blocking agent to minimize unwanted erythrocyte binding

Abstract

Abstract Background: CD47 binds to SIRPα on the surface of macrophages and delivers a “do not eat” signal that suppresses phagocytosis. There is strong evidence that many hematopoietic and solid tumors exploit the CD47-SIRPα pathway to escape macrophage-mediated destruction, and blocking CD47 has emerged as a promising cancer immunotherapy. There are several approaches available to achieve CD47 blockade, including wild type SIRPαFc fusion proteins, engineered high affinity SIRPαFcs and monoclonal antibodies. One concern with CD47-based therapies is the expression of the target on the surface of red blood cells (RBCs), which has the potential to act as a large antigen sink and cause hematological toxicity. Indeed, anemia has been reported in animals treated with high affinity SIRPαFcs variants and CD47-specific antibodies. Methods: Various CD47-specific biologics, including SIRPαFc fusion proteins containing wild type or mutant SIRPα sequences and CD47-specific antibodies were evaluated for binding to erythrocytes and other cell types by flow cytometry. Results: SIRPαFc fusion proteins containing wild type SIRPα sequences, which bind to CD47 with nanomolar affinity, bound very poorly to human RBCs compared to both blocking and non-blocking CD47 antibodies. This striking difference was highly reproducible and occurred across different blood types, but was not seen with other cells (e.g., tumor cell lines), indicating an erythrocyte-specific phenomenon. The presence of wild type SIRPα sequences was critical to achieving a low RBC binding profile, as mutated SIRPαFc with enhanced CD47 affinity bound strongly to human erythrocytes. Interestingly, wild type SIRPαFc binding to human RBCs was dramatically increased when CD47 was first clustered with a non-blocking antibody, suggesting that wild type SIRPαFc normally lacks the ability to aggregate CD47 on the erythrocyte surface. Finally, the lack of appreciable SIRPαFc binding to erythrocytes is unique to humans, as non-mutated mouse SIRPαFc could bind mouse RBCs and wild type human SIRPαFc cross reacts with CD47 on the surface of non-human primate and porcine RBCs. Conclusions: Wild type SIRPαFc, unlike other CD47 blocking agents, exhibits very low binding to human erythrocytes. This predicts that wild type SIRPαFc will have superior pharmacokinetic properties and less toxicity in cancer patients. Furthermore, the strong binding of wild type SIRPαFc to monkey RBCs suggests that preclinical studies in non-human primates may overestimate the risk of hematological toxicity in humans. Citation Format: Robert A. Uger, Karen Dodge, Xinli Pang, Penka S. Petrova. Cancer immunotherapy targeting CD47: Wild type SIRPαFc is the ideal CD47-blocking agent to minimize unwanted erythrocyte binding. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 5011. doi:10.1158/1538-7445.AM2014-5011