Titrating CD47 by mismatch CRISPRi reveals incomplete repression can eliminate IgG-opsonized tumors but CD47 heterogeneity limits induction of anti-tumor IgG

Phagocytic elimination of solid tumors is an attractive mechanism for immunotherapy - particularly because of the possible induction of anti-cancer immunity. The phagocytic potential of macrophages is limited, however, by the CD47-SIRP checkpoint, and how much CD47 disruption is needed for efficacy remains unclear, even when tumors are opsonized by a pro-phagocytic antibody. Here, CRISPR-interference (CRISPRi) is applied with a large set of sgRNAs to produce a broad range of CD47 knockdowns in B16F10 melanoma, which is generally found to be resistant to the heavily studied PD-1 blockade. Guided by 3D immuno-tumoroid results, we identify a critical CD47 density below which macrophage-mediated phagocytosis dominates proliferation in the presence of an otherwise ineffective pro-phagocytic antibody (anti-Tyrp1). Growing tumors and immuno-tumoroids generally show selection for CD47-positive cells, but some mice reject tumors having >97% mean repression of CD47 or even having 80% repression - unless mixed with 50% repressed cells. Interestingly, long-term survivors have de novo pro-phagocytic IgG antibodies that increase in titer with depth of repression and also with early accumulation of tumor macrophages. Given well-known limitations of antibody permeation into solid tumors, our studies set a benchmark for anti-CD47 blockade and suggest deep disruption favors acquired immunity.