Table S1 shows the affinity and kinetics of AO-176 and AO-104 to human and cynomolgus monkey CD47 determined by surface plasmon resonance

Figure S1 depicts the sequence and analytical characteristics of AO-176

Figure S2 shows that AO-176 does not bind to mouse CD47

Figure S4 demonstrates the minimal reduction in RBC and hemoglobin levels in cynomolgus monkeys treated with AO-176

Figure S5 demonstrates that AO-176 does not mediate antibody dependent cell cytotoxicity of tumor cells

Figure S3 shows the specificity of AO-176 for human CD47

Table S2 is a summary of AO-176 tumor cell killing in solid and hematologic tumor cell lines

Table S3 is a summary of AO-176 mediated phagocytosis of solid and hematologic tumor cells

Abstract Inhibitors of adaptive immune checkpoints have shown promise as cancer treatments. CD47 is an innate immune checkpoint receptor broadly expressed on normal tissues and overexpressed on many tumors. Binding of tumor CD47 to signal regulatory protein alpha (SIRPα) on macrophages and dendritic cells triggers a “don't eat me” signal that inhibits phagocytosis enabling escape of innate immune surveillance. Blocking CD47/SIRPα interaction promotes phagocytosis reducing tumor burden in numerous xenograft and syngeneic animal models. We have developed a next-generation humanized anti-CD47 antibody, AO-176, that not only blocks the CD47/SIRPα interaction to induce tumor cell phagocytosis, but also induces tumor cytotoxicity in hematologic and solid human tumor cell lines, but not normal noncancerous cells, by a cell autonomous mechanism (not ADCC). AO-176 also binds preferentially to tumor versus many normal cell types. In particular, AO-176 binds negligibly to RBCs in contrast to tumor cells, even at high concentrations up to 200 μg/mL and does not agglutinate RBCs up to 1 mg/mL in vitro. These properties are expected not only to decrease the antigen sink, but also to minimize on-target clinical adverse effects observed following treatment with other reported RBC-binding anti-CD47 antibodies. When tested in cynomolgus monkeys, AO-176 was well tolerated with no adverse effects. Finally, we show that AO-176 demonstrates dose-dependent antitumor activity in tumor xenograft models. Taken together, the unique properties and antitumor activity of our next-generation anti-CD47 antibody, AO-176, distinguishes it from other CD47/SIRPα axis targeting agents in clinical development.

CD47 is a cell surface glycoprotein that interacts with signal regulatory protein alpha (SIRPα) on macrophages and dendritic cells triggering a “don9t eat me” signal that inhibits phagocytosis. Many tumors evade immune surveillance by overexpressing CD47, thereby preventing their recognition by phagocytes. Blocking the interaction of SIRPα/CD47 promotes phagocytosis and tumor cell destruction leading to a reduction in tumor burden. We have developed a humanized anti-CD47 antibody, AO-176, that blocks the interaction between CD47 and SIRPα and exhibits several additional novel functional characteristics. These characteristics include the induction of cell death in multiple human tumor cell lines in a cell autonomous manner (not ADCC), assessed by an increase in phosphatidylserine/7AAD positive staining. A second novel characteristic is enhanced binding to tumor cells at acidic pH. AO-176 binds to human tumor cell lines in the high pM to low nM range at physiologic pH, however, binding is enhanced up to 20-fold at an acidic pH of 6.5. The acidic pH of the tumor microenvironment which ranges from 6.4-7.2 is characteristic of solid tumors and correlates with tumor progression and metastasis. As a result of this enhanced binding at acidic pH, AO-176 has the potential added advantage of tumor-specific targeting. A third novel characteristic exhibited by AO-176 is its selective binding to tumor cells while exhibiting reduced binding to normal cells including red blood cells (cynomolgus monkey and human), endothelial, epithelial and skeletal muscle cells. In addition to these novel characteristics, AO-176 also exhibits dose-dependent efficacy in multiple mouse tumor models. Taken together, the unique combination of functional characteristics of AO-176, including induction of cell-autonomous killing, enhanced binding to tumor cells at acidic pH, significantly reduced binding to normal cells and potent in vivo efficacy provides the preclinical rationale for further development. Citation Format: Robyn Puro, Katherine Liu, Benjamin Capoccia, Michael Donio, Ronald Hiebsch, Myriam Bouchlaka, Alun Carter, Pamela Manning, Kathleen Crowley, Robert Karr. A humanized anti-CD47 monoclonal antibody that directly kills human tumor cells and has additional unique functional characteristics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 1765.

Inhibitors of adaptive immune checkpoints have shown promise as cancer treatments. CD47 is an innate immune checkpoint receptor broadly expressed on normal tissues and over-expressed on several tumors. Binding of tumor CD47 to signal regulatory protein alpha (SIRPalpha) on macrophages and dendritic cells triggers a “don't eat me” signal that inhibits phagocytosis enabling escape of innate immune surveillance. Blocking CD47/SIRPα interaction promotes phagocytosis reducing tumor burden in numerous xenograft and syngeneic animal models.

TPS3109 Background: The cell surface protein CD47 is expressed or over-expressed on many tumor types. CD47 binds to signal regulatory protein alpha (SIRPα) on macrophages resulting in a “don’t eat me” signal that blocks host cell phagocytosis of the tumor cells, thus allowing them to escape removal by the innate immune system. Recent data indicate that anti-CD47 antibodies also contribute to an effective anti-tumor T cell response in immune-competent mice. Therefore, anti-CD47 antibodies are a new class of immune checkpoint inhibitors that modulate both the innate and adaptive immune systems. Ti-061 is a novel IgG4 humanized monoclonal antibody that specifically binds to CD47 with Kd values range from 100 – 500 pM. Ti-061 exhibits cross-species binding to cynomolgus monkey, mouse and rat CD47, enabling efficacy and toxicity testing across species. Ti-061 binds to CD47 on RBCs; however, it does not cause agglutination of RBCs in vitro from any of the species tested. Ti-061 exhibits anti-tumor activity in several in vivo mouse tumor models. This ongoing Phase 1-2 study will assess the safety, efficacy, pharmacokinetics (PK) and pharmacodynamics (PD) of Ti-061 alone and in combination with other anti-cancer agents in patients with advanced malignancies. Methods: Part A is an open-label, dose-escalation study of Ti-061 administered as a weekly 1-hour IV infusion at doses ranging from 1 to 20 mg/kg. Once the MTD/RP2D or “active dose” is determined, patients with specific solid tumors and high CD47 expression will be enrolled in 4 or more expansion cohorts. Up to 160 patients with histologically confirmed solid tumors, ECOG PS 0-1, adequate blood counts (Hb≥ 10 g/dL), organ function, and archival or fresh tumor tissue will be enrolled in Part A, and will be treated until disease progression, unacceptable toxicity, or withdrawal. Primary endpoint is safety, which will be assessed using NCI-CTCAE v4.03. Secondary endpoints include PK, PD, objective response rate (ORR) and progression-free survival (PFS), which will be assessed using RECIST v1.1. The results of this study will support further development of Ti-061 in combination with checkpoint inhibitors (Part B) and other anti-cancer agents.

Abstract Antibodies vs CD47 (CD47mAbs) that block the CD47-SIRPalpha interaction promote the phagocytosis of cancer cells and have efficacy in several tumor models. A select few CD47mAbs also directly kill cancer cells by lowering cellular cAMP levels leading to mitochondrial damage and cell death. These CD47mAbs are thus referred to as “dual-function” mAbs. Activation of protein kinase A prevents CD47mAb-mediated death suggesting that phosphorylation of one or more target proteins in the cancer cell can block the death mechanism. BNIP3, a member of the BH3-only family, is induced by hypoxia and oncogenes and is necessary for induction of cell death by dual-function CD47mAbs. BNIP3 can activate autophagy, a pro-survival function, and can also induce cell death by damaging mitochondria. The role of BNIP3 in cancer is controversial and context-dependent with some cancers over-expressing BNIP3 compared to low levels of expression in normal tissue, while other cancers cannot tolerate BNIP3 expression and silence the gene, often by methylation of the BNIP3 promoter. This suggests that cancers that tolerate BNIP3 expression employ an as yet unknown mechanism to protect themselves from its toxic effects. The C-terminal transmembrane (TM) domain of BNIP3 (residues 164-184) penetrates the outer mitochondrial membrane allowing the extreme C-terminal ten residue tail of BNIP3 (RRLTTSTSTF, residues 185-194) to extend across the intermembrane space to bind OPA1 on the inner mitochondrial membrane, a key step in the death mechanism. We used a phosphosite-specific antibody to detect phosphorylation of BNIP3 immunoprecipitated from Jurkat leukemia cells at residue T188 which resides in a canonical protein kinase A site (RRLT, amino acids 185-188). Using mass spectrometry of 6His-tagged BNIP3 isolated from HEK293 cells treated with 8BrcAMP, we identified as many as 4 additional phosphorylated sites in the C-terminal tail sequence (residues 189-194, TSTSTF). We generated phosphomimetic (S/T to D) and unphosphorylated (S/T to A or N) mutations at these residues and expressed the mutant BNIP3 proteins in 293 cells. All of the BNIP3 mutants associated with mitochondria but only the phosphomimetic mutants prevented BNIP3-induced mitochondrial damage and cell death. In contrast, mutation of the phosphorylated S/T residues to unphosphorylated residues resulted in rapid and extensive cell death. Importantly, phosphomimetic C-terminal BNIP3 residues blocked cell death without preventing autophagy, providing evidence that the two roles of BNIP3 can be regulated independently. We replicated these results in one lung cancer and three breast cancer cell lines. These findings suggest that phosphorylation at the C-terminus of BNIP3 is a switch that determines the pro-survival vs pro-death effects of BNIP3. Dual-function CD47mAbs may act by dephosphorylating BNIP3 expressed in tumor cells thus unleashing its killing potential. Citation Format: William Frazier, Katherine Liu, Julie Dimitry, Benjamin Capoccia, Pamela Manning, Robert Karr. Phosphorylation of BNIP3 is a switch between life and death in cancer cells. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 2. doi:10.1158/1538-7445.AM2015-2

Recent success in immunomodulation of cancer has targeted immune checkpoints such as CTLA-4, PD-1 and PDL-1 to enhance adaptive immunity by stimulating production of tumor-selective, cytotoxic T cells. Anti-CD47mAbs enhance innate immunity by increasing the phagocytosis of tumor cells by macrophages leading to processing and presentation of tumor antigens to prime the adaptive T cell response. Many cancers, including hematologic cancers, up-regulate the expression of CD47 presumably to avoid immune destruction. Increased CD47 expression protects cancer cells from phagocytosis by sending a “don9t eat me” signal to macrophages via SIRPalpha, an inhibitory receptor that prevents phagocytosis of CD47-bearing cells. CD47mAbs that block the CD47/SIRPalpha interaction (“blocking-only” mAbs) enhance phagocytosis of cancer cells in vitro. We have identified two CD47mAbs, Vx-1000 and Vx-1004, both of which block the CD47/SIRPalpha interaction and promote phagocytosis of tumor cells by macrophages equally well. However, Vx-1004 also has the unique property of killing cancer cells, but not normal blood cells, via a direct, cell-autonomous, cytotoxic mechanism. Therefore, Vx-1004 is a dual-function antibody. Vx-1004 selectively kills a variety of hematologic cancer cells in vitro, while Vx-1000, the blocking-only mAb, does not as assessed by annexin V staining and flow cytometry (Figure 1). In dose-response studies, cell death in leukemia cells was induced in 2 hrs by To determine if the tumor-toxic activity of Vx-1004 confers enhanced efficacy in vivo compared to Vx-1000, we compared them in two mouse hematologic cancer models: murine acute promyelocytic leukemia (APL) and B cell lymphoma (BCL). Briefly, 1x106 GFP-labeled C57BL/6 APL cells were injected IV into wild-type C57BL/6 mice that were then treated IP with 0.4 mg/kg of either Vx-1000 or Vx-1004 on the day of tumor injection and on days 3 and 6 following tumor injection, a very low dose and limited dosing regimen. On day 25, the blood of these mice was analyzed for the number of circulating APL cells. As shown in Figure 2, Vx-1000 did not significantly reduce tumor burden compared to the control group. In contrast, Vx-1004 significantly reduced tumor burden compared to controls, demonstrating greater efficacy of the dual-function CD47mAb. In addition, enhanced efficacy of Vx-1004 compared to Vx-1000 was demonstrated in BCL (Figure 3). In this model, NSG mice were injected with 1x106 murine A20 lymphoma cells subcutaneously and then treated with 0.4mg/kg/day of the CD47mAbs IP for the first five days following tumor injection. In this model, Vx-1000 also failed to inhibit tumor growth compared to controls while Vx-1004 significantly reduced tumor burden at 35 days compared to both the control and Vx-1000 groups, nearly four weeks after treatment was stopped. These data demonstrate increased anti-cancer efficacy with a dual-function CD47mAb that not only blocks the CD47/SIRPalpha interaction to increase phagocytosis of cancer cells, but also selectively kills cancer cells. These studies indicate that dual-function CD47mAbs may have better anti-tumor activity in vivo and support their use in human clinical trials. Disclosures Manning:Corvus Pharmaceutical: Employment, Equity Ownership. Capoccia:Corvus Pharmaceutical: Employment, Equity Ownership. Hiebsch:Corvus Pharmaceutical: Employment, Equity Ownership. Karr:Corvus Pharmaceutical: Employment, Equity Ownership. Frazier:Corvus Pharmaceutical: Consultancy, Equity Ownership.