TTI-621 (Sirpafc): a CD47-Blocking Innate Immune Checkpoint Inhibitor with Broad Antitumor Activity and Minimal Erythrocyte Binding Penka S

Published OnlineFirst November 17, 2016; DOI: 10.1158/1078-0432.CCR-16-1700

Cancer Therapy: Preclinical Clinical Cancer Research TTI-621 (SIRPaFc): A CD47-Blocking Innate Immune Checkpoint Inhibitor with Broad Antitumor Activity and Minimal Erythrocyte Binding Penka S. Petrova1, Natasja Nielsen Viller1, Mark Wong1, Xinli Pang1, Gloria H.Y. Lin1, Karen Dodge1,Vien Chai1, Hui Chen1,Vivian Lee1,Violetta House1, Noel T.Vigo1, Debbie Jin1, Tapfuma Mutukura1, Marilyse Charbonneau1, Tran Truong1, Stephane Viau1, Lisa D. Johnson1, Emma Linderoth1, Eric L. Sievers1, Saman Maleki Vareki2,3, Rene Figueredo2,3, Macarena Pampillo2, James Koropatnick2,3, Suzanne Trudel4, Nathan Mbong4, Liqing Jin4, Jean C.Y. Wang4,5, and Robert A. Uger1

Abstract

Purpose: The ubiquitously expressed transmembrane glyco- Results: TTI-621 enhanced macrophage-mediated phagocy- protein CD47 delivers an anti-phagocytic (do not eat) signal by tosis of both hematologic and solid tumor cells, while sparing binding signal-regulatory protein a (SIRPa) on macrophages. normal cells. In vivo, TTI-621 effectively controlled the growth CD47 is overexpressed in cancer cells and its expression is asso- of aggressive AML and B lymphoma xenografts and was ciated with poor clinical outcomes. TTI-621 (SIRPaFc) is a fully efficacious in a syngeneic B lymphoma model. The IgG1 Fc human recombinant fusion protein that blocks the CD47–SIRPa tail of TTI-621 plays a critical role in its antitumor activity, axis by binding to human CD47 and enhancing phagocytosis of presumably by engaging activating Fcg receptors on macro- malignant cells. Blockade of this inhibitory axis using TTI-621 has phages. Finally, TTI-621 exhibits minimal binding to human emerged as a promising therapeutic strategy to promote tumor erythrocytes, thereby differentiating it from CD47 blocking cell eradication. antibodies. Experimental Design: The ability of TTI-621 to promote mac- Conclusions: These data indicate that TTI-621 is active rophage-mediated phagocytosis of human tumor cells was across a broad range of human tumors. These results further assessed using both confocal microscopy and flow cytometry. In establish CD47 as a critical regulator of innate immune sur- vivo antitumor efficacy was evaluated in xenograft and syngeneic veillance and form the basis for clinical development of TTI- models and the role of the Fc region in antitumor activity was 621 in multiple oncology indications. Clin Cancer Res; 23(4); evaluated using SIRPaFc constructs with different Fc tails. 1068–79. 2016 AACR.

Introduction regulatory protein alpha (SIRPa) on the surface of macrophages. Engagement by CD47 triggers tyrosine phosphorylation of the The phagocytic activity of macrophages is regulated by both cytoplasmic tail of SIRPa, leading to recruitment of the Src activating ("eat") and inhibitory ("do not eat") signals. CD47, a homology-2 domain containing protein tyrosine phosphatases widely expressed transmembrane glycoprotein, serves as a critical SHP-1 and SHP-2 and prevention of myosin-IIA accumulation at inhibitory signal, suppressing phagocytosis by binding to signal- the phagocytic synapse (1). CD47 is believed to regulate the natural clearance of senescent erythrocytes and platelets by splenic – 1Trillium Therapeutics Inc., Mississauga, Ontario, Canada. 2London Regional macrophages (2, 3). In addition, the CD47 SIRPa interaction Cancer Program, London Health Sciences Centre, Lawson Heath Research may represent an important mechanism by which malignant cells Institute, London, Ontario, Canada. 3Department of Oncology, University of escape immune-mediated clearance. Western Ontario, London, Ontario, Canada. 4Princess Margaret Cancer Center, CD47 has been shown to be overexpressed in numerous 5 University Health Network (UHN), Toronto, Ontario, Canada. Division of Medical hematologic malignancies, including acute myeloid leukemia Oncology and Hematology, Department of Medicine, UHN, and Department of (AML), acute lymphoblastic leukemia (ALL), chronic lymphocyt- Medicine, University of Toronto, Toronto, Ontario, Canada. ic leukemia (CLL), multiple myeloma, myelodysplastic syndrome Note: Supplementary data for this article are available at Clinical Cancer (MDS), and in multiple types of non-Hodgkin lymphoma (NHL), Research Online (http://clincancerres.aacrjournals.org/). including diffuse large B-cell lymphoma (DLBCL), mantle cell Corresponding Author: Robert A. Uger, Trillium Therapeutics Inc., Mississauga, lymphoma, and marginal cell lymphoma (4–10). Similarly, ele- Ontario, Canada. Phone: 416-595-0627; Fax: 416-595-5835; E-mail: vated CD47 expression has been demonstrated on solid tumors, [email protected] including bladder, brain, breast, colon, esophageal, gastric, kid- doi: 10.1158/1078-0432.CCR-16-1700 ney, leiomyosarcoma, liver, lung, melanoma, ovarian, pancreatic, 2016 American Association for Cancer Research. and prostate tumors (11–15). CD47 has been found to be an

1068 Clin Cancer Res; 23(4) February 15, 2017

TTI-621 Is a Novel Antitumor Immune Checkpoint Inhibitor

circulating platelets and leukocytes, TTI-621 shows only minimal Translational Relevance binding to human erythrocytes, thereby mitigating concerns of SIRPaFc (TTI-621) is a novel innate immune checkpoint anemia associated with anti-CD47 mAbs. These results further inhibitor designed to: (i) bind human CD47 on tumor cells establish CD47 as a critical regulator of innate immune surveil- and prevent it from delivering inhibitory signals to macro- lance and form the basis for clinical development of TTI-621 in phages and (ii) engage Fc gamma receptors (FcgR) expressed multiple oncology indications. by macrophages to further enhance phagocytosis. Here we show that TTI-621 avidly binds to CD47 on a wide range of Materials and Methods human tumor cells and selectively promotes macrophage- SIRPaFc proteins mediated phagocytosis of both hematologic and solid tumor TTI-621 consists of the N-terminal V domain of human SIRPa cells. Furthermore, these in vitro effects translate into signifi- (GenBank AAH26692) fused to the human IgG1 Fc region (hinge- cant antitumor activity in mouse models of leukemia and CH2-CH3, UniProtKB/Swiss-Prot, P01857). Variant proteins lymphoma. Importantly, unlike anti-CD47 antibodies, TTI- were generated in which the identical human SIRPa domain was 621 binds minimally to human erythrocytes, minimizing linked to a human IgG4 Fc region (hinge-CH2-CH3, UniProtKB/ potential toxicity related to hemolytic anemia. On the basis Swiss-Prot, P01861) or an IgG4 Fc region which was mutated to of the demonstrated antitumor activity in the context of remove residual Fc interactions (20). Both IgG4-based fusion minimal erythrocyte binding, two phase I, multicenter studies proteins contained a hinge-stabilizing mutation that prevents the have been initiated to evaluate TTI-621 in subjects with formation of intrachain disulfide bonds (21). Two mouse surro- relapsed/refractory hematologic malignancies and solid gate SIRPaFcs were constructed, one using the N-terminal V tumors (NCT02663518 and NCT02890368, respectively). domain from NOD mouse SIRPa (22) and the second using a mutated (CV1) N-terminal V domain of human SIRPa (23). In both mouse surrogates, the SIRPa domains were linked to a mouse IgG2a Fc (hinge-CH2-CH3, UniProtKB/Swiss-Prot, P01863). All constructs were generated by overlapping PCR using adverse prognostic factor where high CD47 expression correlates standard molecular biology techniques and expressed in stably with more aggressive disease and poorer clinical outcomes. For transfected CHO-S cells (Invitrogen). Proteins were purified from example, overall survival was significantly lower for DLBCL or culture supernatant using protein A and hydrophobic interaction mantle cell lymphoma patients who had elevated CD47 expres- chromatography, concentrated, and residual endotoxin removed. sion, and higher CD47 expression on tumor cells was associated Control human IgG1 and mouse IgG2a Fc proteins lacking the with significantly poorer event-free survival in patients with CLL SIRPa domain were also generated and similarly purified. All (7). Similar trends have been reported in other hematologic proteins displayed >99% purity by HPLC and <0.4 EU/mg malignancies (5, 6, 8) and solid tumors (11, 13, 16, 17). In endotoxin. addition, there is evidence to suggest that increased CD47 expression is associated with the transition from low-risk to high-risk MDS and subsequent transformation to AML (10). These findings Antibodies are consistent with tumor cells exploiting the suppressive CD47– The anti-CD47 antibodies BRIC126 (Serotec), 2D3 SIRPa axis to evade macrophage-mediated destruction. Blocking (eBioscience), and CC2C6 (BioLegend) were obtained from com- CD47 has thus emerged as a promising therapeutic strategy and mercial sources. Clones B6H12.2 (ATCC HB:9771) and 5F9 (19) several studies have shown that interrupting the CD47–SIRPa were generated internally based on publicly available sequences. signaling pathway using anti-CD47 mAbs promotes antitumor Rituximab (Hoffman-La Roche Ltd) was obtained from the Lon- activity against human cancers both in vitro and in vivo (6–9, 18). don Health Sciences Centre pharmacy (London, Ontario, However, the expression of CD47 on erythrocytes raises concerns Canada). about the potential for anti-CD47 mAbs to cause hemolytic anemia, as seen in preclinical studies (19). Furthermore, eryth- Cells rocyte CD47 constitutes a massive antigen sink that may limit the The following human tumor cell lines were used: DLBCL (HT, ability of CD47-targeting agents to reach tumor cells. Finally, Ly1, Pfeiffer, NUDUL1, SUDHL4, SUDHL6, SUDHL8, SUDHL16, CD47-targeting agents bound to erythrocytes may cause interfer- Toledo), multiple myeloma (MM1.s, 8226, H929, U266), non- ence with blood typing tests. DLBCL B-cell lymphomas (Raji, Namalwa, C1R, Ly5), AML (HL- TTI-621 (SIRPaFc) is a novel innate immune checkpoint inhib- 60, KG-1, THP-1 and TF-1, AML-2), chronic myeloid leukemia itor that binds human CD47 and prevents it from delivering a "do (K562, KU812), acute T-cell leukemia (ENL-1, Jurkat), T-cell not eat" signal to macrophages. It is designed to function as a lymphoma (HH, H9, SUPT1, DERL7), lung cancer (A549), epi- decoy receptor, binding CD47 on the surface of tumor cells and dermoid cancer (A431), ovarian cancer (OVCAR-3), colon cancer blocking its anti-phagocytic "do not eat" signal, thereby allowing (DLD-1), breast cancer (SKBR3), melanoma (A375, SK-MEL-1, G- macrophages to phagocytose malignant cells. Moreover, the IgG1 361, WM-115, SK-MEL-28), and Merkel cell carcinoma (MCC26, Fc region of SIRPaFc can interact with human Fcg receptors on MKL-1). All cell lines were obtained from ATCC except MCC26 macrophages to further enhance phagocytosis, tumor antigen and MKL-1 (Sigma-Aldrich), AML-2 (kindly provided by Mark presentation, and effective antitumor activity. Here, we fully Minden, University Health Network, Toronto, Canada), EN-1 characterize this novel agent and demonstrate that TTI-621 (kindly provided by Aaron Schimmer, University Health Net- strongly binds to a wide range of human tumors and induces work, Toronto, Canada) and Ly5 (kindly provided by David potent phagocytosis of human tumor cells in vitro and in vivo while Spanner, Sunnybrook Research Institute, Toronto, Canada). The sparing most normal cells. Although the decoy receptor binds mouse A20 B lymphoma line was obtained from ATCC.

www.aacrjournals.org Clin Cancer Res; 23(4) February 15, 2017 1069

Petrova et al.

Viably frozen primary tumor cells from the peripheral blood or (BioLegend), B6H12 (in-house), 5F9 (in-house)]. Cells were bone marrow of patients with B-cell ALL, T-cell ALL, MDS, and washed and subsequently stained with biotin-conjugated anti- AML were obtained from the University Health Network (UHN) human IgG Fc PAN (Hybridoma Reagent Laboratory), followed BioBank (Toronto, Canada) according to the procedures by detection with PE-conjugated streptavidin (eBioscience). approved by the Research Ethics Board of UHN. Flow cytometry was performed on a FACSVerse flow cytometer Human macrophages were prepared from heparinized whole (BD Biosciences). blood obtained from normal healthy human donors (Biolog- ical Specialty Corporation); informed consent was obtained Hemagglutination assays from all donors. Peripheral blood mononuclear cells (PBMC) Titrated amounts of TTI-621 or anti-CD47 mAbs (up to were isolated over Ficoll-Paque Plus density gradient (GE 3 mmol/L) were added to wells containing erythrocytes þ Healthcare) and CD14 monocytes were isolated from PBMCs diluted in PBS, and the plates were incubated overnight at by positive selection using CD14 antibody-coated MicroBead 37 Cin5%CO2. The extent of hemagglutination was assessed separation (Miltenyi Biotec). Monocytes were differentiated by scoring each well on a scale of 1 to 6, with 1 representing the into macrophages by culturing for at least 10 days in X-Vivo- absence of hemagglutination and 6 representing complete 15 media (Lonza) supplemented with M-CSF (PeproTech). hemagglutination. One day prior to phagocytosis assays, the monocyte-derived macrophages were primed with IFNg (PeproTech) to generate Phagocytosis assays M1 macrophages or with IL4 (Peprotech) to generate M2 Confocal-based phagocytosis assay. Tumor cells were labeled with fi macrophages. Unless otherwise speci ed, all phagocytosis CellTrace CFSE (Life Technologies) and added to primed assays were carried out using M1 macrophages. When required, macrophages in 24-well plates at a 1:5 effector:target ratio. macrophages were harvested using Enzyme-Free Cell Dissoci- Macrophages and tumor cells were cocultured for 2 hours at ation Buffer (ThermoFisher). 37 Cin5%CO2 in the presence of TTI-621 or control Fc protein and subsequently stained with Alexa Fluor 555–con- Tumor cell binding jugated Wheat Germ Agglutinin (Invitrogen). Phagocytosis was Cell lines or primary patient samples were added in duplicate to assessed by confocal microscopy on a Quorum Wave FX-X1 96-well plates and incubated with titrated amounts of biotiny- Spinning Disc Confocal System and images were analyzed lated TTI-621 or biotinylated isotype-matched control IgG Fc, using Velocity software (PerkinElmer). A phagocytosis index together with Near-IR LIVE/DEAD Fixable Dead Cell Stain (Invi- was calculated as: (number of tumor cells inside macrophages/ trogen) for 30 minutes on ice. Cells were washed, stained with number of macrophages) 100; counting at least 200 macro- phycoerythrin (PE)-conjugated streptavidin (eBioscience), phages per sample. All tumor cells counted were confirmed to washed, and resuspended in Stabilizing Fixative (BD Biosciences). be internalized using z-stack images. Statistical significance was Flow cytometry was performed on a FACSVerse flow cytometer calculated by unpaired t test versus isotype control using (BD Biosciences). Data were analyzed using FlowJo software GraphPad Prism software. (Treestar Inc.). Half-maximal effective concentration (EC50) values were calculated using a sigmoidal dose–response curve in Flow cytometry–based phagocytosis assay. Tumor cells were labeled GraphPad Prism software. with Violet Proliferation Dye 450 (BD Biosciences) and added to primed macrophages in 96-well plates at a 1:5 effector:target ratio. Erythrocyte binding Macrophages and tumor cells were cocultured for 2 hours at 37C Erythrocytes were isolated from sodium-heparinized whole in 5% CO2 in the presence of TTI-621 or control Fc protein and blood from healthy human donors (Biological Specialty Cor- subsequently stained with Near-IR LIVE/DEAD Fixable Dead Cell poration) by centrifugation followed by several washes with Stain (Invitrogen), APC-conjugated anti-human CD14 (61D3, PBS. The resulting packed erythrocytes were diluted in PBS and eBioscience), and PE-conjugated anti-human CD11b (ICRF44, added in duplicate to 96-well plates. Binding was performed by eBioscience), washed and resuspended in Stabilizing Fixative (BD incubating erythrocytes with titrated amounts of TTI-621, anti- Biosciences). Cells were acquired on a FACSVerse flow cytometer, CD47 mAbs [BRIC126 (Serotec), 2D3 (eBioscience), CC2C6 and data were analyzed using FlowJo software (Treestar Inc.).

Figure 1. Structure of TTI-621. TTI-621 consists of the N-terminal domain of human SIRPa (shown in red) linked to a human IgG1 Fc region (shown in blue). The hinge and inter-chain disulﬁde bonds are shown as black lines.

1070 Clin Cancer Res; 23(4) February 15, 2017 Clinical Cancer Research

TTI-621 Is a Novel Antitumor Immune Checkpoint Inhibitor

Figure 2. TTI-621 promotes macrophage-mediated phagocytosis of human tumor cells in vitro. A, Representative scanning confocal microscopy images after macrophages were cocultured with a primary AML patient sample for 2 hours in the presence of 10 mmol/L TTI-621 or control IgG1 Fc protein. Tumor cells and macrophages are stained green and red, respectively. B, Macrophage-mediated phagocytosis of established human tumor cell lines from patients with B-cell malignancies (n ¼ 17), myeloid malignancies (n ¼ 7), T-cell malignancies (n ¼ 6), skin cancers (n ¼ 7), and other solid cancers (n ¼ 5) in the presence of 1 mmol/L TTI-621 (black bars) or control IgG1 Fc protein (white bars). Phagocytosis was quantified by determining a phagocytosis index (number of engulfed tumor cells per 100 macrophages) using confocal microscopy or measuring percentage phagocytosis by flow cytometry, as described in the Materials and Methods section. C, Macrophage-mediated phagocytosis of primary human tumor samples from patients with hematologic malignancies (n ¼ 33) in the presence of 1 mmol/L TTI-621 (black circles) or control IgG1 Fc protein (white circles). D, Representative titration of TTI-621 (black circles) on a primary AML patient sample. Control Fc protein (white circle) was tested at 1 mmol/L. E, Macrophage-mediated phagocytosis of primary AML tumor sample or normal monocytes was assessed by confocal microscopy in the presence of 1 mmol/L TTI-621 or control IgG1 Fc. Statistical significance was assessed by unpaired t test versus Fc control (, P < 0.05; , P < 0.01; , P < 0.001; NS, not significant).

www.aacrjournals.org Clin Cancer Res; 23(4) February 15, 2017 1071

Petrova et al.

A Injected bone marrow Non-injected bone marrow Spleen

P = 0.003 P % AML Engraftment AML % P = 3×10–8 Engraftment AML % Engraftment AML % = 0.02

TTi-621 Control Fc TTi-621 Control Fc TTI-621 Control Fc

B Injected bone marrow Non-injected bone marrow Spleen 8 100 100 7

75 75 6 5

50 50 4 3 P 25 = 0.008 25 2 P = 0.003 1 P = 0.0004 % AML Engraftment % AML Engraftment % AML Engraftment % AML 0 0 0 TTi-621 Control Fc TTi-621 Control Fc TTI-621 Control Fc

C Burkitt Lymphoma D Burkitt Lymphoma (Raji) (Namalwa) 2,000 2,000

) Dosing window ) 3 3 Control Fc Dosing window 1,500 *** 1,500 Control Fc SIRPαFc *** *** SIRPαFc Rituximab *** 1,000 1,000 Rituximab

500 500 Tumor volume (mm Tumor volume (mm 0 0 0 20 40 60 0 20 40 60 Day post-engraftment Day post-engraftment

E Diffuse large B-cell lymphoma F B Lymphoma (Toledo) (A20) 2,000 2,000

) Control Fc ) 3

3 α Dosing window SIRPαFc SIRP Fc 1,500 Dosing window *** Rituximab 1,500 Vehicle

1,000 1,000

500 500 Tumor volume (mm Tumor volume (mm 0 0 0 20 40 60 0 10 20 30 40 Day post-engraftment Day post-engraftment

Figure 3. TTI-621 and its mouse surrogate are efficacious in vivo. A and B, NOD.SCID mice were preconditioned with sublethal irradiation and anti-CD122 antibody (to deplete residual NK cells) and then transplanted with AML cells from patient #0905443 (A) or patient #090191 (B) by intrafemoral injection. Treatment with TTI- 621 (8 mg/kg i.p. 3/week for 4 weeks) or control IgG1 Fc protein was initiated 21 days post-transplantation. The percent AML engraftment (% cells expressing human CD45 and CD33 markers) was assessed by flow cytometry. Each symbol represents one mouse, bars indicate mean values. P values were determined by t test versus Fc control protein. Data shown are representative of 9 separate AML patient xenografts. C–E, SHrN mice (n ¼ 5 per group) received subcutaneously implanted Raji (C), Namalwa (D), or Toledo (E) cells. Three days after implantation (Namalwa and Raji) or 10 days after implantation (Toledo), mice were dosed intraperitoneally with either a mouse surrogate SIRPaFc (10 mg/kg), control mouse IgG2a Fc (6.67 mg/kg), or rituximab (8 mg/kg) five times a week for 3 weeks (indicated by the arrow heads). Tumor volumes were estimated by caliper measurement from both flanks and the means for those measurements were calculated in mm3.(Continued on the following page.)

1072 Clin Cancer Res; 23(4) February 15, 2017 Clinical Cancer Research

TTI-621 Is a Novel Antitumor Immune Checkpoint Inhibitor

þ þ Macrophages were identified as live, single, CD14 CD11b cells. volumes were monitored until they reached the maximum vol- Doublets were excluded by SSC-W and SSC-H discrimination. ume of approximately 1,500 mm3 or maximum permissible Percent phagocytosis was assessed as the percent of macrophages markers of discomfort in the mice were reached (i.e., mouse þ that were VPD450 . The gating strategy and representative dot discomfort or body weight loss reached maximum allowable plots are shown in Supplementary Fig. S1. Statistical significance levels), at which time the mice were sacrificed. All studies were was calculated by unpaired t test versus isotype control using conducted according to the animal care guidelines described by GraphPad Prism software. the Canadian Council on Animal Care (CCAC) and monitored by The Western University Animal Use Subcommittee. Statistical AML xenografts significance was calculated by two-way ANOVA using GraphPad AML xenografts were performed in 10-week-old female Prism software. NOD.SCID mice bred and maintained in the Barrier Unit at the UHN Animal Facility (Toronto, Canada). One day prior to Syngeneic B-cell lymphoma model transplantation, mice were sublethally irradiated (275 cGy) Female BALB/c mice (6–8 weeks old) were purchased from and pretreated with anti-CD122 antibody (0.2 mg/mouse) to Charles River Laboratories and housed in the University of Tor- deplete residual host NK cells. On the day of transplantation, onto animal facility. A20 cells (2 106) were injected subcuta- viably frozen mononuclear cells collected from AML patients neously into the right hind flank of 8-week-old BALB/c female 90543 and 90191 were thawed, counted, and transplanted mice in a volume of 0.1 mL. When the tumors were palpable intrafemorally into the preconditioned mice at a dose of 5 3 6 (approximately 60 mm ), they were randomized and injected 10 cells/mouse in a total volume of 30 mL. Twenty-one days intratumorally with 200 mg (10 mg/kg) of a mouse SIRPaFc after engraftment, mice were dosed with TTI-621 (8 mg/kg) or surrogate (CV1 SIRPa) fusion protein in a 50 mL volume of PBS. equimolar amount of control human IgG1 Fc (5.4 mg/kg) at Control groups were injected with vehicle alone in a 50-mL 0.3 mL/mouse, 3 times/week for 4 weeks. Upon euthanization, volume. Animals were dosed twice weekly for a total of five doses. bone marrow from injected and noninjected bones was Tumors were monitored three times a week and tumor volume collected and stained with mouse anti-human antibodies 1 2 was calculated as /2 length width . Tumor volumes were including CD47-FITC, CD33-PE, CD19-PC5, CD45-APC, monitored until one or more tumor dimensions reaches the CD34-APCCy7, CD38-PECy7. After staining, washed cells were maximum permissible measure (15 mm), or when maximum fl runonanLSRII ow cytometer (BD Biosciences). Events permissible markers of discomfort were observed, at which time – (10,000 20,000) were collected for each sample. Collected the mice were sacrificed. All animal procedures were approved by data were analyzed by FlowJo software to assess AML engraft- the animal care committee of the University of Toronto in ment levels in the injected femur, noninjected bones, and in accordance with the CCAC. Statistical significance was calculated the spleen as determined by the percentage of human þ þ by two-way ANOVA using GraphPad Prism software. CD45 CD33 cells. Results B-cell lymphoma xenografts Structure of TTI-621 Lymphoma xenografts were performed in 6- to 7-week-old TTI-621 (SIRPaFc) was generated by directly linking the female NOD.Cg-PrkdcscidHrhr/NCrHsd (SHrN) mice, a hairless sequences encoding the N-terminal CD47 binding domain of SCID strain, obtained from Harlan Laboratories (Montreal, human SIRPa with the Fc domain of human IgG1 (Fig. 1). The Canada) and maintained at the Victoria Research Laboratories SIRPa region interacts with CD47, while the Fc region binds to Fcg Vivarium (London Health Sciences Centre). Raji and Namalwa receptors. TTI-621 is secreted by a genetically engineered Chinese cells (1 106 per injection) were injected subcutaneously into hamster ovary (CHO) cell line as a 77-kDa disulfide-linked, N- each flank of SHrN mice in a volume of 0.1 mL PBS (i.e., 2 tumor glycosylated homodimer consisting of two identical 345 amino injection sites per mouse). Toledo cells (1 107 per injection) acid chains. were injected in 50% Matrigel (ECM gel, Sigma-Aldrich) in a volume of 0.1 mL into the left flank for SHrN mice. Mice were kept under isoflurane-mediated anesthesia during the injections. Three TTI-621 binds to CD47 and enhances macrophage phagocytosis days after Raji and Nawalma tumor cell implantation and 10 days of tumor cells in vitro after Toledo tumor cell implantation, animals received either 10 The binding of TTI-621 to CD47 on malignant human cells was mg/kg of mouse SIRPaFc (NOD SIRPa) or 6.75 mg/kg control assessed by flow cytometry. TTI-621 was found to bind strongly to IgG2a Fc, daily 5 times per week, for 3 weeks by intraperitoneal a panel of 19 tumor cell lines derived from patients representing a injection. Tumor volumes were estimated twice weekly by stan- wide range of both hematologic and solid tumors (Supplemen- dard caliper measurements of length and width then calculated as tary Table S1). TTI-621 also exhibited strong binding to primary follow: p/6 (longest diameter) (shortest diameter)2. Tumor tumor samples obtained from the blood of patients with B-cell

(Continued.) Mice were sacrificed when tumor volumes exceeded 1,500 mm3 or when there was extensive ulceration. Mean tumor volumes are recorded only for time points in which 1 mouse per group was sacrificed. Data shown are representative of 4 (Raji), 2 (Namalwa), and 2 (Toledo) independent experiments. Mice were terminated at a tumor volume of 1,500 mm3. F,2 106 A20 cells were implanted subcutaneously into the right flank of Balb/c mice on day 0. Mice were randomized (n ¼ 9–10 mice per treatment arm) when the mean tumor size was palpable at which time the tumors were approximately 60 mm3 in volume. A mouse surrogate SIRPaFc (10 mg/kg) or vehicle was given bi-weekly by intratumoral administration. Mice were sacrificed when at least one tumor dimension exceeded 15 mm. Mean tumor volumes SEM were recorded only for time points in which 2 mice per group were sacrificed. Data shown are representative of two independent experiments. Statistical significance was assessed by two-way ANOVA (, P < 0.001).

www.aacrjournals.org Clin Cancer Res; 23(4) February 15, 2017 1073

Petrova et al.

acute lymphoblastic leukemia (B-ALL), T-ALL, and AML, and constructed using the mouse IgG2a Fc region, allowing for full bone marrow samples from patients with MDS, with an average effector function, analogous to the human IgG1 Fc region in TTI- binding EC50 value of 197 182 nmol/L (Supplementary Table 621. Treatment of mice with mSIRPaFc may thus more closely S2). CD47 is widely expressed on normal cells, and TTI-621 also mimic the anticipated pharmacokinetic and Fc effector activity þ þ demonstrated binding to human CD4 T cells, CD8 T cells, B profile of TTI-621 in human subjects. The in vivo efficacy of cells, platelets, natural killer (NK) cells, granulocytes, monocytes, mSIRPaFc was assessed in three aggressive B-cell lymphoma and NK T cells from the peripheral blood of healthy donors xenograft models: Namalwa and Raji (Burkitt lymphomas) and (Supplementary Table S3). Toledo (DLBCL). Hairless NOD.SCID (SHrN) mice were The ability of TTI-621 to promote macrophage-mediated implanted subcutaneously with tumor cells and treated with phagocytosis of human tumor cells was assessed using both mSIRPaFc five times/week for 3 weeks starting either 3 days after confocal microscopy and flow cytometry. Monocyte-derived engraftment (Namalwa and Raji) or 10 days after engraftment macrophages were cocultured with tumor cells for two hours, (Toledo). mSIRPaFc treatment markedly reduced the growth of and in cultures left untreated or treated with a control Fc fragment, Raji tumors (Fig. 3C) and completely ablated Namalwa and macrophages exhibited a low level of phagocytosis, consistent Toledo tumors (Fig. 3D and E); in the latter two models, most with CD47-mediated suppression. In contrast, blockade of CD47 mice remained tumor-free 60 days after inoculation. Moreover, on the target cells using TTI-621 significantly increased macro- mSIRPaFc was superior to rituximab therapy in both Namalwa phage phagocytosis of tumor cells (Fig. 2A). Compared with a and Toledo xenografts. control Fc protein, TTI-621 promoted macrophage phagocytosis To overcome the limitations inherent with xenograft models, of 77% (23/30) of tumor cell lines established from patients with we also assessed whether mSIRPaFc could reduce tumor burden hematologic malignancies and 67% (8/12) of human solid cancer in an immunocompetent syngeneic system. BALB/c mice were cell lines (Fig. 2B). A marked prophagocytic effect of TTI-621 was subcutaneously inoculated with A20 B-cell lymphoma cells, and also observed on primary samples from patients with AML, MDS, mSIRPaFc was administered by intratumoral injection twice multiple myeloma, B-ALL, and T-ALL (Fig. 2C). TTI-621 enhanced weekly starting 7 days postengraftment. As shown in Fig. 3F, macrophage-mediated killing of 97% (32/33) of primary blood mSIRPaFc treatment significantly reduced the growth of A20 cancer samples tested. Drug activity was further characterized by tumors, confirming that CD47 blockade with mSIRPaFc is also titrating TTI-621 on selected human tumor cell lines (n ¼ 13) and efficacious in animals with an intact immune system. primary tumor samples (n ¼ 4) (representative data in Fig. 2D). As Collectively, these in vivo data suggest that blockade of the summarized in Supplementary Table S4, TTI-621 treatment CD47–SIRPa axis using SIRPaFc has broad applicability across a resulted in a saturable, dose-dependent phagocytic response with variety of malignancies. an average EC50 of 10 14 nmol/L. We then assessed the effect of TTI-621 on macrophage-medi- Blockade of CD47 using SIRPaFc requires an IgG1 Fc tail for ated phagocytosis of normal cells in vitro. As shown in Fig. 2E, TTI- maximum potency 621 potently increased phagocytosis of primary AML tumor cells, Engagement of Fcg receptors (FcgR) on macrophages by while sparing normal peripheral blood monocytes, indicating SIRPaFcmaydeliveraprophagocytic signal that could augment that TTI-621-enhanced phagocytosis is tumor cell-specific. the effect of CD47 blockade. TTI-621 possesses an IgG1 Fc tail, Collectively, these in vitro data demonstrate that TTI-621 allowing for binding to the high-affinity receptor FcgRI (CD64) induces potent, tumor-specific macrophage phagocytosis across as well as to the low-affinity receptors FcgRII (CD32) and a broad range of hematologic and solid tumors. In fact, we have FcgRIII (CD16). To determine whether the IgG1 Fc tail is not observed a tumor type that is refractory to TTI-621 treatment, required for maximum potency, we compared the in vitro consistent with prior data demonstrating that the CD47 immune activity of TTI-621 with a variant SIRPaFc in which the IgG1 checkpoint is widely used by malignant cells to escape immune Fc region of TTI-621 was replaced with an IgG4 Fc tail. IgG4 Fc surveillance (5, 24). regions bind well to CD64 but have weaker interactions than IgG1 with CD32 and CD16 (25). We compared the propha- TTI-621 and mouse surrogate SIRPaFc have potent antitumor gocytic activity of both SIRPaFcs using classically activated activity in vivo (M1) and alternatively activated (M2) macrophages. We To determine whether the potent effects of TTI-621 in vitro have previously shown that M1 macrophages are CD32hi translated into in vivo antitumor activity, we employed an AML CD64hiin vitro, whereas M2 macrophages are CD32hi CD64lo xenograft model using primary patient samples. Engrafted mice (26). TTI-621 enhanced phagocytosis by both macrophage were treated with TTI-621 or an Fc fragment control three subsets equally well. In contrast, SIRPaFc with an IgG4 tail times/week for 4 weeks. Although control-treated animals induced significantly less phagocytosis by M2 macrophages exhibited significant engraftment, particularly in the injected (Fig. 4A). These data suggest that an IgG1 tail is necessary for bone marrow, TTI-621 treatment significantly reduced the SIRPaFc's enhancement of phagocytosis by both M1 and M2 tumor burden in bone marrow and spleen (Fig. 3A and B). In macrophages. fact, tumor cells were undetectable in most animals following We next compared the in vivo activity of TTI-621 and the variant TTI-621 therapy. IgG4-containing SIRPaFc in the AML xenograft model. We also The presence of CD47 on nontumor tissue has the potential to tested a SIRPaFc with a mutated IgG4 Fc region that is completely bind SIRPaFc and remove it from circulation, potentially resulting devoid of Fc effector functions. As shown in Fig. 4B, treatment in a significant antigen sink effect. As TTI-621 does not bind to with all three SIRPaFc constructs reduced tumors to undetectable mouse CD47 (data not shown), TTI-621 treatment of xenograft levels in the spleen. In the injected femur and noninjected bone recipient does not model this antigen sink effect. To overcome this marrow, TTI-621 treatment completely ablated tumor growth in limitation, mouse surrogate fusion proteins (mSIRPaFc) were all but one mouse. SIRPaFc with an IgG4 tail reduced tumor

1074 Clin Cancer Res; 23(4) February 15, 2017 Clinical Cancer Research

TTI-621 Is a Novel Antitumor Immune Checkpoint Inhibitor

150 A M1 Mϕ M2 Mϕ ** 100

50 % Max Phagocytosis

0 ) 1) G4 62 IgG4 Ig I- T (T G1 Ig IgG1 (TTI-621

Injected femur Non-injected BM Spleen B P < 0.001 P < 0.001 P < 0.001 P < 0.01 P < 0.05 100 100 8 P < 0.001 P < 0.001 7 75 75 6 P < 0.01 5 P < 0.01 50 50 4 3 25 25 2

% AML Engraftment % AML Engraftment % AML Engraftment % AML 1 0 0 0 ) ) l ) ) ) l t 4) 1 t 4 ol t ol G tro G tr trol tr tro -mu n -mu mu 4 ontrol o 4 (Ig 4- IgG4) on c con c G (TTI-621) c (TTI-621) c IgG Fc (Ig g Fc IgG Fc ( F ( a 1) (TTI-62 (I a F ( t Fc con c G c 1 Pa 1 F F IgG1) mut Fc con Fc R IgG1) mu a SIRP (Ig IgG1 Fc c a SIRP ( - IgG SI - IgG P c Pa Fc F G4 Fc ( G4 SIRP a IgG4-mut Fc SIR a Ig SIR a Ig P IRP SIRP SIR S

80 C 60

20 % Phagocytosis

0 3 2 Fc D 1 G1 l 2 H g o 6 odrug mI B TI-621 N T Contr

Blockade of the CD47–SIRPa interaction: No Yes Yes

Figure 4. SIRPaFc with an IgG1 Fc tail has potent antitumor efficacy. A, M1 and M2 monocyte-derived macrophages were generated by priming for 24 hours with IFNg or IL4, respectively. Macrophage phagocytosis of a DLBCL cell line (Toledo) was assessed by flow cytometry (% phagocytosis) in the presence of SIRPaFc with an IgG1 Fc tail (TTI-621) or an IgG4 Fc tail (both at 1 mmol/L concentration). Data shown represent n ¼ 5 donors. B, NOD.SCID mice were preconditioned with sublethal irradiation and anti-CD122 antibody (to deplete residual NK cells) and then transplanted with AML cells from patient #090191 by intrafemoral injection. Treatment with SIRPaFc (8 mg/kg i.p. 3/week for 4 weeks) or control IgG1 Fc protein was initiated 21 days post-transplantation. The percent AML engraftment (% cells expressing human CD45 and CD33 markers) was assessed by flow cytometry. Each symbol represents one mouse, bars indicate mean values. P values were determined by one-way ANOVA. Data shown are representative of two independent experiments. C, Monocyte-derived macrophages were generated as described and primed for 24 hours with IFNg. Macrophage phagocytosis of a DLBCL cell line (Toledo) was assessed by flow cytometry (% phagocytosis) in the presence of TTI-621, anti-CD47 mAbs B6H12 or 2D3, or isotype-matched controls (all at 1 mmol/L).

www.aacrjournals.org Clin Cancer Res; 23(4) February 15, 2017 1075

Petrova et al.

burden in the noninjected bone marrow, but not in the injected TTI-621 induces anemia in non-human primates but binds femur compared with controls, whereas the mutated IgG4 fusion minimally to human erythrocytes protein was unable to control tumor burden in either bone A significant concern with CD47-blocking agents is related to marrow compartment (Fig. 4B). the high expression of CD47 on human erythrocytes and the The contribution of the Fc region raises the question of whether potential for such agents to cause anemia, as seen in preclinical TTI-621 activity requires neutralization of the CD47 "do not eat" studies (19). To assess the risk of anemia and other adverse events, signal, or whether it simply opsonizes CD47-expressing cells for primate repeat-dose toxicology studies of TTI-621 were conducted Fc receptor–mediated destruction, similar to antibodies that in non-human primates. Cynomolgus monkeys were selected as trigger classical antibody-dependent cellular phagocytosis relevant species based on the high CD47 sequence homology (ADCP). To address this, we compared the in vitro activity of (97.6% identity to human CD47) and cross-reactivity studies TTI-621 to two isotype-matched (mouse IgG1) anti-CD47 anti- The principal dose-limiting toxicity observed in cynomolgus bodies: clone B6H12, which blocks the CD47–SIRPa interaction monkeys was anemia, which occurred at repeat doses of 3 mg/kg and the non-neutralizing clone 2D3. As shown in Fig. 4C, 2D3 or greater. In addition to anemia, other cytopenias, including induces a low level of phagocytosis, attributable to opsonization thrombocytopenia, lymphopenia, neutropenia, and monocyto- of CD47 without blockade of the CD47–SIRPa interaction. penia were observed, although these were reversible and without B6H12 is more effective than 2D3 at inducing the phagocytosis clinical sequelae (see Supplementary Fig. S2 for representative of a human B lymphoma target, indicating that efficient phago- hematology values). The bone marrow exhibited evidence of cytosis in this system requires blockade of the CD47–SIRPa regenerative responses, notably erythropoiesis. No effects were interaction. TTI-621 exhibits even greater activity than B6H12, observed on neurologic, cardiovascular, or other systems. which presumably reflects the combined effect of CD47 blockade Despite the strong binding of TTI-621 to monkey erythrocytes, and more effective Fc receptor engagement by the TTI-621 human we observed only minimal binding to human erythrocytes (Fig. IgG1 Fc region. 5A), which may be due to species-specific differences in the Collectively, these data show that SIRPaFc with an IgG1 tail mobility of CD47 in erythrocyte membranes (data not shown). (TTI-621) is significantly more potent at promoting phagocytosis Importantly, the low binding profile of TTI-621 to human ery- in vitro and controlling tumor burden in vivo, and that both CD47 throcytes is in contrast to the strong binding demonstrated by five blockade and Fc-mediated effector functions contribute to the different anti-CD47 antibody clones (Fig. 5A). Minimal binding mechanism of action of TTI-621. of TTI-621 was observed on erythrocytes from all 43 healthy

TTI-621 BRIC126 2D3 CC2C6 B6H12 5F9 A

Fluorescence intensity

B C 300,000

200,000 7 100,000 6 2D3 5 B6H12 1500 4 BRIC126 5F9 3 1000 TTI-621

(mean SE) ± 2 500 1 Mean fluorescenceintensity

Hemagglutination score Hemagglutination 0 0 0.01 0.1 1 10 100 1,000 10,000 3 6 2 9 1 b 2 126 2D H1 5F 1Fc gG G I-621 C 6 I g Concentration (nmol/L) T CC2C B G m T g mI BRI hI CD47 mAbs Controls

Figure 5. TTI-621 exhibits minimal binding to human erythrocytes. A, Human erythrocytes were stained with saturating concentrations of TTI-621 or CD47-specific antibodies (clones BRIC126, 2D3, CC2C6, B6H12, or 5F9) and analyzed by flow cytometry. Representative histograms are shown, with specific staining shown in black and isotype control staining in gray. B, Summary data showing the mean fluorescence intensity for 43 erythrocyte donors. C, Hemagglutination assays were conducted with human erythrocytes and titrated amounts of TTI-621- or CD47-specific antibodies. The extent of hemagglutination was assessed by blinded scoring on a scale of 1 to 6, with 1 representing the absence of hemagglutination and 6 representing complete hemagglutination.

1076 Clin Cancer Res; 23(4) February 15, 2017 Clinical Cancer Research

TTI-621 Is a Novel Antitumor Immune Checkpoint Inhibitor

donors tested regardless of gender, ABO blood group, or rhesus through blockade of the CD47 "do not eat" signal and simul- antigen status (Fig. 5B). Consistent with these binding data, TTI- taneous delivery of a prophagocytic ("eat") signal through 621 did not induce hemagglutination of human erythrocytes in macrophage FcgRs.Inlinewiththis,ourdatasuggestthat vitro (Fig. 5C). The lack of significant binding of TTI-621 to human TTI-621 does not work simply by opsonization of CD47 and erythrocytes thus offers a significant advantage over CD47-block- ADCP, but triggers phagocytosis through CD47 blockade as ing mAbs. well as simultaneous activation through FcgRs. Although CD47 has recently emerged as a promising immuno-oncology target, concerns have been raised regarding Discussion the potential for anemia and an erythrocyte antigen sink, due to Approved immune checkpoint inhibitors have extended the the expression of high levels of CD47 on human red blood cells survival of multiple subgroups of cancer patients and thus trans- (30, 31). In this regard, TTI-621 exhibits an advantage over anti- formed modern oncology. Although the focus thus far has been CD47 antibodies, in that it binds only minimally to human red on blockade of checkpoints that suppress T-cell responses (e.g., blood cells. A similar observation has recently been reported by PD-1 or PD-L1), there is growing recognition that the innate an independent group (32). The minimal binding of TTI-621 to immune system plays an important role in the initiation and human erythrocytes may be due to the association of CD47 propagation of enduring antitumor responses and CD47 has with the erythrocyte spectrin cytoskeleton (30), which results in recently emerged as a key checkpoint of innate immunity. Our reduced membrane mobility (33) and a consequent failure to findings demonstrate that SIRPaFc (TTI-621) is an effective decoy cluster CD47 effectively. Consistent with this theory, we have receptor that enhances macrophage-mediated phagocytosis in a previously shown strong binding of TTI-621 to human ery- broad spectrum of human hematologic and solid tumors, both in throcytes when CD47 is first preclustered using a nonblocking vitro and in xenograft models. More than 97% of primary patient CD47 antibody (34). samples tested were sensitive to the antitumor effects of TTI-621, While it is acknowledged that TTI-621 binds to human suggesting that this therapeutic approach will have broad appli- platelets and leukocytes (and thus may be associated with the cability in human cancer. development of thrombocytopenia and/or leukopenia), the Importantly, blockade of CD47 by TTI-621 selectively induced extremely low erythrocyte–binding profile of TTI-621 offers phagocytosis of malignant cells over normal cells, providing a several potential advantages over anti-CD47 mAbs that strongly therapeutic window for treatment of patients in the clinic. Pref- bind to erythrocytes. First, treatment with TTI-621 is less likely erential macrophage phagocytosis of AML cells over normal cord to result in anemia. CD47 is thought to protect erythrocytes blood/bone marrow cells has also been reported for an IgG4- from macrophage-mediated clearance (2), and CD47-blocking based SIRPaFc fusion protein, even when nonmalignant cells antibodies are known to trigger anemia in non-human pri- outnumbered the AML cells by a 2:1 ratio (27). In addition, a mates, a finding that may limit their clinical utility despite the mouse anti-human CD47-neutralizing antibody did not induce employment of a priming strategy (19). Second, minimal phagocytosis of normal peripheral blood B cells (7) or normal erythrocyte binding permits the use of an IgG1-based fusion human pancreatic ductal epithelial cells and pancreatic stellate protein, and thus maximizes macrophage phagocytosis of cells (15). tumor cells, without concern for opsonizing red blood cells The specificity for tumor cells is thought to result from the and targeting them for destruction. Third, CD47-targeting expression of prophagocytic signals such as calreticulin on malig- agents that bind erythrocytes may interfere with transfusion nant cells but not on normal cells. Calreticulin is known to trigger typing and cross-matching tests, as seen with other agents that macrophage-mediated phagocytosis, and the phagocytosis of bind erythrocytes (35, 36). Finally, TTI-621 is likely to have a cancer cells induced by CD47 blockade can be completely inhib- superior pharmacokinetic profile compared with anti-CD47 ited by antagonizing the interaction between calreticulin and its mAbs by avoiding the significant antigen sink created by dense receptor (28). It is hypothesized that tumor cells evade phago- cell surface expression of CD47 on erythrocytes, enabling more cytosis because the inhibitory CD47 pathway counterbalances the comprehensive engagement of tumor-expressed CD47. prophagocytic calreticulin signal. Selectively targeting CD47 with We demonstrated that CD47 blockade with SIRPaFc is effi- TTI-621 promotes killing of tumor cells while sparing low calre- cacious in AML and B lymphoma xenograft models, as well as ticulin-expressing normal cells. There are likely to be other as yet in a B lymphoma syngeneic model. Macrophages, in addition unidentified prophagocytic signals on tumor cells that may vary to their direct tumoricidal properties, function as antigen-pre- depending on the tissue type from which the tumor is derived. The senting cells, and thus it is possible that enhancement of broad efficacy of TTI-621 across tumor types suggests that target- phagocytosis by TTI-621 treatment may also result in an ing the CD47–SIRPa axis exploits the reliance of tumor cells on enhanced adaptive immune response. In support of this, CD47 CD47-mediated suppression of phagocytosis regardless of their antibody blockade has been shown to augment tumor antigen þ specific underlying prophagocytic signals. presentation and priming of an antitumor cytotoxic CD8 T- The potent in vivo effects of TTI-621 were attenuated when the cell response in immunocompetent mice (29). In addition, IgG1 Fc tail of the fusion protein was substituted by an IgG4 tail CD47 blockade using a high-affinity SIRPa-variant-human Ig with reduced Fc-mediated effector function, or with an inert fusion protein has also been shown to promote tumor-specific þ mutated IgG4 tail, indicating that Fc effector function is critical CD8 T-cell responses through a dendritic cell–based mecha- for achieving maximum potency of SIRPaFc. These observa- nism (37). These studies provide compelling evidence to sup- tions are consistent with a prior report demonstrating that port the hypothesis that TTI-621 has the potential to generate engineered high affinity SIRPa monomers that bind strongly an enduring antitumor response by acting at the nexus of the to CD47 but lack an Fc region are inactive on their own (29) innate and adaptive immune systems. We propose a mecha- and suggest that maximal antitumor activity is obtained nism in which TTI-621 blocks the CD47 "do not eat" signal on

www.aacrjournals.org Clin Cancer Res; 23(4) February 15, 2017 1077

Petrova et al.

Figure 6. Proposed mechanism of action of TTI-621–mediated CD47 antitumor activity. A, CD47 sends an inhibitory signal to macrophages by binding to SIRPa. B, TTI-621 binds to CD47 on tumor cells and blocks this interaction, (C) while engaging FcgR on macrophages, (D) leading to macrophage-mediated phagocytosis of tumor cells. E, Macrophages that have phagocytosed target cells can present tumor peptides in the context of MHC to tumor-speciﬁc CD8þ T cells, (F) activating the adaptive immune response and leading to destruction of tumor cells by cytotoxic CD8þ T cells.

tumor cells while simultaneously delivering prophagocytic Acquisition of data (provided animals, acquired and managed patients, signals to macrophages through FcgRs, leading to tumor cell provided facilities, etc.): P.S. Petrova, N.N. Viller, M. Wong, X. Pang, K. Dodge, phagocytosis, enhanced antigen presentation, and stimulation V. Chai, H. Chen, V. Lee, V. House, N.T. Vigo, D. Jin, T. Mutukura, – ﬁ M. Charbonneau, T. Truong, E. Linderoth, S.M. Vareki, R. Figueredo, of a tumor antigen speci c T-cell response (Fig. 6). M. Pampillo, J. Koropatnick, N. Mbong, L. Jin, J.C.Y. Wang In summary, these data afﬁrm CD47 as a critical regulator of Analysis and interpretation of data (e.g., statistical analysis, biostatistics, immune surveillance and provide a strong rationale for thera- computational analysis): P.S. Petrova, N.N. Viller, M. Wong, X. Pang, K. Dodge, peutic targeting of CD47. Simultaneous blockade of the inhibi- V. Chai, H. Chen, V. Lee, V. House, N.T. Vigo, M. Charbonneau, T. Truong, tory signal of CD47 with an associated engagement of FcgRon E. Linderoth, E.L. Sievers, S.M. Vareki, R. Figueredo, J. Koropatnick, L. Jin, R.A. Uger macrophages form the basis for clinical development of TTI-621. Writing, review, and/or revision of the manuscript: P.S. Petrova, N.N. Viller, M. Wong, X. Pang, L.D. Johnson, E.L. Sievers, S.M. Vareki, S. Trudel, J.C.Y. Wang, Two phase I, open label, multicenter studies are currently ongoing R.A. Uger to evaluate TTI-621 in patients with relapsed/refractory hemato- Administrative, technical, or material support (i.e., reporting or organizing logic malignancies (NCT02663518) and solid tumors data, constructing databases): P.S. Petrova, X. Pang (NCT02890368). Study supervision: P.S. Petrova, X. Pang, S.M. Vareki, J. Koropatnick, J.C.Y. Wang, R.A. Uger

Disclosure of Potential Conﬂicts of Interest Acknowledgments E.L. Sievers holds ownership interest (including patents) in Trillium Ther- Eilidh Williamson provided medical writing assistance, under the sponsor- apeutics Inc. J. Koropatnick reports receiving commercial research grants from ship of Trillium Therapeutics Inc. Trillium Therapeutics Inc. S. Trudel and J.C.Y. Wang report receiving other commercial research support from Trillium Therapeutics Inc. No potential Grant Support conﬂicts of interest were disclosed by the other authors. These studies were sponsored by Trillium Therapeutics Inc., Mississauga, Canada. Authors' Contributions The costs of publication of this article were defrayed in part by the payment advertisement Conception and design: P.S. Petrova, N.N. Viller, M. Wong, G.H.Y. Lin, of page charges. This article must therefore be hereby marked in T. Mutukura, E.L. Sievers, J. Koropatnick, S. Trudel, J.C.Y. Wang, R.A. Uger accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Development of methodology: P.S. Petrova, M. Wong, X. Pang, G.H.Y. Lin, K. Dodge, V. Chai, H. Chen, V. Lee, V. House, N.T. Vigo, T. Mutukura, Received July 5, 2016; revised October 19, 2016; accepted October 23, 2016; M. Charbonneau, T. Truong, S. Viau published OnlineFirst November 17, 2016.

1078 Clin Cancer Res; 23(4) February 15, 2017 Clinical Cancer Research

TTI-621 Is a Novel Antitumor Immune Checkpoint Inhibitor

References 1. Tsai RK, Discher DE. Inhibition of "self" engulfment through deactivation 20. Armour KL, Clark MR, Hadley AG, Williamson LM. Recombinant human of myosin-II at the phagocytic synapse between human cells. J Cell Biol IgG molecules lacking Fcgamma receptor I binding and monocyte trigger- 2008;180:989–1003. ing activities. Eur J Immunol 1999;29:2613–24. 2. Oldenborg PA, Zheleznyak A, Fang YF, Lagenaur CF, Gresham HD, Lind- 21. Angal S, King DJ, Bodmer MW, Turner A, Lawson AD, Roberts G, et al. A berg FP. Role of CD47 as a marker of self on red blood cells. Science single amino acid substitution abolishes the heterogeneity of chimeric 2000;288:2051–4. mouse/human (IgG4) antibody. Mol Immunol 1993;30:105–8. 3. Olsson M, Bruhns P, Frazier WA, Ravetch JV, Oldenborg PA. Platelet 22. Takenaka K, Prasolava TK, Wang JC, Mortin-Toth SM, Khalouei S, Gan OI, homeostasis is regulated by platelet expression of CD47 under normal et al. Polymorphism in Sirpa modulates engraftment of human hemato- conditions and in passive immune thrombocytopenia. Blood 2005;105: poietic stem cells. Nat Immunol 2007;8:1313–23. 3577–82. 23. Weiskopf K, Ring AM, Ho CC, Volkmer JP, Levin AM, Volkmer AK, et al. 4. Zhan F, Hardin J, Kordsmeier B, Bumm K, Zheng M, Tian E, et al. Global Engineered SIRPalpha variants as immunotherapeutic adjuvants to anti- gene expression profiling of multiple myeloma, monoclonal gammopathy cancer antibodies. Science 2013;341:88–91. of undetermined significance, and normal bone marrow plasma cells. 24. Jaiswal S, Chao MP, Majeti R, Weissman IL. Macrophages as mediators of Blood 2002;99:1745–57. tumor immunosurveillance. Trends Immunol 2010;31:212–9. 5. Jaiswal S, Jamieson CH, Pang WW, Park CY, Chao MP, Majeti R, et al. CD47 25. Bruhns P, Iannascoli B, England P, Mancardi DA, Fernandez N, Jorieux S, is upregulated on circulating hematopoietic stem cells and leukemia cells to et al. Specificity and affinity of human Fcg receptors and their polymorphic avoid phagocytosis. Cell 2009;138:271–85. variants for human IgG subclasses. Blood 2009;113:3716–25. 6. Majeti R, Chao MP, Alizadeh AA, Pang WW, Jaiswal S, Gibbs KDJr, et al. 26. Lin GHY, Chai V, Lee V, Dodge K, Truong T, Wong M, et al. SIRPaFc, a CD47 is an adverse prognostic factor and therapeutic antibody target on CD47-blocking cancer immunotherapeutic, triggers phagocytosis of lym- human acute myeloid leukemia stem cells. Cell 2009;138:286–99. phoma cells by both classically (M1) and alternatively (M2) activated 7. Chao MP, Alizadeh AA, Tang C, Myklebust JH, Varghese B, Gill S, et al. Anti- macrophages [abstract]. In: Proceedings of the AACR 107th Annual Meet- CD47 antibody synergizes with rituximab to promote phagocytosis and ing; 2016 Apr 16–20, 2016; New Orleans, LA; Philadelphia (PA): AACR; eradicate non-Hodgkin lymphoma. Cell 2010;142:699–713. 2016. Abstract nr 2345. 8. Chao MP, Alizadeh AA, Tang C, Jan M, Weissman-Tsukamoto R, Zhao F, 27. Theocharides AP, Jin L, Cheng PY, Prasolava TK, Malko AV, Ho JM, et al. et al. Therapeutic antibody targeting of CD47 eliminates human acute Disruption of SIRPalpha signaling in macrophages eliminates human lymphoblastic leukemia. Cancer Res 2011;71:1374–84. acute myeloid leukemia stem cells in xenografts. J Exp Med 2012;209: 9. Kim D, Wang J, Willingham SB, Martin R, Wernig G, Weissman IL. Anti- 1883–99. CD47 antibodies promote phagocytosis and inhibit the growth of human 28. Chao MP, Jaiswal S, Weissman-Tsukamoto R, Alizadeh AA, Gentles AJ, myeloma cells. Leukemia 2012;26:2538–45. Volkmer J, et al. Calreticulin is the dominant pro-phagocytic signal on 10. Pang WW, Pluvinage JV, Price EA, Sridhar K, Arber DA, Greenberg PL, et al. multiple human cancers and is counterbalanced by CD47. Sci Transl Med Hematopoietic stem cell and progenitor cell mechanisms in myelodys- 2010;2:63ra94. plastic syndromes. Proc Natl Acad Sci U S A 2013;110:3011–6. 29. Tseng D, Volkmer JP, Willingham SB, Contreras-Trujillo H, Fathman JW, 11. Nagahara M, Mimori K, Kataoka A, Ishii H, Tanaka F, Nakagawa T, et al. Fernhoff NB, et al. Anti-CD47 antibody-mediated phagocytosis of cancer Correlated expression of CD47 and SIRPA in bone marrow and in periph- by macrophages primes an effective antitumor T-cell response. Proc Natl eral blood predicts recurrence in breast cancer patients. Clin Cancer Res Acad Sci U S A 2013;110:11103–8. 2010;16:4625–35. 30. Mouro-Chanteloup I, Delaunay J, Gane P, Nicolas V, Johansen M, Brown 12. Edris B, Weiskopf K, Volkmer AK, Volkmer JP, Willingham SB, Contreras- EJ, et al. Evidence that the red cell skeleton protein 4.2 interacts with the Rh Trujillo H, et al. Antibody therapy targeting the CD47 protein is effective in membrane complex member CD47. Blood 2003;101:338–44. a model of aggressive metastatic leiomyosarcoma. Proc Natl Acad Sci U S A 31. Burger P, Hilarius-Stokman P, de KD, van den Berg TK, van BR. CD47 2012;109:6656–61. functions as a molecular switch for erythrocyte phagocytosis. Blood 13. Willingham SB, Volkmer JP, Gentles AJ, Sahoo D, Dalerba P, Mitra SS, et al. 2012;119:5512–21. The CD47-signal regulatory protein alpha (SIRPa) interaction is a thera- 32. Piccione EC, Juarez S, Tseng S, Liu J, Stafford M, Narayanan C, et al. SIRPa- peutic target for human solid tumors. Proc Natl Acad Sci U S A antibody fusion proteins selectively bind and eliminate dual antigen- 2012;109:6662–7. expressing tumor cells. Clin Cancer Res. 2016 Oct 15. [Epub ahead of print]. 14. Wang Y, Xu Z, Guo S, Zhang L, Sharma A, Robertson GP, et al. Intravenous 33. Subramanian S, Tsai R, Sen S, Dahl KN, Discher DE. Membrane mobility delivery of siRNA targeting CD47 effectively inhibits melanoma tumor and clustering of Integrin Associated Protein (IAP, CD47) - major differ- growth and lung metastasis. Mol Ther 2013;21:1919–29. ences between mouse and man and implications for signaling. Blood Cells 15. Cioffi M, Trabulo S, Hidalgo M, Costello E, Greenhalf W, Erkan M, et al. Mol Dis 2006;36:364–72. Inhibition of CD47 effectively targets pancreatic cancer stem cells via dual 34. Uger RA, Dodge K, Pang X, Petrova PS. Cancer immunotherapy targeting mechanisms. Clin Cancer Res 2015;21:2325–37. CD47: wild type SIRPaFc is the ideal CD47-blocking agent to minimize 16. Zhao XW, van Beek EM, Schornagel K, Van der Maaden H, Van HM, Otten unwanted erythrocyte binding [abstract]. In: Proceedings of the AACR MA, et al. CD47-signal regulatory protein-a (SIRPa) interactions form a Annual Meeting; 2014 Apr 5–9; San Diego, CA; Philadelphia (PA): AACR; barrier for antibody-mediated tumor cell destruction. Proc Natl Acad Sci 2014. Abstract nr 5011. U S A 2011;108:18342–7. 35. Oostendorp M, Lammerts van Bueren JJ, Doshi P, Khan I, Ahmadi T, Parren 17. Steinert G, Scholch S, Niemietz T, Iwata N, Garcia SA, Behrens B, et al. PWHI, et al. When blood transfusion medicine becomes complicated due Immune escape and survival mechanisms in circulating tumor cells of to interference by monoclonal antibody therapy. Transfusion 2015;55: colorectal cancer. Cancer Res 2014;74:1694–704. 1555–62. 18. Chao MP, Tang C, Pachynski RK, Chin R, Majeti R, Weissman IL. Extranodal 36. Chapuy CI, Nicholson RT, Aguad MD, Chapuy B, Laubach JP, Richardson dissemination of non-Hodgkin lymphoma requires CD47 and is inhibited PG, et al. Resolving the daratumumab interference with blood compati- by anti-CD47 antibody therapy. Blood 2011;118:4890–901. bility testing. Transfusion 2015;55:1545–54. 19. Liu J, Wang L, Zhao F, Tseng S, Narayanan C, Shura L, et al. Pre-Clinical 37. Liu X, Pu Y, Cron K, Deng L, Kline J, Frazier WA, et al. CD47 blockade development of a humanized anti-CD47 antibody with anti-cancer ther- triggers T cell-mediated destruction of immunogenic tumors. Nat Med apeutic potential. PLoS ONE 2015;10:e0137345. 2015;21:1209–15.

www.aacrjournals.org Clin Cancer Res; 23(4) February 15, 2017 1079

TTI-621 (SIRPαFc): A CD47-Blocking Innate Immune Checkpoint Inhibitor with Broad Antitumor Activity and Minimal Erythrocyte Binding

Penka S. Petrova, Natasja Nielsen Viller, Mark Wong, et al.

Clin Cancer Res 2017;23:1068-1079. Published OnlineFirst November 17, 2016.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-16-1700

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2016/11/16/1078-0432.CCR-16-1700.DC1

Cited articles This article cites 34 articles, 20 of which you can access for free at: http://clincancerres.aacrjournals.org/content/23/4/1068.full#ref-list-1

Citing articles This article has been cited by 11 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/23/4/1068.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/23/4/1068. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.