Table S1: KTN0158 Dosing Cohorts Table S2: Summary of Anti-KTN0158 Antibody Titers in Dogs with Mast Cell Tumors Following KTN0158 Treatment

Legends for Supplementary Figures S1-S4 and Supplementary Table S1.

Cross-species reactivity of anti-MET antibodies.

Anti-MET antibodies demonstrated no agonistic activities in proliferation assay (S1); Sequence alignment of Sema domain β-subunit of human, monkey, canine and murine MET receptors (S2); Effect of KTN0073-IgG1, IgG2 and IgG4 antibodies on HGF-dependent MET phosphorylation in U87MG cell (S3); Differential effect of IgG2 isotype switching on KTN0070, KTN0071 and KTN0075 (S4).

Serum KTN0158 concentration over time

Description of additional methods and procedures used in the study. Also includes Supplementary References.

Abstract Purpose: KTN0158 is a novel anti-KIT antibody that potently inhibits wild-type and mutant KIT. This study evaluated the safety, biologic activity, and pharmacokinetic/pharmacodynamics profile of KTN0158 in dogs with spontaneous mast cell tumors (MCT) as a prelude to human clinical applications. Experimental Design: Cell proliferation, KIT phosphorylation, and mast cell degranulation were evaluated in vitro. KTN0158 was administered to 4 research dogs to assess clinical effects and cutaneous mast cell numbers. Thirteen dogs with spontaneous MCT were enrolled into a prospective phase I dose-escalating open-label clinical study of KTN0158 evaluating 3 dose levels and 2 schedules and with weekly assessments for response and clinical toxicities. Results: KTN0158 was a potent inhibitor of human and dog KIT activation and blocked mast cell degranulation in vitro. In dogs, KTN0158 was well tolerated and reduced cutaneous mast cell numbers in a dose-dependent manner. Clinical benefit of KTN0158 administration in dogs with MCT (n = 5 partial response; n = 7 stable disease) was observed regardless of KIT mutation status, and decreased KIT phosphorylation was demonstrated in tumor samples. Histopathology after study completion demonstrated an absence of neoplastic cells in the primary tumors and/or metastatic lymph nodes from 4 dogs. Reversible hematologic and biochemical adverse events were observed at doses of 10 and 30 mg/kg. The MTD was established as 10 mg/kg. Conclusions: KTN0158 inhibits KIT phosphorylation, demonstrates an acceptable safety profile in dogs, and provides objective responses in canine MCT patients with and without activating KIT mutations, supporting future clinical evaluation of KTN0158 in people. Clin Cancer Res; 23(10); 2565–74. ©2016 AACR.

Head and neck squamous cell carcinoma (HNSCC) accounts for 3-5% of all tumor types and remains an unmet medical need with only two targeted therapies approved to date. ErbB3 (HER3), the kinase-impaired member of the EGFR/ErbB family, has been implicated as a disease driver in a number of solid tumors, including a subset of HNSCC. Here we show that the molecular components required for ErbB3 activation, including its ligand neuregulin-1 (NRG1), are highly prevalent in HNSCC and that HER2, but not EGFR, is the major activating ErbB3 kinase partner. We demonstrate that cetuximab treatment primarily inhibits the ERK signaling pathway and KTN3379, an anti-ErbB3 monoclonal antibody, inhibits the AKT signaling pathway, and that dual ErbB receptor inhibition results in enhanced anti-tumor activity in HNSCC models. Surprisingly, we found that while NRG1 is required for ErbB3 activation, it was not sufficient to fully predict for KTN3379 activity. An evaluation of HNSCC patient samples demonstrated that NRG1 expression was significantly associated with expression of the EGFR ligands amphiregulin (AREG) and transforming growth factor α (TGFα). Furthermore, NRG1-positive HNSCC cell lines that secreted high levels of AREG and TGFα or contained high levels of EGFR homodimers (H11D) demonstrated a better response to KTN3379. Although ErbB3 and EGFR activation are uncoupled at the receptor level, their respective signaling pathways are linked through co-expression of their respective ligands. We propose that NRG1 expression and EGFR activation signatures may enrich for improved efficacy of anti-ErbB3 therapeutic mAb approaches when combined with EGFR-targeting therapies in HNSCC.

Abstract Activated ALK by mutation or amplification is a validated therapeutic target in a subset of patients with high-risk neuroblastoma, and tyrosine kinase inhibitors are being developed to abrogate ALK signaling (Bresler et al, 2014). Native ALK is expressed on the surface of the majority of neuroblastoma tumors, but not on normal tissue, giving it properties of a tumor antigen. We hypothesized that ALK-targeted antibodies may be useful in neuroblastoma as single-agent immunotherapy or in combination with small-molecule ALK inhibitors - especially where mutations reduce kinase inhibitor sensitivity. In this work, an anti-ALK monoclonal antibody (anti-ALK2) was conjugated to a cytotoxic agent (ALK2-ADC) and its anti-tumor activity was investigated in a panel of extensively characterized human neuroblastoma cell lines (n = 10) harboring wild type and mutant ALK. Cell surface ALK was quantified using flow cytometry and revealed differential antigen expression. Using in vitro growth inhibition assays, there was evidence for a dose-dependent cytotoxicity to neuroblastoma cells at subnanomolar concentrations that correlated with ALK expression and was independent of underlying mutation status. In order to evaluate the therapeutic potential of ALK2-ADC in vivo, mice bearing Felix-patient-derived xenograft (PDX) tumors, containing the third most common ALK mutation (R1245C), were used to demonstrate that ALK2-ADC led to a significant reduction in tumor growth compared to unconjugated antibody and a control ADC (p<0.0001). Although ALK2-ADC binds potently to mouse ALK as well as human ALK, doses of up to 10 mg/kg in the rodent model appeared to be well tolerated with no overt toxicity noted. Additional in vivo studies are ongoing to assess the antitumor potential of the ADC in a broader range of ALK-expressing neuroblastoma models. Thus, targeting human ALK-expressing with an ALK antibody-drug conjugate demonstrated early signs of favorable efficacy and tolerability supporting future development of this approach as a potential novel therapeutic for patients with neuroblastoma. Citation Format: Renata Sano, Kateryna Krytska, Colleen Larmour, Gwenda F. Ligon, Jay S. Lillquist, Ulisse Cucchi, Paolo Orsini, Simona Rizzi, Gerald McMahon, Diego Alvarado, Yael P. Mosse. A novel antibody-drug conjugate directed to the ALK receptor tyrosine kinase demonstrates efficacy in models of neuroblastoma. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 2690.

MET gene amplification and exon 14 mutations have been reported and implicated as oncogenic drivers in several types of cancer representing approximately 5-7% of NSCLC, 4-10% of gastric cancer, and 5-10% of colon cancers. Preclinical research and clinical data derived from patients with tumors containing MET gene amplification and exon 14 mutations showed significant growth inhibition and meaningful clinical benefit after treatment with MET-targeting tyrosine kinase inhibitors (TKIs), supporting the hypothesis that MET gene amplification and exon 14 mutations are actionable biomarkers. We have identified several anti-MET monoclonal antibodies that block MET-dependent signaling and tumor growth through: (1) binding to the Sema/PSI domain, (2) preventing HGF interaction with MET and (3) inducing receptor ubiqutination and degradation. One of the anti-MET antibodies, KTN0073, induces potent degradation of oncogenic exon 14 mutant MET, which has been reported to propagate prolonged MET signaling due to a defect in receptor degradation. Our data suggest that KTN0073 engages a pathway different from HGF-induced receptor degradation and protease-mediated shedding, which is independent of the 47 amino acids that are encoded by exon 14 in the intracellular juxtamembrane region and only partially inhibited by the TKI crizotinib. In summary, we report the use of an anti-MET monoclonal antibody to elucidate an exon 14-independent mechanism of MET degradation that could provide a novel way to design anti-MET therapeutics to inhibit growth of human tumors that are driven by HGF, MET amplification and exon 14 mutations. Based on this work, KTN0073-IgG2 anti-MET antibody has been humanized for further preclinical development to explore its potential as a therapeutic for the treatment of human MET expressing cancers. Citation Format: Brett S. Robinson, Sreekala Mandiyan, Gerald McMahon, Yan Yang. An anti-MET IgG2 monoclonal antibody degrades both wild-type and exon 14 mutant MET receptor tyrosine kinase through a novel exon 14-independent mechanism and inhibits tumor growth. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 3835.

Abstract A sound rationale exists for antibody targeting of the MET receptor tyrosine kinase, but therapeutic agents that can broadly block HGF ligand binding and exon 14–mutated or amplified MET to induce receptor degradation have yet to be reported. Here we report the identification of several MET monoclonal antibodies (mAb) that block MET-dependent signaling and tumor growth. In particular, the MET mAb KTN0073 and KTN0074 bind the Sema/PSI domain, at overlapping but distinct epitopes, preventing HGF interaction with MET and triggering receptor ubiquitination and degradation. Notably, both mAbs also triggered degradation of oncogenic MET exon 14 mutants, which propagate more durable MET signals due to a defect in receptor degradation. Mechanistic investigations showed that both mAbs engaged a pathway distinct from HGF-induced receptor degradation and protease-mediated shedding, independently of signaling driven by the exon 14–encoded sequences in the intracellular juxtamembrane region of the MET receptor. Grafting the mAb variable regions onto the IgG2 constant region dramatically enhanced the tumor inhibitory activities of KTN0073 but not KTN0074, suggesting a specific influence of antibody isotype of the epitopes for these two MET mAbs. Overall, our results highlight KTN0073 as a novel IgG2-based MET mAb that acts through exon 14–independent mechanisms to degrade the MET receptor, potentially offering a therapeutic tool to treat a broader range of human tumors where MET is exon 14 mutated or amplified. Cancer Res; 76(19); 5788–97. ©2016 AACR.

Abstract Aberrant activation of MET kinase can be mediated by ligand-dependent mechanisms through paracrine or autocrine stimulation by its ligand, HGF, or by ligand-independent mechanisms through gene amplification, over-expression or mutation. Both mechanisms of activating MET have been associated with promoting tumor growth and progression. The discovery of antagonistic MET antibodies has been challenging as many anti-MET antibodies promote kinase activation that leads to cell proliferation and migration. We have systematically screened MET antibodies using cell based functional assays and identified a number of antibodies that potently block MET-dependent cell proliferation, cell scattering as well as tumor growth in vivo. One such antibody, KTN-MET-IgG1, binds to the Sema/PSI domain and prevents HGF-MET interaction; additionally it induces receptor ubiqutination and degradation. To evaluate if the IgG isotype influences activity on MET activation and tumor growth we grafted the binding domains of this antibody onto 2 other IgG backbones: KTN-MET-IgG2 and KTN-MET-IgG4. Surprisingly, grafting the antibody variable regions onto the IgG2 framework dramatically enhanced the anti-MET activity in vitro and in vivo. Comparing to KTN-MET-IgG1 and KTN-MET-IgG4 antibodies with KTN-MET-IgG2 antibody, it demonstrated significantly better inhibition of receptor phosphorylation and cell proliferation in MET amplified tumor cell lines (EBC1, SNU5) and comparable activity in HGF-dependent tumor cell lines (A549, U87MG). KTN-MET-IgG2 treatment of tumor bearing mice achieved potent and prolonged suppression of tumor growth using the U87MG glioblastoma xenograft model and tumor growth regression with tumors remaining below detection for over 90 days post last dosing using the SNU5 gastric cancer xenograft model. In summary, we have found a novel IgG2 anti-MET antibody that inhibits tumor growth driven by HGF-dependent MET activation and MET amplification. A model to help explain the contribution of IgG2 isotype to enhanced blocking activity will be presented. The humanized version of KTN-MET-IgG2, KTN0216, is currently in preclinical development to evaluate tumor types best suited for KTN0216 treatment. Citation Format: Sreekala Mandiyan, Brett S. Robinson, Lida Kimmel, Gerald McMahon, Yaron Hadari, Yan Yang. Blocking activity of a novel anti-MET humanized monoclonal antibody, KTN0216, is enhanced by IgG2 isotype in HGF-dependent and Met-amplified tumors. [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 1696. doi:10.1158/1538-7445.AM2015-1696