Hangil Park

Monoclonal Antibodies to Fibroblast Growth Factor Receptor 2 Effectively Inhibit Growth of Gastric Tumor Xenografts

Purpose: Overexpression of fibroblast growth factor receptor 2 (FGFR2) may be a causative factor of a number of human tumors, especially gastric tumors of the poorly differentiated type. We investigated whether monoclonal antibodies (mAbs) directed against FGFR2 can inhibit the growth of tumors in xenograft models. Experimental Design: We generated and characterized 3 mAbs that recognize different epitopes on FGFR2: GAL-FR21, GAL-FR22, and GAL-FR23. The ability of the mAbs to recognize the FGFR2IIIb and FGFR2IIIc isoforms of FGFR2 was determined, as was their ability to block binding of FGF ligands to FGFR2. The capability of the mAbs to inhibit FGF-induced FGFR2 phosphorylation and to downmodulate FGFR2 expression was also investigated. Finally, the ability of the anti-FGFR2 mAbs to inhibit tumor growth was determined by establishing xenografts of SNU-16 and OCUM-2M human gastric tumor cell lines in nude mice, treating with each mAb (0.5–5 mg/kg intraperitoneally twice weekly) and monitoring tumor size. Results: Of the 3 mAbs, GAL-FR21 binds only the FGFR2IIIb isoform, whereas GAL-FR22 and GAL-FR23 bind to both the FGFR2IIIb and FGFR2IIIc forms, with binding regions respectively in the D3, D2-D3, and D1 domains of FGFR2. GAL-FR21 and GAL-FR22 blocked the binding of FGF2, FGF7 and FGF10 to FGFR2IIIb. GAL-FR21 inhibited FGF2 and FGF7 induced phosphorylation of FGFR2, and both mAbs downmodulated FGFR2 expression on SNU-16 cells. These mAbs effectively inhibited growth of established SNU-16 and OCUM-2M xenografts in mice. Conclusions: Anti-FGFR2 mAbs GAL-FR21 and GAL-FR22 have potential for the treatment of gastric and other tumors. Clin Cancer Res; 16(23); 5750–58. ©2010 AACR.

Abstract 3256: Highly potent monoclonal and bispecific anti-angiogenic antibodies to VEGF and angiopoietin-2

The angiogenic factor Vascular Endothelial Growth Factor (VEGF) is a well-established target for cancer therapy. The humanized monoclonal antibody (mAb) bevacizumab (Avastin®), which binds and neutralizes VEGF, is widely prescribed for the treatment of colon, lung and certain other tumors, but typically provides a relatively modest survival benefit of several months. It is unknown whether the limited benefit is due to redundancy in angiogenic pathways or to the inability of bevacizumab at the doses used to fully neutralize VEGF in the tumor microenvironment. However, the exploration of higher doses of bevacizumab in the clinical setting has been very limited due to lack of incentive to perform dose-ranging studies for an already approved drug, the expense of the drug, and concern about toxicity. To determine whether more effective neutralization of VEGF would be possible and useful, we have therefore used an intensive immunization protocol to generate a new murine mAb designated VE1 that binds VEGF with very high affinity. A humanized form of this mAb, HuVE1, retains the full affinity of VE1 and binds VEGF about 5-fold better than bevacizumab. In side-by-side comparisons, HuVE1 also blocked the binding of VEGF to its receptor with an IC50 that is 5 to 10-fold lower than for bevacizumab, and it correspondingly inhibited VEGF-induced proliferation of HUVEC about 5-fold better than bevacizumab. HuVE1 almost completely blocked growth of primary human hepatocellular carcinoma (HCC) xenografts in one model, and showed a trend toward better efficacy than bevacizumab in a second HCC xenograft model. For potentially even greater anti-angiogenic efficacy, we developed a humanized mAb HuA2T that potently neutralizes angiopoietin-2 (Ang-2), a cytokine that acts through the Tie-2 receptor to promote angiogenesis, especially in combination with VEGF. The synergy between VEGF and Ang-2 provides a strong rationale to develop a bispecific mAb that neutralizes both these factors. We therefore developed such a mAb using the IgG-like bispecific antibody format Bs(scFv)4-IgG consisting of a homodimer of two monomers, each monomer having a single chain Fv from HuVE1 linked to a light chain constant region, and a single chain Fv from HuA2T linked to a heavy chain constant region. While, as is common in construction of bispecific antibodies, this bispecific HuA2T/HuVE1 mAb did lose some binding affinity for VEGF and Ang-2 relative to the original mAbs, its VEGF binding activity was still greater than that of bevacizumab, illustrating the importance of starting with an extremely high activity mAb. Based on its potency in vitro and in animal models, we believe that HuVE1, in either natural or bispecific form, has the potential for greater clinical efficacy than bevacizumab and thus merits further investigation. Citation Format: Hangil Park, April Zhang, Yi Ding, Lihong Wang, Zhengran Wang, Maximiliano Vasquez, Cary Queen, Jin Kim. Highly potent monoclonal and bispecific anti-angiogenic antibodies to VEGF and angiopoietin-2. [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 3256.

Abstract 3493: New, highly potent antibodies to death receptors having Fc mutations to increase antitumor activity

Death receptors DR4 (TRAIL-R1) and DR5 (TRAIL-R2 or Apo2) are TNF receptor superfamily members that are receptors for TRAIL (Apo2 ligand), an immunomodulatory cytokine. Binding of TRAIL to DR4 or DR5 can activate the extrinsic apoptotic pathway selectively in tumor cells. Agonist antibodies to death receptors DR4 and DR5 thus have potential for treatment of cancer and may be better than derivatives of TRAIL itself because of their superior pharmacokinetics and receptor selectivity. However, despite dramatic effects in vitro and in preclinical xenograft models, several agonist antibodies to DR4 and especially DR5 have not provided significant therapeutic benefit in clinical trials. While resistance mechanisms in cancer cells are undoubtedly one reason, another factor may be inadequate potency of the antibodies tested. We therefore used intense immunization and screening protocols to develop very potent anti-DR4 and anti-DR5 monoclonal antibodies (mAbs), which were humanized to make the mAbs denoted HuD114 and HuG4.2 respectively. These mAbs were substantially more effective at killing tumor cells in vitro than the mAbs that have been tested in clinical trials. In addition, we introduced one or two mutations into the Fc (constant) regions of these and other potent anti-DR4 and anti-DR5 mAbs in order to increase affinity for the Fc gamma receptor IIb. Cross-linking, as provided by binding of anti-DR4 and anti-DR5 mAbs to Fc gamma receptors on immune cells, is required for effective transmission of an apoptotic signal through the death receptors. Accordingly, introduction of these mutations greatly increased the ability of the mAbs to kill tumor cells in vitro in the presence of human peripheral blood mononuclear cells, and to inhibit the growth of tumor xenografts in mouse models, with two mutations generally more effective than a single mutation. As a complementary approach to increase cross-linking, we also developed bispecific antibodies containing two binding domains from an anti-DR4 mAb and two binding domains from an anti-DR5 mAb. These bispecific mAbs were made in the IgG-like Bs(scFv)4-IgG format consisting of a homodimer of two monomers, each monomer having a single chain Fv from an anti-DR4 mAb linked to a heavy chain constant region, and a single chain Fv from an anti-DR5 mAb linked to a light chain constant region. Such a bispecific mAb was more effective at killing tumor cells than an anti-DR4 or anti-DR5 mAb alone, or even a mixture of anti-DR4 and anti-DR5 mAbs. Based on their potency in vitro and in animal models, we believe that HuD114 and HuG4.2, enhanced by the Fc mutations and/or in bispecific form, have the potential for greater clinical efficacy than previously tested anti-DR4 and anti-DR5 mAbs and thus merit further investigation. Citation Format: Lihong Wang, Yi Ding, Hangil Park, April Zhang, Zhengran Wang, Maximiliano Vasquez, Cary Queen, Jin Kim. New, highly potent antibodies to death receptors having Fc mutations to increase antitumor activity. [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 3493.

Abstract 5056: HuGAL-FR21, a humanized monoclonal antibody to Fibroblast Growth Factor Receptor 2, effectively inhibits the growth of gastric tumor xenografts

Proceedings: AACR 102nd Annual Meeting 2011‐‐ Apr 2‐6, 2011; Orlando, FLnnThe family of 22 fibroblast growth factors include potent mitogenic and angiogenic factors believed to play a role in the growth of many tumors. The various FGFs bind to only four tyrosine kinase signaling receptors, Fibroblast Growth Factor Receptors 1-4 (FGFR1-4). FGFR2 is a major cellular receptor for FGF2 (basic FGF), FGF7 (keratinocyte growth factor, KGF) and FGF10; and mutation or overexpression of FGFR2 is associated with a number of cancers, notably endometrial cancer and gastric carcinoma. In this study, we explored the therapeutic potential of the HuGAL-FR21 humanized monoclonal antibody (mAb) derived from our blocking anti-FGFR2 mAb, GAL-FR21.nnWe previously showed that GAL-FR21 binds to only the FGFR2IIIb isoform of FGFR2, which is expressed on epithelial tissues and thus more commonly on carcinomas. Using well-established humanization techniques, GAL-FR21 was humanized to create HuGAL-FR21, which binds human FGFR2IIIb with at least as high affinity as GAL-FR21. HuGAL-FR21, like GAL-FR21, also bound well to monkey and mouse FGFR2IIIb, facilitating study of this mAb in animal models. In functional assays, HuGAL-FR21 inhibited binding of FGF2, FGF7 and FGF10 to FGFR2IIIb as completely as GAL-FR21. And HuGAL-FR21 inhibited FGF7 or FGF10-induced proliferation of the Ba/F3 cells transfected with FGFR2IIIb, actually better than GAL-FR21. Finally, HuGAL-FR21 inhibited phosphorylation of cellular FGFR2 induced by FGF7 and FGF10 as well as GAL-FR21.nnThere are two major types of gastric cancer, well-differentiated and poorly differentiated, and published results show that many poorly differentiated tumors overexpress FGFR2, which is in fact the oncogenic driver for these tumors. We therefore tested the ability of HuGAL-FR21 to inhibit growth of xenografts of two cell lines, SNU-16 and OCUM-2M, derived from FGFR2-overexpressing, poorly differentiated gastric tumors. HuGAL-FR21, administered i.p. at 2.5 or 5 mg/kg twice per week, strongly inhibited the growth of established xenografts of both these cell lines. We conclude that FGFR2 is a promising molecular target, and that HuGAL-FR21 is a viable clinical candidate for the treatment of gastric and possibly other cancers.nnCitation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 5056. doi:10.1158/1538-7445.AM2011-5056