Development of new antibody-based cancer therapies

点击翻译按钮获取中文摘要

Heparan sulfate proteoglycans (HSPGs) regulate numerous cell surface signaling events. They are extracellular modulators of signal transduction pathways during development and disease. HSPGs are cell-surface proteins that mainly consist of glycosylphosphatidylinositol (GPI)-anchored glypicans and transmembrane syndecans. Several HSPGs are currently being evaluated as potential targets for cancer therapy because of their relatively high expression in certain tumor types. In recent years, we have studied glypicans as a new family of cancer targets [reviewed by Li et al. Trends in Cancer, PMID: 30352677, 2019]. Glypican-3 (GPC3) is a new therapeutic target in hepatocellular carcinoma (HCC), the most common form of primary liver cancers. We produced several antibodies targeting GPC3 either by hybridoma and phage display technologies. To isolate high affinity antibodies (e.g. YP7) to the native form of cell surface antigens such as GPC3, we developed a new high-throughput method combining functional cell binding screening by flow cytometry and conventional hybridoma technology [Phung et al., MAbs, PMID 22820551, 2012]. Furthermore, we have developed a new approach to humanize non-human antibodies (including mouse and rabbit antibodies) for clinical development in FY2017-2018 [Zhang and Ho, Scientific Reports, 2016; Zhang and Ho, MAbs, 2017]. In FY2019, we created antibody-drug conjugates (ADCs) using humanized YP7 (hYP7) and showed single treatment of hYP7 ADC induced tumor regression in multiple mouse models [Fu et al. Hepatology, PMID: 30353932, 2018]. In addition, we used phage display technology to generate two human monoclonal antibodies (HN3 and HS20). HN3 is a human single-domain antibody that recognizes a novel functional site in the core protein of GPC3 and inhibits proliferation of HCC cells via blocking Wnt and Yap cancer signaling [Feng et al., PNAS, PMID: 23471984, 2013; Gao et al., Nature Communications, PMID: 25758784, 2015]. In FY2019, using computational modeling and gene editing technology, we have established a structural model of GPC3 containing a putative cysteine-rich domain at its N-terminal lobe and published our finding in Hepatology [Li et al. Hepatology, PMID: 30963603, 2019]. F41 and its surrounding residues in GPC3 forms a Wnt-binding groove that interacts with the middle region located between the lipid thumb domain and the index finger domain of Wnt3a. Mutating residues in this groove significantly inhibits Wnt3a binding, beta-catenin activation, and the transcriptional activation of Wnt-dependent genes. Specifically, blocking this domain using an antibody (HN3) inhibits Wnt activation. In HCC cells, mutating residue F41 on GPC3 inhibits activation of beta-catenin in vitro and reduced xenograft tumor growth in mice compared with cells expressing wild-type GPC3. Our investigation demonstrates a detailed interaction of GPC3 and Wnt3a, reveals the precise mechanism of GPC3 acting as a Wnt coreceptor, and provides a potential target site on GPC3 for Wnt blocking and HCC therapy. HS20 recognizes the heparan sulfate chains of GPC3. The human antibody disrupts the interaction of Wnt3a and GPC3 and inhibits Wnt/beta-catenin signaling [Gao et al., Hepatology, PMID: 24492943, 2014; Gao et al., PLoS One, PMID: 26332121, 2016; Gao et al., Scientific Reports, PMID: 27185050, 2016]. Our antibodies exhibit significant inhibition of HCC xenograft tumor growth in mice and show potential for use as therapeutic candidates. In addition, we found that GPC3 was efficiently internalized from the cell surface and that the HN3-PE38 immunotoxin brought the toxin into the cell, resulting in inhibition of protein synthesis. The immunotoxin caused regression of liver cancer in mice. Interestingly, Its novel mechanism involved both inhibition of cancer signaling (Wnt/Yap) and reduction in protein synthesis [Gao et al. Nature Communications, PMID: 25758784, 2015]. Our strategy combining both antibody and toxin functions could be applicable generally to other immunotoxins and antibody-toxin/drug conjugates. To pursue clinical development of our anti-GPC3 immunotoxin for the treatment of liver cancer, we generated a new version of the anti-GPC3 immunotoxin (HN3-mPE24) and found that the second generation greatly reduced side effects and had better anti-tumor activity [Wang et al., Oncotarget, 2017]. In addition to the immunotoxin therapy, along with our collaborators, we used our anti-GPC3 antibodies to construct various clinical formats for targeted therapy of liver cancer including chimeric antigen receptor (CAR) T cell immunotherapy and photoimmunotherapy [Hanaoka et al. Mol Pharm, 2015; Hanaoka et al. Nanomedicine, 2015]. We filed a patent application regarding our CAR T cells targeting GPC3 for treating liver cancer through the NCI Technology Transfer office in FY2019. In addition to GPC3, we reported GPC2 as a new therapeutic target in pediatric cancers in particular neuroblastoma [Li et al., PNAS, 2017] and in FY2019, we presented GPC1 as a potential target in pancreatic cancer in the 2019 AACR annual meeting. In the mesothelin project, we previously used rabbit monoclonal antibody technology to identify a panel of high affinity antibodies that bind novel sites in mesothelin. We have humanized one of the best candidates (YP218) for the treatment of mesothelioma and other mesothelin-positive cancers [Zhang et al., Scientific Reports, 2015; Zhang and Ho, MAbs, 2017]. In FY2019, we reported the creation of a phage displayed nanobody library [Feng et al. Antibody Therapeutics, PMID: 30627698, 2019] and isolated a panel of cross-species nanobodies from our nanobody libraries that bind human and mouse mesothelin and filed a patent application for their clinical applications for treating mesothelioma, lung cancer, pancreatic cancer, ovarian cancer and other cancers through the NCI Technology Transfer Office.