Molecular oncology. Causes of cancer and targets for treatment. Eds Edward P. Gelmann et al., Cambridge University Press (2014)
The prevalence of HGF/Met pathway activation in human malignancies has driven the rapid development of HGF/Met pathway inhibitors, which can be broadly subdivided into biological (protein-based) agents and low-molecular-weight synthetic compounds. Biologicals usually possess target selectivity and pharmacokinetic properties that are predictable and desirable, but their size and complexity impact drug manufacture, mode of administration, and shelf life; thus, low-molecularweight TK inhibitors (TKIs) presently outnumber biologicals as HGF/Met therapeutics (40; Figure 17.1 and Table 17.1). We summarize below pathway inhibitors currently in human clinical trials for cancer.
Biological HGF/Met pathway antagonists
Monoclonal antibodies (mAbs) directed against either HGF or Met block ligand-receptor binding, Met activation, and biological responses.
mAbs directed against HGF
Rilotumumab is a fully human mAb directed against the mature HGF light chain, thereby potently blocking HGF-Met binding (41). Rilotumumab is under evaluation as monotherapy in Phase Ib/II trials for ovarian and renal cancer, and in combination with avastin in glioma, erlotinib in NSCLC, and platinum-based chemotherapy in SCLC, mesothelioma, and gastric cancer, as well as mitoxantrone in prostate cancer. Rilotumumab monotherapy did not show significant anti-tumor activity in patients with recurrent glioblastoma who had previously received bevacizumab compared with bevacizumab-naive patients (42). Rilotumumab combined with panitumumab in patients with wild-type KRAS metastatic colorectal cancer showed an increased response rate over panitumumab alone (43). Rilotumumab in combination with epirubicin, cisplatin, and capecitabine (ECX) as a first-line treatment for metastatic gastric cancer showed improved progression-free (PFS) and overall survival (OS) over ECX alone, especially in patients with high Met expression (NCT00719550; 40), prompting an ongoing randomized, placebo-controlled Phase III study of rilotumumab in this indication.
Figure 17.1. HGF and Met domain structures and routes to antagonize the HGF/Met pathway. (a) Schematic representation of the domain structure of HGF protein isoform 1 (728 residues). The signal peptide (SP; residues 1–31) is cleaved from the mature two-chain protein. The amino-terminal α–chain contains the heparan sulfate (HS)-binding domain (N) and four kringle domains (K1-K4), the first of which contains the primary Met binding site. The β-chain contains a serine protease-like domain and a secondary Met binding site. Gray areas between named domains represent structurally undefined regions. The lengths of all regions are proportional to their sequence length. b) Schematic of Met domain structure; domain lengths are proportional to sequence length. Mature Met is a disulfide-linked two-chain heterodimer with an extra-cellular amino terminal -chain (45 kDa) and a carboxyl terminal -chain (145 kDa) containing extra-cellular, transmembrane (TM), and intra-cellular domains. The signal peptide (SP) is cleaved from mature protein. The extra-cellular domain contains a Sema homology region organized in seven blades; a cysteine-rich region (PSI); and four immunoglobulin-like repeats (Ig-like). The intra-cellular domain contains juxtamembrane (JM), tyrosine kinase (TK), and carboxyl terminal (CT) domains. Within the TK domain are the ATP binding site (orange), catalytic loop (cat, yellow), activation loop (act, red) and p+1 loop (p+1, green). (b) Three primary routes of pathway intervention have been followed as Met anti-cancer drug development strategies: (i) monoclonal antibodies that neutralize HGF-Met binding; (ii) a monovalent monoclonal antibody that also blocks Met-HGF binding; and (iii) tyrosine kinase inhibitors classified as ATP competitive (TKI1 ) or allosteric (TKI2 ). The major proximal effectors of HGF/Met signaling are shown recruited to Met carboxyl-terminal tyrosyl residues 1349 and 1356 that are phosphorylated upon HGF binding.
Ficlatuzumab (NCT00969410, NCT01039948; 40) and TAK-701 (44) are humanized anti-HGF mAbs now in Phase I and II trials; both potently block HGF-Met binding. These mAbs are well tolerated, show dose-proportional pharmacokinetics, and, like rilotumumab, reduce free plasma HGF to undetectable levels. Interim results suggest that the combination of ficlatuzumab and gefitinib has clinical activity in Asian patients with unresectable NSCLC (45).
mAbs directed against Met
Onartuzumab is a monovalent anti-Met mAb that blocks HGF binding (46). A Phase II study evaluating onartuzumab or placebo in combination with erlotinib showed improved outcome in patients with Met-positive, advanced-stage NSCLC (NCT01456325; 40). Patients whose tumors had high levels of Met protein as determined by IHC that were treated with onartuzumab plus erlotinib showed significantly improved PFS, OS, and nearly threefold
reduction in risk of death over those treated with erlotinib alone, prompting a Phase III trial for Metpositive advanced NSCLC patients (NCT01456325; 40). Onartuzumab is also in randomized double-blinded Phase II trials in combination with paclitaxel and bevacizumab in triple-negative breast cancer or in combination with folinic acid, oxaliplatin, and fluorouracil (FOLFOX) and bevacizumab in colorectal carcinoma (NCT01418222; 40).
Table 17.1. Experimental and US FDA-approved therapeutics targeting the HGF/Met signaling pathway in cancer
* Status: Active, NLR: active, but no longer recruiting; Active, NYR: active, but not yet recruiting.
** Trials of Crizotinib (PF02341066) where ALK, but not MET, involvement is indicated are not included.
Small synthetic Met kinase inhibitors
Most Met TKIs competitively antagonize occupancy of the intra-cellular ATP binding site, preventing TK activation and downstream signaling. They are discussed here starting with early inhibitors now in Phase III trials, which target multiple TKs including Met, and ending with the most Met-selective candidates now entering Phase I rials.
Cabozantinib, now in Phase III trials, targets primarily Met, VEGFR2, and Ret. Cabozantinib was recently US FDA-approved (47) for the treatment of progressive metastatic medullary thyroid cancer based on improved PFS observed in an international, multi-center, randomized (2:1), placebo-controlled trial enrolling 330 patients. Median PFS was 11.2 vs. 4.0 months for the cabozantinib and placebo arms, respectively. Overall response rate was significantly higher in the cabozantinib arm (27 vs. 0%) and all responses were partial (48). In a Phase II study of patients with metastatic castrationresistant prostate cancer, cabozantinib treatment reduced or stabilized soft-tissue lesions, bone metastases, bone pain, and narcotic use (NCT01599793; 40). Cabozantinib is also in trials for the treatment of breast, hepatocellular, melanoma, NSCLC, ovarian (49), brain (50), and kidney cancers (51).
Crizotinib potently inhibits Met and anaplastic lymphoma kinase (ALK) TKs (52). Crizotinib is highly effective against the activated products of ALK gene translocations (most frequently EML4-ALK) that occur in a subset of NSCLC patients (53), and has been US FDA-approved to treat that group on the basis of a companion diagnostic test for ALK rearrangement. Other active efficacy trials of crizotinib target NSCLC (not restricted to ALK translocations), type 1 papillary renal cancer, alveolar soft part sarcoma, rhabdomyosarcoma, glioma, and inflammatory myofibroblastic tumor, where Met and/or ALK pathways are thought to be involved (Table 17.1).
Tivantinib is the only Met-directed TK inhibitor currently in human clinical trials that is not ATP-competitive; it reportedly binds to the Met TK domain near the ATP binding site and acts allosterically (54). A randomized, double-blinded Phase III study of previously treated patients with locally advanced or metastatic, non-squamous NSCLC treated with tivantinib plus erlotinib, or placebo plus erlotinib was recently terminated when interim analysis indicated that its primary endpoint, OS, could not be met (55). Recent results of a multi-center Phase II trial of tivantinib in 47 patients with microphthalmia transcription-factor-associated tumors showed modest efficacy overall, primarily in the form of stable disease (60% of patients); baseline Met content was locally or focally positive in tumors from 74% of patients (56). Adding tivantinib to erlotinib significantly improved PFS in this trial (57). Tivantinib has shown promise in a Phase II trial for HCC (58) and a Phase III in this indication is planned (Table 17.1).
Foretinib targets Met, VEGFR2, Axl, Ron, and Tie-2 with high affinity. In the largest clinical trial devoted to papillary renal cell carcinoma, foretinib demonstrated anti-tumor activity, modulation of several target indicator plasma proteins, and a manageable toxicity profile (59). Prompted by evidence of synergy between foretinib and HER1 (EGFR) TKIs on tumor cells with Met and HER1/2 amplification or over-expression (60), a Phase I/II study of foretinib in combination with lapatinib in patients with HER2 over-expressing metastatic breast cancer is underway (Table 17.1).
Other Met TKIs in Phase II trials for safety and efficacy include MK8033, golvatinib, amuvatinib, BMS777607, MGCD265, and MK2461 (Table 17.1). Many of these agents are more Met-selective than their predecessors and Phase I trials indicate they are well tolerated. MK8033 (NCT00559182; 40) targets Met and Ron, and golvatinib targets Met and VEGFR2 (61). Ongoing Phase Ib/II trials combine golvatinib with sorafenib for hepatocellular carcinoma, lenvatinib for recurrent glioblastoma or unresectable stage III/IV melanoma, or cetuximab for platinum-resistant squamous cell carcinoma of the head and neck (Table 17.1). Amuvatinib inhibits PDGFR, Kit, and Met (62,63) and is now in Phase II for SCLC in combination with platinum and etoposide. MGCD265 (64,65), targeting Met, VEGFR1–3, Ron, and Tie2 is currently in Phase I/II studies in combination with erlotinib or standard-of-care (SOC) treatments.
Late-generation Met TKIs now in Phase I clinical trials include, AMG 337, AMG 208, INCB28060 (66), LY2801653 (67), EMD1214063, and EMD1204831 (Table 17.1); these agents are highly Met selective. A Phase II trial of INCB28060 with gefitinib for treatment of NSCLC in EGFR-mutated, METamplified, and EGFR-inhibitor insensitive patients is underway (NCT01610336; 40). The clinical development programs for JNJ38877605, SGX523, and PF04217903, among the forerunners of highly selective Met TKIs, were discontinued in Phase I; renal toxicity was observed in patients receiving either JNJ38877605 or SGX523 (68), which are structurally similar, and PF04217903 showed diminished potency against certain MET mutations (69).