C-MET tyrosine kinase receptor: mechanisms, clinical applications and future perspectives in cancer therapy

Abstract

The high incidence and mortality rates of tumors have resulted in significant social and economic burdens, posing a major global threat to human life and societal development. In recent years, molecular targeted therapy for tumors has become a research hotspot. C-MET, the receptor for hepatocyte growth factor (HGF), plays a crucial role in the HGF/C-MET signaling pathway, which is involved in various processes such as tumor cell growth, invasion, migration, angiogenesis, epithelial-mesenchymal transition, tumor microenvironment remodeling and therapeutic resistance. Several C-MET-targeting strategies have been developed, including small-molecule tyrosine kinase inhibitors (TKIs), monoclonal antibodies (mAbs) against C-MET or HGF, antibody–drug conjugates (ADCs), nucleic acid aptamers, soluble receptors, natural compounds, and proteolysis targeting chimeras (PROTACs) targeting MET. These inhibitors have demonstrated encouraging anti-tumor effects in both preclinical and clinical studies, with several already available on the market. However, further research is still needed on the activation mechanisms of the HGF/C-MET signaling pathway and its interactions with other receptor tyrosine kinases, which will aid in identifying suitable patients for these treatments. This review provides a comprehensive overview of the structure, regulation, signaling pathways, and functions of C-MET, along with recent advances in C-MET inhibitors, offering valuable insights for cancer therapy.

Introduction

Cancer is the second leading cause of mortality in the United States and the primary cause among individuals under the age of 85, with an estimated 2,041,910 new cases and 618,120 cancer-related deaths projected nationwide in 2025^1,2. Although traditional chemotherapy might prolong the survival of cancer patients, its low selectivity, non-specific systemic toxicity, and resistance to chemotherapy remain significant challenges³. Patients often die from tumor relapse and metastasis. The incorporation of cancer biomarkers into oncology has transformed cancer treatment and patient prognosis, enabling personalized medicine that tailors therapies to the unique molecular profiles of each patient’s tumor⁴. This approach, known as molecular targeted therapy, has become a research hotspot in recent years, particularly for patients with advanced or recurrent cancer or those who are resistant to chemotherapy drugs. For example, sorafenib, which targets KIT, FLT-3, RET, and vascular endothelial growth factor receptor (VEGFR), has been shown to have a potent clinical effect in the treatment of liver cancer⁵; lapatinib, which targets human epidermal growth factor receptor-2 (HER-2) and epidermal growth factor receptor (EGFR), is used to treat breast cancer⁶; and sunitinib, which targets VEGFR and platelet-derived growth factor receptor (PDGFR), is used to treat renal cancer^7,8.

The cellular-mesenchymal epithelial transition factor (C-MET) is a receptor tyrosine kinase (RTK) essential for normal cell survival and function. C-MET signaling promotes cell proliferation, migration, and survival, playing a key role in developmental morphogenesis, wound healing, and organ maintenance^9–12. Dysregulated C-MET activation is associated with cancer progression, metastasis, and therapy resistance, occurring through both ligand-dependent and ligand-independent mechanisms¹³. Currently, there are inhibitors targeting C-MET, including small-molecule tyrosine kinase inhibitors (TKIs), monoclonal antibodies (mAbs) against C-MET, and nucleic acid aptamers, which play a certain role in the treatment of acute myeloid leukemia (AML), thyroid cancer, breast cancer, non-small cell lung cancer (NSCLC), glioblastoma, malignant melanoma, liver cancer, colorectal cancer (CRC), head and neck cancer (HNC) and other solid tumors in preclinical stages and are gradually being used in clinical practice^14–17. Figure 1 provides an overview of key milestones in MET pathway research and small‑molecule drug development.

Fig. 1.

Key milestones in HGF/MET pathway research and MET‑targeted therapy development. This timeline summarizes landmark discoveries in the HGF/MET signaling axis and related therapeutic advances. Major events include the identification of HGF (1984) and its receptor MET (1991), and subsequent insights into MET signaling regulation and splicing variants such as METΔex14. Clinical milestones include approvals of MET‑targeted agents (e.g., crizotinib, cabozantinib, capmatinib, tepotinib, savolitinib) and key trial outcomes (e.g., onartuzumab, tivantinib)

The purpose of this review is to comprehensively clarify the structure, regulation, and function of C-MET, its role in tumor development, and various strategies for its inhibition. We also discuss the medical implications and challenges of targeting C-MET, offering insights into its potential for cancer treatment.

The structure and function of C-MET and its ligand

The natural endogenous ligand for C-MET, hepatocyte growth factor (HGF), was originally identified in rat serum^18,19 and later characterized in human plasma²⁰. HGF, also called scatter factor (SF)^21–23, hepatopoietin A (HPTA)²⁴, fibroblast-derived tumor cytotoxic factor (F-TCF)²⁵, and DFNB39²⁶, is encoded by the human HGF gene located at 7q21.11, comprising 20 exons and producing a protein of ~71 kDa²⁷. HGF is synthesized as an inactive single-chain precursor known as pro-HGF, which undergoes proteolytic cleavage to form an active heterodimer which comprises a heavy chain (α-chain) and a light chain (β-chain) linked by disulfide bonds. The α-chain consists of an N-terminal (N) domain and four kringle domains (K1–K4) that mediate receptor binding and activation. Moreover, the β-chain harbors a catalytically inactive serine protease homology (SPH) domain required for efficient HGF/C-MET signaling²⁸. Secreted predominantly by mesenchymal stromal cells²⁹, it belongs to the plasminogen subfamily within the peptidase S1 family and acts as the sole ligand for the C-MET receptor utilizing both low-affinity binding sites in the SPH domain and high-affinity sites in the N and K1 domains, triggering numerous cellular processes³⁰ (Fig. 2).

Fig. 2.

Schematic diagram of the basic structure of MET and HGF: a The MET protein comprises an extracellular α-subunit and a transmembrane β-subunit. Its extracellular region contains the SEMA domain, the plexin-semaphorin-integrin (PSI) domain, and four immunoglobulin-plexin-transcription (IPT1–4) domains. The intracellular part includes a juxtamembrane (JM) domain, a tyrosine kinase (TK) domain, and a multifunctional docking site (MFDS) at the C-terminal end. b HGF consists of a heavy α-chain and a light β-chain linked by disulfide bonds, with the α-chain containing an N-terminal domain and four kringle domains (K1–K4). c Schematic diagram of the interaction between MET and HGF. d The MET gene is located on chromosome 7q31. The schematic diagram illustrates focal MET amplification, defined by a high MET/CEP7 ratio (MET/CEP7 = 5), and chromosome 7 polysomy, characterized by proportional increases in MET and CEP7 signals with a MET/CEP7 ratio of approximately 1

C-MET, also known as MET, AUTS9, RCCP2, DFNB97 and hepatocyte growth factor receptor (HGFR)³¹, is encoded by the human MET gene (MET proto-oncogene). The MET gene was first identified as a proto-oncogene in a chemically transformed osteosarcoma cell line in the mid-1980s, where it was found to be fused to the nuclear pore complex subunit TPR³². This fusion, TPR-MET, links the oligomerization domain of TPR to the intracellular kinase domain of MET, leading to constitutive tyrosine kinase activity and oncogenic transformation. Through molecular cloning, the full MET oncogene was sequenced and mapped to chromosome 7q21-31, consisting of 24 exons^26,31,33,34. Its product, the MET receptor, is a 170 kDa single-chain precursor protein (pro-MET) that is translated and then cleaved to form a disulfide-linked dimer comprising a 50 kDa extracellular α-subunit and a 145 kDa single-pass transmembrane β-subunit³⁵. The extracellular region of MET includes three main functional domains: the semaphorin (SEMA) domain, encompassing the full α-subunit and part of the β-subunit; the plexin-semaphorin-integrin (PSI) domain; and the four immunoglobulin-plexin-transcription (IPT1–4) domain³⁶. The SEMA domain of MET contains the specific binding site for its ligand HGF, and the interaction between HGF and MET in this domain is crucial for activating various downstream signaling pathways³⁷. The intracellular region of MET comprises three key domains: a juxtamembrane (JM) domain, which negatively regulates kinase activity by promoting MET ubiquitination and degradation through the phosphorylation of S985 and Y1003^38,39; a tyrosine kinase (TK) domain, where the phosphorylation of Y1234 and Y1235 increases kinase activity via trans-autophosphorylation; and a multifunctional docking site (MFDS), which recruits downstream signaling molecules and adapter proteins including SHP2, growth factor receptor-bound protein 2 (GRB2), GRB2-associated binding protein 1 (GAB1), and phosphatidylinositol 3-kinase (PI3K), upon the phosphorylation of Y1349 and Y1356^40,41 (Fig. 2).

Regulation of MET gene expression

Epigenetic and post-transcriptional regulation

The expression of MET gene is regulated through multiple mechanisms, including epigenetic modifications such as DNA methylation and histone acetylation, transcription factors (such as Sp1 and Ets-1), alternative splicing, microRNAs, protein translation, etc⁴². Understanding MET regulation is vital for developing targeted therapies, as MET gene amplification or dysfunction is associated with pathological conditions such as tumor occurrence, progression, metastasis, and drug resistance⁴³.

Increased MET expression is often linked to hypomethylation and histone acetylation. DNA methylation involves transferring a methyl group to cytosines in CpG islands, which blocks transcription factor accessibility and inhibits transcription⁴⁴. It has been reported that MET gene is hypomethylated and overexpressed in pancreatic ductal adenocarcinoma (PDAC), which is correlated with poor survival⁴⁵. Histone acetylation, catalyzed by histone acetyltransferases, neutralizes the positive charges of lysine residues at the histone N-terminus. This reduction in charge decreases interactions between histones and DNA, allowing greater access for transcription factors and RNA polymerases to the target regulatory regions⁴⁶.

After transcription, MET mRNA is selectively spliced into many different mRNA species before entering the cytoplasm^47,48, and its translation is regulated by more than 30 microRNAs²⁰. Among these splice variants, the predominant 8 kb transcript encodes a 170 kDa single‑chain pro‑MET protein, which is subsequently cleaved into the α and β chains. These two chains are connected by disulfide bonds to form a dimer that undergoes N- or O-linked glycosylation in the Golgi apparatus and is transported to the membrane^49–52. Other splice variants have been reported, some of which (e.g., the 7 kb transcript) do not produce a protein product⁴⁷, whereas others may have biological significance. For example, MET mRNA lacking exons 7 and 8 (MetΔ7–8) encodes a cytosolic MET isoform that is constitutively active in certain high‑grade gliomas⁵³. Δ13Met, lacking exon 13, encodes a truncated extracellular protein that inhibits MET signaling⁵⁴.

Upon binding to its ligand HGF, MET kinase activity is transmitted downstream through phosphorylation⁵⁵. Phosphorylated MET is afterwards internalized and degraded, and can be recycled through proteasome, endosome or lysosome pathways⁵⁶. In addition, the steady‑state level and activity of MET protein are regulated by autoregulation and by signaling from other membrane receptors, such as epidermal growth factor receptors and G‑protein‑coupled receptors, through crosstalk and transactivation mechanisms⁵⁷.

Genetic alterations and ligand-dependent activation

Approximately 3% of lung adenocarcinoma patients in Western and Eastern countries harbor MET exon 14 skipping mutations (METΔex14)⁵⁸. The exon 14 encodes part of the intracellular transmembrane domain of the MET receptor, such as tyrosine 1003 (Tyr 1003, Y1003)³⁸. Phosphorylation of MET protein at Y1003 can recruit the E3 ubiquitin ligase c-casitas B-lineage lymphoma (Cbl) in the Cbl family⁵⁹, causing MET ubiquitination, internalization, and degradation. Therefore, individuals with METΔex14 mutations may exhibit increased MET protein expression and activity, as the mutation protects MET from degradation^26,58. Interestingly, METΔex14 has also been found in other types of lung neoplasms (2.3%), brain gliomas (0.4%), and cancers of unknown primary origin (0.4%)^58,60.

In addition to HGF, MET can also be activated by its natural isoform NK1⁶¹. Unlike HGF, which induces receptor activation through an asymmetric 2:2 MET:HGF complex⁶², NK1 forms a stable dimer that bridges two MET molecules in a symmetric assembly. This symmetrical interaction brings the two MET receptors into close proximity, enabling optimal receptor activation and subsequent trans-autophosphorylation of key tyrosine residues in the intracellular kinase domain⁶³. This, in turn, initiates downstream signaling pathways essential for cell proliferation, migration, and survival.

Non-canonical regulation by circMET

Recent studies have revealed important regulatory roles of circular RNA (circRNA) in the regulation of MET pathway. Circular RNAs are a class of covalently closed, non-coding RNAs generated by back-splicing, lack 5′ caps and 3′ poly(A) tails, are capable of regulating gene expression through mechanisms such as microRNA sponging or protein scaffolding^64–69, and participate in diverse oncogenic processes such as epithelial–mesenchymal transition (EMT), drug resistance, and immune evasion^70–73.

Among them, circMET, a circular RNA derived from the MET gene, has recently attracted attention for its role in cancer biology. CircMET is upregulated in a variety of cancers, including hepatocellular carcinoma (HCC), renal cell carcinoma (RCC), CRC, and NSCLC^74–77, and acts as a microRNA sponge modulating MET signaling and other oncogenic pathways. For instance, in HCC, circMET (hsa_circ_0082002) promotes EMT and immune evasion through the miR-30-5p/Snail/DPP4/CXCL10 axis⁷⁷.

In addition, circMET serves as a protein-coding RNA. A study by Zhong et al. identified a novel protein isoform, MET404, encoded by circMET, which consists of 404 amino acids and is capable of directly interacting with the MET β subunit to form a constitutively active MET receptor complex, independent of HGF stimulation. This translation is mediated by N⁶-methyladenosine (m⁶A) modification on circMET and depends on the m⁶A reader YTHDF2, which drives cap-independent translation of MET404^78–83, providing a non-canonical mechanism for MET activation and contributing to tumor progression. Furthermore, the development of a neutralizing mAb against MET404 (anti-MET404 mAb) showed potent antitumor effects in preclinical glioblastoma models and demonstrated synergistic efficacy when combined with a conventional MET inhibitor⁸⁴.

Together, these findings highlight a new layer of complexity in MET regulation via circRNAs and expand potential therapeutic avenues targeting both canonical and non-canonical MET signaling.

HGF/MET signaling pathway

The HGF/MET signaling axis and its downstream effectors

Upon binding with HGF, MET undergoes homodimerization and trans-autophosphorylation of Y1234 and Y1235 in the activation loop, along with phosphorylation of Y1003 in the juxtamembrane region, inducing secondary phosphorylation of docking sites Y1349 and Y1356^85,86. These phosphorylated sites recruit key signaling proteins such as GRB2, GAB1, SRC, the p85 subunit of PI3K, phosphatidylinositol-specific phospholipase Cγ (PLCγ), signal transducer and activator of transcription 3 (STAT3) and Shc36–38 through SH2-mediated interactions^87,88. This, in turn, activates many downstream signaling pathways, such as the three mitogen‑activated protein kinase (MAPK) pathways, including extracellular signal‑regulated kinase 1/2 (ERK1/2), Jun amino‑terminal kinase (JNK), and p38 MAPK pathways, as well as signal transducer and activator of transcription 3 (STAT3), nuclear factor kappa B (NF‑κB), phosphatidylinositol 3‑kinase/protein kinase B/mammalian target of rapamycin (PI3K/AKT/mTOR), Rho‑like GTPase, and SRC/focal adhesion kinase (SRC/FAK) cascades^86,87,89,90. These pathways drive cellular responses like proliferation, survival, migration, and differentiation, which are essential for physiological processes such as early embryonic development, wound healing, and tissue remodeling under normal conditions⁸⁵, but contribute to tumorigenesis when dysregulated^{26,40,91–93}.

The wingless-related integration site (Wnt)/β-catenin pathway is also functionally integrated downstream of MET signaling. MET activation inhibits GSK3β-mediated phosphorylation of β-catenin, leading to the accumulation of β-catenin in the cytoplasm and translocation into the nucleus, thereby enhancing Wnt/β-catenin transcriptional activity^94–97. This in turn induces upregulation of downstream targets such as Myc, Cyclin D1, and MMP-7⁹⁸. In HCC, MET cooperates with β-catenin activation or AXIN1 loss to drive tumor growth through YAP/TAZ signaling. Combined inhibition of MET (cabozantinib) and Wnt/Hippo pathways (G007-LK) shows synergistic antitumor effects in HCC models⁹⁹.

Co-receptors and crosstalk with other RTKs

The HGF/MET signaling pathway can be enhanced by various signaling amplifiers and co-receptors. For example, GAB1, a multi-adapter protein, is recruited and phosphorylated to maintain MET signaling by either binding directly to MET or indirectly via GRB2^100,101. Phosphorylated GAB1 recruits effectors like PI3K, SH2-domain containing phosphatase 2 (SHP2), and PLCγ, amplifying downstream signals^{100,102–108}. The co-receptor CD44v6 is essential for activating the MET-dependent RAS/MAPK pathway, linking MET to the F-actin cytoskeleton and recruiting son of sevenless (SOS), a guanine nucleotide exchange factor that activates RAS¹⁰⁹. Moreover, plexin B1, when bound to semaphorin 4D, can transactivate MET independently of HGF, promoting invasiveness and angiogenesis^110–112. Upon activation, MET can also initiate phosphorylation of the α6β4 integrin complex, which serves as a scaffold for the recruitment of key signaling adapters (SHC, PI3K, and SHP2). This molecular platform amplifies HGF-mediated activation of downstream effectors, including PI3K, RAS, and SRC, ultimately driving invasive behavior and anchorage-independent proliferation^113,114.

The extensive cross-talk between MET and other RTKs, including transforming growth factor-β (TGF-β), EGFR, HER-2, HER-3, AXL, RET, RON, erb-b2 receptor tyrosine kinase (ERBB2), vascular endothelial growth factor receptor-2 (VEGFR-2), and insulin-like growth factor (IGFR), leads to additional signaling response modulation^37,115–118 (Fig. 3). In MET-amplified cancers like NSCLC, MET can form heterodimeric complexes with these receptors^57,119–123, which often become highly phosphorylated in a MET-dependent manner and contribute to sustained oncogenic signaling¹²⁴. Furthermore, MET has been shown to interact with RTKs such as AXL through receptor heterodimerization and reciprocal phosphorylation, where MET kinase activity mediates the phosphorylation of AXL in response to HGF stimulation^125,126. Beyond direct physical interactions, lateral signaling between RTKs has also been reported. For instance, EGFR activation can cause increased MET expression and phosphorylation through downstream signaling cascades, such as the c-Src pathway, even in the absence of HGF^57,127–129. These interactions expand the diversity of MET-driven oncogenic pathways, enhance resistance to targeted therapies, and contribute to tumor aggressiveness^130,131.

Fig. 3.

The signaling pathways of MET. HGF binds to MET, leading to MET homodimerization and autophosphorylation, activating various downstream pathways such as MAPK, PI3K/AKT, FAK, and STAT3, promoting cell proliferation, survival, migration, invasion, and metastasis. MET downregulation occurs via CBL recruitment and ubiquitin-mediated degradation, as well as extracellular shedding. The HGF/MET signaling pathway can be enhanced by various signaling amplifiers and co-receptors such as CD44v6, plexins, integrins, EGFR, FAS, and IGF-IR

HGF/MET signaling in tumor development

Dysregulation of the HGF/MET axis is associated with the occurrence and metastasis of epithelial-derived cancers such as lung cancer, pancreatic cancer, ovarian cancer, liver cancer, CRC, breast cancer, and gastric cancer^132–135. Various molecular mechanisms can cause this dysregulation, including gene amplification, which increases MET copy number and enhances signaling; point mutations, particularly exon 14 skipping mutations, which impair CBL-mediated degradation and prolong MET receptor stability; and gene fusions, which generate constitutively active chimeric proteins through intra- or inter-chromosomal rearrangements, often involving chromosome 7^136–139. Notably, although MET fusions are relatively rare, they have been identified in several malignancies, including glioblastoma and NSCLC, and have shown anecdotal but promising responses to MET inhibitors such as crizotinib in clinical case reports^140–142. Additionally, protein overexpression and aberrant ligand production, such as excessive HGF secretion via autocrine or paracrine mechanisms, can lead to sustained activation of the MET pathway. Moreover, transactivation of MET by other RTKs, including EGFR, VEGFR-2, IGFR, and RON^{26,40,117,143–145}, further contributes to ligand-independent MET signaling (Fig. 4). Collectively, these mechanisms promote tumor cell proliferation, migration, invasion, angiogenesis, EMT, immune evasion, cancer stemness and therapeutic resistance^43,146 (Fig. 5).

Fig. 4.

The activation modes of MET: a ligand binding: excessive HGF production through autocrine or paracrine mechanisms leads to sustained MET activation, b receptor overexpression: increased MET copy number or transcriptional upregulation enhances signaling output, c aberrant splicing: for example, MET exon 14 skipping mutations impair CBL-mediated ubiquitination and degradation, prolonging receptor stability, d transactivation: cross‑talk with other receptor tyrosine kinases (e.g., EGFR, VEGFR‑2, IGFR, and RON) activates MET in a ligand‑independent manner, e MET gene fusion: intra‑ or inter‑chromosomal rearrangements generate constitutively active fusion proteins that drive oncogenic signaling, and f circular MET RNA: circMET can encode MET404, which activates MET signaling in an HGF‑independent manner

Fig. 5.

The multifaceted roles of HGF/MET signaling in tumor development. Upon activation, MET signaling enhances tumor cell proliferation, migration, invasion, and distant metastasis. It promotes epithelial–mesenchymal transition (EMT), facilitating loss of cell–cell adhesion and increased motility. HGF/MET activation also stimulates angiogenesis by upregulating pro‑angiogenic factors, and modulates the tumor immune microenvironment, aiding immune evasion. Furthermore, this pathway supports the maintenance of cancer stem‑like properties, and contributes to resistance to chemotherapy, targeted therapy, and radiotherapy

MET promotes tumor proliferation

Diverse activation modes of MET can result in the dysregulation of HGF/MET signaling in cancer, thereby inducing tumor cell proliferation. Hypoxia-induced overexpression and activation of the MET proto-oncogene, along with amplified HGF signaling, collectively encourage aggressive tumor cell growth^147,148. Increased MET transcription, MET amplification, and MET overexpression are all activation modes that lead to the dysregulation of HGF/MET signaling, subsequently fostering the proliferation, regeneration, and survival of HCC^{134,149–154}. In clear cell sarcoma (CCS), the EWS/ATF1 fusion gene, resulting from a chromosomal translocation, activates the melanocyte master transcription factor MITF, which in turn contributes to MET expression and triggers HGF/MET signaling, driving CCS cell growth, survival, and invasion^155,156. Additionally, HGF-induced activation of the MET receptor in medulloblastoma (MB) cells leads to the expression of tissue factor (TF), which, by facilitating the formation of a provisional fibrin matrix, further promotes tumor proliferation¹⁵⁷.

MET mediates migration and invasion

The HGF/MET pathway is pivotal in tumor invasion and metastasis through various mechanisms. Hypoxia triggers MET signaling by activating HIF-1α-mediated pathways, which induces EMT¹⁵⁸ and boosts the invasive capabilities of PDAC and HCC¹⁵⁹. In CRC, MET amplification and overexpression are observed in colon adenomas and primary tumors¹⁶⁰, with MET expression escalating from normal epithelium to adenoma, carcinoma, and metastasis. This pathway is essential for CRC distant metastasis^161,162, as MET amplification increases metastatic potential¹³². MET overexpression propels invasion of CRC liver metastases¹⁶³, and MET transactivation fosters metastasis via SOX13 mediation¹⁶⁴. Furthermore, GAPLINC, upregulated in CRC tissues, regulates MET signaling to facilitate cell migration and invasion¹⁶⁵. In NSCLC, TIGAR protects cells from ROS-induced apoptosis by enhancing flux through the pentose phosphate pathway, while genetic ablation of TIGAR suppresses invasive and metastatic potential in NSCLC models, both in vitro and in vivo, through modulation of MET receptor signaling¹⁶⁶. In gastric cancer, comparative analyses reveal elevated co-expression of MACC1 and MET in primary tumors with lymph node metastasis, suggesting that MACC1 strengthens metastasis via the HGF/MET pathway. This pathway also contributes to lymphangiogenesis in gastric cancer and promotes lymph node metastasis through the VEGF-C/VEGF-D/VEGFR-3 axis¹⁶⁷. Beyond these, HGF/MET signaling contributes to MB cell migration and invasiveness through TF upregulation and TF-mediated non-hemostatic signaling¹⁵⁷. Dysregulated HGF/MET pathways also mediate invasion and metastasis in lung adenocarcinoma (LUAD)¹⁶⁸ and breast cancer⁹⁴, influencing cell-matrix adhesion, stem cell properties, and the tumor microenvironment (TME), respectively.

MET promotes angiogenesis

MET promotes angiogenesis by activating the PI3K/AKT and MAPK/ERK pathways, stimulating endothelial cell proliferation, migration, and matrix metalloproteinase (MMP) expression to support neovascularization^169–172. HGF/MET signaling, working synergistically with VEGF under hypoxic conditions, drives tumor angiogenesis by enhancing vascular formation and endothelial migration¹⁷³. In a mouse model of lung adenocarcinoma, a dual inhibition of VEGFR and MET pathways has shown potential in delaying resistance and maintaining vascular integrity¹⁷⁴.

MET affects EMT process

Studies have demonstrated the involvement of MET in EMT across many types of cancer. The HGF/MET signaling pathway stimulates tumor cell motility, migration, and invasion. MET regulates E-cadherin/vimentin expression, promoting cell proliferation and migration, through PI3K and MAPK pathways. For example, Xu et al. found that MET/AKT/GSK-3β/Snail pathway fosters bladder cancer EMT via miR-433¹⁷⁵. MET also contributes to EMT through its interaction with the TGF-β signaling axis. In muscle-invasive bladder cancer, HGF activates c-Src, which phosphorylates SMURF2 and disrupts its interaction with SMAD7. This leads to stabilization of the TGF-β receptor and sustained TGF-β signaling, ultimately driving EMT and invasion. Blocking TGF-β signaling in vivo has been shown to prevent tumor invasion, suggesting a potential benefit of combining MET and TGF-β inhibitors in cancer therapy¹⁷⁶.

Moreover, MET activation by autocrine or paracrine HGFs, often derived from carcinoma-associated fibroblasts (CAFs), has been linked to EMT and therapeutic resistance in gastric, CRC, and lung cancers^177–184. In NSCLC, CAF-derived HGF, together with IGF-1, activates the ANXA2 signaling axis, inducing EMT and resistance to anti-EGFR therapy¹⁸⁵, whereas silencing ANXA2 can reverse this EMT phenotype^185,186. In CRC, CAF-secreted HGF enhances EMT and resistance to anti-EGFR therapy in RAS wild-type tumors^{85,132,162,177,187–191}. Moreover, in BRAF-mutant melanoma, HGF secretion by CAFs activates MET and gives rise to reactivation of the MAPK/PI3K signaling cascade, resulting in innate resistance to BRAF inhibitors¹⁹².

MET remodels the tumor microenvironment

MET plays a dual role in immunity, acting either as an immunosuppressive agent or as a stimulator of immune response, with a predominantly immunosuppressive effect over the antitumor immune response. The HGF/MET pathway drives EMT, alters tumor cell morphology, and modulates intercellular adhesion molecules, enabling tumor cells to adapt to and thrive within the TME^193–196. Additionally, it regulates immune cell infiltration and contributes to immune suppression, thereby shaping the immunosuppressive nature of the TME and dampening effective antitumor immune responses.

For instance, MET overexpression in cancer cells can be identified by CD8+ cytotoxic T cells as a tumor-associated antigen (TAA), triggering immune system activation¹⁹⁷. However, in most cases, HGF/MET signaling pathway inhibits the tumor-killing activity of cytotoxic T lymphocytes (CTLs). By promoting a glycolytic phenotype in cancer cells, this signaling pathway increases lactic acid secretion that inhibits CTL function¹⁹⁸. Inhibition of MET can therefore increase the number and activity of tumor-infiltrating T cells, boost their activity, and improve immunotherapy efficacy¹⁹⁹.

Many studies have focused on the effects of HGF and MET on dendritic cells (DCs). HGF/MET signaling enhances DC adhesion, migration, and pro-tolerogenic functions, while inhibiting their antigen-presenting capabilities^200,201. The PI3K/AKT pathway and Bruton’s tyrosine kinase play crucial roles in mediating the inhibitory effects of HGF on DC activation^202,203. In addition, HGF facilitates the release of IL-10 and upregulates indoleamine 2,3-dioxygenase 1 (IDO1) expression in DCs, thereby suppressing T-cell clonal spread, enhancing Treg expansion, and contributing to an immunosuppressive TME^204–206. Elevated IL-10 levels within the TME can further disrupt DC differentiation and increase DC apoptosis, protecting tumor cells from CD8 + T-cell cytotoxicity^207,208.

The HGF–MET axis in neutrophils also exhibits both pro- and antitumor activities, with a primary role in suppressing the antitumor immune response. Studies have indicated that inhibiting MET signaling reduces neutrophil-mediated immunosuppression and enhances immunotherapy efficacy, especially in combination with immune checkpoint blockade, as observed in bladder cancer models, where it strengthens CD8 + T cell responses²⁰⁹. Conversely, activation of HGF/MET in neutrophils, often triggered by inflammatory stimuli such as tumor-derived TNF-α, stimulates chemotaxis toward tumors and nitric oxide release, aiding tumor cell destruction^210,211.

Likewise, HGF/MET signaling inhibits tumor-associated M1 macrophages and polarizes monocytes toward the immunosuppressive M2 macrophages²¹².

Moreover, the HGF/MET pathway has also been shown to sustain PD-L1 expression through mechanisms such as the PI3K/Akt, MAPK, and MET/ERK pathways, leading to T-cell exhaustion and immune tolerance^213,214. In NSCLC, MET overexpression correlates with elevated PD-L1 expression and the increased expression of immunosuppressive genes like PDCD1LG2 and SOCS1^215,216. Similarly, in HCC, renal cancer, pancreatic cancer, and glioblastoma, MET activation boosts PD-L1 expression via multiple pathways, such as the MAPK/NF-κBp65, Ras-PI3K, and STAT1 pathways^217–220. This interplay between MET and PD-L1 not only facilitates tumor immune evasion but is also associated with poor prognosis across various cancer types^221,222.

The roles of HGF/MET signaling pathway in cancer stemness

Accumulating evidence highlights the critical role of HGF/MET signaling pathway in regulating cancer stem cell (CSC) properties across multiple malignancies, including breast, CRC, HCC, pancreatic, prostate, and glioblastoma^223–228. CSCs are a subpopulation of highly tumorigenic cells defined by their abilities to self-renew, differentiate, recapitulate tumor heterogeneity, and resist therapy^229–231. Unlike normal adult tissues where MET expression is low or conditionally inducible during tissue repair, MET is constitutively overexpressed in many advanced cancers and is enriched in CSCs, contributing to their aggressiveness and therapeutic resistance^92,231–233.

Functionally, MET activation engages key downstream effectors such as PI3K/AKT, STAT3, Wnt/β-catenin, Notch, and c-MYC, which are essential for CSC maintenance and survival^234–236. For example, in head and neck squamous cell carcinoma (HNSCC) and colon cancer models, MET activation drives CSC properties via β-catenin–dependent transcription, while MET inhibition with PF-2341066 suppresses CSC populations^{116,232,237–239}. Notably, frizzled class receptor 8 (FZD8), a receptor in the Wnt pathway, is upregulated downstream of MET via the ERK/c-Fos cascade and can partially rescue cells from MET inhibition, indicating feedback activation within the Wnt pathway^237,240. In addition, in the TME, spatial and cellular interactions between MET-positive tumor cells and stromal components such as astrocytes and CAFs contribute to the maintenance of CSC traits^241,242. For instance, in pancreatic cancer, HGF/MET signaling between stellate cells and cancer cells induces stemness markers (e.g., NANOG, OCT-4, SOX-2) and sustains glycolytic metabolism through HK2 induction²⁴³. In HCC, CAF-derived HGF promotes liver tumor-initiating cell self-renewal and HCC progression through the MET/ERK/FRA1/HEY1 signaling axis²⁴⁴. In colon adenocarcinomas, HGF secreted by stromal myofibroblasts has been shown to activate β-catenin-dependent transcription, thereby enhancing clonogenicity and restoring CSC traits in more differentiated tumor cells²³⁸. Follow-up replication studies confirmed the general trend, although some results did not reach statistical significance^245,246.

Recent single-cell RNA sequencing studies have revealed that a macrophage subpopulation (Mac_1) fosters skeletal muscle stem cell proliferation via HGF/MET signaling, a process that can be impaired with aging due to reduced HGF expression²⁴⁷. In pancreatic cancer, chimeric antigen receptor (CAR)-macrophages engineered to target MET can selectively eliminate MET-positive CSCs, inhibit VEGFA/FGF2/ANGPT secretion, and suppress tumor growth and angiogenesis in vivo²⁴⁸. These findings highlight the potential of MET-positive CSCs as viable targets for precision immunotherapy.

MET is associated with therapeutic resistance in cancer treatment

Deregulation of HGF/MET signaling plays a pivotal role in driving drug resistance across various cancer types, significantly impacting treatment efficacy. In HCC, the HGF/MET axis promotes chemoresistance by modulating metabolism and autophagy, which can be reversed by combining MET inhibitors with autophagy suppressors²⁴⁹. Moreover, hypoxia-induced MET/β1 integrin complex formation facilitates invasive resistance in breast cancer and glioblastoma by enhancing fibronectin interactions²⁵⁰. Additionally, the TME drives innate resistance to RAF inhibitors via HGF-dependent MET activation in BRAF-mutant melanoma, CRC, and glioblastoma¹⁹². In EGFR-mutant NSCLC, MET amplification accounts for about 20% of acquired resistance to EGFR inhibitors, mainly through the MET/ERBB3/PI3K signaling pathway^251–254. Wnt/β-catenin signaling has also been implicated as a key contributor to resistance in this context²⁵⁵. MET-driven activation of Wnt signaling upregulates multidrug resistance proteins and stimulates β-catenin nuclear accumulation, forming a positive feedback loop that fosters tumor progression and reduces the efficacy of both EGFR-TKIs and MET inhibitors, which explains the limited success of dual EGFR-MET inhibition and highlights the potential benefit of co-targeting the Wnt/β-catenin pathway to overcome resistance^256,257.

The HGF/MET axis is implicated in radioresistance as well. Specific MET mutation, such as Y1253D, is associated with resistance to radiotherapy in oropharyngeal squamous cell carcinoma²⁵⁸. In breast cancer cells, the activation of the MET signaling pathway by HGF secreted from CAFs promotes EMT, cell growth, and radioresistance²⁵⁹. Similarly, HGF has been shown to increase both chemo- and radioresistance in rhabdomyosarcoma cell lines^41,260.

The relationship between MET expression and patient prognosis

Previous studies have demonstrated the prognostic value of MET across multiple cancer types, including NSCLC^89,261–267, triple-negative breast cancer (TNBC)^268–271, CRC^272–277, gastric cancer^278,279, gastroesophageal adenocarcinoma (GEC)²⁸⁰, esophageal cancer²⁸¹, HNC^282,283, HCC^284,285, RCC²⁸⁶ and cervical cancer^287–289.

In NSCLC, aberrant MET signaling pathway can result from a series of events, such as overexpression of MET or HGF proteins, MET amplification, germline or somatic MET mutations, MET gene rearrangements, or variations in downstream pathways or regulatory factors^89,290–293, relating to the poor prognosis of patients. Moreover, various clinical trials have indicated the therapeutic benefits of targeting MET pathway in NSCLC patients²⁹⁴. NSCLC patients with MET alterations exhibit sensitivity to MET inhibitors, with the level of amplification positively correlated with the therapeutic efficacy of these drugs, and those with higher levels of MET amplification might achieve extended progression-free survival (PFS)^262,263.

In breast cancer, higher MET expression is positively correlated with tumor size, lymph node involvement, and poorer survival outcomes²⁷⁰. A Phase III clinical trial (NCT01908426) found that elevated MET levels were linked to reduced overall survival (OS) in patients with HCC²⁸⁴. The results were in accordance with a systematic meta-analysis encompassing 1408 HCC patients, which revealed that those with elevated MET expression had significantly poorer PFS and OS²⁸⁴.

In gastric cancer patients, MET alterations are linked to an unfavorable prognosis. A meta-analysis incorporating 14 independent studies involving 2258 stage I–IV gastric cancer patients showed that increased MET amplification was associated with decreased OS (hazard ratio (HR) = 2.82, 95% confidence interval (CI) = 1.86–4.27, p < 0.05)⁴⁰. An et al. reported that patients with MET overexpression had a significantly shorter median OS and PFS compared to those without¹³⁵. Additionally, Jin et al. reported that MET-altered patients were prone to have a higher incidence of immune-related adverse events (irAEs) when undergoing PD-1 immunotherapy^295,296.

The prognostic role of MET in colon cancer patients remains controversial. Some studies have linked high MET expression to poor prognosis, including shorter OS and advanced tumor stages^{274,275,297–299}, whereas others have found no significant association^272,300. The discrepancy may stem from variations in study design, patient cohorts, and methods of MET assessment. Notably, some studies suggest that the prognostic impact of MET may depend on additional factors, such as tumor localization²⁹⁷, co-expression with other markers like RON²⁷⁴, or its role in metastasis and immune modulation²⁹⁹. Further research is needed to clarify the role of MET and refine its use as a prognostic biomarker in CRC, particularly through standardized assessment methods and comprehensive molecular profiling.

MET-targeted therapies

Currently, several types of MET inhibitors exist, including small-molecule inhibitors, mAbs against MET or HGF, antibody–drug conjugates (ADCs), nucleic acid aptamers, soluble receptors, natural compounds, and proteolysis-targeting chimeras (PROTACs) targeting MET^301–305 (Fig. 6).

Fig. 6.

Various therapies targeting MET have been developed, including: a selective type I inhibitors that bind to active drugs, b non-selective MET kinase inhibitors, such as type II and type III inhibitors, which bind to the receptor’s inactive non-phosphorylated conformation and allosteric sites, c monoclonal antibodies against MET or HGF, d antibody-drug conjugates (ADCs), e nucleotide aptamers, f soluble receptors, g natural compounds, and h proteolysis-targeting chimeras (PROTACs) targeting MET

MET small-molecule inhibitors

Based on the role of MET in tumor growth, angiogenesis, invasion, and metastasis, MET could serve as an ideal target for cancer therapy. Currently, thousands of Phase I, II, and III clinical trials of MET inhibitors are registered in the Global Clinical Trial Database (www.clinicaltrials.gov). Most small-molecule inhibitors are ATP-competitive inhibitors that can simultaneously target multiple kinases, with only a small subset selectively inhibiting MET. These inhibitors not only inhibit the proliferation of MET-driven tumor cell lines at low concentrations (in the nM range) in vitro but also exhibit antitumor and anti-metastatic effects in relevant tumor xenograft models. In clinical settings, they have also demonstrated good efficacy in patients with sarcoma, RCC, and NSCLC, and are thus marketed³⁰⁶.

MET inhibitors are classified into three categories based on their structural characteristics and binding modes to kinases. Type I inhibitors are ATP-competitive and bind to the ATP binding pocket of MET protein in a U-shaped structure, interacting with hinge region amino acid residues (Pro 1158, Tyr 1159, and Met 1160) through hydrogen bonds. Their selectivity for MET kinase is further enhanced by additional hydrogen bonding and π-π stacking with residues in the DFG motif and the activation loop, resulting in limited off-target activity^307,308 (Supplementary Table 1). Type II inhibitors are also ATP-competitive but function as multi-target inhibitors, binding to MET protein in a more extended conformation with greater molecular weight and lipophilicity. Typically, there are four parts of this type of inhibitor with two distinctive features: first, it includes a five-atom bridge linker with six chemical bonds; second, the linker consists of hydrogen, oxygen, and nitrogen atoms, allowing it to act as either a hydrogen bond donor or acceptor^309–311 (Supplementary Table 2). Type III inhibitors are non-ATP competitive that bind to the inactive form of MET, stabilizing MET in the auto-inhibited state^312–315 (Table 1).

Table 1.

Summary of the basic profile of type III MET small-molecule inhibitors and the related ongoing clinical trials

Drug	Developer	Target (s)	IC50 for MET	Clinical trial no.	Phase of approval	Indications	Monotherapy/combinations	Adverse events	Status
Tivantinib (ARQ 197)	ArQule, Daiichi Sankyo, and Kyowa Hakko Kirin	MET (selective)	In vitro IC50=290 nM	NCT00988741	II	Unresectable+failed one prior systemic therapy+HCC	Monotherapy	Neutropenia, anemia, leukocytopenia, ascites, abdominal pain	Completed
				NCT01755767	III	MET-high+inoperable+one prior sorafenib containing systemic therapy+HCC	Monotherapy		Completed
				NCT02029157	III	MET-high+inoperable+one prior sorafenib containing systemic therapy+HCC	Monotherapy		Completed

Type I MET inhibitors

Crizotinib (PF-02341066, XALKORI, developed by Pfizer)

Crizotinib is a multi-target protein kinase inhibitor that targets MET, ALK, and ROS³¹⁶. Compared to chemotherapy based on pemetrexed plus platinum (median survival time: 7 months), it significantly extended the PFS of patients (median survival time: 10.9 months), with an HR of 0.45 (95% CI: 0.35–0.6, p < 0.001). Crizotinib demonstrates significant anti-tumor efficacy in NSCLC patients with MET exon 14 mutations^317–320 and advanced metastatic papillary renal cell carcinoma type 1 (PRCC1) with MET alterations^318,321. Crizotinib is the only FDA-approved type I MET inhibitor for the treatment of locally advanced or metastatic NSCLC that is positive for anaplastic lymphoma kinase (ALK) or rearranged ROS1^319,320. Phase II/III clinical trials are ongoing in papillary renal cell carcinoma (PRCC), NSCLC, neurofibromatosis type 2, and progressive vestibular schwannomas (NCT02761057, NCT02737501, NCT04084717, NCT04322578, NCT05642572, and NCT04283669).

Capmatinib (INC280, INCB028060, developed by Incyte Corporation and Novartis Pharmaceuticals Corporation)

Capmatinib is a highly selective MET inhibitor (IC50 = 0.13 nM) with strong central nervous system (CNS) penetration. It inhibits phosphorylation of major downstream effectors of the MET pathway, including ERK1/2, AKT, FAK, GAB1, and STAT3/5, and results in the regression of MET-dependent (amplified/autocrine) tumor xenograft models at well-tolerated doses³²². In PDAC cell lines, irradiation can induce MET expression, leading to enhanced malignant traits such as invasion and migration. Radiation has also been shown to elevate the level of phosphorylated MET in a tumor mouse model, which can be suppressed by oral administration of capmatinib³²³.

Capmatinib exhibited both anti-tumor activity and an acceptable safety profile³²⁴. Studies have revealed noteworthy anti-tumor efficacy in advanced NSCLC patients with MET exon 14 skipping mutations^325,326, particularly in treatment-naive individuals^327,328. It also possessed favorable activity in the CNS, with an intracranial response rate of 54%. Consequently, capmatinib received FDA approval on May 6, 2020, for its strong clinical efficacy (overall response rate (ORR) of 41%) in patients harboring MET exon 14 skipping mutations^327,329. Moreover, capmatinib displayed anti-tumor activity in NSCLC with MET alterations following prior crizotinib treatment, possibly attributed to overlapping resistance mechanisms³³⁰. Dual inhibition of EGFR and MET has demonstrated anti-tumor activity in NSCLC patients with acquired EGFR mutations^326,331. Capmatinib is currently under investigation in clinical trials for a wide range of solid tumors, such as breast cancer, genitourinary tumors, NSCLC, metastatic castration-resistant prostate cancer (CRPC), HCC, thyroid cancer and soft tissue sarcomas (NCT02414139, NCT03611595, NCT04677595, NCT04926831, NCT03866382, NCT03149822, NCT04322955, NCT04413123, NCT04586231, NCT03333343, NCT04310007, NCT04471428, NCT05243641, NCT04631744, NCT03539822, NCT04588051, NCT04511455, NCT03690388, NCT01896479, NCT04200443, and NCT04551430). The most common adverse reactions were mild nausea, peripheral edema, rash, constipation, fatigue, and increased lipase³³².

Savolitinib (AZD 6094, HMPL-504, Volitinib, developed by AstraZeneca, Hutchison Medipharma Limited, NCI, and The First Affiliated Hospital of Guangzhou Medical University)

Savolitinib is a potent, highly selective MET TKI (IC50 = 4.0 nM). Treatment with savolitinib led to pharmacodynamic modulation of MET signaling and tumor stasis in all three MET-dysregulated gastric cancer patient-derived xenograft (PDX) models¹⁴³. Savolitinib also demonstrated effective antitumor activity in a human glioma xenograft model using athymic nude mice³³³. A Phase Ib study showed that savolitinib in combination with gefitinib/osimertinib had an acceptable risk-benefit profile and encouraging anti-tumor activity in patients with MET-amplified, EGFR mutation-positive advanced NSCLC^334,335. Phase II studies indicated that savolitinib exhibited promising clinical efficacy in NSCLC patients harboring MET exon 14 skipping mutations. Consequently, savolitinib was approved by the National Medical Products Administration of China on June 22, 2021³³⁶. Moreover, savolitinib poseessed anti-tumor activity in patients with MET-driven PRCC in Phase II/III studies^337,338. Savolitinib is currently in clinical trials for the treatment of multiple solid tumors, including gastric cancer and esophagogastric junction adenocarcinoma, PRCC, and NSCLC (NCT04923932, NCT03091192, NCT05043090, NCT02143466, NCT03778229, NCT04606771, and NCT04923945).

Foretinib (XL880, EXEL-2880, GSK1363089, developed by GSK)

Foretinib is a multi-kinase inhibitor targeting MET, AXL, VEGFR, ROS, RON, and TIE-2^173,339,340. The initial Phase I clinical trial established a recommended dose of 240 mg of foretinib, which was administered during the first 5 days of a 14-day cycle. This regimen resulted in 3 partial response and 22 stable disease cases among 40 patients, confirming the safety and antitumor activity of foretinib³⁴¹. Several Phase I/II studies also provided evidence of tumor regression in patients with advanced PRCC, TNBC, NSCLC, and HCC following treatment with foretinib^342,343. However, a Phase II study in gastric cancer revealed insufficient efficacy of foretinib as a single-agent in MET-amplified metastatic gastric cancer patients, emphasizing the need for a deeper understanding of gastric cancer oncogenesis and refined selection criteria of patients. The toxicity profile of foretinib was generally manageable, with fatigue and hypertension being the most common side effects³⁴³.

Tepotinib (Tepmetko, MSC2156119J, developed by Criterium, Inc./Merck KGaA, Darmstadt, Germany/Chungbuk National University Hospital/EMD Serono Research & Development Institute, and Inc./M.D. Anderson Cancer)

Tepotinib is a highly selective, potent ATP-competitive inhibitor of MET, with >1000-fold selectivity for MET compared with 236 out of 241 other kinases tested, and >200-fold selectivity over the remaining five kinases³⁴⁴. It effectively suppressed the growth of MET-positive HCC cell lines and xenograft models^345,346. In a first-in-human study in US/European patients with various solid cancers (including HCC), tepotinib was generally well tolerated and demonstrated anti-tumor activity in tumors harboring MET alterations^347,348. In a Phase II study conducted on NSCLC patients with MET exon 14 skipping, tepotinib displayed consistent clinical activity and maintained its tolerability, notably highlighting its sustained effectiveness in the subgroup of Asian patients. The results of this study led to the approval of tepotinib and its companion diagnostic test (ArcherMET CDx) by Japan’s Pharmaceuticals and Medical Devices Agency on 25 March 2020³⁴⁹.

The combined therapy of tepotinib and an EGFR TKI overcame resistance to EGFR TKIs in EGFR-mutated NSCLC with MET amplification or high MET expression. Several Phase I/II trials suggested that the combination of tepotinib and gefitinib in advanced NSCLC patients harboring EGFR mutations and acquired MET amplification resulted in superior antitumor activity compared to chemotherapy²⁵². Tepotinib also exhibited promising efficacy and tolerability in HCC patients who had previously responded unfavorably to sorafenib treatment³⁵⁰. Tepotinib is currently in clinical trials for multiple solid tumors, including NSCLC, gastric, and gastroesophageal junction carcinoma (NCT05782361, NCT03940703, NCT06083857, NCT04647838, and NCT05439993). The most common adverse reactions were peripheral edema, increased amylase, and lipase concentrations^351,352.

BPI-9016M (developed by Betta Pharmaceuticals Co., Ltd, Hangzhou, China)

BPI-9016M, a novel small-molecule inhibitor, targets both MET and AXL tyrosine kinases. In preclinical studies, BPI-9016M inhibited multiple kinases in vitro at a concentration of 0.2 μM, including MET, AXL, KDR, DDR2, and Ron, with inhibition rates of 88–100%³⁵³. In a PDX model of lung adenocarcinoma with high MET expression, BPI-9016M demonstrated anti-tumor activity, potentially through the up-regulation of miR-203, leading to reduced DKK1 expression³⁵⁴. A Phase I clinical trial of BPI-9016M in NSCLC is currently ongoing (NCT02929290). Common toxicities encompassed elevated levels of ALT, AST, and bilirubin, as well as dysgeusia, constipation, hypertension, and palmar-plantar erythrodysesthesia syndrome³⁵⁵.

PHA-665752 (developed by Pfizer)

PHA-665752 is a highly selective inhibitor of MET, with selectivity for MET (IC50 = 9 nM) over 50 times greater than that for various other tyrosine kinases and serine-threonine kinases. PHA-665752 effectively inhibited MET autophosphorylation and phosphorylation of downstream signaling proteins such as AKT, STAT3, PLC-γ, Gab-1, ERK, and FAK, leading to suppressed cell proliferation and motility^356,357. Treatment with PHA-665752 on 5 gastric cancer cell lines with high MET expression led to significant apoptotic effects, while no inhibitory effect on proliferation was observed in 12 cell lines without MET gene amplification (p = 0.00016)³⁵⁸. The combination of PHA-665752 and cetuximab resulted in greater suppression of CRC cell growth both in vitro and in vivo compared to using either agent alone³⁵⁹.

ABN401 (developed by ABION Inc. R&D Center, Seoul, Korea)

ABN401 is a next-generation, highly selective MET TKI with promising therapeutic potential in MET-dysregulated cancers³⁶⁰. Unlike earlier quinoline-based inhibitors prone to aldehyde oxidase-mediated degradation, ABN401 demonstrates metabolic stability in human liver cytosol^360–362. It selectively targets MET among 571 kinases and exhibits potent cytotoxicity in MET-addicted cancer cells by inhibiting MET autophosphorylation and downstream signaling pathways³⁶³. In preclinical models, including NSCLC patient-derived xenografts, ABN401 displayed significant antitumor activity, particularly when combined with erlotinib, resulting in up to 90% tumor suppression^360,363. It also achieved favorable pharmacokinetic profiles across multiple species and received Investigational New Drug (IND) approval in South Korea in June 2019³⁶⁰. ABN401 is currently being evaluated in a completed Phase I clinical trial (NCT04052971) in patients with advanced solid tumors, although the results have not yet been published.

MK-2461 (developed by Merck)

MK-2461 is a novel ATP-competitive multi-targeted inhibitor of activated MET, with IC50 values ranging from 0.4 to 2.5 nmol/L. The inhibitory activity was also significant against other receptor tyrosine kinases, including FGFR, PDGFR and other RTKs³¹³. In vitro, MK-2461 inhibited HGF/MET dependent cellular processes like mitogenesis, migration, and tubulogenesis, particularly in tumor cell lines with MET or FGFR2 amplification. In murine gastric cancer xenograft models, a twice daily oral regimen of 100 mg/kg MK-2461 effectively suppressed tumor growth, supporting its progression into preclinical development³¹³.

Type II MET inhibitors

Cabozantinib (XL184, BMS-907351, Cometriq®, developed by Exelixis/Ipsen)

Cabozantinib is a non-selective MET kinase inhibitor that can act on multiple RTKs, including VEGFR2 (IC50 = 0.035 nM), MET (IC50 = 1.3 nM), RET (IC50 = 4 nM), KIT (IC50 = 4.6 nM), and AXL (IC50 = 7 nM), among others. In vitro studies have revealed that cabozantinib inhibited the phosphorylation of MET and VEGFR2 and their downstream pathways, suppressing migration and invasion. In mouse models of breast cancer, lung cancer, and glioma, cabozantinib inhibited the proliferation of tumor and endothelial cells in a dose-dependent manner³⁶⁴. Treatment with 30 mg/kg cabozantinib in RIP-Tag2 mice carrying spontaneous pancreatic islet tumors was sufficient to result in an 80% reduction in tumor vasculature, decreased pericytes, extensive hypoxia and tumor cell apoptosis, with significantly delayed regrowth of tumor vessels even after treatment cessation³⁶⁵. In a model of malignant peripheral nerve sheath tumor (MPNST), cabozantinib inhibited tumor growth and metastasis capabilities³⁶⁶. Numerous Phase I/II trials have been conducted in patients with RCC, melanoma, metastatic breast cancer, NSCLC, CRPC, metastatic urothelial carcinoma, and HCC with promising anti-tumor activities^367–374. Cabozantinib was also shown to reduce tumor volume and alleviate pain in NF1-associated peripheral neuropathy (PN) patients in a Phase II trial (NCT02101736)³⁷⁵. A Phase II trial (NCT03729297) indicated modest efficacy of cabozantinib in managing patients with MET-positive recurrent and/or metastatic salivary gland cancer, but reported serious wound complications, particularly in previously irradiated areas³⁷⁶. However, Phase III studies involving cabozantinib for the treatment of prostate cancer, RCC, HCC, and medullary thyroid carcinoma did not identify wound complications as a prominent concern^377–380. The capsule formulation of cabozantinib, Cometriq®, has been approved by the U.S. FDA for the treatment of advanced metastatic medullary thyroid carcinoma, RCC and HCC^379–382. Some studies reported fistula/perforation as severe side effects^383–386, other common adverse events included hypertension, hand-foot syndrome, and fatigue^386–388.

BMS-777607 (ASLAN002, developed by Bristol-Myers Squibb/Aslan Pharmaceuticals)

BMS-777607 is a small molecule inhibitor that acts on AXL, RON, MET, and TYRO3 with IC50 values of 1.1 nM, 1.8 nM, 3.9 nM, and 4.3 nM, respectively, in in vitro cell-free assays. At higher concentrations, BMS-777607 can also act on other TKIs, such as MER (IC50 = 14.0 nmol/L), FLT-3 (IC50 = 16 nmol/L), Aurora B (IC50 = 78 nmol/L), LCK (IC50 = 120 nmol/L), and VEGFR2 (IC50 = 180 nmol/L)³⁸⁹. BMS-777607 selectively inhibited the proliferation and colony formation of MET-overexpressing tumor cells, such as GTL-16, H1993, U87, PC-3, and DU145 cells, by suppressing MET autophosphorylation and the phosphorylation of downstream signaling molecules including ERK, AKT, p70S6K, and S6^{306,390–392}. In vivo studies showed that oral administration of a low dose of BMS-777607 (6.25 mg/kg) in nude mice with MET/RON-overexpressing GTL-16 human gastric cancer xenografts only partially inhibited tumor growth, whereas a high dose of BMS-777607 (50 mg/kg) demonstrated significant therapeutic effects. This might be due to low concentrations of BMS-777607 primarily acting on RON/MET, while higher concentrations of the drug could also inhibit other RTKs³⁹². In a KHT fibrosarcoma model, daily oral administration of 25 mg/kg BMS 777607 to female C3H/HeJ mice aged 6-8 weeks reduced the formation of metastatic lung nodules without significant toxic side effects³⁰⁶. Phase I/II clinical trials of BMS-777607 in patients with advanced solid tumors (NCT01721148, NCT00605618) have been completed. A Phase I clinical trial suggested favorable tolerability of BMS-777607 at a dose of 300 mg BID, establishing it as the recommended dose for subsequent Phase II studies³⁹³. However, the results of Phase II studies have not yet been released.

Merestinib (LY2801653, developed by Eli Lilly and the Dana-Farber Cancer Institute)

Merestinib (LY2801653) is a small-molecule inhibitor that targets multiple receptor tyrosine kinases, including RON, MET, AXL, and FLT-3^394,395. Merestinib inhibited the proliferation of Hs746t gastric cancer cells harboring a MET exon 14 skipping mutation (IC50 = 34 nM), and completely eliminated p-MET at 100 nM³⁹⁶. Phase I/II clinical trials demonstrated acceptable safety and potential anti-tumor activity of merestinib as monotherapy and/or in combination with other therapies in patients with metastatic colorectal cancer (mCRC) and cholangiocarcinoma^15,397–400. LY2801653 is currently being investigated in patients with AML, BTC, CRC, and other solid tumors (NCT02745769, NCT03125239, NCT02711553, and NCT02791334).

MGCD265 (Glesatinib, developed by Mirati Therapeutics)

MGCD265, a small molecule multi-targeted TKI, selectively inhibits MET, VEGFR1/2/3, RON, TIE-2, and AXL⁴⁰¹. In TNBC tumor xenografts, the combination of MGCD265 and erlotinib, which jointly inhibited MET and EGFR, showed more potent tumor growth inhibition than monotherapy⁴⁰². Preliminary efficacy was observed in patients with solid tumors harboring MET aberrations when treated with the combination of MGCD265 and either erlotinib or docetaxel⁴⁰². Phase I/II clinical trials of MGCD265 demonstrated antitumor activity, particularly in NSCLC patients with MET mutations or amplification. The recommended dose was 750 mg BID (SDD tablet), with common adverse events like diarrhea, nausea, and elevated liver enzymes. Although initial results were promising, the Phase II trial was terminated early due to modest clinical activity^403,404.

Type-III MET inhibitors

Tivantinib (ARQ 197, developed by ArQule, Daiichi Sankyo, and Kyowa Hakko Kirin) is the first non-ATP competitive MET inhibitor that selectively inhibits MET. It not only suppresses MET phosphorylation and downstream signaling pathways, but also results in G2-M phase cell cycle arrest, affecting tumor cell proliferation and migration⁴⁰⁵. Tivantinib induced caspase-dependent cell apoptosis in MET-driven CRC, gastric cancer, and breast cancer cells, and reduced tumor growth in xenograft models⁴⁰⁵. A Phase II study (NCT00988741) demonstrated that oral tivantinib significantly improved PFS and OS in a selected cohort of patients with highly expressed MET proteins²⁸⁹. This observation led to further investigations in two Phase III trials, METIV-HCC and JET-HCC, designed to validate the therapeutic benefit of tivantinib in HCC patients with high MET expression^406,407. However, both trials failed to confirm the initial hypothesis. The JET-HCC trial, conducted in Japanese patients with MET-high HCC, showed only a modest improvement in PFS (2.8 vs. 2.3 months, HR = 0.74, p = 0.082) and no statistically significant improvement in OS (10.3 vs. 8.5 months, HR = 0.82)⁴⁰⁶. Similarly, the METIV-HCC trial, conducted in Western populations, reported no OS benefit for tivantinib compared with placebo (8.4 vs. 9.1 months, HR = 0.97, p = 0.81)⁴⁰⁷. Both trials also revealed a high incidence of grade ≥3 adverse events, particularly neutropenia⁴⁰⁸. The failure of these trials might be due to differences in patient selection methods. In the Phase II study, MET expression was evaluated retrospectively, allowing for a broader patient population, whereas in Phase III, only preselected MET-high patients were enrolled, potentially altering the study population and excluding certain responders. Additionally, MET overexpression alone might not be a reliable predictor of the response to tivantinib, as other molecular factors, such as MET amplification or coexisting mutations, could influence treatment outcomes. Lastly, the prolonged biomarker screening process in Phase III might have given rise to the exclusion of rapidly progressing patients, further affecting the trial results. These findings suggest that future trials should refine patient selection criteria and incorporate additional biomarkers to better identify those most likely to benefit from MET-targeted therapy.

Anti-MET monoclonal antibodies

Several well-known anti-MET mAbs have been developed to target MET (Table 2). Emibetuzumab (LY2875358) is a bivalent antibody that not only binds to MET, interfering with the interaction between the ligand HGF and MET to inhibit ligand-dependent MET signaling, but also induces secondary internalization and degradation of the MET receptor, reducing membrane MET expression and thereby inhibiting ligand-independent MET signaling pathways⁴⁰⁹.

Table 2.

Anti-MET or anti-HGF monoclonal antibodies, antibody–drug conjugates (ADCs), nucleic acid aptamers, natural compounds, and proteolysis targeting chimeras (PROTACs) targeting MET are currently being investigated

Name	Type	Target	Indications	Phase of clinical trials
Emibetuzumab (LY2875358)	Monoclonal antibody	MET	NSCLC, GC	I/II
Onartuzumab (OA-5D5, MetMAb, OAM4558g)	Monoclonal antibody	MET	HCC, NSCLC, BRCA, CRC, GEC, TNBC	I/II/III
Anti-MET404	Monoclonal antibody	MET	Glioblastoma	Preclinical
Ficlatuzumab	Monoclonal antibody	HGF	HNSCC	III
Rilotumumab	Monoclonal antibody	HGF	GC/GEC	III
YYB-101	Monoclonal antibody	HGF	AST	I
TAK-701	Monoclonal antibody	HGF	AST	I
Telisotuzumab vedotin (ABBV-399, Teliso-V)	ADC	MET	NSCLC, AST	II/III
REGN5093-M114	ADC	MET	NSCLC	I/II
MM-131	ADC	MET	MET-positive tumors	Preclinical
SL1	DNA nucleotide aptamer	MET	Multiple myeloma	Preclinical
CLN64	RNA nucleotide aptamer	MET	Lung Cancer	Preclinical
Curcumin	Natural compound	MET	CRC, PC, CC, Glioblastoma, pancreatic cancer, BRCA, CRC, AST	Preclinical, I/II/III
Celastrol	Natural compound	MET	HCC	Preclinical
Epigallocatechin-3-gallate	Natural compound	MET	BC, PC, Skin cancer, BRCA, CRC, SCLC, CC	Preclinical, I/II/III
MET-DD4	PROTAC	MET	NSCLC, GC	Preclinical

ADC antibody-drug conjugate, NSCLC non-small cell lung cancer, GC gastric cancer, HCC hepatocellular carcinoma, BRCA breast cancer, CRC colorectal cancer, GEC:gastroesophageal cancer, TNBC triple negative breast cancer, HNSCC head and neck squamous cell carcinoma, AST advanced solid tumor, PC prostate cancer, CC cervical cancer, BC Bladder cancer, SCLC small cell lung cancer, PROTAC proteolysis targeting chimera

The highly specific MET antibody Onartuzumab (OA-5D5, MetMAb, OAM4558g) binds to the MET receptor, preventing its activation by the ligand HGF, inhibiting downstream pathways, and promoting the internalization and degradation of MET. In a Phase II clinical trial involving MET-positive NSCLC patients, the combination of erlotinib and onartuzumab resulted in a significantly longer OS (12.6 months) compared to erlotinib alone (3.8 months)⁴¹⁰. Unfortunately, this result was not replicated in a similarly designed Phase III trial (METLung), which was stopped early due to futility⁴¹¹. Potential contributing factors include suboptimal biomarker selection, such as reliance on archival tissue for IHC analysis, which may not accurately reflect current MET expression levels, as well as inconsistent cutoff definitions across studies. Additionally, the IHC assay used in METLung targeted an epitope different from that of onartuzumab, potentially compromising its diagnostic specificity^412,413.

ADCs such as telisotuzumab vedotin, MM-131, and REGN5093-M114 have the potential to enhance anti-tumor activity and minimize toxicity through targeted delivery of potent chemotherapeutic agents. Initial findings from ongoing clinical investigations of these ADCs were promising. Telisotuzumab vedotin (ABBV-399, Teliso-V) is a first-in-class ADC that combines a recombinant MET-targeting humanized mAb (ABT-700) with monomethyl auristatin E (MMAE), a potent inhibitor of microtubule polymerization, via a cleavable linker^414,415. Promising results included significant tumor growth inhibition and regression in MET- amplified cell lines and PDX models. Telisotuzumab vedotin also demonstrated efficacy against xenograft tumors resistant to other MET inhibitors and synergized with standard-of-care chemotherapy⁴¹⁶. A Phase Ib clinical trial (NCT02099058) indicated that Telisotuzumab vedotin in combination with erlotinib exhibited favorable antitumor activity and manageable toxicity in EGFR TKI-pretreated patients with an EGFR-activating mutation, MET-positive NSCLC^417,418.

MM-131 is a monovalent bispecific antibody targeting MET and epithelial cell adhesion molecule (EpCAM). This purely antagonistic antibody blocked both ligand-dependent and ligand-independent MET signaling by obstructing HGF binding and triggering receptor down-regulation. These mechanisms suppressed MET-driven cancer cell proliferation and migration in vitro and impeded tumor growth in HGF-dependent and HGF-independent mouse xenograft models⁴¹⁹.

Additionally, as aforementioned, the development of an anti-MET404 mAb, targeting the HGF-independent activation of MET by the circular form of C-MET RNA (circMET), offers another promising strategy for MET-targeted therapies.

Anti-HGF monoclonal antibodies

Ficlatuzumab (AV-299, AVEO) is a humanized IgG1 mAb that inhibits HGF-induced MET signaling by blocking HGF-MET binding^420,421. Although its single-agent activity is limited, combinatorial strategies have shown clinical promise^422,423. In a randomized Phase II trial in advanced NSCLC patients (NCT01039948), ficlatuzumab combined with gefitinib did not improve outcomes in the overall NSCLC population, but demonstrated significant benefits in PFS and OS among patients classified as VeriStrat poor⁴²⁴. In recurrent/metastatic HNSCC, a Phase II trial (NCT03422536) suggested that ficlatuzumab combined with cetuximab achieved an ORR of 19% and a median PFS of 3.7 months in cetuximab-resistant, recurrent/metastatic HPV-negative patients, suggesting the need for further Phase III trials⁴²⁵. A Phase III trial (NCT06064877) is currently underway in this setting. In patients with metastatic PDAC, ficlatuzumab combined with gemcitabine and nab-paclitaxel yielded a 29% PR rate and median OS of 16.2 months⁴²⁶. Across studies, common adverse events included edema, hypoalbuminemia, diarrhea, and fatigue⁴²⁵.

Rilotumumab (AMG 102), a humanized IgG2 mAb targeting HGF^427–430, has been evaluated across various cancer types with mixed clinical outcomes. In advanced gastric or gastroesophageal junction cancer, an early Phase Ib/II study (NCT00719550) of rilotumumab combined with ECX chemotherapy (epirubicin, cisplatin, and capecitabine) revealed manageable safety and a signal of improved PFS, supporting further investigation in MET-positive gastric cancer patients⁴³¹. However, the subsequent Phase III RILOMET-1 trial (NCT01697072) failed to confirm this benefit and was terminated early due to increased mortality in the rilotumumab arm⁴³². Median OS was significantly shorter in the rilotumumab group than in the placebo group (8.8 vs. 10.7 months; HR = 1.34, p = 0.003), with no identifiable subgroups deriving benefits. Several factors may explain this failure: (1) treatment-related side effects such as edema and low albumin may have caused early chemotherapy discontinuation, reducing exposure; (2) rilotumumab may have interfered with chemotherapy; (3) the MET IHC assay may not have accurately identified MET-driven tumors; and (4) blocking HGF may be ineffective in tumors driven by MET amplification or not dependent on MET. Similar disappointing results were observed in ovarian cancer⁴³³, castration-resistant prostate cancer⁴³⁴ and NSCLC⁴³⁰. These findings collectively highlight the challenges of targeting the MET/HGF axis with ligand-blocking antibodies and underscore the need for refined biomarker-driven patient selection and a deeper understanding of MET pathway biology before advancing this class of agents in clinical practice.

YYB-101, a humanized rabbit-derived mAb targeting human HGF, effectively neutralizes HGF activity and inhibits MET pathway activation^435,436. Preclinical evidence supports its anti-tumor efficacy, particularly in combination with chemotherapy agents such as paclitaxel or temozolomide, in models of ovarian cancer and glioblastoma^437–440. TAK-701, another humanized mAb derived from the murine mAb L2G7 that targets HGF, has been shown to overcome HGF-induced gefitinib resistance in EGFR-mutant NSCLC models when used in combination with gefitinib⁴⁴¹, but has demonstrated limited efficacy as monotherapy in pediatric solid tumor xenografts⁴⁴² (Table 2).

Nucleotide aptamers

Nucleic acid aptamers are short single-stranded RNA or DNA sequences with a defined structure that can bind to targets as ligands with high affinity and low toxicity^443,444. Nucleic acid aptamers are typically cost-effective to produce, easy to manufacture, exhibit low immunogenicity, and possess high stability^445,446. They can also serve as delivery vehicles for transporting nanoparticles, chemotherapy drugs, or small interfering RNAs, making them promising candidates for cancer therapy^447,448. The RNA aptamer CLN64 targeting MET could inhibit HGF-dependent activation and signaling of MET. The DNA aptamer SL1 bound to MET with high affinity and specificity, inhibiting the HGF/MET downstream signaling pathway in SNU-5 cells⁴⁴⁹. SL1 selectively suppressed the growth, invasion, and migration of MET-positive multiple myeloma cells, and could work synergistically with bortezomib to inhibit the growth of CD138-positive multiple myeloma cells⁴⁵⁰. However, due to the relatively recent discovery of nucleic acid aptamers, further experiments are required to explore their clinical applications (Table 2).

Soluble MET (sMET)

A considerable body of evidence suggested that the extracellular fragments of MET could be shed from cell surface both under normal physiological conditions and in tumors^54,451,452. Consequently, these soluble MET ectodomains might act as decoy receptors, downregulating the biological effects of HGF and MET^{54,171,451,453,454}. In PC3 and 4T1 cells, both human and murine MET receptor-Fc fusion proteins effectively inhibited MET phosphorylation and HGF-induced cellular invasion in a dose-dependent manner, and hampered the growth of tumor xenografts with manageable toxicities⁴⁵³. Plasma sMET levels have also been suggested as biomarkers to predict prognosis in advanced NSCLC, multiple myeloma, and gastric cancer patients⁴⁵⁵.

Natural compounds

The inhibitory effects of certain natural compounds on the MET pathway have been explored in both preclinical and clinical studies (Table 2). Curcumin (diferuloylmethane), a bioactive compound from Curcuma longa^456,457, has been widely studied for its anticancer effects. It reveals a good safety profile but limited bioavailability⁴⁵⁸. A Phase I trial recommended a dose of up to 3.6 g/day for the treatment of various cancers, excluding those of the gastrointestinal tract⁴⁵⁹. Curcumin has been shown to inhibit HGF-induced MET activation and suppress tumor growth, invasion, migration, and EMT in several cancer types, including liver, lung, prostate, and oral squamous cell carcinoma^460–463. In vivo xenograft models of PC9 lung tumors indicated that curcumin, alone or in combination with agents like erlotinib, could inhibit tumor progression⁴⁶⁴. Additionally, curcumin derivatives, such as hydrazinobenzoylcurcumin (HBC), exhibited similar inhibitory effects on glioblastoma cell progression⁴⁶⁵.

Celastrol, a quinine methide triterpenoid from Tripterygium wilfordii^466,467, has displayed antitumor effects in combination with the MET inhibitor PHA-665752, suppressing tumor growth and metastasis in HCC models⁴⁶⁸. While the safety of celastrol is being evaluated in clinical trials (NCT05494112), its effects on cancer patients remain unstudied⁴⁶⁹.

Epigallocatechin gallate (EGCG), a key polyphenol from green tea^470,471, has demonstrated antitumor effects by modulating pro-inflammatory pathways and acting as an antioxidant⁴⁷⁰. It inhibited HGF-stimulated MET signaling in various cancers, including breast, hypopharyngeal, CRC, prostate, NSCLC and melanoma^{469,472–478}. Studies revealed that EGCG suppressed tumor growth, invasion, and metastasis both in vitro and in vivo, often enhancing the effects of MET inhibitors like erlotinib or crizotinib^476,479.

PROTACs targeting MET

PROTACs are heterobifunctional molecules designed to induce targeted protein degradation via the ubiquitin-proteasome system (UPS)⁴⁸⁰. They consist of two ligands connected by a linker, with one ligand binding to a protein of interest (POI), while the other recruits an E3 ubiquitin ligase, leading to POI ubiquitination and subsequent degradation. Unlike traditional inhibitors, PROTACs eliminate disease-associated proteins rather than merely inhibiting them^481,482, offering advantages such as catalytic activity, lower required doses, better tolerability, the ability to target previously “undruggable” proteins, and the potential to overcome acquired drug resistance^483–487. They have been extensively utilized to give rise to the degradation of a broad range of pathogenic proteins, including transcription factors^488–490, kinases^{482,491–499}, epigenetic regulators^500–503, and immune checkpoint molecules^504–506.

As mentioned before, although MET inhibitors hold promising initial efficacy, resistance often develops due to kinase domain mutations and MET exon 14 amplification⁵⁰⁷. To overcome this, MET-targeting PROTACs have been developed by conjugating MET inhibitors with E3 ligase recruiters such as cereblon (CRBN) and Von Hippel-Lindau (VHL).

PRO-6E is the first orally bioavailable PROTAC for MET-targeted degradation, designed by conjugating the MET inhibitor crizotinib with a linker to recruit cereblon E3 ubiquitin ligase, resulting in efficient MET degradation and enhanced antiproliferative activity⁵⁰⁸.

D10 and D15 are potent, orally active MET-targeted PROTACs developed based on thalidomide and tepotinib. They effectively inhibited cell growth in vitro and in xenograft models. Both compounds also confer efficacy against MET mutations (Y1230H and D1228N) resistant to tepotinib⁵⁰⁹.

48-284, another PROTAC targeting METΔex14, possesses over 15-fold higher potency than the previously reported MET-PROTAC SJF8240⁵¹⁰, exhibits fewer off-target effects, and effectively degrades MET in multiple cell lines^510,511

MET-DD4 is a novel and potent PROTAC that selectively targets MET. It was developed through structure–activity relationship (SAR) studies by optimizing the warhead, linker, and E3 ligase ligand. MET-DD4 achieves strong MET degradation (DC50 = 6.21 nM) and effectively inhibits the growth of MET-dependent cancer cells (IC50 = 4.37 nM). In animal studies, MET-DD4 displays good oral bioavailability, potent antitumor activity, and low toxicity. Compared to earlier MET PROTACs such as MET PROTAC 7 and PRO-6E, it offers improved selectivity and pharmacokinetic properties⁵¹² (Table 2).

The AS1411-SL1 chimeras are novel dual aptamer-functionalized PROTACs designed to target and degrade MET. They are constructed by conjugating the MET-targeting aptamer SL1 with the aptamer AS1411, which binds to nucleolin overexpressed on the surface of tumor cells. Intracellularly, nucleolin interacts with the E3 ubiquitin ligase MDM2, thereby enabling the chimeras to recruit MDM2 for MET ubiquitination and degradation. The AS1411-SL1 chimeras effectively suppress tumor growth and overcome drug resistance in osteosarcoma cells, with no detectable toxicity³⁰⁵.

Although current MET-PROTACs have demonstrated improved antitumor efficacy, the ability to overcome acquired resistance, and better tolerability compared to conventional inhibitors, they still face significant challenges, including non-selectivity and potential off-target toxicity. Therefore, the development of tumor-specific MET-PROTAC degraders with enhanced safety and therapeutic profiles remains a critical goal in cancer therapy^513,514.

Future perspectives

MET signaling remains a highly promising but complex target for cancer therapy. Although numerous MET inhibitors, including MET or HGF antagonists, mAbs, and small-molecule TKIs, have entered clinical trials with some positive results, several key challenges remain unresolved. Addressing these issues will be essential for optimizing MET-targeted therapies.

Biological complexity and heterogeneity of MET activation

MET activation can occur via various mechanisms, including gene amplification, exon 14 skipping mutation, gene fusion, protein overexpression, and aberrant ligand production. These alterations lead to varying levels and types of MET signaling activity, which may influence the therapeutic response. Furthermore, the prevalence and functional consequences of these activation mechanisms vary across different tumor types. This complexity suggests that a “one-size-fits-all” approach may not be effective, and future research must focus on tailoring therapeutic strategies to the specific biology of MET alterations in each cancer type.

Need for robust biomarkers and standardized diagnostics

The accurate identification of patients who are likely to benefit from MET-targeted therapies remains a significant challenge. Immunohistochemistry (IHC) for MET protein expression and fluorescence in situ hybridization (FISH) for MET amplification have been widely used, but both assays have important limitations, including technical variability, lack of standardization, and inconsistent interpretation.

IHC assesses MET protein expression based on staining intensity and the proportion of positive tumor cells. However, cutoff definitions differ among studies: the TATTON and SAVANNAH trials defined MET overexpression as IHC 3+ (≥50% of tumor cells with strong staining)^334,515, whereas other studies used IHC 2+ (≥50% of tumor cells with moderate to strong staining, but <50% with strong staining)⁵¹⁶, or IHC 2 + /3+ combined²⁵². An H-score ≥150, calculated by multiplying staining intensity (0–3) by the percentage of positive cells (range: 0–100%), has also been applied⁴¹⁵. Despite its accessibility, IHC is semi-quantitative and subject to inter-observer variability, limiting its reproducibility and predictive value across clinical settings⁵¹⁷.

FISH offers the advantage of distinguishing true MET amplification from chromosome 7 polysomy by calculating the MET/CEP7 ratio, with CEP7 referring to a centromeric probe used to normalize MET copy number (Fig. 2). However, there is no consensus on FISH-based thresholds^14,518–520. Commonly cited criteria include a MET/CEP7 ratio ≥2.0 (e.g., PathVysion kit, Tanaka et al.)^521–523 or ≥5 MET signals per cell (Cappuzzo scoring system)^524–526. The GEOMETRY mono-1 trial defined high-level amplification as MET gene copy number (GCN) ≥ 10³²⁷, while GCN ≥ 5 was used in the INSIGHT and TATTON studies^252,334. A three-tiered classification has also been proposed: MET/CEP7 ratio low (≥1.8 to ≤2.2), intermediate (>2.2 to <5.0), and high (≥5.0)⁵²⁷, with higher ratios more strongly associated with MET dependency and therapeutic response⁵²⁸.

Next-generation sequencing (NGS) platforms offer broader genomic profiling and have emerged as more reliable predictors of therapeutic response, especially for detecting MET exon 14 skipping mutations^529,530. These mutations are now more readily detectable with NGS, but other MET alterations, such as atypical splice variants or low-level amplification, remain underexplored⁵³¹.

Given these challenges, future efforts should focus on establishing harmonized, validated companion diagnostics that integrate IHC, FISH, and NGS data. A multimodal diagnostic approach, guided by standardized and biologically meaningful cutoffs, will be essential for improving precision in patient selection and maximizing the clinical benefits of MET-targeted therapies.

Importance of treatment history and context in MET inhibitor efficacy

The clinical context in which MET activation occurs, either as a primary oncogenic event or as a resistance mechanism, has profound implications for treatment response. MET amplification is a well-established mechanism of acquired resistance to EGFR-TKIs in NSCLC. In such cases, MET inhibitors may restore sensitivity to EGFR blockade when used in combination. However, the biology of de novo MET-driven tumors may be different, with unique dependencies and vulnerabilities. Clinical trials must therefore stratify patients not only by MET alteration status, but also by treatment history, prior exposure to targeted therapies, and co-occurring resistance mutations. Furthermore, treatment endpoints may need to be adapted accordingly, for instance, PFS and ORR may be more appropriate in the salvage setting, while OS and time to resistance may be more relevant in first-line treatment.

Emerging resistance mechanisms and paradoxical effects of therapy

Resistance to MET inhibition, either intrinsic or acquired, is a major barrier to long-term efficacy. Tumors that initially respond to MET-targeted therapies frequently develop resistance via secondary mutations in the MET kinase domain, activation of bypass pathways (e.g., HER2, AXL), or accumulation of MET protein due to disrupted degradation. Compounding this, certain therapeutic interventions may paradoxically increase MET activity. For example, Sorafenib (Nexavar), a well-known oral multikinase inhibitor that suppresses tumor growth and angiogenesis by targeting Raf-MAPK signaling and inhibiting receptor tyrosine kinases, such as VEGF and PDGF receptors, has demonstrated efficacy in various cancers and is approved by the FDA for the treatment of unresectable HCC. However, it may potentially enhance MET activity through inhibiting the activity of DEP-1 phosphatase which dephosphorylates MET at Tyr-1349^532,533. Similarly, treatment of human lung cancer cell lines H1993 and NIH3T3 with MET inhibitors SU11274 and PHA-665752 inhibited phosphorylation at the MET Tyr-1003 site, preventing subsequent MET ubiquitination. This results in reduced internalization and degradation of MET, leading to the accumulation of MET on the cell membrane⁵³⁴. These phenomena illustrate the intricacies of MET regulation and the need for greater caution when designing therapeutic regimens. Preclinical models should carefully assess whether MET-targeted agents could unintentionally potentiate tumor growth under specific conditions.

Limited efficacy of MET monotherapy and rationale for combination approaches

While MET inhibitors such as capmatinib and tepotinib have exhibited efficacy in METΔex14-altered NSCLC, responses in tumors with MET overexpression or MET amplification alone are less consistent. The limited effectiveness of MET monotherapy in broader populations may be due to biological redundancy in signaling pathways or insufficient pathway addiction. For instance, the selective MET inhibitor PHA-665752 has been shown to inhibit proliferation only in gastric cancer cell lines with high MET expression^535–537. However, since MET gene amplification is observed in only 10–20% of gastric cancer patients, the clinical utility of such inhibitors is limited by narrow patient applicability, resistance development, and modest monotherapy efficacy^{357,538–540}. Consequently, combination therapies are increasingly being investigated to overcome resistance and enhance therapeutic durability. Co-targeting MET and EGFR in tumors with dual pathway activation has demonstrated clinical benefit⁵⁴¹. In addition, MET inhibition may synergize with agents targeting VEGFR, MEK, or PI3K/AKT pathways. The design of such combinations should be grounded in strong mechanistic rationale, supported by preclinical evidence, and carefully matched to tumor-specific signaling profiles. Prospective trials should also incorporate biomarkers that reflect pathway co-dependencies.

MET’s role in immune cells and the tumor microenvironment

Recent studies have revealed a dual and context-dependent role of MET in cancer biology. While MET mutations or amplifications drive tumor cell proliferation, survival, and metastasis in various cancers, MET is also expressed in tumor-associated neutrophils. In this immune cell population, MET is upregulated by inflammatory stimuli such as TNF-α and is required for neutrophil infiltration into tumors and the production of inducible nitric oxide synthase (iNOS) upon HGF stimulation. This nitric oxide release contributes to neutrophil-mediated tumor cell killing. Deletion of MET in neutrophils reduces their recruitment and cytotoxic function, resulting in accelerated tumor growth and metastasis. These findings suggest that systemic MET inhibition may inadvertently compromise antitumor immunity by affecting neutrophil function. Therefore, more refined therapeutic approaches are needed, which selectively target MET in cancer cells while preserving or even enhancing its immune-regulatory role in neutrophils. Additionally, this immunomodulatory role of MET points to potential utility of MET inhibitors in certain inflammatory diseases²¹⁰.

Complexity of MET variants and co-mutations in therapy design

Truncated MET variants such as MetΔ7–8, which lack extracellular domains, present unique therapeutic challenges⁵³. These forms are constitutively active but cannot be effectively targeted by mAbs due to the absence of the extracellular epitope. Additionally, co-occurring genetic alterations, such as loss of p53, may modulate MET dependency and impact treatment response. For instance, MET-driven tumors with wild-type p53 are more likely to undergo apoptosis upon MET inhibition^542,543, suggesting a possible combinatorial biomarker. Furthermore, MET signaling has been implicated in epigenetic reprogramming through stabilization of the Ets2–MLL complex, which drives histone H3K4 methylation and the upregulation of metastasis-associated genes like MMP1 and MMP3⁵⁴⁴. This layered regulation indicates that future MET-targeted therapies may benefit from a multi-dimensional approach incorporating mutational, transcriptional, and epigenetic data.

Novel targeting strategies and genome editing prospects

Innovative approaches to MET inhibition are currently being explored beyond traditional kinase inhibitors and antibodies. One promising direction involves disrupting protein–protein interactions that activate MET. For example, blocking the FasL–MET interaction with peptides or small molecules can significantly reduce MET phosphorylation and downstream signaling⁵⁴⁵. Caspase-mediated cleavage of MET at specific aspartate sites (D1000 and D1374) has also been shown to generate pro-apoptotic fragments⁵⁴⁶, offering a potential dual therapeutic benefit. Additionally, advances in CRISPR-based gene editing offer the theoretical possibility of correcting oncogenic MET mutations at the genomic level⁵⁴⁷, although this remains in the early stages. Furthermore, combinatorial regimens that pair MET inhibitors with epigenetic agents such as SAHA (vorinostat) may overcome resistance. SAHA has been reported to activate MET via integrin α5β1 in an HGF-independent manner, suggesting that co-targeting could enhance its efficacy in solid tumors⁵⁴⁸. All these findings may support further investigation of combination strategies.

Conclusions

Targeting MET holds promising prospects in anti-tumor therapy due to its roles in tumor biology and drug resistance. High expression of MET has been observed in various tumors, correlating with poor prognosis in cancer patients and resistance to chemotherapy, molecular targeted therapy, and immunotherapy. Inhibiting MET can impact multiple aspects of tumor progression, including growth, survival, invasion, metastasis, stem cell characteristics, angiogenesis, drug resistance, and the EMT process, ultimately improving patient survival. While early clinical trials of MET inhibitors alone or in combination with other therapies have yielded encouraging results, further research is needed on the activation mechanisms of HGF/MET signaling pathways, downstream pathways, and the interactions between MET and other RTKs. More comprehensive multicenter clinical studies are required to evaluate the efficacy and safety of MET inhibitors, their potential impact on other molecular targeted drugs, and most importantly, to identify suitable patients for these inhibitors in advance. This will help optimize patient care and achieve true personalized therapy for cancer patients.

Author contributions

Z.C.J. and H.X. provided the idea. Z.C.J. and L.Y. wrote the article. X.H.Z., S.L., W.Y.C. helped with the literature search, M.C.X. and C.Q.J. participated in figure painting. All authors reviewed the manuscript and approved the final manuscript.

Data availability

All data supporting the findings of this study are available within the article.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41392-026-02682-9.