Gallbladder cancer(GBC) is a relatively rare primary malignancy, but it is the most common malignant neoplasm of the biliary tract, and one of the most aggressive cancers of the biliary tract with poor prognosis [1]. A variety of genetic and epigenetic alterations are associated with GBC, involving tumor suppressor genes, oncogenes, and DNA repair genes. Previous studies have focused on the importance of oncogenes (epidermal growth factor receptor [EGFR], K-Ras, and ERBB2), tumor suppressor genes (p53), cell cycle regulators (cyclin D1, cyclin E), and micro-RNAs in the development and prognosis of GBC. However, the precise information regarding genetic and molecular alterations in GBC is still poorly understood [2]. Li et al. [3] investigated genetic alterations in 57 GBC cases by whole exome and targeted sequencing, and found frequent mutations of TP53, KRASK, and ERBB3. Moreover, ErbB signaling, including EGFR, ERBB2, ERBB3, ERBB4, and their downstream genes is the most extensively mutated pathway, affecting 36.8% of patients with GBC. Although cytotoxic chemotherapy with the combination of gemcitabine and cisplatin has been accepted as the first-line therapy, the patient outcome is still disappointing. After ErbB signaling pathway was found to be one of the pathways contributing to gallbladder carcinoma, target-oriented agents, including monoclonal antibodies and tyrosine kinase inhibitors against EGFR and vascular endothelial growth factor have entered clinical trials [4,5]. However, as those clinical trials revealed no, or marginal benefits, novels targets are needed to improve the patient outcomes.

The mesenchymal epithelial transition (MET) receptor tyrosine kinase pathway is one of the most commonly activated signaling pathways in primary human cancers. MET has a heterodimeric structure composed of an extracellular α subunit and a β chain that contains an extracellular domain, a transmembrane segment, and an intracellular tyrosine kinase region [6]. Activation of MET signaling has a critical role not only in tumor formation, progression, angiogenesis, and metastasis, but is also involved in mediating resistance against other targeted treatments [7,8]. Hepatocyte growth factor (HGF), a ligand of MET, stimulates the tyrosine kinase activity of MET and activates multiple downstream signaling pathways, including the Grb2-Ras–mitogen-activated protein kinase (MAPK) cascade, the phosphatidylinositol-3 kinase (PI3K) pathway, the Wnt/β-catenin pathway, and the signal transducer and activator of transcription (STAT) pathway [9]. Dysregulation of MET protein has been reported in a variety of human malignancies and premalignant lesions, including hepatocellular carcinoma (HCC) and non-small cell lung cancer (NSCLC) [10,11]. Accordingly, the possibility of targeting the MET gene in different cancer types is now being evaluated in clinical trials. Several drugs have been developed that target MET or HGF. These agents are divided into small molecule inhibitors and monoclonal antibodies. The small molecule tyrosine kinase inhibitors are further subdivided into multiple kinases and selective MET inhibitors. Crizotinib and cabozantinib are examples of multikinase MET inhibitors. Selective MET inhibitors include capmatinib and tepotinib. Monoclonal antibody therapy is divided into anti-MET antibodies, such as onartuzumab and emibetuzumab, and anti-HGF antibodies, including ficlatuzumab and rilotumumab [12].

As MET is a potential therapeutic target, information regarding its prevalence and clinicopathological relevance is important. However, the prevalence of MET overexpression and gene amplification in GBCs and its clinicopathological significance is not yet known precisely. Therefore, this study evaluated the expression level and gene copy number of MET in GBC tissues and their correlations with various clinicopathological characteristics, including patient’s prognosis.

MET is a transmembrane protein, which acts as a receptor of HGF. The binding of HGF to the MET protein triggers several downstream signaling pathways, including PI3K/AKT, MAPK, Wnt/β-catenin, and STAT signaling pathways [19]. Activation of these pathways can drive various biological processes related to cancer, such as cell proliferation, migration, invasion, inhibition of apoptosis, epithelial to mesenchymal transition, and angiogenesis [8,19]. MET was first identified in a human osteosarcoma as an oncogene, in the form of TPR-MET fusion [20]. Then, dysregulation of MET protein and amplification had been found in several human malignancies including gastric [21], colorectal [22], breast [23], and NSCLC [10].

The oncogenic role of MET was also investigated in biliary tumors, including intra- and extrahepatic cholangiocarcinomas and GBCs. MET overexpression has been observed in the neoplastic glandular epithelium of furan-induced rat cholangiocarcinomas [24]. Previous studies with clinical samples have investigated MET expression by IHC and demonstrated that the prevalence of MET overexpression in GBCs was variable and ranged from 5% to 74% (Table 5) [25-29]. This variability might be due to the difference in primary antibodies used in IHC and interpretation criteria for overexpression. Moon et al. [28] revealed that MET overexpression was observed in 74% of the GBC cases and associated with invasion depth and increased staining intensity of MET at the invasive front of GBCs, which suggested their important role in tumor invasion [28]. However, Sanada et al. [27] found diffuse membranous staining of MET in the intraductal neoplastic components and no staining in the invasive components, suggesting that MET overexpression might be related to early events of carcinogenesis rather than tumor progression [27]. Yang et al. [26] found that MET overexpression correlated with adverse clinicopathological factors, such as lymph node metastasis, invasion, and decreased OS. In our study, MET overexpression was observed in 39.8% of GBCs. The interpretation criteria of overexpression used in this study was adopted from a previous clinical trial, which revealed that dual MET/EGFR inhibition was clinically beneficial in NSCLC only in the MET overexpression group. However, no significant correlation was observed between MET overexpression and clinicopathological features in our GBC cohort. We performed IHC and SISH on whole tissue sections in subset of cases. Although the different MET immunoreactivity and gene copy number is observed at different areas within tumor, there is no tendency of increased MET immunoreactivity or gene copy number at invasive front as compared to tumor center. Although some previous studies have found no clinicopathological significance of MET overexpression [25,29], other studies have shown significant correlation with worse prognostic factors [26,28]. These discrepancies might be due to different ethnicities of cohorts, the definition of overexpression used, and relatively small number of tumor samples.

Only one ISH study on the MET gene copy number in GBCs was found in the literature, and it revealed that no amplification was detected in 89 cases of GBCs [25]. The previous studies using NGS data reported that MET amplification was rare event in GBCs [30,31]. In our study, no MET amplification was detected by NGS method in 17 available cases. However, we found that 18.3% of GBCs showed MET amplification by ISH. This discrepancy between ISH and NGS methods for MET copy number evaluation may be resulted from intratumoral heterogeneity, contamination of non-neoplastic stromal and immune cells, and different threshold for detecting copy number alteration in high throughput data. However, the most important strength of tissue section based ISH method is that we can discriminate tumor cells from other non-neoplastic components. The tumor cellularity of GBC used in sequencing is generally low and usually less than 50%. In the previous study, there have been reported intratumoral heterogeneity of MET expression in solid tumors [32,33]. We also identified intratumoral heterogeneity of MET expression and amplification on whole tissue sections.

MET amplification, regardless of interpretation criteria, showed significant association with the adverse clinicopathological features, including higher histological grade, advanced pT category, frequent lymph node metastasis, and advanced AJCC stage. Moreover, MET amplification showed a tendency to shorten OS and RFS, which were not statistically significant.

We investigated the correlation between MET protein expression and gene amplification. However, there was no correlation between MET protein overexpression and amplification, although the incidence of MET overexpression tended to be higher in tumors from patients with MET amplification than that in tumors from patients with no amplification. Some studies have shown lack of correlation between MET overexpression and gene amplification in HCC and NSCLC [34,35]. These studies and our results suggested that other than gene amplification, there are different mechanisms of protein overexpression. It can be inferred that the other mechanisms, including autocrine and paracrine HGF, increased expression of HGF activator, ligand-independent interactions with other receptors, interactions with other active cell-surface receptors and/or epigenetic regulation might play an important role in MET expression without genomic amplification [19,36,37]. Additionally, activation of other oncogenic factors, inactivation of tumor suppressors, such as TP53, and upregulated microRNAs have been known to enhance MET transcription [38].

In conclusion, the present study found that MET amplification was observed in approximately 20% of the GBCs, and that it was associated with adverse clinicopathological characteristics, such as histological grade, T category, lymph node metastasis, and AJCC stage. Further studies are needed to elucidate the significance of MET amplification as a potential therapeutic target and predictive biomarker for antiMET-targeted therapy in GBC patients.