Mesenchymal-epithelial transition factor amplification correlates with adverse pathological features and poor clinical outcome in colorectal cancer

BACKGROUND Colorectal cancer (CRC) is the third most common cancer and the second most common cause of cancer-related mortality worldwide. Mesenchymal-epithelial transition factor (MET) gene participates in multiple tumor biology and shows clinical potential for pharmacological manipulation in tumor treatment. MET amplification has been reported in CRC, but data are very limited. Investigating pathological values of MET in CRC may provide new therapeutic and genetic screening options in future clinical practice. AIM To determine the pathological significance of MET amplification in CRC and to propose a feasible screening strategy. METHODS A number of 205 newly diagnosed CRC patients undergoing surgical resection without any preoperative therapy at Shenzhen Cancer Hospital of Chinese Academy of Medical Sciences were recruited. All patients were without RAS/RAF mutation or microsatellite instability-high. MET amplification and c-MET protein expression were analyzed using fluorescence in situ hybridization (FISH) and immunohistochemistry (IHC), respectively. Correlations between MET aberration and pathological features were detected using the chi-squared test. Progression free survival (PFS) during the two-year follow-up was detected using the Kaplan-Meier method and log rank test. The results of MET FISH and IHC were compared using one-way ANOVA. RESULTS Polysomy-induced MET amplification was observed in 14.4% of cases, and focal MET amplification was not detected. Polysomy-induced MET amplification was associated with a higher frequency of lymph node metastasis (LNM) (P < 0.001) and higher tumor budding grade (P = 0.02). In the survival analysis, significant difference was detected between patients with amplified- and non-amplified MET in a two-year follow-up after the first diagnosis (P = 0.001). C-MET scores of 0, 1+, 2+, and 3+ were observed in 1.4%, 24.9%, 54.7%, and 19.0% of tumors, respectively. C-MET overexpression correlated with higher frequency of LNM (P = 0.002), but no significant difference of PFS was detected between patients with different protein levels. In terms of concordance between MET FISH and IHC results, MET copy number showed no difference in c-MET IHC 0/1+ (3.35 +/- 0.18), 2+ (3.29 +/- 0.11) and 3+ (3.58 +/- 0.22) cohorts, and the MET-to-CEP7 ratio showed no difference in three groups (1.09 +/- 0.02, 1.10 +/- 0.01, and 1.09 +/- 0.03). CONCLUSION In CRC, focal MET amplification was a rare event. Polysomy-induced MET amplification correlated with adverse pathological characteristics and poor prognosis. IHC was a poor screening tool for MET amplification.