Alteration status and prognostic value of MET in head and neck squamous cell carcinoma

Abstract

The MET pathway plays a key role in various cancers, and its inhibition represents a potential treatment target. However, appropriate biomarkers are needed to facilitate the selection of patients who would benefit from MET inhibiting therapy. We herein conducted a robust confirmatory evaluation of the MET copy number alteration status and prognostic significance of c-Met expression in a large series of patients (n = 396) who underwent standard surgical resection and adjuvant chemoradiotherapy for head and neck squamous cell carcinoma (HNSCC). Surgically resected HNSCC samples were subjected to immunohistochemical and H-score analysis of c-Met expression and silver in situ hybridization analysis of MET amplification and copy number gains. c-Met expression varied, with mean and median H-scores (scale: 0-300 scale) of 61.2 and 60.0, respectively. The lowest and highest expression levels were observed in SCC of the larynx and oral cavity, respectively. MET copy number gains were observed in 16.9% of cases (67/339) and were associated with c-Met protein expression. High c-Met expression, determined according to MET gain status, was associated with an inferior overall survival rate, especially among completely resected cases. In conclusion, our robust analysis revealed that c-Met expression in HNSCCs varied according to anatomical site, correlated with MET copy number gains, and was associated with poor prognosis. This c-Met expression analysis method, which is based on the MET gain status, appears to appropriately predict high-risk HNSCC patients in the context of anti-MET therapeutic decisions.

Keywords: Head and neck squamous cell carcinoma; MET gene; Prognosis; c-Met protein.

Figure 1

Immunohistochemical analysis of c-MET protein expression and silver in-situ hybridization (SISH) to determine MET copy number alterations. Representative images of immunohistochemical staining for c-Met demonstrate negative staining (intensity score 0, A), weak or barely detectable membranous staining (intensity score 1, B), distinct brown membranous staining (intensity score 2, C), and strong dark brown membranous staining (intensity score 3, D). Each intensity score was multiplied by the percentage of positive nuclear cell staining (0-100%) to calculate the H-score (possible scores: 0-300). In a SISH dataset, the black signals indicate MET copies and red signals indicate CEP7. The representative case harbored 3-4 MET signals per nucleus in ≥20% of counted tumor cells, indicating a MET copy number gain. Arrows point to cells that show 4 MET signals in the nucleus (E). Another representative case exhibits normal disomy of the MET and CEP7 signals (F). Selective magnification of cells is shown inside the black box (E and F).

Figure 2

Relationships of c-Met expression with affected anatomic site and MET copy number. The c-MET expression levels were lowest and highest in laryngeal squamous cell carcinoma (SCC) and SCC of the oral cavity, respectively (A). Intermediate c-Met levels were observed in oropharyngeal and hypopharyngeal SCC (P = 0.003). The c-MET protein expression level was higher in cases exhibiting MET copy number gains than in those without MET gains (B, P = 0.002). A receiver operating characteristic curve analysis based on the MET copy number gain was conducted to determine the c-Met protein expression cut-off value (C).

Figure 3

Kaplan-Meier analysis of overall survival according to c-Met expression status. In the overall cohort of head and neck squamous cell carcinoma (HNSCC) cases, a high c-Met protein expression level tended to correlate with inferior overall survival in a marginally significant relationship (A). Among HNSCC cases that achieved complete (R0) resection, a high c-MET protein expression level correlated significantly with inferior overall survival (B).