Snail as a novel marker for regional metastasis in head and neck squamous cell carcinoma

ObjectivePrevious studies have shown Snail expression integral to the epithelial-mesenchymal transition during tumor progression. However, its behavior in clinical head and neck squamous cell carcinomas (HNSCCs) is yet undefined. We therefore sought to (1) investigate clinical and histopathologic characteristics of Snail-positive HNSCC and (2) understand the link between Snail and other commonly used HNSCC tumor markers.Study DesignA retrospective case-control study was conducted.SettingThis study was conducted in a large-scale academic center.Study SubjectsOf 51 consecutive HNSCC, 42 surgical resections were included.MethodsTwo separate pathologists performed standard histopathologic reviews along with immunohistochemistries (Snail, E-cadherin, p16, epidermal growth factor receptor [EGFR]) and in situ hybridization (human papilloma virus [HPV]). Medical review for all cases was performed.ResultsTwenty-two (52%) of 42 cases stained 4+ Snail (>75% staining). The remaining 20 cases were considered negative. Snail was strongly inversely related to E-cadherin expression (ρ = −0.69, P < .001), but statistically independent from HPV, p16, or EGFR expression. Snail(+) tumors were equally represented from each anatomic subsite. Snail(+) tumors were strongly associated with poor differentiation (P < .001) and basaloid classification (P = .004). Snail(+) tumors were also strongly associated with lymphovascular invasion (P = .02), but not perineural invasion. Ultimately, 11 (50%) of 22 of Snail(+) tumors demonstrated positive nodal metastasis and 11 (79%) of 14 node-positive cases were Snail(+) (P = .02).ConclusionThis pilot study provides promising evidence of Snail' role as a molecular prognostic marker for HNSCC. Snail positivity is significantly predictive of poorly differentiated, lymphovascular invasive, as well as regionally metastatic tumors. Because Snail positivity appears independent of HPV, p16, and EGFR expression, Snail may prove to improve upon these markers' predictive limitations.