Immunohistochemical expression of the high-mobility group box-1 protein in head and neck squamous cell carcinoma

Aim of the work The aim of the present study was to investigate high-mobility group box-1 (HMGB1) expression and subcellular distribution in head and neck squamous cell carcinoma (HNSCC) tissue. Materials and methods Using immunohistochemistry, HMGB1 protein expression in 72 HNSCC specimens and its association with the clinicopathological characteristics were investigated. Tumour proliferation indices (PIs) were also assessed using Ki-67 immunoreactivity and examined for the presence of a possible correlation with HMGB1 expression. Results Nuclear HMGB1 expression was detected in 84.8% of the tumours, and among them, 39 (54.2%) cases showed high expression scores. A significantly higher expression was found in HNSCC than in the adjacent non-neoplastic epithelium ( P =0.001). In HNSCC cases, nuclear HMGB1 overexpression was significantly associated with lymphatic spread, tumour classification, the clinical stage and a high tumour PI ( P <0.05). Among positive cases, nuclear and cytoplasmic HMGB1 coexpression (coexpression pattern) was observed in 16/61 (26.2%), whereas nuclear-only expression existed in 45/61 (73.8%) of the HNSCC cases, and it existed in all sections of the HMGB1 -positive non-neoplastic adjacent epithelium. Interestingly, the coexpression pattern of HMGB1 was significantly associated with lymphatic spread, tumour classification, the clinical stage and a high PI ( P <0.05). On multivariate analyses, lymph node metastasis and a high tumour PI were independently associated with nuclear HMGB1 overexpression. A high tumour PI was independently associated with the HMGB1 coexpression pattern. Conclusion This study showed that HMGB1 is abundantly expressed in HNSCC, suggesting its significant participation in the carcinogenesis of this type of cancer. HMGB1 overexpression and/or nuclear and cytoplasmic coexpression are closely associated with aggressive tumour features, indicating that HMGB1 expression and distribution patterns might contribute to the progression of HNSCC.