Screening diagnostic biomarkers of OSCC via an LCM-based proteomic approach.

The current standard for the diagnosis of oral squamous cell carcinoma (OSCC) is based on the histologic examination of hematoxylin and eosin-stained sections; however, the discrimination among normal tissue, pre-cancerous lesions and cancerous lesions can be difficult. The aim of the present study was to identify proteins with diagnostic significance in differentiating or predicting oral mucosal carcinogenesis. Proteomic profiling based on the laser capture microdissection of formalin-fixed, paraffin-embedded samples was performed, followed by liquid chromatography-tandem mass spectrometry (LC/MS) analysis. Immunohistochemistry (IHC) was used to evaluate the results. IHC of cytokeratins (CKs) was performed in neck dissection treatment cases. The accuracy rate and 95% confidence intervals (CIs) were used to evaluate the value of CKs as biomarkers of OSCC. A lymph node metastasis mouse model was used to validate the selected biomarkers. Among the proteins identified using LC/MS, several CKs exhibited significant differential expression patterns between the cancerous and para-cancerous tissues. The IHC results showed that negative staining of CK4 and CK10/13 distinguished cancerous from para-cancerous tissues with an accuracy of 90% (95% CI, 0.68-0.99) and 75% (95% CI, 0.51-0.91), respectively. Furthermore, the positive staining of CK14 and CK17 clearly distinguished cancerous from para-cancerous lesions with an accuracy of 100% (95% CI, 83-100%) and 90% (95% CI, 0.68-0.99), respectively. There was also CK14-positive staining in micro-metastases of lymph nodes in the clinical samples and in an animal model.