Prediction of lymph node metastasis in oral squamous cell carcinoma based on protein profile

Objective: Lymph node metastasis leads to high mortality rates of oral squamous cell carcinoma (OSCC). However, it is still controversial to define clinically negative neck (cN0) and positive neck (cN1-3). Methods: We retrieved candidate biomarkers identified by proteomic analysis in OSCC from published works of literature. In training stage, immunohistochemistry (IHC) analysis was used to determine the expression of proteins and logistic regression models with stepwise variable selection were used to identify potential factors that might affect lymph node metastasis and life status. Furthermore, the prediction model was validated in validating stage. Results: We screened eight highly expressed proteins related to lymph node metastasis in OSCC and found that the expression levels of SOD2, BST2, CAD, ITGB6, and PRDX4 were significantly elevated in patients with lymph node metastasis compared to the patients without lymph node metastasis. Furthermore, in training and validating stages, the prediction model base on the combination of CAD, SOD2 expression levels, and histopathologic grade was developed and validated in patients with OSCC. Conclusions: Our findings showed that the developed model well predicts the lymph node metastasis and life status in patients with OSCC, independent of TNM stage.

研究目的：淋巴结转移会导致口腔鳞状细胞癌（oral squamous cell carcinoma, OSCC）患者死亡率升高，但目前对于临床阴性颈（cN0）与临床阳性颈（cN1-3）的定义仍存在争议。
研究方法：我们从已发表文献中检索得到经蛋白质组学分析筛选出的口腔鳞状细胞癌候选生物标志物。在训练阶段，采用免疫组织化学（immunohistochemistry, IHC）检测蛋白表达水平，并通过逐步变量选择的逻辑回归模型，筛选可能影响淋巴结转移与生存状态的潜在因素。此外，本研究在验证阶段对构建的预测模型进行了验证。
研究结果：我们筛选出8种与口腔鳞状细胞癌淋巴结转移相关的高表达蛋白，且与无淋巴结转移患者相比，淋巴结转移患者的SOD2、BST2、CAD、ITGB6及PRDX4表达水平显著升高。此外，在训练与验证阶段，我们基于CAD、SOD2表达水平及组织学分级构建了预测模型，并在口腔鳞状细胞癌患者群体中完成了模型验证。
研究结论：本研究所构建的预测模型可有效预测口腔鳞状细胞癌患者的淋巴结转移情况与生存状态，且该预测效果不受TNM分期的影响。