Multimodal Prediction of Cervical Lymph Node Metastasis and Recurrence in Oral Squamous Cell Carcinoma.

Abstract

Oral squamous cell carcinoma (OSCC) is the most common malignancy in the head/neck region, and cervical lymph node (CLN) metastasis is a strong poor-prognosis factor. In addition, many patients with OSCC experience recurrence despite multidisciplinary treatment. We sought to identify factors associated with CLN metastasis and recurrence in patients with OSCC. We evaluated a total of 45 patients and 233 target CLNs. The longest diameter of the target CLN, the shortest diameter of the target CLN (LS), the area of the target CLN, and the relative computed tomography (CT) values of the target CLNs calculated based on the CT values of the internal jugular vein (LCT) were obtained from preoperative CT images, and the maximum standardized uptake values of the primary tumor (pSUV) and target CLN (nSUV) were obtained from preoperative 18F-fluorodeoxyglucose-positron emission tomography/CT images. We performed immunohistochemical staining for cytokeratin 13 (CK13) and 17 (CK17) on neck dissection tissues. A discrimination equation was used that can predict CLN metastasis with a 92.2% discrimination rate using LS, LCT, pSUV, and nSUV. The CLNs were divided into discrimination and non-discrimination groups based on discriminant equations and CK13 and CK17 were used as the objective variables. A significantly higher recurrence rate was observed in the non-discrimination group (CK13: 5-year recurrence rate 28.6% vs. 64.3%, p<0.01; CK17: 5-year recurrence rate 28.0% vs. 76.0%, p<0.01). CLN metastases in OSCC can be assessed by combining preoperative imaging. The combined use of CK13 and CK17 expression with imaging findings offers an integrated approach to predict OSCC recurrence.