A Prediction Model for Lymph Node Metastasis of Oral Squamous Cell Carcinoma Based on Multiple Risk Factors

Abstract

Objectives: Cervical lymph node metastasis (CLNM) frequently occurs in oral cancer patients. This study aims to investigate risk factors associated with CLNM and predict CLNM preoperatively in patients with oral squamous cell carcinoma (OSCC).

Materials and methods: This population-based, hospital retrospective cohort study included 158 patients with oral cancer. We performed regression analysis to determine risk factors and establish a model for predicting CLNM in patients with OSCC. To distinguish and validate the prediction model, we used the area under the receiver operating characteristic (ROC) curve (AUC).

Results: Lymph node size, tumor size, degree of differentiation, and LVI were risk factors for cancer metastasis. The OR values were 1.245, 2.847, 2.527, and 6.945, respectively. The AUC value for the clinical prediction model was 0.8736 (95% CI: 0.8043-0.9429).

Conclusions: The prediction model for OSCC patients predicts CLNM and provides a new method for preoperative assessment of whether cervical lymph nodes are metastatic, as well as a guide for surgical treatment, including whether to carry out neck dissection and which neck dissection procedure to use.

Keywords: cervical lymph node metastasis; oral squamous cell carcinoma; prediction model; risk factors.

Figure 1

The nomogram prediction model for lymph node metastasis in oral cancer patients.

Figure 2

The ROC curve of the clinical prediction model.

Figure 3

The ROC curve of the clinical prediction model, with tumor size and lymph node size. The red line represents the clinical prediction model, the blue line represents the tumor size model, and the yellow line represents the lymph node size model.

Figure 4

The ROC curve of the validation group.