Prediction model for lymph node metastasis and recommendations for elective neck dissection in lip cancer.

INTRODUCTION

In patients with squamous cell carcinoma (SCC) of the lip, occurrence of lymph node metastasis (LNM) is more frequent than in other cutaneous head and neck SCCs. The aim of this study was to identify predictive factors for LNM in SCC of the lip and to establish a prediction model identifying patients at high LNM risk.

MATERIALS AND METHODS

Tumor characteristics of 326 patients with lip SCC were analyzed retrospectively to assess differences between the LNM group and controls. Using binary logistic and Cox regression analysis, a prediction model for LNM was calculated.

RESULTS

Lymph node metastasis occurred in 26 (8%) patients. Regression analysis revealed tumor extent, tumor depth and grading as the most important factors in the correct classification of LNM in 94.2% of patients. A prediction model taking tumor depth and grading into account allowed for stratification of patients into high and low risk groups (sensitivity 92.3%, specificity 78.3%, negative predictive value 99.2%).

CONCLUSIONS

Our new prediction model was able to identify patients with lip cancer who had a high risk of LNM with a good level of accuracy. This algorithm is easy to apply as part of the decision process for elective and selective lymph node dissection in SCC of the lip.