Prediction Model of Distant Metastasis in Oral Cavity Squamous Cell Carcinoma With or Without Regional Lymphatic Metastasis

Abstract

Patients with oral cavity squamous cell carcinoma (OCSCC) who develop distant metastasis (DM) face poor outcomes, and effective prediction models of DM are rare. A total of 595 patients with OCSCC were retrospectively enrolled in this study. Because pathological N staging significantly influences the development and mechanisms of DM, the patients were divided into nodal-negative (pN−) and -positive (pN+) groups. Clinical outcomes, prognoses, and prediction models were analyzed separately for both groups. Overall, 8.9% (53/595) of these patients developed DM. Among the DM cases, 84.9% (45/53) of them developed DM within the first 3 years. The median overall survival, locoregional recurrence-free survival, time until DM development, and postmetastatic survival were 19.8, 12.7, 14.6, and 4.1 months, respectively. Distinguishing patients who only developed locoregional recurrence from those with DM according to locoregional conditions was difficult. Age, surgical margin, and early locoregional recurrence were predictors of DM that were independent of time until DM in the pN− group; the lymphocyte-to-monocyte ratio, presence of lymphovascular invasion, and early locoregional recurrence in the pN+ group were determined. If one point was scored for each factor, then two scoring systems were used to classify the patients into low- (score = 0), intermittent- (score = 1), or high- (score = 2 or 3) risk for the pN− and pN+ groups. According to this scoring system, the 3-year DM rates for the low, intermittent, and high risk subgroups were 0.0%, 5.9%, and 17.8% for the pN− group and 7.1%, 44.9%, and 82.5% for the pN+ group, respectively. These systems also effectively predicted DM, and the areas under the curve predicted DM occurring within the first 3 years were 0.744 and 0.820 for the pN− and pN+ groups, respectively. In conclusion, effective scoring models were established for predicting DM.