Prognostic Nomograms for Predicting Overall Survival and Cancer-Specific Survival in Patients with Head and Neck Mucosal Melanoma

Abstract

Background: Accurate forecasting of the risk of death is crucial for people living with head and neck mucosal melanoma (HNMM). We aimed to establish and validate an effective prognostic nomogram for HNMM. Methods: Patients with HNMM who underwent surgery between 2010 and 2015 were selected from the Surveillance, Epidemiology, and End Results (SEER) database for model construction. After eliminating invalid and missing clinical information, 288 patients were ultimately identified and randomly divided into a training cohort (199 cases) and a validation cohort (54 cases). Univariate and multivariate Cox proportional hazards regression analyses were performed in the training cohort to identify prognostic variables. Independent influencing factors were used to build the model. Through internal verification (training cohort) and external verification (validation cohort), the concordance indexes (C-indexes) and calibration curves were used to evaluate the predictive value of the nomogram. Results: For the training cohort, five independent risk predictors, namely age, location, T stage, N stage, and surgery, were selected, and nomograms with estimated 1-and 3-year overall survival (OS) and cancer-specific survival (CSS) were established. The C-index showed that the predictive performance of the nomogram was better than that of the TNM staging system and was internally verified (through the training queue: OS: 0.764 vs 0.683, CSS: 0.783 vs 0.705) and externally verified (through the verification queue: OS: 0.808 vs 0.644, CSS: 0.823 vs 0.648). The calibration curves also showed good agreement between the prediction based on the nomogram and the observed survival rate. Conclusion: The nomogram prediction model can more accurately predict the prognosis of HNMM patients than the traditional TNM staging system and may be beneficial for guiding clinical treatment.