A Prognostic Model for Cancer-Specific Survival Among Patients with Nodular Melanoma in Head and Neck.

Abstract

This study aims to evaluate the correlation between risk factors and treatment methods affecting nodular melanoma (NM) in the head and neck, as well as cancer-specific survival (CSS), and provide personalized predictive tools for clinical physicians. The retrospective study data of 1848 patients were extracted from the Surveillance, Epidemiology, and End Results (SEER) database. All variables were included in the correlation analysis using the Spearman method. Significant prognostic factors were extracted and integrated through Cox risk regression analysis to construct a nomogram. To assess the performance of the nomogram, Harrell's concordance index (C-index) and a receiver operating characteristic (ROC) curve analysis were employed. Spearman's correlation analysis revealed a positive correlation between radiotherapy and lymph node metastasis, whereas chemotherapy showed a stronger association with distant metastasis. However, Cox risk regression analysis demonstrated that Mohs surgery and wide excision with margins exceeding 1 cm yielded substantial therapeutic advantages. Five independent risk prognostic factors (Breslow thickness, ulceration, N classification, M classification, and surgery type) were employed to construct a nomogram. The C-index for this nomogram was 0.713 for the training set and 0.720 for the validation set. In the training set, the 3-, 5-, and 8-year areas under the curve (AUCs) for CSS were 0.752, 0.723, and 0.720, whereas the validation set's AUCs were 0.754, 0.763, and 0.760, respectively. Calibration curves indicated the nomogram's strong discriminative ability for predicting CSS. In this study, we identified independent prognostic factors for patients with NM in head and neck and developed a relatively accurate model to predict the survival probability of them, which could contribute to the tumor assessment and clinical decision-making. Level 3 Laryngoscope, 2024.