The prognostic nomograms in patients with nodular melanoma: a SEER population-based study

Abstract Nodular melanoma (NM) is one of the most common subtype of malignant melanoma, usually with bad prognosis. In this study, we aim to develop a nomogram to predict 3-year and 5-year overall survival (OS) and cancer-specific survival (CSS) in patients with nodular melanoma. Patients diagnosed as NM were extracted from the Surveillance, Epidemiology, and End Results (SEER) database between 2010–2015, and then they were randomly assigned to the training and validation cohorts. Several independent risk factors were found and merged to generate a nomogram using univariate and multivariate COX risk regression methods. To determine the nomogram's predictive ability, the receiver operating characteristic (ROC) curve, concordance index (C-index), area under curve (AUC) and calibration curves were calculated. In clinical practice, decision curve analysis (DCA) was employed to measure nomograms. A total of 8050 eligible patients were randomly assigned to one of two groups: 70% training (n = 5635) or 30% validation (n = 2415). The Nomogram's construct was applied with the selected independent factors, and the validation indexes revealed that the nomogram had strong discriminative power. In the training cohort of OS, the C-index was 0.746, and in the validation cohort of OS, it was 0.745. In the training cohort of CSS, the C-index was 0.777, whereas in the validation cohort, it was 0.775. The AUCs for the 3- and 5-year OS rates, as well as the CSS rates, were above 0.78, and the calibration curves exhibited strong nomogram predictive power. The created nomograms could provide net clinical benefit, according to DCA. Our study has developed a nomogram to more accurately and comprehensively predict 3-year and 5-year OS and CSS in patients with nodular melanoma, which is critical for clinicians to make early assessments, improve individualized treatment, identify relevant prognostic factors, and improve survival of the affected population.