Abstract
Melanoma is the most common malignant tumor, with a large patient population. Metastatic melanoma typically occurs in sites such as lymph nodes, skin, lungs, liver, brain, and bones. This metastatic spread often indicates advanced disease, with significantly decreased treatment efficacy and patient survival rates. Accurately predicting the early death (six months) rates of patients with metastatic melanoma to guide the selection of optimal treatment strategies is an urgent clinical issue in need of resolution. Patient demographic and clinical information for this study was extracted from the SEER database. Univariate and multivariate logistic regression analyses were used to screen for independent risk factors. Operating Characteristic curve (ROC) and calibration curve were used to validate the accuracy of the model. Decision curve analysis (DCA) was used to assess the model's benefit to patients. Finally, Kaplan-Meier survival curves were plotted to show differences in patient survival across risk groups. A total of 1109 patients were included in the study and randomly allocated to a training cohort (n = 777) and a validation cohort (n = 332). Logistic regression analysis identified six variables as independent factors influencing short-term survival in patients with metastatic melanoma, which were used to construct a nomogram. The Receiver Operating Characteristic (ROC) curve demonstrated good performance of the nomogram in both the training cohort (AUC = 0.755) and the validation cohort (AUC = 0.694). The calibration curve showed good fit. Decision curve analysis (DCA) indicated that patients can derive significant benefit from using this nomogram. We have successfully developed and well validated a nomogram that accurately predicts early death (6 months) in patients with transformational melanoma. This can assist clinicians in choosing better clinical strategies for their patients.