Abstract
OBJECTIVE: Uveal melanoma (UM) is a highly malignant intraocular tumor with a higher incidence in white populations. This study aims to develop a nomogram model to predict the 3-year, 5-year, and 8-year overall survival (OS) of UM patients. The model is based on data from the SEER database and incorporates independent prognostic factors such as gender, age, marital status, AJCC stage, surgery, and radiotherapy, aiming to improve the accuracy of survival predictions and support individualized treatment strategies. METHODS: This study analyzed data from 1,036 white UM patients in the SEER database from 2004 to 2015. Significant prognostic factors were identified using multivariate Cox regression analysis and incorporated into the nomogram model. The model's performance was evaluated using the C-index, net reclassification improvement (NRI), decision curve analysis (DCA), and calibration curves, and was internally validated in both the training and validation cohorts. RESULTS: The nomogram model demonstrated strong predictive power, with C-index values of 0.714 and 0.728 in the training and validation cohorts, respectively, outperforming the traditional AJCC staging system. Calibration curves showed high concordance between the model's predictions and actual survival rates. NRI and DCA analyses indicated that this model offers superior clinical utility in survival prediction. CONCLUSION: This study provides a nomogram specifically for predicting survival in white UM patients, significantly improving the accuracy of 3-year, 5-year, and 8-year survival predictions compared to the traditional AJCC staging system. This model may serve as a valuable clinical tool to guide individualized treatment planning, improve prognosis management, and enhance the quality of clinical decision-making.