Precision prognostic model for uveal melanoma: nomogram prediction and comparison with AJCC system

Area of Science:

• Ophthalmology
• Oncology
• Biostatistics

Background:

• Uveal melanoma (UM) is a rare but aggressive intraocular cancer, predominantly affecting individuals of white ethnicity.
• Accurate prognostication is crucial for effective management and treatment planning in UM patients.

Purpose of the Study:

• To develop and validate a nomogram model for predicting the 3-year, 5-year, and 8-year overall survival (OS) of white UM patients.
• To enhance the accuracy of survival predictions beyond the traditional AJCC staging system.
• To provide a tool for individualized treatment strategies and prognosis management.

Main Methods:

• Analysis of 1,036 white UM cases from the SEER database (2004-2015).
• Identification of independent prognostic factors using multivariate Cox regression.
• Development and internal validation of a nomogram model using C-index, NRI, and DCA.

Main Results:

• The nomogram demonstrated strong predictive performance with C-indices of 0.714 (training) and 0.728 (validation).
• The model showed superior accuracy and clinical utility compared to the AJCC staging system.
• Calibration curves confirmed high concordance between predicted and actual survival rates.

Conclusions:

• The developed nomogram provides a reliable tool for predicting survival in white UM patients.
• This model significantly improves the accuracy of long-term survival predictions.
• It can aid clinicians in personalized treatment planning and enhance clinical decision-making for UM.