Development and validation of prognostic nomogram in pediatric melanoma: a population-based study

Area of Science:

• Oncology
• Epidemiology
• Biostatistics

Background:

• Pediatric melanoma is rare but requires accurate prognostic tools.
• Existing prognostic models may not adequately address the unique characteristics of childhood melanoma.
• The Surveillance, Epidemiology, and End Results (SEER) database offers a valuable resource for population-based cancer research.

Purpose of the Study:

• To develop and validate a clinical nomogram for predicting the survival probability of pediatric melanoma patients.
• To assess the nomogram's accuracy, calibration, and clinical utility in decision-making.
• To leverage the SEER database for robust prognostic model development in pediatric oncology.

Main Methods:

• Utilized the SEER database (2000-2018) to identify pediatric melanoma cases.
• Randomly allocated 890 patients into training (70%) and validation (30%) cohorts.
• Employed Kaplan-Meier analyses, nomogram construction, discrimination and calibration assessments, and decision curve analysis.

Main Results:

• The developed nomogram demonstrated strong discriminative ability with C-index values of 0.817 (training) and 0.832 (validation).
• Calibration plots confirmed a high concordance between predicted and observed survival probabilities.
• Clinical decision curve analysis indicated a favorable net benefit for clinical application in pediatric melanoma.

Conclusions:

• An effective and accurate nomogram for predicting survival in pediatric melanoma patients was successfully developed.
• The nomogram serves as a valuable tool to support clinical decision-making for childhood melanoma.
• Further external validation is recommended to confirm the nomogram's generalizability.