Development and validation of a nomogram for elderly patients with ulcerative melanoma

Melanoma Res. 2024 Jun 1;34(3):207-214. doi: 10.1097/CMR.0000000000000940. Epub 2023 Dec 13.

Abstract

The current state of survival prediction models for elderly patients with ulcerative melanoma (uCM) is limited. We sought to develop a nomogram model that can predict overall survival of geriatric patients with uCM. The Surveillance, Epidemiology, and End Results (SEER) database served as a source for patients diagnosed with uCM between 2004 and 2015. Statistical analyses were conducted to determine the significant prognostic elements affecting overall survival using multivariate and univariate Cox proportional risk regression models. Subsequently, an independent forecasting nomogram was developed on the basis of these identified predictors. The predictive model was then assessed and validated through the utilization of receiver operating characteristic curves, calibration curves as well as decision curves. The study included a total of 5019 participants. Univariate and multivariate analyses revealed age, sex, marital status, primary site, tumor size, N stage, M stage, histological type, and surgery were independent prognostic factors. A nomogram was developed using the findings from both univariate and multivariate Cox analyses ( P < 0.05). The receiver operating characteristic curves, which vary over time, and the area under the curve (AUC) for the training and validation cohorts, demonstrated the nomogram's strong discriminatory ability. Additionally, the calibration curves indicated satisfactory agreement between the predicted values from the nomogram and the practical outcomes observed in both cohorts. Furthermore, the decision curve analysis curves displayed favorable positive net gains at all times, when the critical value is most likely to occur. In this study, age, sex, marital status, primary site, tumor size, N stage, M stage, histologic type and surgery were determined as independent predictors for elderly patients with uCM. Then, a predictive model with good discriminatory ability was constructed to predict 12-, 24-, and 36-month overall survival in geriatric patients with uCM, which facilitates patients' counseling and individualized medical decision.

Publication types

  • Validation Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Aged, 80 and over
  • Female
  • Humans
  • Male
  • Melanoma* / mortality
  • Melanoma* / pathology
  • Nomograms*
  • Prognosis
  • SEER Program
  • Skin Neoplasms* / mortality
  • Skin Neoplasms* / pathology