Validation of a nomogram to predict the presence of sentinel lymph node metastases in melanoma

Ann Surg Oncol. 2008 Oct;15(10):2874-7. doi: 10.1245/s10434-008-0077-x. Epub 2008 Jul 22.

Abstract

Background: Lymph node involvement is a very important prognostic factor for cutaneous melanoma. In this paper we try to validate a nomogram that was created at the Memorial Sloan-Kettering Cancer Center, New York, to predict the probability of metastases in the sentinel nodes of patients with cutaneous melanoma.

Methods: Values of the following variables were collected in 218 patients with cutaneous melanoma and sentinel lymph node: age, thickness, level of Clark, location of the lesion, and ulceration or not, and the nomogram was applied to assess the probability of sentinel node involvement in each patient. The discrimination of the nomogram was assessed by calculating the area under the receiver operating characteristics (ROC) curve, and to assess the accuracy of the nomogram actual probabilities were plotted against the nomogram-calculated predicted probability.

Results: The overall predictive accuracy of the nomogram was 0.869 (95% confidence interval 0.813-0.925). Mean predicted probability of sentinel node metastasis was highly correlated to the observed risk (r = 0.953; P < 0.012).

Conclusion: The nomogram is a useful diagnostic tool that provides an adequate accurate prediction of the probability of sentinel lymph node metastases in patients with cutaneous melanoma.

Publication types

  • Validation Study

MeSH terms

  • Female
  • Humans
  • Lymph Nodes / pathology*
  • Lymph Nodes / surgery
  • Lymphatic Metastasis
  • Male
  • Melanoma / secondary*
  • Melanoma / surgery
  • Middle Aged
  • Nomograms*
  • Predictive Value of Tests
  • Prognosis
  • Prospective Studies
  • ROC Curve
  • Sentinel Lymph Node Biopsy
  • Skin Neoplasms / pathology*
  • Skin Neoplasms / surgery