Estimation of Prognosis in Invasive Cutaneous Melanoma: An Independent Study of the Accuracy of a Gene Expression Profile Test

Dermatol Surg. 2018 Dec;44(12):1494-1500. doi: 10.1097/DSS.0000000000001588.

Abstract

Background: Cutaneous melanomas (CMs) with similar clinical and histopathologic features can harbor differing capacities for metastasis. A validated gene expression profile (GEP) test offers prognostic information by classifying CMs as low risk (Class 1A/1B) or high risk (Class 2A/2B) for metastasis.

Objective: The authors sought to perform an independent study of the predictive accuracy of the GEP test, to determine what clinical and histopathologic features predict high-risk classification, and to evaluate how intermediate classes (1B & 2A) performed clinically.

Materials and methods: Using our institution's prospectively collected melanoma registry, the authors identified patients who had been treated for CM within the last 5 years and undergone GEP testing. Clinical, histopathologic, and outcomes data were analyzed. A subcohort of patients with known metastatic disease were identified and tested.

Results: The GEP test accurately identified 77% of metastatic CMs as high risk (Class 2). The GEP had a negative predictive value of 99% for Class 1 CMs. Class 2 CMs were 22 times more likely to metastasize.

Conclusion: The GEP test's performance in our independent cohort corresponded with previous industry-sponsored studies and proved to be a helpful clinical prognostic tool with the potential to direct patient care protocols.

MeSH terms

  • Aged
  • Aged, 80 and over
  • Female
  • Gene Expression Profiling*
  • Humans
  • Male
  • Melanoma / classification*
  • Melanoma / complications
  • Melanoma / genetics*
  • Melanoma / secondary
  • Middle Aged
  • Neoplasm Invasiveness
  • Predictive Value of Tests
  • Prognosis
  • Registries
  • Risk Assessment / methods
  • Skin Neoplasms / classification*
  • Skin Neoplasms / complications
  • Skin Neoplasms / genetics*
  • Skin Neoplasms / pathology
  • Skin Ulcer / etiology
  • Transcriptome