Identification of a gene signature for discriminating metastatic from primary melanoma using a molecular interaction network approach

Sci Rep. 2017 Dec 11;7(1):17314. doi: 10.1038/s41598-017-17330-0.

Abstract

Understanding the biological factors that are characteristic of metastasis in melanoma remains a key approach to improving treatment. In this study, we seek to identify a gene signature of metastatic melanoma. We configured a new network-based computational pipeline, combined with a machine learning method, to mine publicly available transcriptomic data from melanoma patient samples. Our method is unbiased and scans a genome-wide protein-protein interaction network using a novel formulation for network scoring. Using this, we identify the most influential, differentially expressed nodes in metastatic as compared to primary melanoma. We evaluated the shortlisted genes by a machine learning method to rank them by their discriminatory capacities. From this, we identified a panel of 6 genes, ALDH1A1, HSP90AB1, KIT, KRT16, SPRR3 and TMEM45B whose expression values discriminated metastatic from primary melanoma (87% classification accuracy). In an independent transcriptomic data set derived from 703 primary melanomas, we showed that all six genes were significant in predicting melanoma specific survival (MSS) in a univariate analysis, which was also consistent with AJCC staging. Further, 3 of these genes, HSP90AB1, SPRR3 and KRT16 remained significant predictors of MSS in a joint analysis (HR = 2.3, P = 0.03) although, HSP90AB1 (HR = 1.9, P = 2 × 10-4) alone remained predictive after adjusting for clinical predictors.

Publication types

  • Research Support, N.I.H., Extramural
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Biomarkers, Tumor / genetics
  • Biomarkers, Tumor / metabolism*
  • Gene Expression Profiling*
  • Gene Expression Regulation, Neoplastic*
  • Humans
  • Melanoma / genetics
  • Melanoma / metabolism
  • Melanoma / pathology*
  • Melanoma / secondary
  • Prognosis
  • Protein Interaction Maps
  • Retrospective Studies
  • Skin Neoplasms / genetics
  • Skin Neoplasms / metabolism
  • Skin Neoplasms / secondary*
  • Survival Rate
  • Transcriptome

Substances

  • Biomarkers, Tumor