Association between miR-200c and the survival of patients with stage I epithelial ovarian cancer: a retrospective study of two independent tumour tissue collections

Lancet Oncol. 2011 Mar;12(3):273-85. doi: 10.1016/S1470-2045(11)70012-2. Epub 2011 Feb 21.

Abstract

Background: International Federation of Gynecology and Obstetrics stage I epithelial ovarian cancer (EOC) has a significantly better prognosis than stage III/IV EOC, with about 80% of patients surviving at 5 years (compared with about 20% of those with stage III/IV EOC). However, 20% of patients with stage I EOC relapse within 5 years. It is therefore crucial that the biological properties of stage I EOCs are further elucidated. MicroRNAs (miRNAs) have shown diagnostic and prognostic potential in stage III and IV EOCs, but the small number of patients diagnosed with stage I EOC has so far prevented an investigation of its molecular features. We profiled miRNA expression in stage I EOC tumours to assess whether there is a miRNA signature associated with overall and progression-free survival (PFS) in stage I EOC.

Methods: We analysed tumour samples from 144 patients (29 of whom relapsed) with stage I EOC gathered from two independent tumour tissue collections (A and B), both with a median follow-up of 9 years. 89 samples from tumour tissue collection A were stratified into a training set (51 samples, 15 of which were from patients who relapsed) for miRNA signature generation, and into a validation set (38 samples, seven of which were from patients who relapsed) for signature validation. Tumour tissue collection B (55 samples, seven of which were from patients who relapsed) was used as an independent test set. The Cox proportional hazards model and the log-rank test were used to assess the correlation of quantitative reverse transcription PCR (qRT-PCR)-validated miRNAs with overall survival and PFS.

Findings: A signature of 34 miRNAs associated with survival was generated by microarray analysis in the training set. In both the training set and validation set, qRT-PCR analysis confirmed that 11 miRNAs (miR-214, miR-199a-3p, miR-199a-5p, miR-145, miR-200b, miR-30a, miR-30a*, miR-30d, miR-200c, miR-20a, and miR-143) were expressed differently in relapsers compared with non-relapsers. Three of these miRNAs (miR-200c, miR-199a-3p, miR-199a-5p) were associated with PFS, overall survival, or both in multivariate analysis. qRT-PCR analysis in the test set confirmed the downregulation of miR-200c in relapsers compared with non-relapsers, but not the upregulation of miR-199a-3p and miR-199a-5p. Multivariate analysis confirmed that downregulation of miR-200c in the test set was associated with overall survival (HR 0·094, 95% CI 0·012-0·766, p=0·0272) and PFS (0·035, 0·004-0·311; p=0·0026), independent of clinical covariates.

Interpretation: miR-200c has potential as a predictor of survival, and is a biomarker of relapse, in stage I EOC.

Funding: Nerina and Mario Mattioli Foundation, Cariplo Foundation (Grant Number 2010-0744), and the Italian Association for Cancer Research.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Aged
  • Aged, 80 and over
  • Biomarkers, Tumor*
  • Carcinoma, Ovarian Epithelial
  • Female
  • Gene Expression Profiling
  • Humans
  • MicroRNAs*
  • Middle Aged
  • Neoplasms, Glandular and Epithelial* / genetics
  • Neoplasms, Glandular and Epithelial* / mortality
  • Neoplasms, Glandular and Epithelial* / pathology
  • Ovarian Neoplasms* / genetics
  • Ovarian Neoplasms* / mortality
  • Ovarian Neoplasms* / pathology
  • Prognosis
  • Retrospective Studies

Substances

  • Biomarkers, Tumor
  • MIRN200 microRNA, human
  • MicroRNAs