Objective: The aim of this study was to evaluate the prognostic significance of cyclooxygenase-2 (COX-2) on survival in patients with ovarian cancer by using a meta-analysis of observational studies.
Methods: We searched Pubmed and Embase to retrieve observational studies evaluating the association between COX-2 status and survival in patients with ovarian cancer. Hazards ratios (HRs) or odds ratios (ORs) with 95% confidence intervals (CIs) were pooled across studies using a random-effects model.
Results: A total of 17 studies were included in this meta-analysis to estimate the association between COX-2 and overall survival (OS), disease-free survival (DFS), response to chemotherapy (RC), and other clinical parameters. In a random-effects meta-analysis of 15 studies, higher COX-2 expression significantly predicted poor OS (death HR, 1.34; 95% CI, 1.05-1.71; I(2)=56.5%). A more prominent association was found between COX-2 expression and poor OS when studies with adjustment for age, stage, and histology were included (death HR, 1.65; 95% CI, 1.25-2.17; I(2)=0%). However, higher COX-2 expression was not significantly associated with poor DFS (recurrence HR, 1.36; 95% CI, 0.79-2.33; I(2)=53.6%) and RC (OR, 1.89; 95% CI, 0.85-4.21; I(2)=17.6%). There was a marginally significant association between COX-2 positivity and several clinical parameters such as age, stage, and histology. The pooled ORs of higher COX-2 expression were 1.75 (95% CI, 1.01-3.04) for advanced stages, 1.34 (95% CI, 0.97-1.85) for old age, and 1.42 (95% CI, 0.98-2.05) for serous cancer in histologic type, respectively.
Conclusions: The present meta-analysis suggests that higher COX-2 expression may be an independent risk factor for poor OS in patients with ovarian cancer.
Copyright © 2013 Elsevier Inc. All rights reserved.