Has Cox-2 a prognostic role in non-small-cell lung cancer? A systematic review of the literature with meta-analysis of the survival results

Abstract

Cyclooxygenase-2 (COX-2) is overexpressed in lung cancer, especially in adenocarcinoma (ADC). Our aim was to determine the prognostic value of COX-2 on survival in patients with lung cancer. Studies evaluating the survival impact of COX-2 in lung cancer, published until December 2005, were selected. Data for estimation of individual hazard ratios (HR) for survival were extracted from the publications and combined in a pooled HR. Among 14 eligible papers, all dealing with non-small-cell lung cancer, 10 provided results for meta-analysis of survival data (evaluable studies). Cyclooxygenase-2 positivity was associated with reduced survival, improved survival or no statistically significant impact in six, one and seven studies, respectively. Combined HR for the 10 evaluable studies (1236 patients) was 1.39 (95% confidence intervals (CI): 0.97-1.99). In stage I lung cancer (six evaluable studies, 554 patients), it was 1.64 (95% CI: 1.21-2.24). No significant impact was shown in ADC. A slight detrimental effect on survival in patients with lung cancer is associated with COX-2 expression, but the statistical significance is not reached. This effect is statistically significant in stage I, suggesting that COX-2 expression could be useful at early stages to distinguish those with a worse prognosis.

Figure 1

Meta-analysis of the 10 evaluable studies assessing COX-2 in NSCLC. Hazard ratio and 95% CI of survival in studies evaluating COX-2 status in NSCLC. HR>1 implies a survival disadvantage for the group with COX-2 expression. The square size is proportional to the number of patients included in the study. The centre of the lozenge gives the combined HR of the meta-analysis and its extremities give the 95% CI. HR=1.39; CI 95% 0.97–1.99. Total number of patients: 1236.

Figure 3

Overall and subgroup analyses. Hazard ratio and 95% CI of survival in studies evaluating COX-2 status in NSCLC. HR>1 implies a survival disadvantage for the group with COX-2 expression. The square size is proportional to the number of patients included in the study and its extremities gives the 95% CI. The Figure 3 shows that there is a trend for a pejorative role of COX-2 as a prognostic survival in NSCLC and that the results become significant (CI not crossing 1) for the subgroups of stage 1 and of RT–PCR.

Figure 2

Meta-analysis of the six evaluable studies assessing COX-2 in stage I NSCLC. Hazard ratio and 95% CI of survival in studies evaluating COX-2 status in NSCLC. HR>1 implies a survival disadvantage for the group with COX-2 expression. The square size is proportional to the number of patients included in the study. The centre of the lozenge gives the combined HR of the meta-analysis and its extremities give the 95% CI. Hazard ratio=1.64; CI 95% 1.21–2.24. Total number of patients: 554.