Three RCTs (n=1,648) were included.
Primary analyses: survival meta-analysis of individual patient data.
Using the individual survival data derived from the original survival graphs, aromatase inhibitors significantly prolonged survival (relative risk of death 0.79; 95% confidence interval: 0.69, 0.91, P=0.0011). Heterogeneity between trials was statistically significant. The analysis was performed on an intention to treat basis, with 803 patients treated with aromatase inhibitors and 845 patients treated with megestrol. The mean survival gain per patient was 4.1 months.
Primary analyses: meta-analysis of aggregate survival data.
The pooled odds ratio for death at 2 years was 0.73 (95% confidence interval: 0.62, 0.86, P<0.001) in favour of aromatase inhibitors. Inter-trial heterogeneity was statistically significant.
Secondary analysis of survival data.
A retrospective non-randomised comparison between letrozole, anastrozole and exemestane showed better survival among exemestane-treated patients, followed by letrozole-treated patients, then anastrozole-treated patients. The megestrol group had a better survival pattern in the exemestane study, compared with the letrozole and anastrozole trials.