The prognostic value of estrogen receptors (ESR1 and ESR2) mRNA expression in patients with non-small cell lung cancer (NSCLC) remains controversial. Therefore, a systematic review with meta-analysis was conducted. A systematic literature search was conducted in both Pubmed and Embase. Studies that reported association of ESR and survival in NSCLC patients in the form of hazard ratio (HR) and 95% confidence interval (CI) were included. Pooled HR was taken as the effect size to reflect the association. Five eligible articles provided six separate studies for ESR1 and four for ESR2. For ESR1, the pooled HR of overall survival was 0.72 (95% CI 0.41–1.27) by univariate analysis and was 0.33 (95% CI: 0.20–0.53) by multivariate analysis. For ESR2, the pooled HR was 0.95 (95% CI 0.73–1.23) by univariate analysis. Sub-group analysis suggested that the disease stages and cut-off point may explain heterogeneity among studies of ESR1. Results of the meta-analysis revealed a potential benefit of positive ESR1 mRNA expression in survival in patients with NSCLC, especially in those of advanced stage. No statistically significant association was found between ESR2 mRNA expression and NSCLC patients' prognosis.