ERβ overexpression may not be a direct prognostic factor in patients with NSCLC: A meta-analysis.

Abstract

Overall survival of non-small cell lung cancer (NSCLC) patients remains disappointingly low. The estrogen receptor (ER) was considered a promising therapeutic target for NSCLC. Numerous studies have linked expression of ERβ to lung cancer outcome. However, results are conflicting regarding the association of ERβ with surviving lung cancer. The aim of this meta-analysis was to evaluate the prognostic aspect of ERβ expression on survival among NSCLC patients. We performed a final analysis of prognostic value of overexpression ERβ on 3500 patients from 18 evaluable studies (from January 1, 2000 to May 1, 2021). The reference category is specified as low ERβ expression levels. Summarized hazard ratios were calculated. Our study showed that the pooled hazard ratios of ERβ overexpression for overall survival in NSCLC was 0.81 (95% confidence interval (CI): 0.64-1.02, P = 0.07) by univariate analysis and 1.06 (95% CI: 0.83-1.36, P = 0.63) by multivariate analysis. Pooled hazard ratio by univariate analysis in Asian studies was 0.73 (95%CI: 0.59-0.89, P = 0.002). Pooled hazard ratio by univariate analysis was 0.75 (95% CI: 0.61-0.93, P = 0.009) from seven studies reported for nuclear ERβ. No significant results were found in subgroups by multivariate analysis. No significant results were found in studies outside Asia or in studies reported for cytoplasmic ERβ. Our results suggested that expression of ERβ might not be a direct prognostic factor for NSCLC patients. More detailed prospective studies are needed to identify direct prognostic factors in these patients.