Occupational-Related Exposure to Benzene and Risk of Breast Cancer: Systematic Review and Meta-analysis

Abstract

Abstract Purpose Benzene is a recognized carcinogen; however, its association with breast cancer is not well established. Hence, a meta-analysis of cohort and case-control studies was performed to determine the association between occupational benzene exposure and the risk of breast cancer. Methods A systematic literature review identified 7221 publications from which 21 cohort and case-control studies were retained and evaluated using meta-analyses (fixed effects model). PRISMA guidelines were followed. Our protocol was registered in the PROSPERO database (Registration No. 379720). Study quality was assessed using a modified Newcastle-Ottawa scale (NOS). Results The summary RR for ever-benzene exposure was 1.08 (95% CI = 1.02–1.14, I2 = 43.6%, n = 21 risk estimates); corresponding RR for cancer incidence and mortality were 1.08 (95% CI = 1.02–1.14, I2 = 58.6%, n = 15) and 1.09 (95% CI = 0.87–1.38, I2 < 0.001%, n = 9), respectively. These main results were confirmed in sub-analyses by geography, industry, publication year, and levels of exposure. No heterogeneity was detected amongst geographical regions (p-het = 0.19) or industry of employment (p-het = 0.05). However, heterogeneity was detected for sex (p-het = 0.002), publication year (p-het < 0.001), study design (p-het < 0.001), and study quality (p-het < 0.001). Publication bias was detected (p = 0.04). Conclusions Our meta-analysis identified an association between occupational benzene exposure and risk of breast cancer. No heterogeneity was detected for occupational cohorts, geographical regions, and durations of exposure. These findings, and an inability to exclude residual confounding, together with publication bias, prevent conclusions of causality.