Association between occupational exposures and sarcoma incidence and mortality: systematic review and meta-analysis

Abstract

Abstract Background Sarcomas are a rare and heterogeneous group of tumors originating from mesenchymal or connective tissue. They represent less than 1% of all adult cancers. The etiology and epidemiology of sarcomas remain understudied and poorly understood. The main objective of our study was to systematically assess the association between various occupational exposures and risk of sarcomas. Methods We performed a systematic literature search using the PubMed, Scopus, EMBASE and Cochrane databases to identify relevant cohort and case–control studies. A meta-analysis method was applied on the incidence and mortality outcomes where the estimate with 95% confidence interval (CI) was obtained. Results We included a total of 50 publications in our systematic review and 35 in meta-analysis. For exposures to phenoxy herbicides and chlorophenols, the pooled odds ratio (OR) for sarcoma was 1.85 (95% CI: 1.22, 2.82), based on 16 studies with 2254 participants, while the pooled standardized mortality ratio was 40.93 (95% CI 2.19, 765.90), based on 4 cohort studies with 59,289 participants. For exposure to vinyl chloride monomers the pooled risk ratios for angiosarcoma of the liver and other STS were 19.23 (95% CI 2.03, 182.46) and 2.23 (95 CI 1.55, 3.22) respectively based on 3 cohort studies with 12,816 participants. Exposure to dioxins was associated with an increased STS mortality; the pooled standardized mortality ratio was 2.56 (95% CI 1.60, 4.10) based on 4 cohort studies with 30,797 participants. Finally, woodworking occupation was associated with an increased risk of STS with the pooled OR of 2.16 (95% CI 1.39, 3.36). Conclusions Our findings suggest a positive association between higher exposure to dioxins and increased mortality from STS, between cumulative exposure to vinyl chloride monomers and increased mortality from angiosarcoma of the liver and STS, and between woodworking occupation and STS incidence. These findings were all statistically significant.