Abstract
Benzene, toluene, ethylbenzene, and xylene (BTEX) have been associated with certain cancers in the occupational population. This study aimed to investigate the associations between low-level ambient BTEX exposure and cancer risks in the general population. We leveraged data from the UK Biobank and included individuals free of cancer at 2006-2010 baseline. Annual concentrations of BTEX were estimated using a chemistry-climate model, and the associations between BTEX and incident overall and 18 site-specific cancers were investigated with Cox proportional hazard models. We also fitted restricted cubic splines to explore the exposure-response relationships. The study sample comprised 409,579 participants [mean age 56.2 (8.11) years; 219,315 (53.5%) females]. Over a mean (SD) follow-up period of 11.2 (2.64) years (4,597,164 person-years), 60,777 overall incident cancer cases occurred. The results showed significant associations between overall cancers and benzene [HR 1.93 (95% CI: 1.89, 1.96)], toluene [1.25 (1.23, 1.26)] and xylene [1.11 (1.10, 1.12)]. Benzene and toluene were associated with a higher risk of 18 site-specific cancers. For xylenes (a summation of ethylbenzene, m/p-xylene, and o-xylene in the model), significant associations with multiple myeloma, hepatobiliary tract, thyroid, or connective soft tissue were not observed. Exposure-response curves suggested a higher risk of overall cancer beyond the benzene threshold. For toluene and xylene, there was no threshold or plateau across the range of exposures. This large-scale prospective cohort study demonstrates that long-term exposure to low-level ambient BTEX could increase the risk of overall and site-specific cancers in the general population.