RESUMO
BACKGROUND: Benzene exposure has been associated with increased leukemia and other cancer risk; however, epidemiological evidence is inconsistent for the latter and confounding from smoking and alcohol was rarely adjusted. METHODS: We investigated associations of occupational benzene exposure and risk of leukemia, lymphoma, myeloma, lung, stomach, liver, and kidney cancers in a population-based cohort of 61,377 men ages 40-74. A job-exposure matrix, constructed by industrial hygienists specifically for the study population, was used to derive cumulative benzene exposure from all jobs held. Cox regressions were performed to estimate adjusted hazard ratios (aHR) and 95% confidence intervals (CI) for benzene-cancer risk associations with adjustment for potential confounders. RESULTS: Over 15-years of follow-up, 1,145 lung, 656 stomach, 445 liver, 243 kidney cancer cases, 100 leukemia, 124 lymphoma, and 46 myeloma cases were identified. Benzene exposure >550mg/m3 was associated with increased leukemia (aHR=2.3, 95%CI=1.1-4.5), lung (aHR=1.2, 95%CI=1.0-1.6), and stomach (aHR=1.4, 95%CI=1.0-1.9) cancer risk; benzene-exposure was associated with early cancer diagnosis age. The benzene-leukemia and -stomach cancer associations followed a linear dose-response pattern (Plinear=0.016 and 0.023), whereas benzene-lung cancer association was evident at higher exposure levels (Pnon-linear=0.027). Alcohol consumption modified the benzene-leukemia association (HR=3.0, 95%CI=1.1-8.3 for drinkers, aHR=0.9, 95%CI=0.4-2.0 for non-drinkers, Pinteraction=0.047). CONCLUSIONS: Benzene exposure was associated with increased leukemia, stomach, and lung cancer risk. Alcohol consumption may modify the benzene-leukemia association, although estimates are imprecise. IMPACT: Our study provides additional evidence that benzene exposure increases cancer risk beyond leukemia, information important for policymakers to develop programs to mitigate cancer risk among benzene-exposed workers.