Association of carbon tetrachloride with lung cancer incidence and mortality, including effects in low-smoking regions.

Abstract

8076 Background: Although cigarette smoking is the main risk factor for lung cancer, 10–20 percent of cases occur in non-smokers, with women disproportionately affected. This pattern suggests that community-level environmental exposures may contribute to lung cancer risk beyond smoking. Methods: In this ecologic, county-level study, average age-adjusted lung cancer incidence from 2017–2021 and sex-specific smoking prevalence were obtained from the NCI State Cancer Profiles. County-level exposure data were derived from the EPA National Air Toxics Assessment, with 73 air toxics screened to identify the most strongly associated pollutant. The leading exposure was evaluated in sex-stratified multivariable spatial error regression models adjusting for smoking prevalence and key county-level covariates, including radon, PM 2.5 , Social Vulnerability Index (SVI). Analyses included 2,042–2,238 U.S. counties. Additional models were fit separately within strata of smoking prevalence to compare associations across lower and higher-smoking counties. Results: Among 73 EPA-tracked air toxins, carbon tetrachloride (CCl 4 ) showed the strongest association with lung cancer incidence (population-weighted r = 0.34, p < 0.001). In fully adjusted spatial models, each standard deviation increase in carbon tetrachloride was associated with a 10.6% increase in lung cancer incidence in women and 10.4% in men, and with a 6.5% and 7.6% increase in lung cancer mortality, respectively. In stratified analyses, associations between CCl 4 and lung cancer incidence were larger in lower-smoking counties (10–12% per SD) than in higher-smoking counties (5–6% per SD), with similar patterns by sex. Conclusions: CCl 4 , a legacy industrial solvent with ongoing emissions, was the most predictive air toxin associated with population-level lung cancer incidence, showing larger adjusted associations than PM 2.5 or radon. Although effects were not sex-specific, incidence increases were greatest in lower-smoking communities, suggesting a greater relative contribution of environmental exposures. Percent change in lung cancer incidence and mortality per 1 standard deviation increase in covariates. Variable Women – Incidence (% per 1-SD) Men – Incidence (% per 1-SD) Women – Mortality (% per 1-SD) Men – Mortality (% per 1-SD) Current Smoking (Women) 6.7 (5.9, 7.6) 7.0 (6.2, 7.9) 7.3 (6.3, 8.3) 8.3 (7.4, 9.3) Former Smoking (Women) 4.6 (3.8, 5.3) 1.7 (1.0, 2.5) 5.0 (4.2, 6.0) 1.1 (0.3, 1.9) Current Smoking (Men) 5.0 (4.2, 6.0) 6.9 (6.1, 7.8) 7.4 (6.3, 8.3) 8.5 (7.6, 9.5) Former Smoking (Men) 4.0 (3.1, 4.9) 3.3 (2.4, 4.1) 4.9 (3.9, 5.9) 3.9 (3.0, 4.8) PM 2.5 0.0 (−0.7, 0.7) 0.3 (−0.4, 1.0) 0.9 (0.1, 1.6) 1.3 (0.5, 2.0) Radon 0.9 (−0.2, 2.0) 1.8 (0.7, 2.8) 1.1 (0.0, 2.3) 1.9 (0.7, 3.0) Carbon Tetrachloride (CCl 4 ) 10.6 (9.1, 12.2) 10.4 (8.9, 12.0) 6.5 (5.1, 7.8) 7.6 (6.2, 9.0) Counties in model (n) 2,170 2,238 2,042 2,182