Outdoor PM2.5 Chemical Components and Risk of Lung Cancer in a Large Prospective U.S. Cohort

Abstract Regional heterogeneity in the association between fine particulate matter (PM2.5) and lung cancer suggests that PM2.5 components may differentially influence risk. We used a spatiotemporal model to estimate historical (2000–2004) residential outdoor PM2.5 component concentrations (sulfate, ammonium, nitrate, black carbon, organic matter, mineral dust, sea salt) for participants from the NIH-AARP Diet and Health Study. We used Cox models to estimate hazard ratios and 95% confidence intervals (HR [CI]) of associations per interquartile range of PM2.5 components and common copollutant nitrogen dioxide (NO2) with lung cancer, overall and by histologic subtype. We fitted single-pollutant models and then specified the multipollutant mixture using elastic net regression. From the start of follow-up in 2010 through 2018, 8565 lung cancer cases were diagnosed among 364,842 participants. In single-pollutant models, sea salt (HR = 1.02 [1.00–1.04]) was associated with overall lung cancer risk; sea salt (HR = 1.04 [1.00–1.07]) and black carbon (HR = 1.05 [1.00–1.10]) with adenocarcinoma; and ammonium (HR = 1.10 [1.00–1.21]), sulfate (HR = 1.15 [1.04–1.27]), organic matter (HR = 1.06 [1.01–1.12]), and black carbon (HR = 1.09 [1.01–1.17]) with squamous cell carcinoma. NO2 was associated with increased risk overall and adenocarcinoma. In mixture models, NO2 (HR = 1.08 [1.03–1.13]) and sea salt (HR = 1.04 [1.01–1.07]) were associated with adenocarcinoma, and sulfate was marginally associated (HR = 1.14 [0.98–1.13]) with squamous cell carcinoma. We found no clear associations with small cell tumors. These findings suggest that certain PM2.5 components play a role in lung cancer subtype etiology. Given the limited number of existing studies, replication of these findings in diverse geographic locations is an important next step.