Mortality Risk of Specific Caners and Long-term Exposure to Particulate Matter in the South Korean National Sample Cohort

Abstract

BACKGROUND AND AIM: Cancer is a major burden of disease worldwide. International Agency for Research on Cancer classified particulate matter (PM) as carcinogenic to human. Long-term exposure to PM has been associated with cancer mortality mainly from lung cancer. Although a few U.S studies demonstrated the association with mortality of various cancer types, evidence for non-lung cancer is limited. The aim of this study is to compare PM associations with cancer mortality across different types of cancer in the Seoul Metropolitan Areas, South Korea using a population-based nationwide cohort. METHODS: Using the National Health Insurance Services-National Sample cohort, we selected the subjects who received national screening during 2005-2007, aged over 30, were non-severely disabled, and had full address information. We defined mortality of seventeen specific cancers based on the International Classification of Disease, 10th reversion. Individual-level long-term PM concentrations were estimated for the previous 5 years of 2002-2015 from a previously developed prediction model. We performed time-dependent Cox proportional hazards model to estimate hazard ratios (HRs) and 95% confidence intervals per 10ug/m3 increase in PM2.5 and PM10 adjusting for sociodemographic characteristics, family history of cancer, health behaviors, and area-level characteristics. RESULTS:Of 206,717 included subjects, 4,122 subjects were died with cancer for 2007-2015 and lung cancer was accounted the most (24.81%). Although none of estimates were statistically significant possibly because of small sample size, HRs of PM2.5 were greater than 1 for lung, stomach, esophagus, non-Hodgikins, oral and pharynx, and brain cancer mortality (HR range: 1.44-7.14). These results were similar for PM10 but with reduced effect estimates. CONCLUSIONS:Although we found no evidence of the association, our findings of positive risk estimates for PM2.5 in lung, oral and pharynx, and brain cancer mortality were consistent with those in U.S. studies. Future studies should re-examine the association in expanded cohorts. KEYWORDS: Ambient air pollution, Particulate matter, Cancer mortality, National sample cohort