Childhood cancer and residential proximity to petrol stations

Abstract

TPS 761: Health effects of pollution sources and components, Johan Friso Foyer, Floor 1, August 28, 2019, 3:00 PM - 4:30 PM Background: Exposure to benzene is a known risk factor for acute myeloid leukaemia among adults. Residential and parental exposure to benzene has also been linked with childhood leukaemia. Children living close to petrol stations are exposed to higher concentrations of benzene compared to background levels. However, few studies have investigated the associated risk of childhood cancer. We aimed to investigate whether children living in close proximity of petrol stations are at greater risk of childhood cancers, leukaemia, and central nervous system (CNS) tumours. Methods: We identified all cases of childhood cancer (<16 years old) diagnosed between 1985 and 2015 from the Swiss Childhood Cancer Registry and selected 10 age and sex matched controls per case from the Swiss national censuses (1990, 2000, 2010 and onward). We extracted exact geocodes of cases and controls from the SCCR and the censuses, respectively, and from the Swiss Business Census for petrol stations. Proximity to petrol stations was measured as the distance from children’s homes to the nearest petrol station recorded in the last business census preceding a case’s diagnosis. Odds ratios were estimated using conditional logistic regression. Results: We identified 6,129 cases for all cancers combined, 1,880 cases of leukaemia, and 1,290 cases of CNS tumours. A total of 27 cancer cases lived in close proximity to petrol stations (<50m). The odds ratio for children living close to a petrol station compared with unexposed children (≥500m) was 1.57 (95%-CI 1.05, 2.36) for all cancers combined. We found no evidence of an association for leukaemia (OR 1.13; 95%-CI 0.52-2.47) and CNS tumours (2.06; 0.79-5.41). Conclusions: We found evidence of an increased risk of childhood cancer (all diagnosis combined) among children living in close vicinity of petrol stations. Our study does not allow any conclusions for specific diagnostic groups.