Pooled analysis of genotoxicity markers in relation to exposure in the Flemish Environment and Health Studies (FLEHS) between 1999 and 2018

• Published in: Environmental research 2020-11, Vol.190, p.110002, Article 110002
• Main Authors: Koppen, G., Franken, C., Den Hond, E., Plusquin, M., Reimann, B., Leermakers, M., Covaci, A., Nawrot, T., Van Larebeke, N., Schoeters, G., Bruckers, L., Govarts, E.
• Format: Article
• Language: English
• Subjects: 8-Oxo-2′-deoxyguanosine; Adolescent; Biomarkers; Comet assay; Cross-Sectional Studies; Environmental Monitoring; Environmental Pollutants - toxicity; Flemish Environment and Health Studies; Genotoxicity; Human biomonitoring; Humans; Micronucleus assay; Particulate Matter; Pooled analysis
• ISSN: 0013-9351; 1096-0953
• DOI: 10.1016/j.envres.2020.110002

The Flemish Environment and Health Studies (FLEHS) are human biomonitoring surveys running in Flanders since 1999. Additionally to biomarkers of exposure, markers of genotoxicity and oxidative stress have been measured, including the alkaline comet and micronucleus assay in peripheral whole blood cells, and urinary concentrations of 8-oxo-2′-deoxyguanosine (8-oxodG). Exposure-effect associations were explored in a pooled dataset of nine different cross-sectional FLEHS surveys. Data of adolescents collected in a time frame of about 20 years (1999–2018) were compiled. The aim of the study was to examine whether increased variation in exposure, lifestyle and environmental factors would lead to more powerful and robust exposure-effect associations. The biomarkers were measured in 2283 adolescents in the age range of 14–18 years. Exposure to polycyclic aromatic hydrocarbons [1-hydroxypyrene (1-OHP)], benzene (tt’-muconic acid), metals (arsenic, cadmium, copper, nickel, thallium, lead, chromium), persistent organochlorines and phthalates were assessed in blood or urine. Furthermore, outdoor air levels of particulate matter (PM10 and PM2.5) at the residences of the youngsters were calculated. Pooled statistical analysis was done using mixed models. Study-specific differences in the genotoxicity markers and in the strength/direction of the association were accounted for. This was done by incorporating the random factor ‘study’ and a random study slope (if possible). The exposure markers were centered around the study-specific mean in order to correct for protocol changes over time. A significant association was observed for the urinary oxidative stress marker 8-oxodG, which was positively associated with 1-OHP (5% increase for doubling of 1-OHP levels, p = 0.001), and with urinary copper (26% increase for doubling of copper levels, p = 0.001), a metal involved in the Fenton reaction in biological systems. 8-oxodG was also associated with the sum of the metabolites of the phthalate di(2-ethylhexyl) phthalate (DEHP) (3% increase for doubling of the DEHP levels, p = 0.02). For those associations, data pooling increased the statistical power. However, some of the associations in the individual surveys, were not confirmed in the pooled analysis (such as comet assay and 8-oxodG vs. atmospheric PM; and 8-oxodG vs. urinary nickel). This may be due to inconsistencies in exposure-effect relations and/or variations in the pollutant mix over time and regions. Pooled analysis