COMPARISON OF ACRYLAMIDE EXPOSURE BIOMARKERS IN CHILDREN AND ADOLESCENTS USING NATIONAL HEALTH AND NUTRITION EXAMINATION SURVEY (NHANES) 2003-04 AND 2015-16

Abstract

Acrylamide (AA) is an important industrial chemical of occupational concern due to its neurotoxicity and probable carcinogenicity; it is also a tobacco burning product and thus contributes to health concerns in smokers. More recently it was discovered to be thermally generated in the cooking of starch-rich foods, creating a potentially wider public health concern. Children and adolescents are a particularly vulnerable group because they consume more acrylamide-rich foods compared to adults. In addition, they are still going through important developmental stages. This study examines AA and its metabolite glycidamide (GA) using hemoglobin adduct biomarkers (HbAA and HbGA respectively) from the U.S. children (6-11) and adolescents’ (12-19) National Health and Nutrition Examination Survey’s 2003-04 (n=2814) and the 2015-16 (n=697). The study investigated changes in exposure over time and examined the contribution of potential modifiers including smoking status, race/ethnicity, poverty-to-income ratio, sex, and age. All HbAA and HbGA are reported as pmoL adduct per gram Hb (pmol/G Hb).Overall, HbAA biomarkers significantly (p<0.0001) declined from 2003-04, GMs (95% CI) (57.9 [55.7, 60.1] pmol/G Hb) versus (42.8 [41.4, 44.2] pmol/G Hb) in 2015-16 for all ages, with similar reductions observed in the individual children and the adolescent groups. Smokers had a higher burden of HbAA biomarkers than non-smokers, and with a significant reduction in numbers of smokers from 2003-04 to 2015-16, this likely contributes to the reduction in overall exposure. When non-smokers only were examined, a significant (p<0.0001) decrease in HbAA was still observed, from 2003-04 GMs (95% CI) (53.4 [52.0, 54.9] pmol/G Hb) versus (41.2 [40.2, 42.2] pmol/G Hb) in 2015-16, suggesting an additional contribution of changes in AA levels in food or frequency of high-risk food consumption. Similar statistically significant reductions were seen for both children and adolescent groups separately. HbGA is a marker of AA biotransformation to GA, which is a more mutagenic metabolite of AA. The ratio is of HbAA:HbGA is a phenotypic marker of mutagenic risk. In non-smokers, there was a significant (p=0.001) difference in the HbAA:HbGA ratio in children GMs (95% CI) (0.8 [0.8, 0.8] pmol/G Hb) at 2003-04 and (0.9 [0.9, 1.0] pmol/G Hb) at 2015-16 versus adolescents (1.0 [1.0, 1.1] pmol/G Hb) at 2003-04 and (1.1 [1.0, 1.2] pmol/G Hb) at 2015-16, respectively, suggesting children may be at greater risk to the mutational effects of AA exposure compared to adolescents. In multivariate regression analysis of non-smokers only, age and race significantly contributed to the HbAA biomarker levels, with higher HbAA in younger age groups and in non-Hispanic black participants, highlighting a disparity in exposure pattern. Overall, AA exposure seems to have reduced from 2003-04 to 2015-16; the reduction is driven by both changes in smoking but also diet. The young and non-Hispanic black participants remain at highest risk of exposure and potential health effects from exposure to AA.