Skip to main content
NUTRITION 2024 Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Back

Nutritional Epidemiology (Poster Session)

(P13-001-24) Serum and Urine Metabolomic Profiles of Long-Term Ultra-Processed Food Intake in the Longitudinal Interactive Diet and Activity Tracking in AARP (IDATA) Study

Sunday, June 30, 2024
12:45 PM – 1:45 PM CT
Location: Poster Board 390
Disclosure(s):
Leila Abar, PhD: No relevant financial relationship(s) with ineligible companies to disclose.
Erikka Loftfield, PhD, MPH: No relevant financial relationship(s) with ineligible companies to disclose.


Objectives: We aimed to identify candidate biomarkers of ultra-processed food (UPF) intake in a longitudinal study of free-living adults.

Methods: IDATA participants (n=718) with serially collected serum, 24-hour and first morning void (FMV) urine samples, two of each collected 6 months a part, and up to six 24-hour dietary recall (ASA-24), collected over the 12-month study period, were selected for metabolomics analysis using ultra-high performance liquid chromatography with tandem mass spectrometry. Average daily UPF intake was defined using the Nova system as energy (%) from category 4 foods. Partial Spearman’s correlations and Elastic Net regression, with ten-fold cross-validation, were used to estimate UPF-metabolites correlations and identify robust metabolomic profiles of UPF intake, respectively.

Results: IDATA participants were majority female (51%) and 97% completed 4 or more ASA-24s. Mean proportion of daily energy from UPF was 50%. Out of 952 serum and 1111 urine metabolites, 187 serum (144 known), 284 24-hour (207 known), and 234 FMV (168 known) urine metabolites were significantly correlated (p.FDR ≤ 0.01) with UPF intake. These included lipid (n=50 serum, 20 24-hour & 21 FMV urine), amino acid (n=34, 59 & 47), carbohydrate (n=3, 8 & 7), xenobiotic (n=30, 49 & 37), cofactor and vitamin (n= 9, 11 & 12), peptide (n=7, 6 &5), and nucleotide (n=5, 8 & 6) metabolites. Elastic net regression models identified 43 serum, 23 24-hour urine and 19 FMV urine metabolites as robust predictors of UPF intake. Of these, 4 metabolites overlapped between serum and both urine types, including two amino acids, (S)C(S)S-S-methyl cysteine sulfoxide (rs: -0.19 to -0.23) and N2-N5-diacetylornithine (rs: -0.26 to -0.27), and a partially characterized molecule, pentoic acid (rs: -0.28, to -0.31), which were negatively correlated with UPF intake, and the carbohydrate, 6-carboxymethylysine (rs: 0.15 to 0.21), which was positively correlated with UPF intake.

Conclusions: Long-term UPF intake is associated with distinct serum and urine metabolomic profiles in free-living adults with varying diets.

Funding Sources: NIH Intramural Research Program