Reproducibility of Plasma Metabolome over 1 Year in a Population-Based Cohort of Black Breast Cancer Survivors.

Abstract

Background: The metabolomics approach using blood samples from epidemiologic studies has the potential to elucidate pathways or uncover biomarkers for breast cancer outcomes. Therefore, understanding the within-person reproducibility of the blood metabolome and the factors that influence metabolite levels over time in breast cancer survivors are crucial, but these remain largely unknown.

Methods: We estimated the within-person reproducibility of plasma metabolites in 61 Black breast cancer survivors from the Women's Circle of Health Follow-Up Study. Samples were collected from each participant at two time points, approximately 2 and 3 years after diagnosis. Untargeted metabolomic profiles were analyzed by Metabolon using ultrahigh-performance LC/MS-MS. We calculated the intraclass correlation coefficients (ICC) for each metabolite by dividing the between-person variance by the total variance. ICCs were compared across preanalytic factors (e.g., fasting) and participant characteristics using the Wilcoxon test.

Results: Among 857 named metabolites, the median ICC was 0.58 (IQR: 0.44-0.70). Of the metabolites, 16.6% showed high within-person reproducibility (ICC ≥ 0.75), spanning all metabolite classes, whereas 65.6% had an ICC within 0.4 to 0.75, and 17.9% had an ICC < 0.4. Reasonable ICCs were also observed for nonfasting samples (median 0.53, IQR: 0.39-0.67), although lower than those for fasting samples (median 0.63, IQR: 0.45-0.77). ICCs were slightly lower in younger, nonobese participants and in women with estrogen receptor-positive breast cancer.

Conclusions: The within-person reproducibility of plasma metabolites over 1 year among breast cancer survivors was generally acceptable.

Impact: A single-timepoint measurement could be useful in evaluating associations between metabolites and breast cancer outcomes.