Association of Active and Sedentary Behaviors with Postmenopausal Estrogen Metabolism

• Authors: Cher M. Dallal, University of Maryland; Louise A. Brinton, National Cancer Institute; Charles E. Matthews, National Cancer Institute; Ruth M. Pfeiffer, National Cancer Institute; Terryl Hartman, Emory University; Jolanta Lissowska, M.Sklodowska-Curie Memorial Cancer Center; Roni T. Falk, National Cancer Institute; Montserrat Garcia-Closas, Institute of Cancer Research; Xia Xu, Frederick National Laboratory for Cancer Research; Timothy D. Veenstra, Frederick National Laboratory for Cancer Research; Gretchen L. Gierach, National Cancer Institute
• Journal Title: Medicine and Science in Sports and Exercise
• Volume: Volume 48, Number 3
• Publisher: American College of Sports Medicine (ACSM) | 2016-03-01, Pages 439-448
• Type of Work: Article | Post-print: After Peer Review
• Publisher DOI: http://doi.org/10.1249/MSS.0000000000000790
• Subjects: Health Sciences, Epidemiology
• Keywords: Science & Technology; Life Sciences & Biomedicine; Sport Sciences; accelerometers; physical activity; sedentary; estrogen metabolites

Abstract

Purpose: Physical activity may reduce endogenous estrogens but few studies have assessed effects on estrogen metabolism and none have evaluated sedentary behavior in relation to estrogen metabolism. We assessed relationships between accelerometer-measured physical activity and sedentary behavior and 15 urinary estrogens and estrogen metabolites (EM) among postmenopausal controls from a population-based breast cancer case-control study conducted in Poland (2000-2003). Methods: Postmenopausal women (N=542) were ages 40 to 72 years and not currently using hormone therapy. Accelerometers, worn for seven days, were used to derive measures of average activity (counts/day) and sedentary behavior (<100 counts/min/day). EM were measured in 12-hour urine samples using liquid chromatography-tandem mass spectrometry. EM were analyzed individually, in metabolic pathways (C-2, -4, or -16), and as ratios relative to parent estrogens. Geometric means of EM by tertiles of accelerometer-measures, adjusted for age and body mass, were computed using linear models. Results: High activity was associated with lower levels of estrone and estradiol (p-trend=0.01) while increased sedentary time was positively associated with these parent estrogens (p-trend=0.04). Inverse associations were observed between high activity and 2-methoxyestradiol, 4-methoxyestradiol, 17-epiestriol and 16-epiestriol (p-trend=0.03). Sedentary time was positively associated with methylated catechols in the 2- and 4-hydroxylation pathways (p-trend≤0.04). Women in the highest tertile of activity had increased hydroxylation at the C-2, -4, and -16 sites relative to parent estrogens (p-trend≤0.02) while increased sedentary time was associated with a lower 16-pathway:parent estrogen ratio (p-trend=0.01). Conclusions: Higher activity was associated with lower urinary estrogens, possibly through increased estrogen hydroxylation and subsequent metabolism, while sedentary behavior may reduce metabolism.