Abstract MP46: Metabolomic Response to Randomized Treatment With Estrogen and Estrogen Plus Progestin Therapy in Postmenopausal Women

Abstract

Introduction: Hormone therapy (HT) has been linked to chronic disease risk in postmenopausal women, but the effect of HT on the metabolome is unclear. Hypothesis: Randomized HT is associated with perturbations in the metabolome. Methods: The discovery set, from the Women’s Health Initiative (WHI), included 288 women in the conjugated equine estrogens-alone (CEE) trial (139 active; 149 placebo) and 244 women in the CEE + medroxyprogesterone acetate (CEE+MPA) trial (133 active; 111 placebo). Blood samples were collected 1 year after intervention. The validation set included 494 women from the Nurses’ Health Study (NHS) and NHS2 (264 non-users, 158 estrogen-alone users, and 72 estrogen + progestogen users). Plasma metabolome was profiled by liquid chromatography-tandem mass spectrometry; 478 known metabolites were used in analyses. We used elastic net regressions to develop metabolomic signatures for CEE and CEE+MPA in the WHI, adjusting for age and race. We applied the signature models to NHS/NHS2 and compared the signatures of current HT users to those of non-users. Pathway analysis was done using MetaboAnalyst. Results: In the WHI, we identified a CEE metabolomic signature comprised of 42 metabolites and a CEE+MPA signature comprised of 34 metabolites ( Fig. A ). The two signatures shared 8 metabolites (4 amino acids, 2 glycerophospholipids, and 2 quaternary amines). In NHS/NHS2, CEE and CEE+MPA signatures were highly correlated ( r =0.85) and both significantly higher in current HT users than non-users ( P ≤2.7х10 -15 ) ( Fig. B ). The CEE+MPA signature metabolites were enriched in 3 pathways (glycerophospholipid metabolism; aminoacyl-tRNA biosynthesis; and glycine, serine and threonine metabolism); no pathway enrichment was found for the CEE signature. Conclusions: We identified and validated metabolomic signatures for CEE and CEE+MPA use in postmenopausal women from WHI and NHS/NHS2. Whether alterations in lipid or amino acid metabolism mediate associations of HT use with chronic disease needs future investigation.