OR01-3 Anti-Mullerian Hormone and Prediction of Trans-Menopausal Bone Loss

Program: Abstracts - Orals, Poster Previews, and Posters
Session: OR01-Osteoporosis: What You Had, What You Lost, and What You Gain
Translational
Friday, April 1, 2016: 11:45 AM-1:15 PM
Presentation Start Time: 12:15 PM
Room 157 (BCEC)
Arun S Karlamangla*1, Albert Shieh1, Sherri-Ann M Burnett-Bowie2, Elaine W. Yu2, Gail A Greendale1, Patrick M. Sluss2, Deborah Martin3 and Joel S Finkelstein2
1University of California, Los Angeles, CA, 2Massachusetts General Hospital, Boston, MA, 3University of Pittsburgh, Pittsburgh, PA
The menopause transition (MT) in women is a period of bone loss, with the most rapid declines occurring in a 3-year period bracketing the final menstrual period (FMP). This period of rapid bone loss has been called the trans-menopause, and the rate of BMD decline over this period varies substantially between women (1).  Circulating levels of Anti-Mullerian Hormone (AMH) made by ovarian granulosa cells also decline as women progress through the MT (2). We hypothesized that serum levels of AMH in women early in the MT will predict the rate of bone loss over the trans-menopause. We tested this hypothesis using data from The Study of Women’s Health Across the Nation, a 7-site, multi-ethnic study of the MT. At baseline, participants had to be 42 to 52 years old, pre- or early peri-menopausal, have an intact uterus with 1 or 2 ovaries, and not be taking exogenous sex steroid hormones. Enrollment began in 1996 and women were asked to return annually.  At each visit, blood was collected between 8:00 and 10:00 AM after a 12-hour fast, during the early follicular phase (cycle days 2–5) whenever possible, and serum was stored at -80F.  In all women who had a natural (non-surgical) MT and a dateable FMP, serum level of AMH was measured from frozen blood samples using a new high-sensitivity monoclonal ELISA with a detection limit of 2 pg/mL (Pico AMH, Ansh Labs, Webster, TX). BMD in the lumbar spine and femoral neck was measured annually in 5 of the 7 study sites.  In 474 women who had AMH and BMD measurements between 2 and 4 years before the FMP, had a 2nd BMD measurement 3-4 years later, and had not taken any medications that affect bone prior to the 2nd BMD measurement, we examined the ability of AMH level to predict the annualized rate of BMD decline between the two visits (% decline per year). AMH inter-quartile range was [11,146] pg/mL. Median rate of BMD decline was 1.3% per year in the spine and 1.0% per year in the femoral neck. Adjusted for age, BMI, smoking, race/ethnicity, and study site, in multivariable linear regression, each 75% (or four-fold) decrement in AMH level was associated with 0.15% per year faster decline in spine BMD (p<0.001) and 0.13% per year faster decline in femoral neck BMD (p=0.005).  These associations persisted even after additional adjustment for time from FMP and serum levels of estradiol and FSH. In multivariable logistic regression, adjusted for age, BMI, smoking, race/ethnicity, and study site, each four-fold decrement in AMH level was also associated with 18% increase in the odds of faster-than-median decline in spine BMD (p=0.02) and 17% increase in the odds of faster-than-median decline in femoral neck BMD (p=0.02). These findings suggest that serum levels of AMH in women going through the MT can indeed predict the rate of trans-menopausal bone loss, and help identify the women at risk of most loss.  AMH levels appear to provide information about the rate of bone loss beyond that provided by serum levels of estradiol and FSH.

(1) Greendale GA et al., JBMR 2012; 27: 111-8. (2) Sowers MFR et al., JCEM 2008; 93: 3478-83.

Nothing to Disclose: ASK, AS, SAMB, EWY, GAG, PMS, DM, JSF

*Please take note of The Endocrine Society's News Embargo Policy at https://www.endocrine.org/news-room/endo-annual-meeting/pr-resources-for-endo

Sources of Research Support: Ansh Labs for measurement of AMH free of charge; NIH support to Study of Women’s Health Across the Nation (SWAN): NR004061, AG012505, AG012535, AG012531, AG012539, AG012546, AG012553, AG012554, AG012495; NIH Support to SWAN Repository: AG017719.