OBJECTIVE Suppression of ovarian hormones in premenopausal women on gonadotropin-releasing hormone agonist (GnRH(AG)) therapy can cause fat mass (FM) gain and fat-free mass (FFM) loss. Whether this is specifically caused by a decline in serum estradiol (E2) is unknown. This study aims to evaluate the effects of GnRH(AG) with placebo (PL) or E2 add-back therapy on FM, FFM, and bone mineral density (BMD). Our exploratory aim was to evaluate the effects of resistance exercise training on body composition during the drug intervention. METHODS Seventy healthy premenopausal women underwent 5 months of GnRH(AG) therapy and were randomized to receive transdermal E2 (GnRH(AG) + E2, n = 35) or PL (GnRH(AG) + PL, n = 35) add-back therapy. As part of our exploratory aim to evaluate whether exercise can minimize the effects of hormone suppression, some women within each drug arm were randomized to undergo a resistance exercise program (GnRH(AG) + E2 + Ex, n = 12; GnRH(AG) + PL + Ex, n = 12). RESULTS The groups did not differ in mean (SD) age (36  and 35  y) or mean (SD) body mass index (both 28  kg/m). FFM declined in response to GnRH(AG) + PL (mean, -0.6 kg; 95% CI, -1.0 to -0.3) but not in response to GnRH(AG) + E2 (mean, 0.3 kg; 95% CI, -0.2 to 0.8) or GnRH(AG) + PL + Ex (mean, 0.1 kg; 95% CI, -0.6 to 0.7). Although FM did not change in either group, visceral fat area increased in response to GnRH(AG) + PL but not in response to GnRH(AG) + E2. GnRH(AG) + PL induced a decrease in BMD at the lumbar spine and proximal femur that was prevented by E2. Preliminary data suggest that exercise may have favorable effects on FM, FFM, and hip BMD. CONCLUSIONS Suppression of ovarian E2 results in loss of bone and FFM and expansion of abdominal adipose depots. Failure of hormone suppression to increase total FM conflicts with previous studies of the effects of GnRH(AG). Further research is necessary to understand the role of estrogen in energy balance regulation and fat distribution.