Young women with anorexia nervosa (AN) have reduced secretion of dehydroepiandrosterone (DHEA) and estrogen contributing to skeletal deficits. In this randomized, placebo-controlled trial, we investigated the effects of oral DHEA + combined oral contraceptive (COC) versus placebo on changes in bone geometry in young women with AN. Eighty women with AN, aged 13 to 27 years, received a random, double-blinded assignment to micronized DHEA (50 mg/day) + COC (20 µg ethinyl estradiol/0.1 mg levonorgestrel) or placebo for 18 months. Measurements of areal bone mineral density (aBMD) at the total hip were obtained by dual-energy X-ray absorptiometry at 0, 6, 12, and 18 months. We used the Hip Structural Analysis (HSA) program to determine BMD, cross-sectional area (CSA), and section modulus at the femoral neck and shaft. Each measurement was expressed as a percentage of the age-, height-, and lean mass-specific mean from an independent sample of healthy adolescent females. Over the 18 months, DHEA + COC led to stabilization in femoral shaft BMD (0.0 ± 0.5% of normal mean for age, height, and lean mass/year) compared with decreases in the placebo group (−1.1 ± 0.5% per year, p = 0.03). Similarly, CSA, section modulus, and cortical thickness improved with treatment. In young women with AN, adrenal and gonadal hormone replacement improved bone health and increased cross-sectional geometry. Our results indicate that this combination treatment has a beneficial impact on surrogate measures of bone strength, and not only bone density, in young women with AN. © 2014 American Society for Bone and Mineral Research.