Effects of Anorexia Nervosa on Peak Bone Mass

Teenage girls with anorexia nervosa (AN) are at risk for low bone density and low rates of bone accrual, raising concerns regarding acquisition of peak bone mass, an important determinant of future bone health and fracture risk. Important factors contributing to low bone density in AN include low levels of estrogen and insulin like growth factor-1 (IGF-1). While estrogen is important for preventing bone loss, IGF-1 is important for optimizing bone formation. We have shown in a previous study that replacement of estrogen is effective in increasing bone density in teenage girls with AN; however, this increase in bone density remains lower than that seen in normal-weight controls over the same duration, and residual deficits persist. Importantly, the impact of administering replacement doses of IGF-1 with estrogen replacement has not been studied in teenagers with AN. This study will examine the impact of administering recombinant human (rh) insulin like growth factor-1 (rhIGF-1) with estrogen (to mimic pubertal levels of these hormones) versus administration of estrogen alone on bone metabolism in adolescent girls with anorexia nervosa (AN). One aim of this proposal is to investigate whether co-administration of insulin like growth factor-1 (rhIGF-1) with physiologic estradiol replacement to adolescent girls with AN will increase BMD (bone mineral density) more than estrogen monotherapy, and whether bone mass will approach that seen in healthy adolescent girls. An additional aim is to determine whether co-administration of rhIGF-1 with estradiol to mimic the normal pubertal milieu stimulates bone formation through an IGF-1 mediated anabolic effect, increases bone density to a greater extent than estrogen monotherapy, and improves bone mass accrual to approach that in healthy controls. The impact of rhIGF-1 +estradiol versus estradiol alone on bone microarchitecture will also be assessed.

Given the increasing prevalence of AN, its profound consequences on bone health, and lack of optimal treatment interventions, these studies will provide critical data needed to identify optimal treatment strategies for this severe co-morbid disease using state- of- the- art endpoints of BMD, bone microarchitecture and strength. Although both low IGF-1 and hypogonadism are associated with increased skeletal fragility in AN, the mechanisms by which these factors interact are incompletely understood. Specifically, the increased skeletal fragility that is associated with AN is poorly reflected by DXA-derived BMD. Furthermore, the magnitude and mechanisms by which IGF-1 deficiency and hypogonadism influence bone microarchitecture are not defined. The growing incidence of eating disorders in adolescent girls and their long-term effects on skeletal health provide strong rationale for studies that will provide a better understanding of these issues and the evaluation of rational therapeutic approaches. The studies described in this proposal utilize both cross-sectional and RCT approaches to achieve this goal. Additionally, our utilization of sophisticated techniques such as high resolution peripheral QCT (HR-pQCT) will improve our understanding of the relationship between IGF-1, gonadal steroids and bone quality and will aid in the development of effective therapies in the treatment of skeletal fragility in Anorexia Nervosa.

RhIGF-1 will be started at a dose of 30mcg/k/dose twice daily, and will be titrated up or down in 25% dose increments to maintain IGF-1 levels in the upper half of the normal range. Estradiol will be delivered transdermally using a 100 mcg patch (Vivelle Dot) changed twice weekly. Subjects will receive cyclic micronized progesterone (Prometrium) 100 mg daily for the first 10 days of each month. All subjects will receive supplemental calcium and vitamin D.

Placebo injections will be administered twice daily. Estradiol will be delivered transdermally using a patch (100 mcg) changed twice weekly. Subjects will receive cyclic micronized progesterone (Prometrium) 100 mg daily for the first 10 days of each month. All subjects will receive supplemental calcium and vitamin D.

Change in lumbar spine BMD z-score over 12 months as assessed by dual energy x-ray absorptiometry (DXA) The z-score indicates the number of standard deviations that BMD is away from the mean for age, sex and race. A z-score of 0 is equal to the mean with negative numbers indicating values lower than the mean and positive values higher values. A positive change in z-scores indicates a favorable outcome whereas a negative change in z-scores indicates an unfavorable outcome.

Change in trabecular number at the ultradistal radius over 12 months as assessed by high resolution peripheral quantitative computed tomography (HRpQCT)

Inclusion Criteria: AN: Age: 14-22 years old Bone age (BA): ≥14 years Should meet DSM IV criteria for AN Subjects at MGH will be evaluated by co-investigator Dr. David Herzog, Director of the Harris Center for Eating Disorders, at MGH, and by Dr. Debra Katzman, co-investigator, and the Hospital for Sick Children, Toronto who directs their Eating Disorders Program, respectively, before enrollment. Inclusion Criteria: Controls: Healthy adolescent girls 14-22 years BA of ≥14 years BMI between the 10th-90th percentiles for age Regular menstrual periods every 28-35 days for subjects ≥ 2 years post-menarche. Exclusion Criteria: Diseases known to affect bone metabolism including untreated thyroid disease, Cushing's syndrome, diabetes, pituitary disease, renal failure and prior bone fracture within six months of the study. Medications known to affect bone metabolism, including gonadal steroids, within three months. Evidence of suicidality, psychosis, or substance abuse. Premature ovarian failure, as demonstrated by an elevated FSH. Abnormal TSH. Hematocrit <30%, Potassium <3.0 mmol/L, Glucose <50 mg/dl Pregnancy History of malignancy Contraindications to estrogen therapy (for girls with AN)