Endometriosis commonly begins during adolescence1 and can be a debilitating disease complicated by pain and activity limitation. Appropriate management involves prompt initiation of therapy and often maintenance of therapy over the long term.
Gonadotropin-releasing hormone (GnRH) agonists are commonly used for patients who fail first-line therapies. Gonadotropin-releasing hormone agonists relieve symptoms and reduce endometrial lesions seen on laparoscopy.2 Long-term GnRH agonist use is problematic as a result of deleterious effects on bone mineral density (BMD).3 Adults lost 5–8% of spine BMD after only 3–6 months of GnRH agonist therapy.4–64–64–6 Bone mineral density may not return to baseline after cessation of treatment.7,87,8
Add-back therapy with daily low doses of steroid hormone appears promising.4,5,94,5,94,5,9 Hornstein et al4 followed adults treated with GnRH agonists for 1 year. Patients receiving no add-back lost 6.3% of BMD, whereas BMD was preserved in all three studied add-back groups. The benefits of add-back extended for at least 8 months after completion of therapy.8 However, the significant BMD loss in the placebo group had not returned to baseline by 24 months after therapy.
The efficacy of add-back therapy must be studied in adolescents and young women given this critical time window for the attainment of peak bone mass.10 Any interference with this process puts patients at risk for lifelong low BMD and possibly increased fracture risk.11,1211,12 However, GnRH agonist therapy remains the only option to relieve debilitating pain for many young patients. Current care for these adolescents and young women includes the use of add-back based on adult experience.13 We were unable to find any data regarding GnRH agonist use with add-back for adolescents and young women (PubMed search, conducted from inception through April 2015, search terms “add-back,” “adolescents,” “endometriosis”).
Thus, our objective was to assess whether norethindrone acetate+conjugated equine estrogens or norethindrone acetate+placebo was superior for maintaining BMD in adolescents and young women with endometriosis treated with a GnRH agonist.
MATERIALS AND METHODS
From 2003 to 2008, 65 young women were screened for study eligibility (Fig. 1). Eligible females were aged 15–22 years, at least 2 years postmenarche, and had surgically confirmed endometriosis diagnosed on the basis of visual inspection14 with the clinical decision made to initiate treatment with leuprolide acetate depot (11.25 mg intramuscularly) every 3 months. Patients were excluded for other medical diagnoses (eg, celiac disease, diabetes mellitus) or medications (eg, glucocorticoids) known to affect BMD. The treating provider made the decision to offer GnRH agonist therapy to each patient on the basis of ongoing pain or intolerance of other medications. First, the patient had to agree to begin GnRH agonist therapy; patients were then approached and enrolled into the trial. Treatment assignment was determined before baseline measurements for logistic reasons. Fifty-three patients underwent randomization; 51 completed baseline measurements and were included in the intention-to-treat analysis. Before initiation of the GnRH agonists, patients were treated with varied medical therapies: combination oral contraceptives (n=27 [53%]), norethindrone acetate monotherapy (n=20 [39%]), depot medroxyprogesterone (n=2 [4%]), or no medical therapy (n=2 [4%]). Patients were not required to undergo a “wash out” before beginning the GnRH agonist. The Boston Children's Hospital institutional review board approved the protocol. Informed consent was obtained with parental consent or participant assent for patients younger than 18 years (clinicaltrials.gov NCT00474851).
The study was a single-site, randomized, double-blind, placebo-controlled 12-month trial conducted at Boston Children's Hospital. A permuted-block randomization design was used to prevent prediction of assignments while preserving balanced allocation of treatments. Treatment allocation was governed by a sequential list of study IDs and corresponding random assignments prepared in advance by the study statistician and followed by the pharmacy as each new participant enrolled and was assigned the next study ID. Only the statistician and pharmacy staff were privy to the assignments until after the trial was completed. The principal investigator, study staff, and patients were blinded throughout the trial. Group 1 received a two-drug add-back: oral norethindrone acetate (5 mg daily) plus conjugated equine estrogens (0.625 mg daily). Group 2 received norethindrone acetate plus placebo. The conjugated equine estrogens and placebo were dispensed by our pharmacy in a gelatin capsule and were identical in appearance. Add-back therapy started 21 days after the first injection of the GnRH agonist. All patients were advised to consume the recommended daily intake of calcium and vitamin D.
The primary study outcomes were areal BMD of the total hip, lumbar spine (L1–L4), and whole body by dual-energy X-ray absorptiometry measured at 0, 6, and 12 months. Measurements were compared with age- and gender-matched norms. Average in vivo precision for areal BMD (expressed as percent coefficient of variation) was 0.62% at the spine and 0.72% at the total hip. Body composition (lean body mass, fat mass) was also measured by dual-energy X-ray absorptiometry.
After randomization, participants returned for assessments at 3, 6, 9, and 12 months. Medication compliance was measured at each visit with pill counts. Pain symptoms were graded using the Pain scale from the 36-Item Short Form Health Survey (SF-36). Height was measured in a standardized fashion using the same wall-mounted stadiometer. Weight was obtained on the same digital scale with patients clothed in a hospital gown after voiding. Measures of quality of life and mood were obtained. The SF-36 measures health-related quality of life under two domains, physical and mental, with lower scores indicating more disability.15 The Beck Depression Inventory-II measures depressive symptoms.16 The Menopausal Rating Scale measures health-related quality of life with a focus on psychological, somatovegetative, and urogenital symptoms.17
Fasting venous blood samples were collected at the baseline, 6-month, and 12-month study visit for measurement of hormone concentrations (estradiol [high-performance liquid chromatography tandem mass spectrometry]) and safety monitoring studies (liver function tests, lipid profiles). Samples were assayed at a central reference laboratory.
The trial was designed with 80% power to detect a rate of change of 0.017 g/cm2 per year in hip BMD. Baseline comparison of continuous measures between trial arms was made by Student's t test, corroborated by Wilcoxon two-sample test for variables with skewed distribution. Dichotomies were compared by Fisher's exact test. Analysis followed the intention-to-treat principle with all data attributed to the patient's assigned treatment group regardless of whether treatment was delivered or completed. The time course of each measurement from baseline to 3, 6, 9, and 12 months was compared between arms by repeated-measures analysis of variance with an autoregressive covariance model to account for visit-to-visit correlation within patients. The primary test of treatment efficacy was time-by-treatment interaction. Adjusted changes over time and differences between trial arms were constructed from parameters of the repeated-measures analysis of variance. Outcome variables with highly skewed distributions were log-transformed for analysis and retransformed for reporting with changes and differences expressed multiplicatively. SAS 9.2 was used for all computations. P=.05 or below was considered a statistically significant difference.
Among the 53 patients randomized, two (one norethindrone acetate plus conjugated equine estrogens, one norethindrone acetate plus placebo) became ineligible or withdrew before completing baseline measurements (Fig. 1), resulting in a final sample of 51 participants (25 norethindrone acetate plus conjugated equine estrogens, 26 norethindrone acetate plus placebo) aged 17.9±1.7 years (mean±standard deviation). The groups did not differ in baseline demographic characteristics (Table 1). All patients had American Society for Reproductive Medicine stage 1 or 2 endometriosis at the time of diagnostic laparoscopy. Vitamin D concentrations were measured at baseline; if deficient or insufficient, patients were supplemented to achieve normal levels.18
Seven patients receiving norethindrone acetate plus conjugated equine estrogens and 10 patients receiving placebo discontinued participation (Fig. 1). Of those terminating study participation, 15 of 17 (88%) did so because of a decision to stop GnRH agonist therapy altogether. Reasons for discontinuation are presented in Table 2. Dropouts did not differ from completers with respect to baseline characteristics.
Twenty-eight percent of patients reported missing at least one dose of add-back medication before the 3-month visit, 33% at 6 months; 26% at 9 months; and 21% at 12 months. Reported compliance did not vary between visits (P=.67) or between trial arms (P=.77). During study participation, seven patients were evaluated in the emergency department for abdominal pain. Thirty-four adolescents and young women completed the trial.
At baseline, the two study groups had similar, normal measurements of BMD and BMD Z-scores at all measured sites (Table 1; P>.50). During 12 months of GnRH agonist therapy, participants receiving norethindrone acetate plus placebo exhibited stabilization of total body, lumbar spine, and hip BMD and BMD Z-scores (Table 3; Pwithin>.30 for all measures). With norethindrone acetate plus conjugated equine estrogens, total body BMD showed a marginally significant increase over time (Fig. 2; Table 3, adjusted mean change +0.012 g/cm2 at 12 months, Pwithin=.05). No losses in BMD at the lumbar spine and hip as well as BMD Z-scores were seen over time (Pwithin>.21 for all). No between-group differences were seen for any BMD or BMD Z-score measures. Body mineral density Z-scores remained within the normal range at all sites in both groups throughout the 12-month trial.
Bone mineral content changed in similar ways to BMD (Table 3). With norethindrone acetate plus conjugated equine estrogens, total body bone mineral content increased over time (Fig. 2; Table 3; mean adjusted change +37 at 12 months; Pwithin<.001), although not significantly changing in the norethindrone acetate plus placebo group (Pwithin=.31), a significant between-arm difference in time course (Pbetween=.02). At the hip and spine, bone mineral content remained stable throughout the trial in both groups (Table 3).
Both groups were overweight at baseline (body mass index [calculated as weight (kg)/[height (m)]2] mean±standard deviation, 25.2±4.6) and gained weight at a nonsignificant rate during the 12-month study (Table 4). Patients receiving norethindrone acetate plus conjugated equine estrogens gained 3.4±1.7 kg by 12 months (mean±standard error, Pwithin=.24), whereas those receiving norethindrone acetate plus placebo gained 2.3±1.8 kg over the same time period (mean±standard error, Pwithin=.73; Pbetween=.77). Body mass index did not change significantly in either group.
Body composition as measured by dual-energy X-ray absorptiometry did vary over the 12-month study (Fig. 2). Lean mass increased only in those receiving norethindrone acetate plus conjugated equine estrogens (Table 4; adjusted mean change +1.4 kg at 12 months, Pwithin=.001), whereas it remained unchanged in the norethindrone acetate plus placebo arm (Pwithin=.07, Pbetween=.006). Fat mass increased similarly over time in both groups (Pbetween=.44).
At baseline, serum estradiol concentrations are difficult to interpret, because the majority of participants were receiving hormonal therapy at the time of the blood draw. Estradiol concentrations did not change significantly during treatment within (P>.29) or between the trial arms (P=.46). Estradiol concentrations remained in the menopausal range starting 12 weeks after first GnRH agonist injection for the course of the study.
Pain subscale scores on the SF-36 improved in both trial arms over time. In patients receiving norethindrone acetate plus placebo, pain scores improved at 6 months (adjusted mean change +14) and then declined slightly by 12 months (+8, Pwithin≤.001). Recipients of norethindrone acetate plus conjugated equine estrogens had sustained improvements in pain scores to 12 months (+12, Pwithin=.002) leading to a significant difference between the treatment groups (Pbetween=.008).
At baseline these young women reported depressed mood and poor quality of life related to physical functioning. Overall scores for physical health (Physical Summary Score) were significantly lower than the United States mean for both add-back groups at baseline, indicating impairment. Although both groups improved (Fig. 3; Pwithin≤.003), patients in the norethindrone acetate plus conjugated equine estrogens group showed greater increases in Physical Summary Score than the norethindrone acetate plus placebo group (Pbetween=.005). Mental Summary Scores for both groups were not lower than the U.S. mean at baseline. Neither group showed significant changes in Mental Summary Scores over time (Pwithin≥.49). Neither add-back regimen was associated with changes in scores on the Beck Depression Inventory-II or Menopausal Rating Scale (Pwithin≥.20).
Mean alanine transaminase and aspartate aminotransaminase were similar between groups at trial initiation and remained normal throughout (Fig. 4). Only one participant (randomized to norethindrone acetate plus placebo) developed a transaminase elevation (alanine transaminase greater than 100 international units/L) while in the study. The elevation occurred at her 12-month visit and was thought to be related to a concurrent viral illness. The alanine transaminase spontaneously resolved within the next 4 weeks. Neither add-back regimen led to deleterious changes in lipid profiles. Seven patients had an elevated total cholesterol greater than 200 mg/dL at baseline; in every case, the total cholesterol decreased over time, whereas the high-density lipoprotein rose. Total cholesterol and triglycerides declined similarly in the two trial arms (Fig. 4; Pbetween≥.34).
In this randomized trial, add-back therapy with either regimen successfully preserved skeletal health in young women with endometriosis treated with 1 year of GnRH agonist therapy. However, combination therapy with oral norethindrone acetate plus conjugated equine estrogens appears to be superior to norethindrone acetate plus placebo. Patients randomized to norethindrone acetate plus conjugated equine estrogens had increases of total body bone mineral content and areal BMD, whereas those receiving norethindrone acetate plus placebo showed no changes. No BMD losses were noted at the hip or lumbar spine with either add-back therapy.
The rationale for add-back is based on the “estrogen threshold hypothesis.”19 Endometriotic tissue growth is stimulated by estrogen. Low circulating estradiol concentrations cause regression of estrogen-sensitive tissues like endometrial implants. Adding back small amounts of hormone increases circulating estradiol levels enough to maintain bone integrity and prevent menopausal symptoms while suppressing other tissues, like the endometrium. Norethindrone acetate has been the most widely studied adjunctive therapy in adults.4 Norethindrone acetate inhibits pituitary gonadotropin secretion, which prevents follicular maturation and ovulation. In vivo, norethindrone acetate is converted into ethinyl estradiol after oral ingestion such that 20 mg norethindrone acetate may be equivalent to taking a pill containing 30 micrograms ethinyl estradiol.20 This conversion may be partly responsible for norethindrone acetate's efficacy for preserving bone health.
The addition of low-dose conjugated equine estrogens add-back appears to offer additional benefit. Estrogen deficiency reduces bone accrual during adolescence and bone loss after achievement of peak bone mass.21 Estrogen deficiency may also slow growth in axial bone size during adolescence, a time when truncal growth typically accelerates.22 Low estrogen states such as anorexia nervosa or menopause are characterized by increased bone resorption.23 In other patient groups, estrogen decreases accelerated bone resorption.24 Unlike norethindrone acetate plus placebo, combined norethindrone acetate plus conjugated equine estrogens not only preserves, but also increases areal BMD and bone mineral content during this important time for bone accrual.
We can only hypothesize regarding the mechanism of action behind the preservation of areal BMD noted in our study. Estrogens inhibit bone resorption while decreasing bone formation.25 Androgens stimulate bone formation.26 Norethindrone acetate is an “estrogen precursor” and is converted in vivo into estrogen. Given the potentially synergistic mechanisms of action, we explored the effect of combination norethindrone acetate plus conjugated equine estrogens during GnRH agonist treatment in adolescence. With dual add-back, both bone formation and resorption may be affected to attenuate bone loss. Interestingly, total body BMD, a site composed primarily of cortical bone rather than trabecular-rich spinal BMD, responded to the norethindrone acetate plus conjugated equine estrogens. Trabecular bone is typically more hormonally responsive and hypothetically should have been more sensitive to these hormonal regimens. Our findings did not support that hypothesis and bear further study.
Participants reported significant disability related to their physical health before initiating GnRH agonist therapy. Over time, improvements in self-perceived physical health were seen, as demonstrated by increases in SF-36 Physical Summary Scores. Although improvements occurred in both trial arms, patients receiving norethindrone acetate plus conjugated equine estrogens maintained greater increases at 12 months than those receiving norethindrone acetate plus placebo. Minimal changes were seen in mental health measures, but patients did not report impairments in this area at baseline. Reassuringly, neither depressed mood nor menopausal symptoms worsened over the trial.
Potential adverse effects of treatment were monitored closely. In adults, treatment with GnRH agonists without add-back led to low-density lipoprotein cholesterol elevations and increased low-density lipoprotein and high-density lipoprotein cholesterol ratios.27 Gonadotropin-releasing hormone agonists plus progestin add-back have negatively affected lipid profiles28; these adverse effects were minimized in patients taking lower doses of steroid hormones such as those utilized in the current study.4,284,28 We did not find negative changes in lipid profiles nor elevation in liver enzymes associated with our hormonal therapy.
Study limitations should be acknowledged. Our sample was limited to women at least 18–24 months postmenarche because further skeletal growth is minimal and skeletal maturity is complete29; results may not be generalizable to younger females with open epiphyses. Our study was not powered to determine the effects of add-back therapy on body mass index, quality of life, estradiol, or safety measures. Measurements of areal BMD provide two-dimensional measurements of BMD and do not yield information regarding skeletal strength or microarchitecture. Our future work will explore the effects of this treatment on the peripheral skeleton, a common fracture site among youth, and use more sophisticated assessment tools to explore mechanisms of action and effects on bone strength.
In summary, a combination regimen of oral norethindrone acetate plus conjugated equine estrogens appears to be safe and effective for increasing areal BMD and bone mineral content in young women with endometriosis during 1 year of GnRH agonist treatment and superior to norethindrone acetate plus placebo. Monitoring of clinical and biochemical safety parameters yielded reassuring results. Given the increasing prevalence of endometriosis, these data suggest norethindrone acetate plus conjugated equine estrogens to be a useful adjunctive therapy to prevent bone loss while these adolescents and young women receive appropriate medical treatment for their underlying disease.
1. Laufer MR, Goitein L, Bush M, Cramer DW, Emans SJ. Prevalence of endometriosis in adolescent girls with chronic pelvic pain not responding to conventional therapy. J Pediatr Adolesc Gynecol 1997;10:199–202.
2. Steingold KA, Cedars M, Lu JK, Randle D, Judd HL, Meldrum DR. Treatment of endometriosis with a long-acting gonadotropin-releasing hormone agonist. Obstet Gynecol 1987;69:403–11.
3. Surrey ES. Gonadotropin-releasing hormone agonist and add-back therapy: what do the data show? Curr Opin Obstet Gynecol 2010;22:283–8.
4. Hornstein MD, Surrey ES, Weisberg GW, Casino LA. Leuprolide acetate depot and hormonal add-back in endometriosis: a 12-month study. Lupron Add-Back Study Group. Obstet Gynecol 1998;91:16–24.
5. Surrey ES, Judd HL. Reduction of vasomotor symptoms and bone mineral density loss with combined norethindrone and long-acting gonadotropin-releasing hormone agonist therapy of symptomatic endometriosis: a prospective randomized trial. J Clin Endocrinol Metab 1992;75:558–63.
6. Matsuo H. Prediction of the change in bone mineral density induced by gonadotropin-releasing hormone agonist treatment for endometriosis. Fertil Steril 2004;81:149–53.
7. Revilla R, Revilla M, Hernández ER, Villa LF, Varela L, Rico H. Evidence that the loss of bone mass induced by GnRH agonists is not totally recovered. Maturitas 1995;22:145–50.
8. Surrey ES, Hornstein MD. Prolonged GnRH agonist and add-back therapy for symptomatic endometriosis: long-term follow-up. Obstet Gynecol 2002;99:709–19.
9. Kiesel L, Schweppe KW, Sillem M, Siebzehnrübl E. Should add-back therapy for endometriosis be deferred for optimal results? Br J Obstet Gynaecol 1996;103(suppl 14):15–7.
10. DiVasta AD, Gordon CM. Bone health in adolescents. Adolesc Med Clin 2006;17:639–52.
11. Golden NH, Abrams SA; Committee on Nutrition. Optimizing bone health in children and adolescents. Pediatrics 2014;134:e1229–43.
12. Bonjour JP, Chevalley T. Pubertal timing, bone acquisition, and risk of fracture throughout life. Endocr Rev 2014;35:820–47.
13. Endometriosis in adolescents. ACOG Committee Opinion No. 310. American College of Obstetricians and Gynecologists. Obstet Gynecol 2005;105:921–7.
14. Becker CM, Laufer MR, Stratton P, Hummelshoj L, Missmer SA, Zondervan KT, et al.. World Endometriosis Research Foundation Endometriosis Phenome and Biobanking Harmonisation Project: I. Surgical phenotype data collection in endometriosis research. Fertil Steril 2014;102:1213–22.
15. Ware JE. SF-36 health survey update. In: Maruish ME, editor. The use of psychological testing for treatment planning and outcomes assessment. Vol. 3. 3rd ed. Mahwah (NJ): Lawrence Erlbaum Associates; 2004. p. 693–718.
16. Beck AT, Brown G, Steer RA. Beck depression inventory II manual. San Antonio (TX): The Psychological Corporation; 1996.
17. Schneider HP, Heinemann LA, Rosemeier HP, Potthoff P, Behre HM. The Menopause Rating Scale (MRS): comparison with Kupperman index and quality-of-life scale SF-36. Climacteric 2000;3:50–8.
18. Holick MF, Binkley NC, Bischoff-Ferrari HA, Gordon CM, Hanley DA, Heaney RP, et al.. Evaluation, treatment, and prevention of vitamin D deficiency: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 2011;96:1911–30.
19. Barbieri RL. Hormone treatment of endometriosis: the estrogen threshold hypothesis. Am J Obstet Gynecol 1992;166:740–5.
20. Chu MC, Zhang X, Gentzschein E, Stanczyk FZ, Lobo RA. Formation of ethinyl estradiol in women during treatment with norethindrone acetate. J Clin Endocrinol Metab 2007;92:2205–7.
21. Gordon CM, Nelson LM. Amenorrhea and bone health in adolescents and young women. Curr Opin Obstet Gynecol 2003;15:377–84.
22. Seeman E, Karlsson MK, Duan Y. On exposure to anorexia nervosa, the temporal variation in axial and appendicular skeletal development predisposes to site-specific deficits in bone size and density: a cross-sectional study. J Bone Miner Res 2000;15:2259–65.
23. Gordon CM, Goodman E, Emans SJ, Grace E, Becker KA, Rosen CJ, et al.. Physiologic regulators of bone turnover in young women with anorexia nervosa. J Pediatr 2002;141:64–70.
24. Lindsay R, Gallagher JC, Kleerekoper M, Pickar JH. Effect of lower doses of conjugated equine estrogens with and without medroxyprogesterone acetate on bone in early postmenopausal women. JAMA 2002;287:2668–76.
25. Liu SL, Lebrun CM. Effect of oral contraceptives and hormone replacement therapy on bone mineral density in premenopausal and perimenopausal women: a systematic review. Br J Sports Med 2006;40:11–24.
26. Gordon CM, Grace E, Emans SJ, Goodman E, Crawford MH, Leboff MS. Changes in bone turnover markers and menstrual function after short-term oral DHEA in young women with anorexia nervosa. J Bone Miner Res 1999;14:136–45.
27. Al-Omari WR, Nassir UN, Izzat B. Estrogen “add-back” and lipid profile during GnRH agonist (triptorelin) therapy. Int J Gynaecol Obstet 2001;74:61–2.
28. Surrey ES, Fournet N, Voigt B, Judd HL. Effects of sodium etidronate in combination with low-dose norethindrone in patients administered a long-acting GnRH agonist: a preliminary report. Obstet Gynecol 1993;81:581–6.
29. Chevalley T, Bonjour JP, Ferrari S, Rizzoli R. The influence of pubertal timing on bone mass acquisition: a predetermined trajectory detectable five years before menarche. J Clin Endocrinol Metab 2009;94:3424–31.