Endometriosis: Long-term treatment with gonadotropin-releasing hormone agonists

INTRODUCTION — Pelvic pain due to endometriosis can be treated with medical therapy, surgical intervention, or a combination of both. In most cases, women with chronic pelvic pain (CPP) thought to be due to endometriosis are initially treated empirically with nonsteroidal anti-inflammatory drugs (NSAIDs) and combined estrogen-progestin contraceptives. (See "Chronic pelvic pain in adult females: Treatment".)

If these medications do not resolve the pain, laparoscopy is usually performed to determine a definitive diagnosis. If the diagnosis of endometriosis is confirmed at laparoscopy, conservative surgery involving excision, fulguration, or laser vaporization of endometriotic implants and adhesiolysis is often attempted. (See "Endometriosis: Surgical management of pelvic pain".)

Pain can continue after conservative surgery. Recurrence of endometriosis is estimated as 21.5 percent at two years and 40 to 50 percent at five years after surgery [1]. For women with persistent pain or selected patients who have not responded to empiric treatment with NSAIDs or oral contraceptives [2], administration of a gonadotropin-releasing hormone (GnRH) agonist is usually effective.

Side effects of GnRH agonists, such as vasomotor symptoms and accelerated bone loss, limit treatment duration to six months. However, treatment can be extended beyond six months if add-back therapy is combined with the GnRH agonist.

BASIS FOR HORMONAL TREATMENT — Compared with eutopic endometrium, endometriotic implants are characterized by overproduction of prostaglandins and local production of estrogens and cytokines, which synergize the activities of each other, promote implantation of ectopic endometrium, and cause the pain associated with endometriosis [3]. In addition, aromatase overactivity results in increased COX2 expression favoring prostaglandin E production, which, in turn, upregulates estrogen synthesis pathways. Interventions that reduce ovarian estrogen production reduce this synergistic process, thereby reducing or eliminating endometriosis-related pain. (See "Endometriosis in adults: Pathogenesis, epidemiology, and clinical impact".)

The most effective methods for suppressing ovarian estrogen production in women of reproductive age are bilateral salpingo-oophorectomy and treatment with a GnRH agonist.

●Salpingo-oophorectomy – Bilateral oophorectomy results in a rapid, deep reduction in estrogen to menopausal levels. Cohort studies indicate that hysterectomy with bilateral salpingo-oophorectomy is highly effective for treatment of pelvic pain caused by endometriosis [4-6]. However, this procedure induces severe menopausal symptoms and results in sterility. Thus, it is not a practical treatment option for most young women.

●GnRH agonists – GnRH agonists are the most widely used hormonal method for suppressing ovarian estrogen production, which is reduced to values found in oophorectomized women. They are highly effective for treatment of pelvic pain caused by endometriosis and have the advantage that their hypoestrogenic effects are largely reversible upon discontinuation of treatment. Menses typically return 60 to 90 days after cessation of intramuscular therapy.

In the United States, three GnRH agonists are approved by the Food and Drug Administration (FDA) for treatment of pelvic pain caused by endometriosis: leuprolide, nafarelin, and goserelin. They are approved for six months of continuous use for this indication but may be repeated or continued for an additional six months. The six-month limitation is due to concern about the significant bone loss that occurs with GnRH agonist therapy. Additional GnRH agonists (eg, triptorelin) are available for treatment of prostate cancer; some of these drugs are approved for therapy of endometriosis in other countries (eg, buserelin in Canada). Recommended doses for GnRH-agonists are listed in the table (table 1).

PHARMACOLOGY OF GnRH AGONISTS — GnRH agonists are derived from native GnRH by substituting a D-amino acid for the native L-amino acid at position 6 of the native decapeptide (table 2). This substitution makes the agonist resistant to degradation by endopeptidases, thereby increasing the half-life and resulting in prolonged receptor occupancy [7].

During the first 10 days of treatment with these agents, binding to the GnRH receptor stimulates the pituitary to produce luteinizing hormone (LH) and follicle stimulating hormone (FSH). Paradoxically, chronic treatment leads to a decrease in pituitary secretion of LH and FSH due to down-regulation of the GnRH receptor and pituitary desensitization. The fall in pituitary LH and FSH secretion suppresses ovarian follicular growth and ovulation, resulting in low levels of circulating estradiol and progesterone. Women chronically treated with clinically effective doses of parenteral GnRH agonists have circulating estradiol concentrations in the range of menopausal women, approximately 15 pg/mL.

In addition, GnRH agonists may have direct effects on the endometrium. GnRH receptors are present in endometrial cells and a study in cultured endometrioma cell lines reported increasing concentrations of leuprolide [1000 ng/mL] resulted in inhibition of cell growth [8].

SIDE EFFECTS

●Hypoestrogenic effects – The main side effects of GnRH agonists result from the hypoestrogenic state; three-quarters of women treated with leuprolide become hypoestrogenic within the first four weeks of therapy; 98 percent are hypoestrogenic by eight weeks [9]. Signs and symptoms include amenorrhea, vasomotor symptoms, sleep disturbance due to hot flashes, urogenital atrophy, and accelerated bone loss. Over 80 percent of women treated with leuprolide report vasomotor symptoms, approximately 30 percent report vaginal symptoms (eg, vaginitis) and 30 percent report headaches [10].

●Bone loss – The potential for bone loss is a major concern, as accelerated bone loss during treatment of women in their 20s and 30s may increase the risk of osteoporotic fractures when these women reach their 60s and 70s.

●Mood changes – Some patients may experience new or worsened depression [11,12].

●Concern for cardiovascular effects – In 2010, the US Food and Drug Administration (FDA) issued a warning that men receiving GnRH agonists for prostate cancer were at a small increased risk for diabetes, heart attack, stroke, and sudden death [13]. There is no evidence that these drugs have similar risks in women being treated for endometriosis. Women being treated for endometriosis are much younger than men being treated for prostate cancer, and the doses of GnRH agonists used to treat endometriosis are lower than those used to treat prostate cancer. Given the low baseline risk of cardiovascular disease in young women, a small increase in risk from use of GnRH agonists would be difficult to detect, even in large epidemiological studies. Nonetheless, given the FDA advisory, treating for no longer than a year with hormonal add-back and six months without hormonal add-back would seem prudent.

EFFICACY — Treatment with a GnRH agonist reduces endometriosis disease activity as objectively measured by pretreatment and posttreatment surgical staging, and subjectively decreases pelvic pain [10,14-19]. As an example, in one trial, 52 women with surgically proven endometriosis and pelvic pain were randomly assigned to receive leuprolide acetate (3.75 mg intramuscularly) or a placebo injection every four weeks for six months [10]. Eighty-two percent of women treated with leuprolide acetate completed the trial and 85 percent of these women reported a significant decrease in dysmenorrhea, pelvic pain, and dyspareunia. By comparison, 95 percent of women treated with placebo left the clinical trial because of continuing pelvic pain.

Danazol is also an effective therapy of endometriosis. A systematic review of 15 randomized trials comparing GnRH agonists with danazol concluded they were similarly effective for reduction in pain and endometriotic implants [17]. However, the side effect profile of danazol is worse than that of GnRH agonists. This was illustrated in a trial of 213 women with surgically proven endometriosis who were randomly assigned to receive danazol (800 mg orally) or one of two doses of nasal nafarelin (400 mcg or 800 mcg) daily for six months [14]. The three regimens had similar efficacy: over 80 percent of subjects had symptomatic relief of pelvic pain and objective improvement of endometriosis lesions (documented by pre- and posttreatment surgical staging). The main difference between the drugs was in the type and severity of side effects: danazol administration was associated with weight gain, edema, myalgia, increased circulating levels of serum aspartate aminotransferase (AST) and alanine aminotransferase (ALT), and decreased levels of high-density lipoprotein (HDL) cholesterol. The most common side effects associated with nafarelin were hot flashes, decreased libido, and vaginal dryness; HDL cholesterol, AST, and ALT were not affected. Bone density was not measured.

An advantage of GnRH agonists over danazol is that the hypoestrogenic side effects associated with GnRH agonists can be minimized by add-back therapy, while androgenic side effects associated with danazol are more difficult to treat.

Gonadotropin-releasing hormone agonists after conservative surgery — Whether GnRH agonist therapy is useful postoperatively in women with stage III-IV endometriosis is controversial. A small retrospective study of 90 women with stage III-IV endometriosis included 16 women who received no postoperative medical therapy, 52 women who received gestrinone, and 22 women who received a GnRH agonist [20]. The time to recurrence in the three groups was 64 months, 63 months, and 47 months, respectively. The cumulative pregnancy rates were 70 percent (7/10), 66.7 percent (10/15), and 60 percent (6/10), respectively. This study did not show improvement in pregnancy rates with endocrine therapy after conservative surgery.

Recurrence of symptomatic endometriosis — While postoperative GnRH agonist therapy does not appear to improve pregnancy rates, there is evidence that it might delay the recurrence of symptoms. It has been appreciated that GnRH agonist therapy is not useful for the treatment of infertility as the lesions are suppressed while on therapy, but "reactivate" soon after ovarian function returns [21]. After surgical management, the recurrence of disease or symptoms may result from recurrence of the treated lesions or further progression of the microscopic lesions that were not treated at the time of surgery. Recurrence rates after either surgical or medical therapy are reported to be up to 45 percent after five years [22]. For young women under the age of 21 years old diagnosed with endometriosis at laparoscopy during a five-year follow-up, 56 percent had recurrence of their symptoms. There was felt to be no association observed between the symptoms with the extent or site of the disease, the type of surgery or whether there was postsurgical medical treatment [23]. After GnRH agonist therapy, there is evidence that continuation of endometrial suppression with oral contraceptives reduces pain scores and the rate of subsequent hysterectomy [24]. In a randomized trial in which 109 women received either a GnRH agonist (nafarelin) or placebo after reductive laparoscopic resection of endometriosis, 31 percent of the GnRH agonist treated women required additional therapy for recurrence of symptoms at a median of 24 months compared with 57 percent of women receiving placebo being retreated at a median of 12 months [25].

Symptomatic rectovaginal endometriosis — Surgical treatment may be effective in women with symptomatic rectovaginal endometriosis but is associated with potentially significant morbidity and complications. A meta-analysis including 217 women with rectovaginal endometriosis evaluated the outcome of medical treatment with vaginal danazol, aromatase inhibitors, GnRH agonists (with add-back), and intrauterine progestin [26]. With one exception (aromatase inhibitors used alone), 60 to 90 percent of patients treated with these therapies reported considerable reduction or complete relief from pain symptoms during the 6- to 12-month treatment period.

GnRH WITH ADD-BACK THERAPY — Many women experience hypoestrogenic side effects during GnRH agonist therapy and recurrence of pain after discontinuation. Investigators have attempted to improve patient acceptance and extend duration of use by giving concomitant medications (called add-back therapy) to reduce the severity of the hypoestrogenic side effects induced by the treatment and improve quality of life (table 3) [27].

Treatment with a GnRH agonist plus add-back appears to be comparable to continuous contraceptives in terms of efficacy and quality of life. This was illustrated in a randomized trial of women with endometriosis-associated pelvic pain who were assigned to receive either leuprolide acetate depot 11.25 mg intramuscularly every 12 weeks and norethindrone acetate 5 mg orally daily or a monophasic oral contraceptive (ethinyl estradiol 35 µg plus norethindrone 1 mg) for 48 weeks [28]. There were no significant differences between the treatments in pelvic pain scores or markers of quality of life such as the Beck Depression Index or the Index of Sexual Satisfaction.

Drugs

Progestin-only add-back — The combination of leuprolide depot 3.75 mg monthly injection plus a high dose progestin (norethindrone acetate 5 mg orally daily) has been approved by the FDA for treatment of pelvic pain caused by endometriosis. In most women, this combination treatment is associated with both a reduction in vasomotor symptoms and preservation of bone mineral density.

Data supporting this add-back regimen were provided by a large trial in which women with pelvic pain and endometriosis were randomly assigned to one of four year-long treatment groups: (1) GnRH agonist alone (leuprolide acetate depot 3.75 mg intramuscularly every four weeks), (2) GnRH agonist plus a progestin (norethindrone acetate 5 mg orally daily), (3) GnRH agonist plus low-dose estrogen (conjugated estrogen 0.625 mg orally daily) plus a progestin (norethindrone acetate 5 mg orally daily), (4) GnRH agonist plus high-dose estrogen (conjugated estrogen 1.25 mg orally daily) plus a progestin (norethindrone acetate 5 mg orally daily) [9]. Daily calcium supplements (1000 mg) were administered to subjects in all four groups.

The major findings from this trial were:

●Women receiving no add-back or progestin add-back with or without low-dose estrogen (0.625 mg) had a similar reduction in pelvic pain.

●Add-back with low dose estrogen (0.625 mg) was more effective than with higher dose estrogen (1.25 mg). The group taking higher dose estrogen was more likely than other groups to discontinue treatment due to pelvic pain, probably because the higher dose estrogen stimulated continued functioning and growth of the endometriosis implants. Other studies have also reported that the use of higher doses of estrogen add-back (eg, estradiol 1 mg daily) is associated with more pelvic pain than regimens that utilize lower doses of estrogen add-back [29].

●Bone density significantly decreased in women who received a GnRH agonist (and calcium) alone, but was preserved in groups in which any hormonal add-back was included (figure 1).

●Vasomotor symptoms were markedly reduced in all groups receiving hormonal add-back.

●In a follow-up report 12 and 24 months posttreatment, women in all treatment groups continued to have symptom scores that were below baseline for at least eight months after completion of active therapy [30].

This trial proved the efficacy of add-back therapy with the combination of a GnRH agonist plus norethindrone acetate. This regimen should be offered to women with chronic pelvic pain caused by endometriosis prior to considering aggressive surgical therapy (hysterectomy and bilateral salpingo-oophorectomy) since it preserves fertility potential.

Estrogen plus progestin add-back — Some women do not tolerate a high dose progestin add-back regimen because of progestin-related symptoms such as mood changes, weight gain, and bloating. The FDA has not approved a regimen of GnRH agonist with low dose estrogen plus low dose progestin add-back for treatment of pelvic pain caused by endometriosis; however, numerous clinical trials have demonstrated that this combination is effective in the treatment of pelvic pain caused by endometriosis and with a reduction in side effects observed with GnRH agonist-only treatment (eg, vasomotor symptoms, reduction in bone mineral density) [27,31-34].

As an example, in one trial, women with endometriosis and pelvic pain were randomly assigned to receive either a GnRH agonist alone or a GnRH agonist plus low dose estradiol (transdermal estradiol 25 mcg per day) plus low dose progestin (medroxyprogesterone acetate 2.5 mg orally daily) [35]. The two groups had equivalent reductions in both pelvic pain symptom scores and endometriosis lesion activity as determined by surgical staging (figure 2). Compared with women who received estrogen-progestin add-back, women who received the GnRH agonist alone reported more vasomotor symptoms and greater bone loss was noted by dual energy radiograph absorptiometry of the spine.

This study suggests that control of symptoms and disease activity can be achieved and induction of hypoestrogenic side effects reduced by administering a GnRH agonist plus low dose estrogen-progestin add-back. Of some concern, bone mineral density loss occurred despite administration of very low doses of these hormones (figure 3). By comparison, the add-back trial described above showed add-back with a high dose progestin (norethindrone acetate 5 mg daily) with or without estrogen was not associated with a significant decline in bone density (figure 1) [9]. While patient compliance with GnRH therapy for women with endometriosis is not high, compliance is improved with the use of add-back therapy [36].

We advise clinicians to use the FDA-approved regimen of a GnRH agonist plus norethindrone acetate (5 mg) as first-line therapy. For women who have significant side effects from the high dose progestin, a trial of estrogen-low dose progestin add-back (eg, conjugated estrogen 0.625 mg plus medroxyprogesterone acetate 5 mg orally) is reasonable. Fewer than 10 percent of women on this regimen will have vaginal bleeding. Both the GnRH-agonist and add-back are initiated with onset of menses; there is no advantage to beginning the agonist without add-back therapy.

Nonsteroidal add-back — An alternative to steroidal add-back therapy is the use of nonsteroidal medicines to treat hypoestrogenic side effects. Many nonsteroidal regimens have been proposed to treat menopausal symptoms and accelerated bone loss. Although estrogen and high dose progestins are the most effective treatments for vasomotor symptoms caused by hypoestrogenism, herbal remedies [37], selective serotonin reuptake inhibitors (SSRIs), and serotonin/norepinephrine reuptake inhibitors (SNRIs) have been reported to have some efficacy. For women with accelerated bone loss, weight bearing exercise, calcium, and vitamin D supplementation have modest efficacy. (See "Menopausal hot flashes" and "Overview of the management of low bone mass and osteoporosis in postmenopausal women".)

Bisphosphonates are effective for restoring bone density and preventing fractures in women with osteopenia. Since most bisphosphonates are slowly released from the bone into the circulation for many years after treatment is discontinued, some experts are concerned that use of these drugs in women who are planning future pregnancies may expose the fetus and adversely affect fetal bone development [38]. Approximately two dozen cases of alendronate exposure prior to or during pregnancy have been described without reports of congenital anomalies [39].

Other — There are minimal data on estrogen-only add-back. This approach has generally been avoided due to concerns about stimulating endometriosis, leading to increased pain, and stimulating eutopic endometrium, leading to endometrial hyperplasia. A double-blind trial randomly assigned 13 women with endometriosis-related pain treated with leuprolide acetate to add-back with oral estradiol 1 mg daily or an identical placebo [29]. The trial was terminated early because of a trend toward increasing endometriosis-related pain in the estradiol group. A retrospective cohort study of 117 women who underwent conservative laparoscopic surgery followed by leuprorelin acetate and short-term add-back with tibolone or estradiol (1 mg) found no difference in pain scores between the two regimens [40]. Low or micro-dose estrogen add-back (eg, transdermal estradiol 0.014 to 0.025 mg/day) is unlikely to cause recurrence of endometriosis pain or endometrial hyperplasia, and can be effective in relieving vasomotor symptoms and urogenital dryness.

Concurrent versus delayed initiation of hormonal add-back — The most common regimen for add-back is to begin GnRH agonist therapy with steroid add-back at the same time. The advantage of this approach is that it will markedly reduce the likelihood that the woman will experience vasomotor symptoms and maximally preserves bone density. However, multiple drugs are initiated at once, a situation that might not be acceptable to all patients.

Alternatively, one can administer a GnRH agonist alone for one to six months; then, if the intent is to continue GnRH agonist therapy, steroid add-back is begun. The advantages of this approach are that it allows the clinician time to determine if treatment with GnRH will be successful and to gauge the severity of the individual woman's hypoestrogenic side effects. Given the reduction in vasomotor symptoms and the preservation of bone density with concurrent initiation of add-back with the GnRH agonist, we favor starting add-back with the agonist in most circumstances.

Duration of therapy — A short course of therapy (eg, three months) can provide effective relief, but pain is likely to return after discontinuation of the medication [41]. Use of GnRH agonist (leuprolide acetate depot 3.75 mg monthly) plus high-dose progestin add-back (norethindrone acetate 5 mg daily) has not been approved for use beyond 12 months by the FDA, as the safety and efficacy of extended therapy have not been evaluated extensively. However, a small number of clinical trials and small cohort studies have demonstrated that a GnRH agonist plus steroid add-back can be effective from 30 months to up to 10 years of treatment [42,43].

We do not recommend extension of GnRH analog therapy with add-back beyond 12 months (or beyond 6 months if GnRH agonist therapy is used alone). As oral contraceptive suppression following GnRH agonist therapy appears to reduce pain scores and GnRH agonist with add-back therapy appears similarly efficacious as continuous oral contraceptive therapy, it is reasonable to convert women to continuous oral contraceptive therapy after a year of combined GnRH agonist/add-back therapy [24,28]. In the trial, norethindrone 1 mg plus ethinyl estradiol 35 mcg was used for continuous oral contraceptive therapy [28]. For women who cannot or choose not to use estrogen, an alternative is to continue norethindrone acetate as used for add-back therapy.

LOW-DOSE GnRH — An alternative to add-back therapy for treatment of the hypoestrogenic state is to decrease the dose of GnRH agonist or increase the interval between doses. This reduces the magnitude of suppression of LH and FSH and allows some ovarian estrogen secretion. We employ this regimen when we need to use GnRH agonists for more than six months without add-back therapy.

Body mass index (BMI) is probably an important variable in predicting whether dose reduction or extended interval dosing will be effective. Women with a BMI less than 25 kg/m² who also exercise intensively are the most likely to reliably respond to these lower dose regimens.

Two options for lowering the GnRH dose are:

Titration method — Initiate therapy with a standard dose of daily GnRH agonist (eg, nafarelin 200 mcg [1 spray into a nostril] in the morning and again at night). After one to three months, suppression of ovarian estrogen secretion and amenorrhea will have been achieved in most patients and pelvic pain should be improved. At this point, the dose of nafarelin can be decreased to two nasal sprays on even days and one nasal spray on odd days of the month. This is an approximately 25 percent reduction in the dose of GnRH agonist.

At this lower dose, many women will note fewer vasomotor symptoms, but with continued amenorrhea and relief of pelvic pain. A small number of women can reduce the dose of nafarelin further to as little as one nasal spray daily and remain amenorrheic. If menses or pain recurs, the dose can be increased to achieve more complete suppression of pituitary LH and FSH secretion, and, in turn, ovarian estrogen secretion [44-46].

Longer interval method — Another approach is to increase the time interval between depot injections of the GnRH agonist from four to six weeks. As an example, in one study, women with endometriosis and pelvic pain were randomly assigned to receive triptorelin depot 3.75 mg intramuscularly every four weeks (standard regimen) or every six weeks (extended interval dosing) [47]. Women in both groups had similar improvement in pelvic pain, with a reduction in cost due to less frequent administration. Others have also reported equivalent relief of dysmenorrhea in women treated with standard monthly injection of triptorelin and those who received an extended-interval dosing regimen of every six weeks over a six-month interval [48].

Addition of an aromatase inhibitor — Ovarian estrogen production is just one of the three sources of estrogen that may stimulate endometrial implants. The other two sources are peripheral conversion of precursors in adipose tissue and local production of estrogens in the implants themselves. In women with suboptimal pain relief after use of GnRH agonists, the addition of an aromatase inhibitor can be helpful. In a six-month trial comparing anastrozole plus goserelin versus goserelin alone, statistically significant increases in the pain-free interval and decreased symptom recurrence rates were observed in the anastrozole/goserelin-treated subjects [49].

GnRH ANTAGONISTS — GnRH antagonists are a newer therapeutic agent for moderate to severe pain related to endometriosis; clinical data are still being collected [50,51]. One potential benefit of these agents is avoiding the initial stimulation of pituitary gonadotropins seen with GnRH agonists. While data on long-term use and outcomes are not yet available, a comprehensive review of oral GnRH antagonists advised a treatment duration of up to 24 months for the 150 mg once-daily dosage (results in partial suppression of estrogen production) and up to six months for the 200 mg twice-daily dosage (results in full suppression of estrogen with resultant bone loss) [52]. It is important to counsel patients that elagolix is not a contraceptive; pregnancies were reported in clinical trials with elagolix. A detailed discussion of the available GnRH antagonists and supporting data is available in related content. (See "Endometriosis: Treatment of pelvic pain", section on 'Antagonists'.)

LEVONORGESTREL-RELEASING IUD — There is a growing body of experience with the levonorgestrel-releasing intrauterine device (LNG IUD) for management of endometriosis. The following preliminary studies suggest the LNG IUD may be an alternative to GnRH agonists and their long-term side effects. More experience is warranted.

●A randomized trial of 40 parous women with moderate or severe dysmenorrhea undergoing operative laparoscopy for treatment of symptomatic endometriosis compared the effect of immediate postoperative insertion of a LNG IUD versus surgery alone [53]. One year postoperatively, moderate or severe dysmenorrhea occurred significantly less often in the LNG IUD group (2/20 subjects [10 percent] versus 9/20 [45 percent]; OR 0.14, 95% CI 0.02-0.75), resulting in greater patient satisfaction (OR 3.00, 95% CI 0.79-11.44) [53,54].

●In another study comparing lipids and other cardiovascular variables (interleukin-6, C-reactive protein, vascular cell adhesion molecules) between LNG IUD subjects and those receiving a GnRH agonist, there were more favorable changes seen in the LNG IUD group [55].

●A study of 22 women experiencing pain related to endometriosis compared the effects of six months treatment with a LNG IUD (n = 11) versus a GnRH agonist (n = 11) [56]. Pre- and posttreatment specimens of eutopic and ectopic endometrium were evaluated for both proliferative and apoptosis markers, and hormone receptors. The cell proliferation index was significantly reduced in the epithelium and stroma of both eutopic and ectopic endometrium in both treatment groups. Only the IUD users had an increased histologic score for Fas (which stimulates cell death through apoptosis [57]) in both tissues. In eutopic endometrium, both treatments lowered expression of estrogen receptor-alpha and progesterone receptor-alpha, whereas in ectopic endometrium, only LNG IUD treatment lowered expression. These results support that, unexpectedly, the "local" application of a levonorgestrel-releasing IUD produced therapeutic effects on the ectopic endometrium and an even greater apoptotic effect than a GnRH agonist.

An overview of the various medical and surgical treatments for endometriosis can be found separately. (See "Endometriosis: Treatment of pelvic pain".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Endometriosis".)

SUMMARY AND RECOMMENDATIONS

●Rationale for hypoestrogenic state – A hypoestrogenic state leads to atrophy of ectopic endometrium and consistently reduces or eliminates endometriosis-related pain. (See 'Basis for hormonal treatment' above.)

●Use of GnRH agonist – For women with chronic endometriosis-related pelvic pain unresponsive to oral contraceptives or progestins, we suggest a gonadotropin-releasing hormone (GnRH) agonist rather than danazol (Grade 2B). GnRH agonists have a more favorable side effect profile and, unlike oophorectomy, do not impair fertility. (See 'Efficacy' above.)

●Role of add-back therapy – Add-back therapy results in fewer vasomotor symptoms and less bone loss than when a GnRH agonist is used as monotherapy. Several add-back options are available (table 3). We do not advise use of GnRH analog therapy with add-back beyond a year or GnRH analog only beyond six months. After discontinuation of the GnRH analog, we advise use of a continuous combined estrogen-progestin oral contraceptive.

•We suggest a GnRH agonist plus norethindrone acetate (5 mg) as first-line therapy (Grade 2B). This regimen appears to preserve bone mineral density better than low-dose estrogen-progestin add-back. (See 'Progestin-only add-back' above.)

•For women who have significant side effects from the high dose progestin, we suggest a GnRH agonist plus low dose estrogen and progestin add-back (Grade 2B). (See 'Estrogen plus progestin add-back' above.)

•If steroidal add-back therapy is not possible, we suggest decreasing the dose of GnRH agonist or increasing the interval between doses (Grade 2C). (See 'Low-dose GnRH' above.)