Gonadotrophin receptor hormone analogues in combination with add-back therapy: an update

Introduction

Gonadotrophin receptor hormone analogues (GnRHa) have been used in a range of sex hormone-dependent disorders in both men and women. In gynaecology, they are used prior to superovulation and in benign estrogen/progesterone-dependent disorders. In the management of premenstrual syndrome (PMS), GnRHa can completely abolish symptoms and give women the confidence that their condition can be treated, often for the first time in many years.

Their use is restricted by unacceptable side-effects associated with the hypo-estrogen state. The addition of other medications to counteract these side-effects, known as ‘add-back therapy’, has enabled wider application, longer durations of treatment and an increase in compliance. This review article will discuss the mechanisms of actions of GnRHa, their side-effects, the theory behind add-back therapy and evidence available for their application in various disease states in women.

Mechanism of action

GnRHa work by down-regulation and desensitization of pituitary GnRH receptors. The desensitization of GnRH receptors results in a reduction in serum gonadotrophins and ovarian steroid levels leading to a reversible hypo-estrogenic state. GnRH analogues have a 100–200 fold higher binding affinity to GnRH receptors. ¹ The effect of the agonist analogue results in an initial estrogen flare often worsening the disease state before the down-regulation of GnRH receptors occur.

Adverse effects of GnRHa therapy

The hypo-estrogenic state induced by GnRHa is associated with significant menopausal symptoms. The sudden drop in estradiol levels results in pronounced climacteric symptoms of hot flushes, vaginal dryness and decreased libido. These symptoms have a significant effect on compliance, particularly when accompanied with concerns regarding bone mineral density (BMD) and increased cardiovascular risk.

The clinically significant effects of GnRHa on BMD is the reason the recommended duration of unopposed GnRHa is restricted. Surrey and Hornstein ² demonstrated in a study of long-term treatment of endometriosis with GnRH agonist therapy for 12 months that women lost 5.4% of bone density and regained only 3% 12 months following the cessation of treatment. The implications of osteopenia have led to a standard treatment course of six months. At this duration, BMD has been demonstrated to partially restore to baseline six months following treatment. ³

The association of the hypo-estrogenic state of menopause and increase in cardiovascular risk is well documented. ^4–6 Concerns over adverse lipid changes and increased atherosclerotic risk are of concern, but to date, there is no conclusive evidence regarding the effects of unopposed GnRHa therapy on lipid profiles. Al-Omari et al. ⁷ published a brief report on the effects of GnRHa therapy over an eight-week period on the lipid profile in patients treated for endometriosis, but little has been published since. Vascular endothelial function has also been demonstrated to decrease in GnRHa-induced hypo-estrogenic states, with consequent improvement in function following the administration of estrogen/progesterone add-back therapy. ⁸ This may have considerable effect on cardiovascular risk in long-term use of GnRHa but more research is required in this area.

It is therefore necessary to use GnRHa in conjunction with adjuvant therapy or alternatively, long-term use of GnRHa alone should be avoided. The unacceptable side-effects that limit the duration of therapy are the rationale behind add-back therapy.

The concept behind add-back therapy administration is the use of additional drugs to reduce the deleterious effects of GnRHa while maintaining therapeutic effect. Add-back therapies are either hormonal or non-hormonal agents primarily aimed at lessening climacteric symptoms and maintaining BMD (Box 1). Some of the earlier trials involved estrogen, progesterone or a combination of both which, on first glance, appears counterintuitive. Barbieri ⁹ explained this with the estrogen threshold hypothesis based on a hierarchy of organ response to estradiol. The level of estradiol required to avoid climacteric side-effects and maintain BMD is far less than required in a number of estrogen-dependent disease states.

Premenstrual syndrome

Approximately 95% of all women of reproductive age experience psychological symptoms associated with the luteal phase of their menstrual cycle. A significant proportion of these women experience symptoms severe enough to significantly impact on their lives and warrant a diagnosis of PMS. ¹⁰ Such severe forms of PMS are classified in psychiatry under the Diagnostic and Statistical Manual of Mental Disorders, fourth edition, as premenstrual dysphoric disorder (PMDD). ¹¹

The cause of PMS is still largely unknown, but evidence suggests an association with an abnormal neurotransmitter with circulating ovarian steroids during ovulation being the trigger for this. ¹² Medical therapies are aimed either at ovulation suppression or modulating the abnormal neurotransmitter. At present, only progestogens are licensed for the treatment of PMS in the UK. Several selective serotonin-reuptake inhibitors such as Fluoxetine and drospirenone-containing oral contraceptives are licensed in the USA for PMDD.

GnRHa have been shown to significantly reduce physical and psychological symptoms of PMS. ¹³ Progestogen appears to be the main precipitating factor in triggering symptoms, but estrogen (debatably) may also play a role. PMS is far more sensitive to these hormones than other ovarian-hormone-dependent disorders, making the application of the estrogen threshold hypothesis difficult. Add-back therapy with low-dose estrogen is effective in decreasing the adverse effects of GnRHa but long-term use is not feasible given the risk of endometrial hyperplasia (EH) ¹⁴ and lack of evidence to support long-term add-back therapy. Some authors have found women with PMS to be completely intolerant of progestogen ¹⁵ while others have suggested long-cycle add-back treatment to be appropriate for symptom relief and endometrial protection. ¹⁴ Tibolone has also been considered and one controlled study supports its use. Although GnRHa are effective in treating PMS, the hypo-estrogenic effects preclude their use in long-term management. Although there is evidence to support various agents as effective add-back therapies, more research is required in this area.

Other gynaecological conditions in which GnRHa have been used include endometriosis, EH, dysfunctional uterine bleeding and fibroids.

Endometriosis

GnRHa with add-back therapy is effective in the treatment of endometriosis. ¹⁶ The prevalence of endometriosis is on the rise, and more of these are complex presentations requiring long-term therapy. Studies have repeatedly confirmed that add-back therapy significantly reduces the hypo-estrogenic side-effects of hot flushes, vaginal dryness, decreased libido and loss of bone density without affecting the improvement in pain scores achieved with GnRHa. ^17–19 Without the success of add-back, GnRHa would not have remained such a valuable tool in the treatment of endometriosis.

A recent Cochrane review ²⁰ of 15 prospective randomized controlled studies concluded that both danazol and estrogen + progestogen add-back therapy are effective in protecting BMD during treatment, and up to six and 12 months after cessation, respectively. Progestogen was not found to be protective of BMD when prescribed with GnRH analogues while there was little evidence to support the use of calcium-regulating agents. Although effective, danazol has limited long-term use given its significant androgenic side-effects and adverse effects on a lipid profile. ²¹

Although successful in the treatment of endometriosis, there is little or no difference in efficacy between GnRHa and other medical treatments. In a randomized controlled trial of leuprolide versus continuous oral contraceptives in the treatment of endometriosis-associated pelvic pain, there were no differences between improvements in pain scores. ²² Given the lower cost, easier administration and generally low side-effects, continuous combined oral contraceptive (COCP) should be considered as first-line treatment of endometriosis. For disease states refractory to, or where oral contraceptives (OCPs)/progestogen are contraindicated, GnRHa + add-back therapy remains a valuable alternative.

It is important to note that when considering add-back therapy, estrogen alone is not suitable as it only serves to stimulate the original disease. Therefore, Tibolone continuous combined hormone replacement therapy (HRT) or OCP are the most appropriate forms of add-back therapy in endometriosis.

Endometrial hyperplasia

There is little evidence to support the use of GnRHa + add-back therapy in the treatment of EH. Agorastos et al. ²³ suggests that GnRHa combined with Tibolone as add-back therapy is a potent alternative in the treatment of EH. In their study, 26 women with histological confirmation of EH were treated for 12 months with monthly injections of 3.75 mg leuprolide acetate and Tibolone 2.5 mg per day. EH resolved in all 26 women but relapsed in 4/21 (19%) at two-year follow-up. This case series was followed up from their initial report ²⁴ of 42 women with EH (simple = 30, complex = 12) who received a six month course of GnRHa (leuprolide acetate or triptorelin) without add-back therapy. At three months of treatment a woman with initial simple hyperplasia (SH) was found to have developed complex hyperplasia (CH) that returned to functional endothelium at six months of treatment. Another patient developed focal CH from initial SH at six months of treatment and this did not resolve at six months after cessation of treatment. Following cessation of treatment, seven patients with an initial diagnosis of SH had progressed to CH (2 at 6 months follow-up, 5 at 12 months follow-up) resulting in a hysterectomy.

There seems to be little room for GnRHa + add-back therapy in the treatment of EH when such high rates of recurrence and progression are suggested, particularly with the availability of well-tolerated, low-cost, effective alternatives (i.e. progestogen-containing intrauterine system). In women who are intolerant to progestogen or where surgery is contraindicated, it may be an alternative but requires further research and close monitoring.

Dysfunctional uterine bleeding

Dysfunctional uterine bleeding (DUB) is a common premenopausal and perimenopausal disorder characterized by prolonged or excessive bleeding of uterine origin in the absence of an organic cause. Traditionally, first-line management of DUB has consisted of a medical approach with antifibrinolytic agents, progestogens, COCPs or danazol. National Institute for Health and Clinical Excellence guidance advises the Lng IUS as the most effective first-line therapy. Endometrial ablation has increased our options in avoiding major surgery. A Cochrane review found that thinning of the endometrium by GnRH analogues prior to endometrial ablation simplified the procedure resulting in shorter operating times, increase in success rates and decreased postoperative dysmenorrhoea. ²⁵

There is very little evidence supporting the use of prolonged GnRHa with add-back therapy in the treatment of DUB. In a randomized control trial of low-dose estradiol/norethisterone as add-back therapy, Franke et al. ²⁶ demonstrated that low-dose estrogen + progestogen did not adversely affect the treatment goal, reduced climacteric symptoms and reduced BMD loss over six months. However, these positive effects did not continue with prolonged treatment, with a return from baseline of climacteric symptoms at 24 months and decrease in BMD from six months. This study suggests that low-dose combined estrogen/progestogen is not suitable for prolonged add-back therapy and highlights the lack of evidence for long-term use. It may have a role in the short term prior to hysterectomy in anaemic patients so that haemoglobin levels can be improved preoperatively.

Fibroids

Uterine leiomyomas are a common benign estrogen-dependent tumour with a lifetime risk of 70% by the age of 50 years. ²⁷ In all, 20–50% of women will have symptoms significant enough to intervene. ²⁸ Symptomatic fibroids are the leading cause for hysterectomy in the UK. This has been greatly reduced over recent years with increasing medical and radiological management of fibroid-related symptoms.

Medically, there are only a few alternatives that specifically target the most common fibroid-associated symptom of menorrhagia (i.e. IUS, COCP, tranexemic acid). Although this has led to a significant reduction in hysterectomies, there are very few options that target pressure symptoms related to size and position.

Uterine fibroids appear in women during their reproductive years and regress during menopause. Although the initiating factor is still unknown, their growth depends on sex hormones, particularly estrogen. ²⁹ Uterine fibroids have been shown to have significantly increased concentrations of estrogen and progestogen receptors compared with normal myometrium, ³⁰ supporting the estrogen threshold hypothesis.

In 1985, Maheux et al. ³¹ first described the use of GnRHa for the treatment of fibroids. Since then, the literature consistently reports a 25–50% reduction in uterine volume following three months of treatment. Despite this, uterine volume returns to pretreatment size within six months of cessation. Given the side-effects, cost, mode of delivery and need for long-term continuous therapy, GnRHa are not indicated for the treatment of fibroids unless in conjunction with an alternative therapy.

A systematic review found that GnRHa given for at least three months prior to fibroid surgery improved preoperative haemoglobin and reduced vertical incisions during laparotomy and increased rate of vaginal rather than abdominal approach for women having a hysterectomy. ³²

Recent investigations suggest that GnRHa also improve the thermoablative effect of magnetic resonance-guided focused ultrasound. ³³

GnRHa is often prescribed for a period of less than six months in conjunction with alternative therapy, and therefore add-back therapy is not commonly required. When used, the majority of data appears to be in support of Tibolone as it has been shown to decrease vasomotor symptoms and loss of BMD without affecting uterine volume size when given at the time of commencement of GnRHa. ³⁴ Progestogen only is ineffective at protecting BMD ³² , while more evidence is required to support estrogen-only treatment. Raloxifene had no effect on improving vasomotor symptoms, but significantly reduced the loss of BMD and caused a significantly greater reduction in fibroid size at six months compared with leuprolide acetate alone. ³⁵

GnRHa remains a valuable tool in adjuvant therapy and may be considered for long-term management if surgery is contraindicated.

Discussion

This review article has outlined the various applications of GnRHa in gonadrophin-dependent disorders. There is an abundance of evidence supporting the efficacy of this agent in treatment of these conditions, but support for long-term therapy with the support of add-back is minimal.

Historically, the success of add-back treatment has been gauged on its protective effects on BMD. Measurement of BMD is the only objective method available to assess fracture risk during menopause and GnRHa therapy. Relying on this as the only indicator may be misleading. GnRHa have been demonstrated to have a direct effect on bone re-modelling regardless of the degree of BMD loss. This may have a profound effect on fracture risk but is yet to be evaluated.

Arguably, a side-effect of equal or greater concern is effects on cardiovascular risk. The effects of menopause on cardiovascular risk are well known but have not been adequately assessed in GnRHa therapy. More research is needed into the effects on lipid profile, vascular function, and whether specific add-back agents safely counter these.

GnRHa are an effective and versatile tool in the treatment of a variety of gynaecological conditions. In PMS these approaches are widely used but require further research. GnRHa's success in the treatment of endometriosis and adjuvant therapy in the management of fibroids is proven. However, the side-effects related to the hypo-estrogen state greatly limit their use in long-term management of these chronic conditions and without add-back therapy could not be entertained.

Competing interests

Prof. Rymer has received consultancy fees from drug companies that produce HRT products.