Description of the condition

Uterine fibroids, also called uterine leiomyomas or myomas, are the most common benign tumours in women of reproductive age. Studies report that about 70% to 80% of women have uterine fibroids, varying with populations and diagnostic method used, but only approximately 25% of them experience symptoms that interfere with daily living (Cramer 1990; Day 2003; Walker 2005). Generally, the severity of symptoms depends on the size, number and location of the fibroids (Cramer 1990). Symptoms associated with uterine fibroids include heavy or irregular bleeding, infertility and recurrent abortion. Fibroids can grow large enough to affect other pelvic structures, resulting in pressure symptoms such as increased urinary frequency, pelvic pain and constipation (Stewart 2001). Many women with significant symptoms require further therapy. Fibroids are the most common indication for gynaecological surgery, accounting for more than 200,000 hysterectomies annually in the United States (Walker 2005). Thus they have a substantial impact on women's health and quality of life.

Description of the intervention

Therapeutic options for women with fibroids need to be guided by the size and location of the fibroids, symptoms and the age and reproductive wishes of the woman. Traditional treatments for fibroids include a variety of surgical techniques. Hysterectomy (removal of the uterus) has long been regarded as the definitive management approach as it resolves bleeding and removes the risk of uterine cancer (Lepine 1997; Wallach 2004). The principal mode of treatment for women with symptomatic fibroids who wish to preserve their fertility is myomectomy (surgical resection of one or more fibroids from the uterine wall), performed during laparotomy (open surgery), by laparoscopy (minimally invasive surgery) or by hysteroscopy (surgery in the uterine cavity by scope with access through the cervix) (Donnez 2002; Stewart 2001). The primary disadvantage of myomectomy is a recurrence rate of 50%. Also, possible postoperative formation of adhesions can cause complications such as pain and infertility. Recently, more conservative approaches have been introduced such as uterine artery embolisation (UAE) (Gupta 2012) and magnetic resonance imaging–guided focused ultrasound surgery (MRgFUS) (Yoon 2013). These are approved by the U.S. Food and Drug Administration (FDA) (Hindley 2004; Pron 2003). However, albeit obtaining a similar patient satisfaction rate, together with a shorter recovery period, relative to hysterectomy and myomectomy (Gupta 2012), UAE is associated with higher rates of minor complications and an increased likelihood that surgical re‐intervention will be required within two to five years of the initial procedure (Moss 2011). Moreover, ongoing concerns surround the impact of UAE on fertility and pregnancy (Firouznia 2009; Singh 2007). In addition, for MRgFUS, published evidence is limited. Further studies are needed to determine the long‐term efficacy of this procedure and to evaluate its efficacy and safety compared with other treatments for uterine fibroids.

Effective medical therapy may reduce the need for surgery and may be cost‐effective. However, the long‐term effectiveness of medical therapy for fibroids may be limited by fibroid recurrence. The only FDA‐approved medical therapy is a gonadotrophin‐releasing hormone (GnRH) agonist that is used preoperatively with iron. GnRH agonists reduce both bleeding and bulk‐related symptoms, but because of the systemic hypoestrogenism caused by this drug, significant menopausal side effects have been reported, and this has limited their application. In contrast to their success in preclinical studies, to date selective oestrogen receptor modulators (a class of compounds that act on the oestrogen receptor) have yielded disappointing results in clinical trials and are not recommended by a Cochrane review (Wu 2007). Other drugs such as danazol or progesterone can shrink fibroids but are not appropriate for long‐term use because of adverse effects (Ke 2009).

Aromatase is an enzyme that catalyses the conversion of androgens via hydroxylation (a chemical process that introduces a hydroxyl group into an organic compound) to oestrogens. It is a microsomal enzyme of the cytochrome P450 superfamily and the product of the CYP19 gene located on chromosome 15q21 (Cole 1990). Aromatase enzyme activity has been identified in a variety of tissues, including ovary, breast, endometrium, myometrium, adipose tissue, brain, skin, bone and vascular endothelium, among others (Simpson 2003). Because estradiol and estrone are the final products in the steroidogenic pathway, aromatase is considered a target for selective inhibition.

Aromatase inhibitors (AIs) represent a class of anti‐oestrogens that block the synthesis of oestrogen. Two types of AIs are available: irreversible steroidal inhibitors, for example, exemestane, which form a permanent bond with the aromatase enzyme complex; and non‐steroidal inhibitors, for example, anastrozole and letrozole, which inhibit the enzyme by reversible competition. AIs fall into three generations of evolution. First‐generation agents, namely, aminoglutethimide and testolactone, have been used extensively in the treatment of oestrogen receptor–positive postmenopausal breast cancer (Gershanovich 1998). The second generation includes formestane and fadrazole, and the third generation consists of anastrazole, letrozole, eximestane and vorozole. Letrozole and anastrazole were the first AIs used commonly in clinical practice. Both suppress up to 97% to 99% of aromatase activity (Goss 2001) with subsequent substantial reductions in circulating levels of oestrogen (Buzdar 2002). Both letrozole and anastrazole are rapidly absorbed after oral administration, are cleared by the liver and have a mean half‐life of approximately 45 hours. Subsequently, their clinical effects are expected to diminish substantially after 10 days, on average.

In premenopausal women, AIs have been reported to be well tolerated and to reduce the symptoms and the size of uterine fibroids (Gurates 2008; Hilário 2009). The most commonly reported adverse effects are hypoestrogenic symptoms, such as hot flushes and osteoporosis (bone mineral density loss). Less commonly reported effects include headache, back pain, leg cramps and joint disorders (Felson 2005). However, these effects are usually mild. More frequent adverse events have been reported after prolonged daily use in older women compared with short‐term use in younger women. The FDA has reported other adverse events associated with dietary supplements illegally containing AIs, such as a decreased rate of bone maturation and growth, decreased sperm production, infertility, aggressive behaviour, adrenal insufficiency, kidney failure and liver dysfunction (FDA 2010).

The first wide‐scale clinical use of the aromatase inhibition concept was seen in the treatment of metastatic breast cancer, and the 'third‐generation' AIs have now become standard adjuvant endocrine treatment for oestrogen receptor–positive postmenopausal breast cancer (Howell 2005). Investigations undertaken to look for other applications are ongoing. A systematic review concluded that AIs might be effective in treating endometriosis‐related chronic pelvic pain in both reproductive‐age and postmenopausal women (Pavone 2012). It has been suggested that AIs may have a role in ovulation induction in anovulatory women (Pavone 2013), as well as in subfertile women with polycystic ovary syndrome (Franik 2012; Misso 2012).

How the intervention might work

Uterine fibroids are sex steroid hormone–dependent, benign tumours that develop during the reproductive period and are usually suppressed with menopause. Besides the ovary, leiomyomal tissue itself is a source of oestrogen. Oestrogen secreted by leiomyomal tissue may reach a sufficient concentration within the local compartment to support its own growth, allowing independence from ovarian oestrogen (Bulun 2005). Moreover, both aromatase and 17b‐hydroxysteroid dehydrogenase (17b‐HSD) type I are overexpressed in uterine fibroids compared with myometrium. Sumitani 2000 indicates that leiomyoma cells convert circulating androstenedione into estrone (via aromatase) and then into the active form of oestrogen, estradiol (via 17b‐HSD type I).

Aromatase seems to be the key enzyme in this process in that inhibition of aromatase activity blocks proliferation. Aromatase activity is significantly stimulated by a cyclic adenosine monophosphate (cAMP) analogue, prostaglandin E2 (PGE2), or by a combination of a glucocorticoid and a cytokine (interleukin (IL)‐1) (Bulun 1994; Han 2008; Shozu 2002; Sumitani 2000). Aromatase mRNA has been detected in more than 90% of fibroids (Bulun 1994) but was undetectable in myometrial tissues from disease‐free uteri. Ishikawa et al found that leiomyoma tissues from African American women contained the highest levels of aromatase expression, which may result in elevated tissue concentrations of oestrogen and may account for their higher prevalence and earlier incidence (Ishikawa 2009).

On the basis of the hypothesis discussed previously, AIs are used as treatment for uterine fibroids. Theoretically, AIs provide several advantages over GnRH agonists because the former drugs directly inhibit oestrogen synthesis in the fibroids and ovary, as opposed to the indirect inhibition induced by GnRH agonists. Serum oestrogen levels begin to decrease as early as the first day of AI intake; therefore, a rapid effect is likely. Also, because AIs create a condition of complete suppression of oestrogen in the fibroids but partial suppression in the ovary, the risk of systemic side effects is relatively low. Gurates et al conducted a prospective clinical trial that enrolled 16 premenopausal women with symptomatic uterine fibroids. Investigators found that letrozole significantly decreased fibroid size and promptly relieved heavy menstrual bleeding without changing bone mineral density (Gurates 2008). Hilário et al reported that anastrozole was effective in reducing the volume of fibroids in a population of 20 participants, thereby controlling symptoms without changing serum follicle‐stimulating hormone or estradiol levels (Hilário 2009). Varelas et al found similar results with anastrozole (Varelas 2007) in 45 participants. Additionally, the value of in situ oestrogen inhibition by AIs becomes more pronounced in postmenopausal women because GnRH agonists are ineffective in this age group. Kaunitz et al reported that for an obese postmenopausal woman with persistent uterine bleeding, use of anastrozole was associated with reduction in fibroid size, endometrial thinning and cessation of bleeding (Kaunitz 2007).

Why it is important to do this review

Traditionally, uterine fibroids have been managed surgically. An increasing demand for medical therapy can eliminate the need for surgery, thereby preserving the uterus and future fertility. A comfortable and non‐invasive approach used to reduce fibroids may be preferable for most women and could reduce the burden on healthcare systems. Several clinical studies have examined AIs as treatment for uterine fibroids (Gurates 2008; Hilário 2009; Kaunitz 2007; Varelas 2007). However, weakness in study design has been a common problem, and systematic evaluation of the risks and benefits of AIs is required.