Abstract
Keywords
A common thread through many aspects of women's reproductive physiology in health and disease focuses on establishing the status of the ovarian follicle reserve and its reproductive potential. Developing sensitive biomarkers of ovarian reserve and oocyte quality would be a benefit in conditions such as polycystic ovarian syndrome (PCOS), premature ovarian failure, IVF, impending menopause and other conditions where fertility is a concern. In recent years, there has been an increasing interest in anti-Müllerian hormone (AMH, otherwise known as Müllerian-inhibiting substance), a member of the TGF-β family, as a suitable biomarker. While AMH appears to show more promise than other markers, it is unclear if it has an improved clinical utility compared with ovarian follicle count and whether there are advantages in combination with other markers. This review focuses on recent findings (2006–2007). For a review of literature up to and including 2006, the reader is referred to [1–3].
Anti-Müllerian hormone
Anti-Müllerian hormone was originally described as a gonadal factor that caused the regression of the Müllerian ducts, precursors to the female reproductive tract in early mammalian embryos [3]. Subsequently, AMH was identified in the ovary as a strong suppressor of the initiation and growth of human primordial ovarian follicles in vivo and in vitro [3–6]. These studies suggest that AMH plays an important role in regulating the rate of folliculogenesis during adult life.
Anti-Müllerian hormone appears useful as a biomarker of ovarian reserve for several reasons. It is an exclusive ovarian product (a product of the granulosa cell) whose serum levels decrease to the level of assay detection once ovarian follicle numbers are depleted. AMH provides a qualitative measure of ovarian antral number and this measure is seemingly uninfluenced by changing patterns of known reproductive hormones (follicle-stimulating hormone [FSH], estradiol or inhibin B) [7]. One consequence is that it shows little, if any, change in circulating levels during the menstrual cycle [8–10], thus, in contrast to other reproductive hormones, it is less variable and hence more reliable as a diagnostic marker.
A number of studies have confirmed that serum AMH provides a good indicator of antral follicle numbers [1–3], although with some caveats. Originally, it was thought that follicle production of AMH was primarily confined to the early–intermediate antral follicle stages. However, studies by Frank's group using quantitative immunocytochemistry have demonstrated that primordial follicles exhibit staining intensity 30–40% of later stages [11], although this conclusion has been questioned [12].
All reports agree that the follicle content and expression of AMH decrease dramatically once the follicle has exceeded 2–5 mm in diameter. This fall is particularly precipitous. In one study, follicular fluid AMH levels fell from 750 ng/ml (2–5 mm) to less than 2 ng/ml with follicles larger than 10 mm [13]. A similar marked fall was observed in granulosa cell production and secretion of AMH in vitro from follicles larger than 10 mm [14]. What is unclear is the basis for this decrease. In vitro production of AMH by granulosa cells from PCOS ovaries was inhibited by FSH in contrast to granulosa cells cultures of normal ovaries [14], suggesting an increased sensitivity to FSH in PCOS. Others have conjectured that under FSH/luteinizing hormone (LH) stimulation, increasing steroid levels produced by the developing preovulatory follicle may be inhibitory [15], by suppressing AMH signaling pathways. In addition, AMH inhibits FSH-stimulated estradiol production by granulosa cells in culture [1–3] and thus would reduce follicle sensitivity to FSH. Nonetheless, no studies to date have provided a satisfactory basis to explain this dramatic decline in AMH production.
IVF treatment
The role of AMH and its ability to predict ovarian response in women undergoing controlled ovarian stimulation is now more clearly defined [1,16–19], with some caveats [20]. In a recent study, La Marca and colleagues showed that AMH measurement on any day of the menstrual cycle strongly predicts ovarian response (in terms of retrieved oocytes) in assisted reproductive technology [21]. This conclusion was based on sensitivity–specificity characteristics of 80 and 93%, respectively, in predicting poor ovarian responses and was similar to other reports [22,23]. Furthermore, its high cycle-to-cycle reproducibility makes the test more flexible as a diagnostic tool in IVF procedures [19].
Other potential markers are inferior to AMH in predicting oocyte yield [20]. AMH levels assessed at the time of human chorionic gonadotropin administration correlated significantly with a greater number of six-cell embryos and a better embryo morphology score, although its diagnostic value has not been assessed [24].
Other studies have explored the effects of lifestyle factors in normal ovulatory women on the relationship between AMH and antral follicle count [25]. Smoking, BMI, alcohol consumption, ethnic origin, chronological age, age at menarche, years since menarche and gravidity were examined. AMH and follicle count correlated with age and number of years since menache but not with other factors. Other studies showed positive correlations with serum testosterone once adjusted for age [26].
Polycystic ovarian syndrome
A number of studies have shown that serum AMH is a useful marker in the diagnosis of women with PCOS [1,27]. PCOS is characterized by ovulatory disturbances, including anovulation, hyperandrogenism and polycystic ovaries, with increased numbers of follicles in the antral follicle range [28]. The increase in antral follicle number is accompanied by an increase in both proliferation and AMH production by granulosa cells in vitro [14], a decrease in atresia as assessed from cultures of small ovarian biopsies [5,6] and an elevation in serum AMH.
The failure of folliculogenesis to proceed in PCOS is related to a decreased sensitivity to FSH as exogenous FSH treatment is effective in inducing follicular development and ovulation, although there is an increased risk of ovarian hyperstimulation. AMH levels are elevated sixfold in patients with ovarian hyperstimulation, suggesting AMH would be a useful biomarker for identifying at-risk patients [29].
AMH levels, either expressed per follicle [30] or per granulosa cell [14], are twice as high in women with anovulatory PCOS compared with healthy women or PCOS women with ovulatory cycles and thus the use of AMH as a diagnostic test has been suggested to be clinically useful [31]. The close relationship between the elevated follicular fluid levels of androgen and AMH levels in women with PCOS suggests that testosterone may be the stimulatory factor [26,31].
However, there are caveats to this conclusion. In a recent study, AMH levels were compared in women with PCOS with or without Type 1 diabetes mellitus [30]. Despite similarly elevated androgen levels, LH/FSH ratio and antral follicle count between groups, the AMH levels (expressed as serum AMH/follicle ratios) in the diabetic group were similar (1.9) to normal controls (2.9) and reduced compared with the PCOS group (6.9). The authors concluded that not all conditions of hyperandrogenic disorders result in elevated AMH levels.
Interestingly, several studies suggest that AMH itself may be a key factor in PCOS [32]. In vitro studies have shown that AMH inhibits FSH-induced aromatase activity and estradiol production by granulosa cells and thus reduces the follicle sensitivity to FSH. Studies exploring the effects of controlled FSH treatment of women with PCOS have shown that serum AMH levels fall during treatment [32]. It is postulated that this fall is associated with a decreased inhibition of aromatase activity resulting in increased estradiol production and continued progression of folliculogenesis through to ovulation. An alternative hypothesis would propose that the fall in serum AMH is primarily a result of a decrease in the number of small antral follicles as a result of FSH treatment and less caused by the effects of steroid inhibition.
Recent studies showed an increased ovarian volume and serum AMH in daughters of women with PCOS compared with age-matched controls [33,34]. These data provide evidence of an inherited trait related to increased follicular development in women with PCOS. Based on these observations, it is interesting to speculate that recruitment of follicles from ovarian stores is more rapid in PCOS-prone women, resulting in an increased follicle number. The resultant elevated AMH levels would inhibit follicle maturation, resulting in anovulation.
What are the advantages of AMH measurements compared with ovarian antral follicle count in the diagnosis of PCOS? In a study of PCOS patients and controls, the diagnostic power of serum AMH in predicting antral follicle count was determined at 92% specificity and 67% sensitivity [26]. These data indicated that AMH was useful as a surrogate test for ovarian follicle count. The advantages of AMH primarily relate to its practicability as a serum test in terms of patient acceptability, practicability and cost.
Serum AMH is being used to monitor ovarian response to weight loss in women with PCOS. Overweight women who showed an improved menstrual response with weight loss had low AMH levels (24 vs 38 pmol/l) [35]. These authors propose that pretreatment AMH is the best marker of responsiveness to lifestyle changes in women with PCOS.
Late-reproductive aged women
AMH has been used to monitor ovarian decline during the menopause transition [1–3,36]. However, recent studies have shown that AMH levels are 65% lower in late-reproductive aged women with obesity [37], although this decease is likely to be greater as the AMH levels in the obese women were below the sensitivity of the assay. A similar decrease has not been noted in women with PCOS.
Premature ovarian failure
In a preliminary study, AMH has been used as a marker of premature ovarian failure, in particular to monitor the presence of developing follicles [12,38]. Serum AMH levels varied markedly between subjects and were mostly below the normal range. A parallel change in serum AMH levels and number of small follicles in ovarian biopsies was noted.
Cancer
A further application of serum AMH as a marker of gonadal function has been used with women treated for Hodgkin's Lymphoma [39,40]. Overall, AMH levels were reduced in these patients despite some patients with FSH levels in the normal range. The authors suggest that AMH is an earlier and more sensitive marker of gonadal damage. In preliminary studies, AMH has been applied to ovarian cancers as a biomarker of granulosa cell tumors, with a sensitivity ranging between 76 and 93% [41]. In earlier studies, but worth mentioning here, AMH was also useful in the diagnosis and management of patients with a variety of intersex and gonadal abnormalities [42].
Conclusion & future perspective
To what extent is serum AMH a simpler and more practical method for detecting antral follicle count in monitoring disease states of ovarian folliculogenesis? At this point, its practicability appears to be its primary advantage as AMH provides a fairly acceptable measure of antral follicle count in a range of clinical conditions with high reproducibility. There are still factors other than age (some types of hyperandrogenism and obesity) that influence AMH levels that require clarification. Furthermore, there is a need for increased understanding of the role of AMH in follicle and oocyte quality. Development of new or re-examination of old markers in conjunction with AMH may provide the needed sensitivity and specificity.
Executive Summary
The literature related to the clinical application of serum anti-Mullerian hormone (AMH) as a diagnostic tool in monitoring ovarian function during assisted reproduction treatment and other reproductive processes was reviewed for the years 2006-2007.
AMH is superior to other markers in predicting oocyte yield in IVF based on its relatively good sensitivity–specificity characteristics and ease of assay.
AMH is useful in the diagnosis of polycystic ovary syndrome (PCOS) where the serum AMH follicle ratio is elevated, although the basis for this elevation is unclear.
AMH has been used to monitor ovarian response to hormonal hyperstimulation and weight loss in women with PCOS and to monitor ovarian decline during the menopause transition, premature ovarian failure and following cancer treatment.
Footnotes
The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
No writing assistance was utilized in the production of this manuscript.