Treatments Targeting Neuroendocrine Dysfunction in Polycystic Ovary Syndrome (PCOS)

Akanksha Garg; Bijal Patel; Ali Abbara; Waljit S. Dhillo


Clin Endocrinol. 2022;97(2):156-164. 

In This Article

Abstract and Introduction


Polycystic ovary syndrome (PCOS) is the most common endocrine disorder in women of reproductive age and is the leading cause of anovulatory subfertility. Increased gonadotrophin releasing hormone (GnRH) pulsatility in the hypothalamus results in preferential luteinizing hormone (LH) secretion from the pituitary gland, leading to ovarian hyperandrogenism and oligo/anovulation. The resultant hyperandrogenism reduces negative feedback from sex steroids such as oestradiol and progesterone to the hypothalamus, and thus perpetuates the increase in GnRH pulsatility. GnRH neurons do not have receptors for oestrogen, progesterone, or androgens, and thus the disrupted feedback is hypothesized to occur via upstream neurons. Likely candidates for these upstream regulators of GnRH neuronal pulsatility are Kisspeptin, Neurokinin B (NKB), and Dynorphin neurons (termed KNDy neurons). Growing insight into the neuroendocrine dysfunction underpinning the heightened GnRH pulsatility seen in PCOS has led to research on the use of pharmaceutical agents that specifically target the activity of these KNDy neurons to attenuate symptoms of PCOS. This review aims to highlight the neuroendocrine abnormalities that lead to increased GnRH pulsatility in PCOS, and outline data on recent therapeutic advancements that could potentially be used to treat PCOS. Emerging evidence has investigated the use of neurokinin 3 receptor (NK3R) antagonists as a method of reducing GnRH pulsatility and alleviating features of PCOS such as hyperandrogenism. We also consider other potential mechanisms by which increased GnRH pulsatility is controlled, which could form the basis of future avenues of research.


Polycystic ovary syndrome (PCOS) is the most common endocrine disorder in females of reproductive age and the leading cause of anovulatory subfertility affecting 3%–13% of women.[1,2] Increased gonadotrophin releasing hormone (GnRH) pulsatility is implicated as a fundamental abnormality in the pathophysiology of PCOS.[3] Increased hypothalamic GnRH pulsatility results in preferential luteinizing hormone (LH) secretion from the pituitary gland, which in turn promotes ovarian hyperandrogenism and ovulatory dysfunction.[3,4] Although several diagnostic criteria exist to characterize PCOS, cardinal features include hyperandrogenism, polycystic ovarian morphology on ultrasound, and menstrual disturbance.[1] Women with PCOS additionally experience varying degrees of metabolic, endocrine, and psychological disturbances, and have an increased risk of pregnancy-related complications including gestational diabetes, preterm birth, and pregnancy-induced hypertension.[4] Current pharmacological treatments for PCOS aim to address the most pertinent presenting clinical feature for each woman, such as hirsutism or subfertility, however, most treatments do not aim to resolve the key pathophysiological abnormality that underlies PCOS. In recent times, novel therapeutic agents are under investigation that aims to specifically target the increased GnRH pulsatility believed to underpin the endocrine abnormalities observed in PCOS. This review highlights the neuroendocrine abnormalities that contribute to the increased GnRH pulsatility in PCOS and summarizes data from pharmacological agents that target this central dysfunction.

Overview of Neuroendocrine Regulation of GnRH Neurons

GnRH is a decapeptide that is a key regulator of the hypothalamic–pituitary–gonadal (HPG) axis (Figure 1). In humans, it is released from the infundibular nucleus of the hypothalamus in a pulsatile manner, and potently stimulates release of LH and follicle stimulating hormone (FSH) from the anterior pituitary gland. FSH promotes ovarian follicle development, whereas LH plays a major role in ovulation and theca cell androgen production. Higher frequency GnRH pulsatility favours LH release from the pituitary gland, whereas reduced GnRH pulsatility promotes FSH-predominant secretion. GnRH release is regulated by a complex neuroendocrine network and responds to negative feedback from sex steroids. Notably, GnRH neurons lack the receptors for androgens, progesterone, and oestrogens to directly mediate this negative feedback.[3] A growing body of evidence suggests that neurons in the infundibular nucleus of the hypothalamus (arcuate nucleus in animals), which co-express Kisspeptin, Neurokinin B (NKB), and Dynorphin (termed KNDy neurons), are integral regulators of the activity of GnRH neurons (Figure 1). Kisspeptin and NKB have a stimulatory action on kisspeptin release, whereas dynorphin has an inhibitory effect.[5] There is redundancy in the quantity of functional KNDy neurons required for reproductive health, with only 20% of neurons being sufficient to maintain folliculogenesis in female mice.[6]

Figure 1.

Legend: Neuroendocrine abnormalities in polycystic ovary syndrome (PCOS). In healthy women, pulsatile gonadotrophin releasing hormone (GnRH) secretion from the hypothalamus stimulates luteinizing hormone (LH) and follicle stimulating hormone (FSH) release from the anterior pituitary gland. This results in ovarian sex-steroid production (oestrogen, progesterone, and androgens), which in healthy women negatively feedback on the hypothalamus and pituitary gland to decrease gonadotrophin secretion (left panel). In PCOS, increased GnRH pulsatility results in increased LH release, which in turn stimulates increased ovarian androgen secretion. Androgens impair the normal negative feedback from sex-steroids on the hypothalamus (Right panel). At the hypothalamic level (centre), Kisspeptin, Neurokinin B and Dynorphin co-expressing neurons (KNDy neurons) secrete kisspeptin to regulate GnRH release. These neuropeptides interact in an auto/paracrine manner to regulate the pulsatile release of kisspeptin from KNDy neurons, with dynorphin having an inhibitory action, and Neurokinin B having a stimulatory action. Kisspeptin regulates GnRH release. At the pituitary gland, high-frequency GnRH pulses favour LH release, whereas low frequency GnRH pulses favour FSH secretion from pituitary gonadotrophs. Sex-steroids feedback the hypothalamo-pituitary-gonadal axis via oestrogen, progesterone and androgen receptors (ER, PR, and AR) on KNDy neurons. Anti-Müllerian hormone (AMH) and γ-aminobutyric acid (GABA) are putative regulators of hypothalamic GnRH release

A second population of kisspeptin neurons in the pre-optic area of the hypothalamus mediates the mid-cycle LH surge, which occurs in response to a switch from negative to positive feedback from oestrogen during the late follicular phase. Nonsynaptic apposition of kisspeptin neurons at GnRH neuronal dendrons is sufficient for GnRH pulsatility, but kisspeptin action at both GnRH neuronal cell body and dendron is needed for the midcycle LH surge.[7]

Abnormality in GnRH Pulsatility in PCOS

LH (a measurable surrogate for GnRH) pulse frequency is increased by approximately 40% in women with PCOS in comparison to healthy controls (22–24 vs. 16 pulses per 24 h), leading to high LH levels and a relative FSH deficiency.[2,8] LH levels are greater than the 95th percentile in 40%–60% of women with PCOS, particularly if associated with menstrual disturbance.[2] The increased LH levels stimulate ovarian androgen synthesis, and the relative FSH deficiency contributes to follicular arrest, polycystic ovarian morphology, and oligo/anovulation.[9] Indeed, women with oligo/anovulatory PCOS have increased LH:FSH ratios compared to ovulatory women with PCOS (2.0 vs. 1.4).[10] Furthermore, raised LH levels contribute to increased ovarian theca cell androgen production, with increased total and free testosterone levels in women with PCOS compared with controls.[4] Androgen excess results in impaired negative feedback by sex-steroids, such that GnRH pulsatility is unrestrained, maintaining raised LH levels, and subsequently androgen levels, in a vicious cycle. Notably, the action of androgens to induce a PCOS-like phenotype appears to be mediated centrally rather than peripherally.[7] Although not often used, sustained administration of GnRH agonists downregulate the GnRH receptor and attenuate the hyperandrogenism associated with PCOS. Thus, accumulating evidence suggests that increased GnRH pulsatility underpins the endocrine abnormalities observed in PCOS. This has generated interest in identifying pathways involved in the dysfunctional GnRH release and identifying potential therapeutic targets to ameliorate this.

Sex Steroids and GnRH Dignalling in PCOS

Between 50% and 80% of women with PCOS have elevated androgens,[11] including testosterone, androstenedione, and dehydroepiandrosterone sulfate (DHEAS).[12] Women with PCOS have reduced sensitivity of GnRH pulsatility to the normal inhibition by oestrogen and progesterone.[13] GnRH neurons lack the receptors to mediate negative feedback from sex-steroids including the androgen receptor (AR), ER α, and the progesterone receptor[14] and thus this sex-steroid mediated negative feedback is thought to occur upstream of GnRH neurons ie KNDy neurons[4] (Figure 1).

Global AR knockout mice models demonstrate abnormal function of the HPG axis, with a subsequent compromise of ovarian folliculogenesis.[14] Many PCOS-like animal models are generated via exposure to androgens either in prenatal, neonatal, or pubertal life.[15] Prenatal dihydrotestosterone (DHT) exposure leads to an increased number of Kisspeptin/NKB reactive neurons and elevated LH levels in keeping with the hypothesis of increased KNDy neuronal activity in PCOS.[16] Prenatal androgen-treated (PNA) mouse models of PCOS have increased AR gene expression on KNDy neurons, reduced progesterone receptor, and dynorphin expression.[17] Prenatal AMH exposure induces a PCOS-like animal model, with disrupted oestrus cyclicity, reduced fertility, and levels of reproductive hormones consistent with increased GnRH pulsatility (increased LH and reduced FSH levels), however, this effect was not present in female mice with kisspeptin-specific AR knockout (KARKO).[18] This indicates that the effects of AMH are mediated via the androgen receptor in kisspeptin expressing neurons.

Androgen blockade has been investigated as a potential therapeutic option in PCOS. Flutamide treatment restores oestrous cyclicity and reduces the number of cyst-like follicles in a PCOS-like mouse model.[19] Low-dose flutamide treatment over three years in women with PCOS increased ovulatory cycles and normalized LH levels.[20] However, data is conflicting with another study of twice daily flutamide administration in women with PCOS not demonstrating a change in LH pulsatility.[21] Although flutamide reduced testosterone levels in healthy women, it did not do so in women with PCOS,[21] however, it did restore the ability of oestradiol and progesterone to suppress LH secretion.[21]

Kisspeptin in PCOS

Kisspeptins are peptides encoded by the KISS1 gene and have been implicated as crucial upstream moderators of GnRH release.[22] Kisspeptin acts via the "kisspeptin receptor" (KISS1R), a G-protein coupled receptor present on GnRH neurons in the preoptic area of the hypothalamus. Loss of function of the kisspeptin receptor in both humans and mouse models results in a failure of GnRH secretion and in hypogonadotrophic hypogonadism.[23,24] Various forms of kisspeptin (KP-54, KP-14, KP-13, and KP-10) have been characterized based on their amino-acid length, with KP54 and KP10 being the most widely studied.[22] Kisspeptin neurons in the infundibular nucleus are responsible for regulating GnRH pulsatility, whilst the population of kisspeptin neurons in the preoptic area mediate the mid-cycle surge.[9] Kisspeptin expressing neurons also express receptors for androgens, oestradiol and progesterone that have an important role in mediating negative feedback from sex steroids.[25] Additionally, kisspeptin signalling within the ovaries is implicated in PCOS, with abnormal expression of both kisspeptin and its receptor demonstrated in granulosa cells from women with PCOS.[26,27]

Recent meta-analyses report higher circulating levels of kisspeptin in women with PCOS compared with controls.[28] However, higher circulating kisspeptin levels were mainly observed in lean PCOS, with no significant difference observed in obese women with PCOS.[28] In this meta-analysis, serum kisspeptin levels were positively associated with anti-Müllerian hormone (AMH), testosterone and dehydroepiandrosterone sulphate (DHEAS),[28] pro-inflammatory cytokines, and vasculo-endothelial growth factors,[29,30] but surprisingly not LH. Interestingly, circulating kisspeptin levels were pulsatile and the frequency of these pulses was increased in women with PCOS.[28,31] Furthermore, kisspeptin pulses were coupled to LH pulses, consistent with a central source of circulating kisspeptin levels,[28,31] but coupling was lost in oligo/amenorrhoeic women with PCOS.[31]

Pre-clinical PCOS-like animal models demonstrate considerable heterogeneity regarding alterations in kisspeptin neuron anatomy and function, due to the method by which the model was generated.[15] Most commonly, androgen exposure is used to generate a PCOS-like model, however prenatal (but not postnatal) exposure to dihydrotestosterone (DHT) results in increased arcuate kisspeptin expression in rats.[15] In addition to AMH-induced models, mifepristone-induced PCOS-like models also have increased hypothalamic kisspeptin expression.[15]

Kisspeptin as a Therapeutic Target in PCOS

Hypothalamic kisspeptin signalling is essential to the occurrence of the mid-cycle LH surge and can be used to trigger ovulation.[22] Indeed, a single bolus of kisspeptin incites an LH surge sufficient to induce oocyte maturation in sub-fertile women with PCOS undergoing in vitro fertilization (IVF) treatment, without causing the dangerous complication "Ovarian Hyperstimulation Syndrome" (OHSS).[32–35] Oocyte maturation induced using kisspeptin was associated with higher expression of FSH receptor and steroidogenic enzymes (CYP19A1 and STAR) in ovarian granulosa cells (GCs) than other hormonal agents, alluding to a more physiological degree of stimulation.[36,37]

Recently, the KISS1R agonist, MVT-602, which has greater stability, solubility and potency than the native peptide, KP54, was shown to induce a longer duration of LH rise in women with PCOS compared with KP54.[38] The LH and FSH rises were similar in women with PCOS as in women without PCOS.[38] Notably, the longer duration of gonadotrophin-rise after MVT-602, which is more akin to that seen during the mid-cycle LH surge, would make this agent suitable for further evaluation as a trigger of oocyte maturation during in vitro fertilization (IVF) treatment.[33,35,38]

In another study of kisspeptin administration in PCOS, the LH response to KP10 correlated to the pre-treatment oestradiol level, and was LH-predominant with little FSH rise.[39] However, an FSH rise could be detected if pre-treated for 1 week with an NK3 Receptor antagonist (NK3Ra) (which slows GnRH pulsatility).[39] Indeed, this is concordant with the FSH response observed in women with hypothalamic amenorrhoea (who have reduced GnRH pulsatility), where the FSH response is also heightened.[38]

Certainly, kisspeptin is well-recognized to induce a greater LH than FSH rise in most physiological circumstances, and thus there is a risk that kisspeptin may exacerbate the relative FSH deficiency already present in women with PCOS, thus potentially making it less suitable for use as a sole ovulation induction agent in PCOS.

Consequently, the ability of kisspeptin to induce a sufficient FSH rise to restore ovulation was assessed by Romero-Ruiz and colleagues.[40] They investigated the efficacy of chronic KP54 administration regimens to induce gonadotrophin release and ovulation in both preclinical PCOS-like rodent models and in anovulatory women with PCOS.[40] In preclinical models, the first dose of KP54 induced a greater FSH response in all PCOS-like models than in controls, but this was particularly augmented in neonatal androgenised and postnatal androgenised rats, whereas the LH rise was greater in prenatal androgenised rats.[40] However, tachyphylaxis occurred after chronic treatment with some of the dosing regimens used.[22] Nevertheless, KP54 administered for 11 days rescued ovulation in neonatal androgenised rats but not in postnatally androgenised rats (prenatally androgenised rats already had already maintained ovulatory status in this study).[40]

Twelve oligo/amenorrhoeic women with PCOS received twice daily subcutaneous doses of kisspeptin-54 for up to 3 weeks at doses between 3.2 and 12.8 nmol/kg, overall resulting in a small rise in LH from before to after KP54 (10.8–13.4 IU/L), but not in FSH or inhibin B.[40] Five of seven women treated twice daily with a fixed dose of 9.6 nmol/kg had an LH response and two (both without hyperandrogenism) of these ovulated.[40] These data highlight the variability in endocrine profile in women with PCOS, with pre-treatment LH levels in this study varying from 2.2 to 23.5 IU/L, and thus the response to kisspeptin can also be expected to vary between individuals, with the potential for both over- (causing tachyphylaxis) and underdosing with kisspeptin, with a differential gonadotrophin response depending on GnRH pulsatility at the time of administration. Overall, it does appear feasible to induce follicular growth using chronic administration of kisspeptin in select patients with PCOS. Further larger scale studies with the ability to individualize treatment protocols may help unravel the complexity and variability in response to kisspeptin in women with PCOS to establish its potential as an ovulation induction agent.

As the response to kisspeptin is usually greater on LH than FSH, it seems feasible to use a kisspeptin antagonist to reduce LH levels, to thus ameliorate the relative FSH deficiency in women with PCOS and high LH levels. Peptide-234 is a kisspeptin antagonist that has been shown to inhibit GnRH and LH levels in animal models.[41] However, its effects were not reproducible in all species,[42] and kisspeptin antagonists have yet to be administered to humans. Furthermore, it would be necessary to titrate the dose to avoid excessive suppression and castrate LH levels.

Neurokinin-B Signalling

Neurokinins belong to the tachykinin family of peptides and include Neurokinin A (NKA) and Neurokinin B (NKB), encoded by the TAC1 and TAC3 genes, respectively. They act through Gq-coupled receptors (NK1R, NK2R, and NK3R), encoded by TACR1, TACR2, and TACR3, respectively.[43] NKB preferentially binds to NK3R,[44] which predominates in humans, and is believed to be a stimulatory input for GnRH pulsatility. Senktide, an NKB receptor agonist, significantly increased pulsatile GnRH release when administered to human GnRH neurons in vitro.[45] Furthermore, NKB and kisspeptin gene expression are increased in some PCOS-like animal models,[46] suggesting that overactivity of KNDy neurons is responsibility for the increase GnRH pulsatility and endocrine abnormalities observed in PCOS. Although inactivating variants in human TAC3 (NKB) and TACR3 (NK3R) genes have been shown to result in hypogonadotrophic hypogonadism,[44] women with functionally null TAC3 can still conceive and deliver successfully.[44] Impairment of GnRH function is less stark than after loss of kisspeptin, with knockout mice lacking tachykinins (including NKB) still able to have LH pulses.[47] Furthermore, administration of NKB in humans did not impact LH, FSH, or sex steroids at maximally tolerated doses.[48]

Nevertheless, this milder action could enable a more tempered inhibition of GnRH pulsatility with NK3R antagonists, to normalize the increased GnRH pulsatility without reducing it excessively. Thus, there has been great interest in using NK3R antagonists to treat the core endocrine abnormalities observed in PCOS.

Targeting NKB Signalling in PCOS

The first randomized, double-blind placebo-controlled phase 2 study trialled MLE4901 (formerly AZD4901), a specific neurokinin 3 receptor antagonist, in women with PCOS.[49] Women treated with MLE4901 (80 mg/day) had a significant decrease in the area under the curve of LH levels (by 52%, 95% confidence interval [CI] 30%–67%), LH pulses per 8 h (by 3.6 pulses/h, 95% CI 2.0–5.1), and total testosterone levels (by 29%, 95% CI 14%–41%).[49] Interestingly, NK3R antagonism (MLE4901) was recently studied over 12 weeks in PCOS-like mouse model displaying metabolic traits including obesity, hyperglycaemia, and raised triglycerides.[50] There was an improvement in the metabolic traits with a reduction in body weight and adiposity following treatment, but without resolution of the reproductive phenotypes.[50] This may in part reflect the postnatal androgenised model used, which does not display overactivity of arcuate KNDy neurons,[50] and thus may not exclude an effect on reproductive phenotype in other PCOS-like mouse models.

Nevertheless, further development of MLE4901 in humans was discontinued due to concerns over hepatic dysfunction. However, this appears to be a drug-specific effect rather than a class effect, which is of significant relevance for women with PCOS who may commonly be affected by metabolic-associated fatty liver disease (MAFLD). Phase I studies of Fezolinetant (previously known as ESN364), another potent NK3R antagonist, demonstrated a reassuring safety profile in healthy males and females.[51] Fezolinetant resulted in a dose-dependent reduction in LH levels at 4 and 6 h after both 60 and 80 mg doses, with recovery of LH to baseline levels by 16 h.[51] Interestingly, there was no effect on FSH levels, consistent with its action via GnRH pulsatility, however oestrogen levels dropped by >30% between 6 and 24 h post-dosing.[51] Healthy women treated with Fezolinetant had a significantly delayed midcycle gonadotrophin surge and an increased menstrual cycle length (from 27.3–28.8 days to 37.7 days).[51] Interestingly, in female monkeys, Fezolinetant could completely prevent ovulation raising the possibility for use as a contraceptive agent if this affect can be achieved reliably in humans.[52]

Fezolinetant is already under phase III development for the treatment of vasomotor symptoms in postmenopausal women, but it could also be a promising candidate to reduce hyperandrogenism in women with PCOS.[53] A Phase II, randomized, double-blind, multicentre study evaluated the effect of Fezolinetant (60 mg or 180 mg) in 73 women with PCOS (as per the Rotterdam criteria) over 12 weeks.[53] Serum testosterone levels decreased by 14%–33% after 12 weeks of treatment in a dose-dependent manner.[53] LH levels were halved over the 12 week period, but there was no change in FSH, progesterone, or oestradiol levels.[53] Fezolinetant did not improve PCOS lifestyle scores, endometrial thickness, follicle development, or menstrual cycle irregularity,[53] however the study was not designed to assess some of these outcomes. This may again in part reflect the heterogeneity in gonadotrophin levels at baseline in women with PCOS, and the potential for excessive endocrine suppression also preventing ovulation in some cases. Thus, treatment with NK3R antagonists for shorter durations, with monitoring of gonadotrophin levels might be more desirable for inducing ovulation. This approach could enable improvement in the relative FSH deficiency in the early follicular phase but avoid any negative impact on the midcycle LH surge by stopping treatment soon after folliculogenesis has commenced.

The NKB/NK3 signalling pathway is present not only in the hypothalamus, but also in ovarian granulosa cells, suggesting scope for direct ovarian action. Indeed, NKB agonists directly induce ovarian oestrogen production, which can be impaired in PCOS.[54] In women with PCOS, mural granulosa cells and cumulus cells demonstrate decreased mRNA expression of TAC3, TAC3R, and KISS1 than in healthy women.[26] However, the predominant effect of Fezolinetant is believed to occur via its action at the hypothalamic level. Overall, these data support the hypothesis that NK3R antagonism can reduce GnRH and LH release and ameliorate hyperandrogenism, however, data on other outcomes such as ovulation is less clear.


Dynorphin is expressed by KNDy neurons and binds to kappa opioid receptors on KNDy neurons acting in an auto/paracrine manner to inhibit the release of kisspeptin onto GnRH neurons [9] (Figure 1). It has been hypothesized that the increase in GnRH pulsatility in PCOS may be a result of reduced hypothalamic opioid inhibitory activity, although this is not reflected by peripheral circulating β-endorphin levels which were increased.[55] Therefore, one could expect an opioid agonist with a central action to be able to ameliorate GnRH pulsatility and have therapeutic potential in PCOS.

Interestingly, naltrexone, an opioid receptor antagonist/weak partial agonist, has been suggested to be beneficial in PCOS. Naltrexone resulted in weight loss, and was associated with a reduction in LH levels and LH:FSH ratio in women with clomiphene resistant PCOS,[56] and normalized the pituitary response to GnRH in women with PCOS.[55] Additionally, co-treatment with naltrexone and pulsatile GnRH in obese women with PCOS improved ovulation induction outcomes compared to GnRH alone.[57] However, this effect may in part be mediated via weight loss, attenuation of insulin resistance, with an associated reduction in androgen levels, rather than via a direct action on hypothalamic GnRH pulsatility, although the precise mechanism remains unclear. Insulin administration stimulates GnRH pulsatility reflected by LH levels in healthy women, but not in women with PCOS,[58] consistent with its predominant action after PCOS has been established being on ovarian androgen production rather than on hypothalamic GnRH release.

Recently, administration of peripherally restricted (but which can still access the hypothalamus via the median eminence) kappa receptor agonists (PRKA) to a murine PCOS-like model resulted in restrained KNDy activity leading to reduced LH and testosterone levels, with normalization of ovarian cyclicity.[59] Thus, it is possible that the opioid system can be targeted in future studies to normalize GnRH pulsatility in women with PCOS.

Gamma Aminobutyric Acid (GABA)

Interestingly, prenatal animal androgen exposure increases GABA innervation to GnRH neurons.[60] Although GABA is typically an inhibitory neurotransmitter, GnRH neurons respond to GABA with excitation due to higher intracellular chloride levels,[61] with DHT increasing GABA mediated stimulation of GnRH neurons. Peripheral circulating GABA levels were found to be reduced,[62] whereas central CSF GABA levels were increased,[63] in women with PCOS. Accordingly, medications that increase GABA, such as anti-epileptic drugs, have been associated with an increased prevalence of PCOS.[64]

Anti-Müllerian Hormone (AMH)

AMH is a member of the transforming growth factor β (TGF-β) superfamily[65] secreted by ovarian granulosa cells (GCs) and regulates follicle selection. AMH is predominantly produced by small and pre-antral antral follicles[65] and binds to AMH type II receptors (AMHR2), present in the ovaries and hypothalamic GnRH neurons. Women with PCOS have 2–4 fold higher serum AMH levels and increased AMHR2 compared with controls,[66] associated with slow follicular growth, follicular arrest, and abnormal GC functioning.[67] Androgens increase GC FSH receptors (FSHR), and increased FSH action stimulates AMH expression from growing follicles in PCOS.[65,67] Higher AMH levels were observed in anovulatory and hyper-androgenic women with PCOS compared with normo-androgenic women.[67,68] In women, AMH levels better predicted oligo/amenorrhoea than AFC.[69] Additionally, AMH was positively associated with LH-predominant gonadotrophin secretion and the number of features of PCOS.[69] AMH directly activates GnRH neuronal activity in a dose-dependent manner in mice, and prenatal exposure to AMH induces a PCOS-like phenotype.[70] A recent systematic review identified increased AMH levels in the umbilical cord blood of new-borns of PCOS mothers.[71]

NK3R antagonism in women with PCOS over 12-weeks resulted in a reduction in AMH levels.[53] Notionally, a reduction in AMH levels via manipulation of KNDy signalling pathways could be beneficial by reducing GnRH pulsatility. Hypothetically, AMH receptor antagonists could be of benefit in PCOS, however, no data is available to date. Work delineating the signalling pathways of AMHR2 is ongoing, which could form the basis of AMH antagonists in future.[72]