Epidemiology, phenotype, and genetics of the polycystic ovary syndrome in adults

INTRODUCTION — Polycystic ovary syndrome (PCOS), a heterogeneous, complex genetic trait of unclear and likely multiple etiologies, is an important cause of cardiometabolic risks, ovulatory and menstrual irregularity, subfertility and infertility, and clinically evident hyperandrogenism. When fully expressed, the manifestations include androgen excess, ovulatory dysfunction, and polycystic ovaries. It is recognized as one of the most common endocrine/metabolic disorders of females. This syndrome was first described by Stein and Leventhal in 1935 [1], although the presence of sclerocystic ovaries had been recognized for at least 90 years prior to their report.

The definition, epidemiology, and pathogenesis (including genetics) of PCOS will be reviewed here. The clinical manifestations, diagnosis, and treatment of PCOS are discussed separately. (See "Clinical manifestations of polycystic ovary syndrome in adults" and "Diagnosis of polycystic ovary syndrome in adults" and "Treatment of polycystic ovary syndrome in adults".)

PHENOTYPES OF PCOS — The principal features of polycystic ovary syndrome (PCOS) include androgen excess, ovulatory dysfunction, and/or polycystic ovaries [2]. Androgen excess or hyperandrogenism can be found either biochemically (ie, hyperandrogenemia) or clinically (eg, hirsutism). There are several proposed diagnostic criteria for PCOS that are reviewed in detail separately. Of note, the National Institutes of Health (NIH) Evidence-based Methodology Workshop Panel on Polycystic Ovary Syndrome in 2012 suggested the following recommendations [3]: (1) renaming the disorder to more adequately reflect the complex metabolic, hypothalamic, pituitary, ovarian, and adrenal interactions that characterize the syndrome (and their reproductive implications, although no specific name was recommended); and (2) maintaining the broad, inclusionary diagnostic criteria of Rotterdam (which includes the "classic NIH" and the Androgen Excess [AE]-PCOS Society criteria), while specifically identifying each of the phenotypes in research and clinical initiatives

The 2023 International Evidence-based Guideline for the Assessment and Management of PCOS supports the use of the Rotterdam criteria [4]. (See "Diagnosis of polycystic ovary syndrome in adults", section on 'Diagnosis'.).

All diagnostic recommendations indicate that to make the diagnosis of PCOS, other causes of ovulatory dysfunction and hyperandrogenism must be excluded. (See "Diagnosis of polycystic ovary syndrome in adults", section on 'Other proposed criteria'.)

In general, four different phenotypes of PCOS have been identified. This classification is perhaps most relevant for clinical research studies:

●Phenotype A (also known as "full PCOS" or "classic PCOS") includes biochemical or clinical hyperandrogenism, oligoovulation, and polycystic ovarian morphology

●Phenotype B (also known as "classic PCOS") includes hyperandrogenism and oligoanovulation

●Phenotype C (also known as "ovulatory PCOS") includes hyperandrogenism and polycystic ovarian morphology

●Phenotype D (also known as "non-hyperandrogenic PCOS") includes oligoanovulation and polycystic ovarian morphology

Of note, the NIH criteria include only phenotypes A and B, the AE-PCOS Society criteria phenotypes A, B, and C, and the Rotterdam criteria all four phenotypes.

EPIDEMIOLOGY — Polycystic ovary syndrome (PCOS) is recognized as one of the most common endocrine/metabolic disorders in females. Its prevalence depends in part upon the diagnostic criteria used to define the disorder since each criteria includes a varying number of PCOS phenotypes.

General population — In a 2016 meta-analysis of 24 population studies performed in Europe, Australia, Asia, and the United States, the rates of PCOS (and 95% confidence interval) according to diagnostic criteria in unselected populations were [5] (see "Diagnosis of polycystic ovary syndrome in adults", section on 'Diagnosis'):

●The National Institutes of Health (NIH) – 6 percent (5 to 8 percent, n = 18 trials)

●Rotterdam criteria – 10 percent (8 to 13 percent, n = 15 trials)

●Androgen Excess and PCOS (AE-PCOS) Society criteria – 10 percent (7 to 13 percent, n = 10 trials)

Thus, the most conservative estimate for the prevalence for PCOS would be approximately 6 percent, but the actual prevalence is probably closer to 10 percent of reproductive age females. However, as the disorder is primarily heritable [6], this prevalence likely translates to the entire female population, regardless of age. (See "Diagnosis of polycystic ovary syndrome in adults", section on 'Other proposed criteria'.)

Onemeta-analysis reported a prevalence of PCOS from 10 to 13 percent [4].

The phenotype of PCOS appears to vary by ethnicity and geography. In a population-based study of Asian and Caucasian females in East Siberia, the upper normal limits (UNLs) for total testosterone and free androgen index varied by ethnicity [7]. In another study, females with PCOS in Alabama were more likely to have hirsutism and metabolic issues, while those in California were more likely to have hyperandrogenemia [8]. While there may be an environmental impact on PCOS phenotype, referral bias likely plays a role as well [9].

High-risk groups — A number of conditions appear to be associated with an increased prevalence of PCOS (table 1):

●Oligoovulatory infertility [10-13].

●Obesity and/or insulin resistance [14-17], although the impact of obesity appears to be relatively modest [17]. (See 'Obesity and energy regulation' below.)

●Type 1 [18,19], type 2 [20,21], or gestational diabetes mellitus [22,23]. (See "Clinical manifestations of polycystic ovary syndrome in adults", section on 'IGT/type 2 diabetes'.)

●A history of premature adrenarche [24,25].

●First-degree relatives with PCOS [26,27].

●Certain racial/ethnic groups (eg, Mexican American, Indigenous Australians), when compared with White or African American females [28-30], although this observation needs to be confirmed.

●Use of antiseizure medications – Females with epilepsy receiving antiseizure medications have an increased frequency of PCOS. While some studies suggest that the association is independent of antiseizure medication [31], most available data now report that the increased rate of PCOS in these females is due to antiseizure medication use, in particular, valproate [32-34].

In one prospective study of females starting valproic acid for bipolar disorder, oligomenorrhea with hyperandrogenism developed in 9 (10.5 percent) of 86 females on valproate, compared with 2 (1.4 percent) of 144 females on a non-valproate anticonvulsant or lithium [35]. A follow-up study in a subgroup of these females suggested that menstrual cycle irregularities and clinical hyperandrogenism improved after stopping valproate, but weight and polycystic ovarian morphology on ultrasound did not change [36]. The best evidence comes from a meta-analysis of 11 studies demonstrating a twofold excess risk of developing PCOS in 556 females treated with valproate, compared with 593 females treated with other antiseizure medications. Of note, valproic acid appears to potentiate androgen biosynthesis in theca cells [37].

PATHOGENESIS

Historical perspective — Stein and Leventhal, in their original report [1], emphasized the dyad of what they called "polycystic" ovaries in association with amenorrhea. Although they also noted that some of their subjects had hirsutism or acne, these features were not viewed as central to the syndrome. They went on to observe that "bilateral polycystic ovaries are most probably a result of some hormonal stimulation and very likely relate to the anterior lobe of the pituitary gland," noting that other investigators achieved similar polycystic ovaries when injecting "Antuitrin-S," a urinary extract of anterior pituitary hormones, notably human chorionic gonadotropin (hCG) [38]. Others subsequently observed excess luteinizing hormone (LH) activity in the urine of females with the Stein-Leventhal syndrome, determined using a bioassay (ie, the ovarian response of immature female rats or the prostatic response of hypophysectomized male rats to urinary extracts) [39-42]. This was later confirmed by plasma levels determined by radioimmunoassay [43].

Prior to the description of polycystic ovaries by Stein and Leventhal [1], the presence of sclerocystic ovaries was felt to be due to a number of disparate etiologies. In 1910, Fogue and Massabuau described three potential mechanisms: inflammation, congestion, and dystrophy [44]. The inflammation theory proposed that the microcystic ovary was the result of infection either of internal or external provenance. The congestion theory suggested that the lesion was the result of pressure, partial torsion, or other interruption in circulatory flow to the ovary. Finally, the dystrophy theory proposed that the abnormalities were caused by modifications or abnormalities in the nutrition of the ovary.

Others suggested that the development of polycystic ovaries was due to the morphological changes observed in the ovaries, including a thickened tunica albuginea that impeded normal ovulation [45]; however, these anatomic changes appear to primarily reflect the endocrine milieu, and effective ovulation is achievable by modulating the endocrine environment with clomiphene, gonadotropins, or insulin sensitizers. (See "Overview of ovulation induction", section on 'Oral agents'.)

Some early investigators proposed an adrenal etiology for polycystic ovary syndrome (PCOS) based upon observations in patients with congenital adrenal hyperplasia or adrenal neoplasms (both of whom often present with irregular menses and hyperandrogenism), and the response of these patients to cortisone therapy [46,47]. Today, while an adrenal component is observed in some females with PCOS [48], it is known that the ovaries are the predominant source of excess androgens in PCOS.

Current perspective

Genetics — PCOS is a complex genetic trait, similar to cardiovascular disease, type 2 diabetes mellitus, and the metabolic syndrome, where multiple genetic variants and environmental factors interact to foster the development and features of the disorder. The inherited basis of PCOS was established by twin studies and reports demonstrating an increased prevalence of PCOS in female first-degree relatives of affected females [6,26,27].

The largest twin study documented a monozygotic correlation of 71 percent and a dizygotic correlation of 38 percent; the study authors estimated that genetic influences account for as much as 70 percent of the variance in the pathogenesis of PCOS [6]. The prevalence of PCOS in mothers and sisters of PCOS females is 20 to 40 percent, considerably higher than that seen in the general population, strongly supporting a genetic basis of PCOS [26,27].

Potential genetic targets include genes regulating gonadotropin secretion and action, ovarian folliculogenesis, insulin secretion and action, weight and energy regulation, and androgen biosynthesis and action (figure 1).

Using candidate gene analysis, a number of studies have demonstrated some evidence of linkage or association with PCOS for a variety of gene variants, but follow-up studies have often failed to replicate results on the same candidate gene in different populations. Studies of PCOS candidate genes have been hampered by a number of factors including:

●A lack of understanding concerning the fundamental pathophysiology of PCOS

●The genetic complexity of the disorder

●The phenotypic heterogeneity of PCOS

●Methodologic issues:

•Only one or two variants in each gene of interest are usually genotyped

•The phenotype is often incompletely characterized in probands and family members

•A lack of appropriate controls

•The small numbers of subjects included in most studies

An alternative approach has been to perform genome-wide association studies (GWAS) to identify putative gene targets. In the first genome-wide association and replication studies of PCOS conducted in Chinese Han individuals, three loci were identified that were significantly associated with PCOS: two loci on chromosome 2 and a third locus on chromosome 9 [49]. One of these loci, on chromosome 2p16.3, contains the gene for the LH/hCG receptor (LHCGR), a logical susceptibility gene for PCOS. Two of the three, 2p21 and 9p33.3, contained multiple single-nucleotide polymorphisms (SNPs) that appeared to be independently associated with PCOS. The chromosome 2p21 locus contained SNPs in THADA, a gene that codes for a thyroid adenoma-associated protein. A report in patients with type 2 diabetes suggested that the THADA gene variant is associated with impaired beta cell function [50].

A subsequent GWAS observed that several of the same variants in DENND1A and THADA associated with PCOS in the Chinese population also affected the odds of PCOS in individuals of European origin [51]. The observation that the same genes influence PCOS risk in a number of different ethnic groups supports the hypothesis that PCOS is an ancient disorder [52]. The potential importance of DENND1A is described below.

Most, but not all, of the genes near the associated loci identified by GWAS of PCOS females are related to the control of hormone production and action, insulin resistance, and organ growth [53]. However, GWAS identifies loci, not genes; the pathophysiologic and clinical relevance of these loci needs to be confirmed.

Various approaches have been used to confirm the genetic significance and pathophysiological relevance of the genetic variants discovered by GWAS. For example, in one study, an association between the G allele of the rs13405728 variant and glucose and insulin metabolism in White females with PCOS was reported [54]. In another study, PCOS patients carrying the SNP rs13429458 near the THADA gene were reported to have higher LH and testosterone levels, while those carrying the rs12478601 locus demonstrated higher low-density lipoprotein levels [55]. Others have reported an association between the rs12468394-A allele and increased testosterone level [56], although not all investigators have observed an association between variants of THADA and metabolic or hormonal parameters [49,57].

GWAS in PCOS females have identified loci of interest near the DENND1A gene, a gene thought to play a role in the hyperandrogenemia of PCOS. Studying ovarian theca cells, PCOS patients have been observed to have low levels of the gene transcription product (a protein) DENND1A.V1, but higher expression of the gene product DENNDIA1A variant 2 (DENND1A.V2) compared with normal cycling females [58,59]. DENND1A immunostaining was more intense in the theca of PCOS ovaries, and PCOS patients express and demonstrate an elevated DENND1A.V2 to DENND1A.V1 ratio in ovarian theca cells [58]. Induced expression of DENND1A.V2 in vitro leads to the development of a PCOS-like phenotype in otherwise normal theca cells, with increased androgen production; while inactivation of DENND1A.V2 in PCOS theca cells with specific blocking antibodies converted the cells to a normal morphology or phenotype [60]. Other investigators have found an association between DENND1A polymorphisms and elevated serum insulin levels after glucose load [55].

Overall, these data begin to provide a genetic, molecular, and physiological context to the GWAS findings in PCOS. Polygenic risk scores to predict the genetic risk of individuals for PCOS are being developed, although their utility remains to be determined [61]. There is also interest in using Mendelian Randomization [62].

Of note, the heritability of PCOS is estimated to be approximately 70 percent, but the proportion of heritability accounted for by the PCOS loci identified so far by GWAS is <10 percent, not dissimilar to that observed in other complex genetic traits.

Gonadotropin secretion and action — Altered LH action appears to be involved in the pathogenesis of PCOS, as illustrated by the following:

●PCOS patients often have higher serum LH concentrations [63,64] and increased LH pulse frequency and amplitude [65] than matched controls. However, serum LH tends to be lower in obese females with PCOS compared with their lean counterparts [66].

●LH action at the ovarian level may be enhanced in PCOS as the LH receptor is overexpressed in thecal and granulosa cells from polycystic ovaries [67].

●Genetic variants of the LH beta-subunit [66,68] and loci near FSHR, LHCGR, and FSHB in GWAS [69] have been reported in patients with PCOS.

The increased LH to follicle-stimulating hormone (FSH) ratio further enhances hypersecretion of androgens in the theca cells in the ovarian follicles. The increase in follicular androgens impairs follicular development and reduces the normal inhibition of gonadotropin-releasing hormone (GnRH) pulse frequency by progesterone, further promoting the development of the PCOS phenotype [70]. In addition, there is evidence of resistance to the effects of FSH at the follicular levels in the ovaries of PCOS females, possibly in part secondary to excess local anti-müllerian hormone (AMH) production [71]. Increased LH pulses and enhanced daytime LH pulse secretion are also observed early during puberty in girls with hyperandrogenism [70], suggesting that abnormalities in the pulsatile release of GnRH might underlie the development of PCOS, at least in some patients.

However, excess LH levels are not required for increased ovarian androgen secretion or polycystic ovarian morphology. As an example, in a study of 45 females with regular ovulatory cycles [72], females with polycystic ovaries on ultrasound (n = 21) had higher serum androgens when compared with females with normal ovarian morphology (n = 24), although the two groups had identical mean LH concentrations on daily and frequent sampling studies.

Dysfunction in ovarian folliculogenesis — In PCOS, the selection of a dominant follicle is abnormal, a consequence of insufficient FSH stimulation and local inhibition of FSH action, possibly due to excess local AMH and other intra-ovarian factors that modulate follicular recruitment and growth [73]. Increased pituitary secretion of FSH alone, for example, through the administration of antiestrogens such as clomiphene citrate or aromatase inhibitors (letrozole), will often result in the resumption of normal follicular growth and ovulation in PCOS. (See "Ovulation induction with clomiphene citrate" and "Ovulation induction with letrozole".)

Furthermore, in PCOS, ovarian theca cells may be more sensitive to the effects of LH. Also, GWAS studies have consistently identified SNPs near the gene DENND1A as potential PCOS loci [69], suggesting that the protein transcript DENND1A.V2 could be, at least in part, responsible for the overproduction of androgens by the theca cells in the ovaries of PCOS females [60].

Insulin secretion and action — It was first observed that patients with PCOS were hyperinsulinemic in response to an oral glucose tolerance test [74]. It is now known that insulin resistance, and the development of compensatory hyperinsulinemia, is a frequent finding in PCOS [75-78]. The insulin resistance and hyperinsulinemia of PCOS patients that underlies many of the features of this disorder is highlighted by the finding that the administration of insulin-sensitizing agents, principally metformin, thiazolidinediones, and d-chiro-inositol, has been found to improve these features in many patients [79-82]. (See "Metformin for treatment of the polycystic ovary syndrome".)

Overall, 50 to 70 percent of females with PCOS demonstrate clinically measurable insulin resistance in vivo, above and beyond that determined by their body weight (ie, degree of obesity) [83,84]. Pancreatic beta cell compensation for the insulin resistance leads to increased levels of insulin, which in turn stimulate theca cell secretion of androgens [85,86] and inhibit hepatic sex hormone-binding globulin (SHBG) production [87,88], resulting in an increase in free androgens.

In this setting, insulin functions as a co-gonadotropin through its cognate receptor to modulate ovarian steroidogenesis, and the theca cells in PCOS females are hyper-responsive to the stimulatory effects of insulin on androgen secretion [86]. A number of variants of genes related to insulin action have been reported to be associated with PCOS.

The etiology for the increased insulin resistance and, consequently, the hyperinsulinism in PCOS, remains unclear. A post-binding defect in receptor signaling that selectively affects metabolic, but not mitogenic, pathways in classic insulin target tissues, such as adipose tissue and muscle (and possibly the ovary), has been described [89]. These defects seem to affect glucose transporter 4 (GLUT4) expression. Additionally, adipose tissue epigenetic dysfunction may play a role in the insulin resistance of PCOS [90].

Although assessing insulin resistance and/or hyperinsulinemia in PCOS is not straightforward, markers for metabolic dysfunction in PCOS may include the severity of menstrual dysfunction and hirsutism. In one study, more severe degrees of menstrual dysfunction were associated with greater insulin resistance [91]. In a second study the degree of hyperinsulinemia. independent of androgen levels, reflected the severity of hirsutism in PCOS [92].

Obesity and energy regulation — The presence of obesity worsens insulin resistance, the degree of hyperinsulinemia, the severity of ovulatory and menstrual dysfunction, and pregnancy outcome in PCOS and is associated with an increasing prevalence of metabolic syndrome, glucose intolerance, cardiovascular risk factors, and sleep apnea [93-97]. Bidirectional Mendelian randomization analyses suggest that increased (body mass index) BMI is causal for PCOS while the reverse is not the case [98]. (See "Clinical manifestations of polycystic ovary syndrome in adults".)

Despite the fact that the risk of PCOS increases modestly with the degree of obesity [15-17] and that the metabolic features of PCOS are worsened by the concomitant presence of obesity, it is still unclear whether obesity itself is causative. In a 2004 study from the United States, 60 percent of females with PCOS were obese [99], approximately twice the rate seen in the general adult population at that time [100]. However, the prevalence of obesity in PCOS varies widely with the population studied [101], suggesting that environmental factors play a significant role in determining the presence of obesity in PCOS. Alternatively, while the prevalence of obesity in the population varies widely throughout the world, the prevalence of PCOS appears to vary only modestly [102-105]. This was illustrated in a study of 675 unselected females seeking an employment history; the prevalence rates of PCOS in underweight, normal weight, overweight, and obese females were 8.2, 9.8, 9.9, and 9.0 percent, respectively [17].

Current data suggest that obesity is not as frequent in PCOS as previously thought, and it appears that the high rate of obesity reported in the disorder may, to a large extent, reflect referral bias in the populations studied. In a study of 292 PCOS patients identified at a tertiary care outpatient facility (referral PCOS) and 64 unselected PCOS subjects identified through the screening of a population of 668 females seeking a pre-employment physical, the referral PCOS subjects had higher mean BMI, hirsutism scores, and degrees of hyperandrogenemia, and they were more likely to be non-Hispanic White persons (84 percent) and to have a more severe PCOS subphenotype than the unselected PCOS or unselected controls [9].

The prevalence of obesity and severe obesity in referral PCOS females was 2.3 and 2.5 times greater than estimates of the same in unselected PCOS and 2.2 and 3.8 times greater than estimates in unselected controls, respectively. Alternatively, unselected PCOS subjects had a prevalence of obesity and severe obesity and a mean BMI similar to those of the general population from which they were derived. A meta-analysis also confirmed that the BMI of PCOS patients seen in referral populations is significantly higher than for PCOS patients identified in unselected populations [106]. This referral bias is not entirely surprising, since obesity is a significant detractor from quality of life and likely drives patient self-referral for medical care. Overall, these newer data suggest that while females with PCOS may appear to be more obese than their peers, much of this increased prevalence may be the result of referral bias.

Although excess adiposity may not be a principal driver of the high prevalence of PCOS, we should note that the mean BMI of females with PCOS seen in the clinic has been gradually increasing over time; the change parallels the increase in BMI seen in the general population [17]. While these data suggest that obesity in PCOS primarily reflects environmental factors, it is also possible that PCOS is associated with a greater propensity for obesity and weight gain, and genes relating to weight and energy regulation have been studied [98]. (See 'Epidemiology' above and "Clinical manifestations of polycystic ovary syndrome in adults", section on 'Factors affecting phenotype'.)

Androgen biosynthesis and action — Hyperandrogenism is a central feature for most forms (phenotypes A through C) of PCOS. The androgens are secreted primarily by the ovaries and secondarily by the adrenals [48]. Although hyperinsulinism is associated with hyperandrogenism in PCOS, insulin resistance alone is not sufficient for the development of PCOS [15], suggesting that an underlying (genetic) predisposition to hyperandrogenism must also be present. Variants of a number of genes involved in the regulation of androgen biosynthesis or action have been described in PCOS, including variants in the DENND1A gene. (See 'Genetics' above.)

Additional information on the disturbances of androgen biosynthesis observed in PCOS is reviewed separately. (See "Steroid hormone metabolism in polycystic ovary syndrome".)

Environmental factors — The most clearly defined environmental factor likely affecting the development of PCOS is diet and its association with obesity. Nonetheless, despite wide variations in the prevalence of obesity and type of diet [107], the prevalence of PCOS appears to be relatively uniform across the globe [5,102-105] (see 'Obesity and energy regulation' above and 'General population' above). Other potential factors may include androgen-mimicking environmental toxins [107-109].

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: Polycystic ovary syndrome".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Polycystic ovary syndrome (The Basics)")

●Beyond the Basics topics (see "Patient education: Polycystic ovary syndrome (PCOS) (Beyond the Basics)")

SUMMARY

●PCOS – The polycystic ovary syndrome (PCOS) is a disorder that includes functional androgen excess (which may include supranormal androgen levels and/or hirsutism, among other hyperandrogenic signs), ovulatory and menstrual dysfunction, polycystic ovarian morphology, and, in many subjects, metabolic dysfunction. (See 'Pathogenesis' above.)

●Epidemiology – The prevalence of PCOS depends to a degree upon the criteria used to define this disorder. Most studies globally have observed a prevalence between 10 and 13 percent in unselected reproductive-aged females, somewhat higher when broader criteria were used (ie, that of the Androgen Excess [AE]-PCOS Society and Rotterdam). (See 'Epidemiology' above.)

The prevalence of PCOS is increased in the presence of obesity, insulin resistance, type 1, type 2, or gestational diabetes mellitus, oligoovulatory infertility, premature adrenarche, a positive family history for PCOS among first-degree relatives, and, possibly, in certain ethnic groups. Regardless of the definition used, clinical evaluations and research reports should clearly denote the type of PCOS phenotypes the subjects have (table 1). (See 'High-risk groups' above.)

●Genetics – PCOS is a complex genetic trait whose development is likely influenced to some degree by environmental factors (eg, diet and the development of obesity) and more significantly by a number of different genetic variants. A susceptible genotype may stem from the inheritance of variants in genes regulating androgen biosynthesis and action, gonadotropin secretion and action, ovarian folliculogenesis, insulin secretion and action, and weight and energy regulation. Further studies of well-defined patients, preferably selected from medically unbiased populations, are necessary to develop an understanding of the etiology(s) of this highly prevalent and morbid disorder. (See 'Genetics' above.)