The genetics of polycystic ovary syndrome

The Genetics of Polycystic RICHARD S. LEGRO, M.D.,

Ovary

Syndrome

Hershey, Pennsylvania

Problems that confound the clinical characterixation of polycystic ovary syndrome (PCOS) also complicate the search for its genetic cause. There is no consensus as to the nature of the clinical phenotype for PCOS, and there is even doubt whether polycystic ovaries are an indispensable part of the syndrome. Few ethnk studies on PCOS have been performed, although it has been reported in ‘most major racial groups. Genetic studies of family clusters and relatives of affective probands have shown a high incidence of affected relatives. A dominant- mode, of inheritance, rather than a recessive one, seems more likely, Multiple genetic causes of adult-onset hyperandrogenism and chronic anovulation have been identified. Chromosomal studies of patients with PCOS have shokn no consistent abnormality. Motecular genetic studies are now ongoing. Future genetic models should consider such problematicai-eas as the heterogeneity of the syndrome, phenotypes in males as well as in females in the nonreproductive years, the exclusion of secondary causes of hyperandrogenism, and the persistence of a syndrome that adversely affects fecundity.

P

roblems that confound the clinical characterization of polycystic ovary syndrome (PCOS) also complicate the search for its genetic causes. There is no consensus as to what the clinical phenotype is, although for the purpose of this article the diagnostic criteria from the 1990 National Institutes of Health-National Institute of Child Health and Human Development (NIH-NICHD) conference on PCOS will be used: hyperandrogenism and/or hyperandrogenemia, oligo-ovulation, and exclusion of other known disorders such as congenital adrenal hyperplasia (CAH), hyperprolactinemia, or Cushing’s syndrome [ll. To date, there is unfortunately no consistent clinical marker or phenotype unique to PCOS that distinguishes it from other forms of hyperandrogenism. Additionally, the criterion that was originally the sine qua non of the syndrome, polycystic ovaries, has become perhaps one of the dispensable criteria. Polycystic ovaries represent a final common phenotype of a wide variety of etiologies, or as Givens has so succinctly stated, they are “a sign, not a diagnosis” [21. The purpose of this article is to review the literature on the ethnicity and genetics of PCOS, discuss some preliminary results from the molecular genetic investigations we have carried out, and discuss potential models for studying the genetics of PCOS. The prevalence of PCOS among the population will vary according to the diagnostic criteria utilized. However, estimates of its frequency suggest that it may be the most common endocrine abnormality in women. Polson et al [3] examined a large group of volunteers from the general population. They found that 22% of 257 women had polycystic ovaries. Of these, 66% had menstrual irregularities. Up to 10% of women of reproductive age may have the full-blown syndrome of hyperandrogenism, chronic anovulation, and polycystic ovaries.

ETHNIC STUDIES06 PCOS

From the Deoartment of Obstetrics and Gvnecolonv. Penn State College of Medicine, Milton ‘S. Hershey Medical Center, Hershey, Pennsylvania. Requests for reprints should be addressed to Richard S. Legro, M.D., Box 850, Division of Reproductive Endocrinology, Department of Obstetrics and Gynecology, M. S. Hershey Medical Center, Hershey, Pennsylvania 17033.

Few studies on the ethnicity of PCOS have been performed, although it has been reported in most major racial groups. Aono et at [4] identified a group of 11 Japanese women with polycystic ovaries identified on laparoscopy or laparotomy who had a significantly elevated mean testosterone and luteinizing hormone:follicle-stimulating hormone (LH:FSH) ratio compared with ethnic controls.

January 16, 1995

The American Journal of Medicine

Volume 98 (suppI 1A)

lA-9S

SYMPOSIUM ON ANDROGENSAND WOMEN’S HEALTH/ LEGRO

FAMILIAL STUDIESOF PCOS

Patients with PCOS had an exaggerated response of LH secretion to both gonadotropin-releasing hormone (GnRH) infusion and conjugated estrogen infusion to the same degree as had previously been reported in PCOS patients from U.S. and European groups. Carmina et al [51 studied a cohort of 75 patients with hyperandrogenic chronic anovulation (HCA) composed of 25 Japanese, 25 Italian, and 25 Hispanic-American women compared with ethnic controls. Participants were characterized on the basis of history, physical, ultrasound appearance of the ovaries, gonadotropins, sex steroids, and insulin sensitivity. The Japanese women were less obese and were not hirsute compared with the other ethnic groups. All groups had similar testosterone and LH levels and a similar incidence of polycystic ovaries on ultrasound. Adrenal androgens were elevated in comparable numbers of patients and to a similar degree. Insulin resistance (IR) as measured by a dissociation constant of an insulin tolerance test was significantly elevated, but similar in all groups. Dunaif et al [6] studied the effect of PCOS and ethnicity on insulin action. A total of 13 Caribbean-Hispanics with PCOS were compared with eight non-Hispanic Caucasians with PCOS as well as controls matched for age, weight, body composition, and ethnicity. Significantly decreased insulinmediated glucose disposal and increased steadystate insulin levels as determined by euglycemic clamp were noted both for PCOS and ethnicity. Caribbean-Hispanics with PCOS tended to be the most insulin resistant.

Previous genetic studies (or more precisely segregation analyses) of PCOS, which have focused on probands and their relatives, have found a high incidence of affected relatives. These studies have used various criteria for identifying probands with the syndrome, as well as for characterizing other affected family members. Cooper et al 1’71performed the first large study of the genetics of PCOS that attempted to identify and characterize other family members. Matched-pair analyses of 18 affected individuals were studied. All were Caucasian. The affected patients were all identified as having Stein-Leventhal syndrome, with both clinical and biochemical abnormalities, although the exact manifestations were not identified. All probands had ovaries diagnosed as polycystic, either on the basis of a wedge resection or culdoscopy. Only firstdegree female relatives of the identified probands were studied and were compared with a control group. History of oligomenorrhea was more common in mothers of PCOS patients (4 of 18) compared with controls (0 of 13) as well as in sisters (9 of 19) compared with controls (1 of 18). Culdoscopy showed that 8 of 12 sisters of probands had SteinLeventhal-appearing ovaries. Although male relatives were not specifically studied, a questionnaire revealed that male relatives were noted to have increased “pilosity.” This was one of the first published inferences that males could also be affected. The proposed mechanism of inheritance was autosomal dominant with decreased penetrance.

Key @ Arteriolar

(> Oligomenorrhea

q q

@I Diabetes

66

Examined

@ Hirsutism

q

mellitus

nephrosclerosis

Figure 1. Partial Proband

@I Hyperlipidemia

identified

by an arrow.

malitles

in the pedigree

M.

IA-IOS

January 16, 1995 The American Journal of Medicine Volume98

(suppl

1A)

pedigree with oligomenorrhea

of a larger family. and hirsubsm IS

Other metabolic

abnor-

are identified.

From

SYMPOSIUM ON ANDROGENSAND WOMEN’S HEALTH/ LEGRO

Givens and his associates [8-lo] from the University of Tennessee in Memphis have reported on multiple kindreds showing affected members in several generations. Ethnicity is not specifically stated for each family, but the largest pedigree studied (>150 members) was identified on the basis of a “black” female. Diagnostic criteria were hirsutism and oligomenorrhea with enlarged ovaries. Some pedigree members, including males, were studied in considerable detail. Controls were not utilized as in other studies. These studies were the first to reveal some of the severe metabolic sequelae that may accompany the syndrome. Sequelae identified by the Memphis group include diabetes mellitus, IR, lipid abnormalities, hypertension, and arteriosclerosis (part of a larger sample pedigree is shown in Figure 1). Additionally, the study of these pedigrees served to underscore the variability of phenotype in PCOS, even within the same kindred. An example frequently cited is the variability of the ovarian morphology, with ovaries from the same pedigree varying in size, the number of follicular versus atretic cysts, and the extent of stromal hyperplasia. Some males were also studied more closely for the first time. In one kindred there were several males with oligospermia and one with Klinefelter’s syndrome (47,XXY) [ll]. Elevated LH levels were discovered in some males. Givens concluded that there is abnormal gonadotropin secretion and testicular function in some male kindred members. When they classified female kindred members on the basis of hirsutism and oligomenorrhea, they found a high percentage of females affected through both maternal and paternal transmission, although the paternal transmission appeared stronger (Table I). This would suggest inheritance in an X-linked dominant manner, although in later publications the group tended to deemphasize the linkage to the X chromosome and stress a probable dominant mode of inheritance. Ferriman and Purdie [12] reported on a larger group of 700 hirsute patients with or without oligomenorrhea. The affected group was classified on the basis of hirsutism and enlarged ovaries (documented by an outmoded air-contrast technique known as gynecography). Results shown in Table II display a significantly higher prevalence of hirsutism, oligomenorrhea, and infertility among firstdegree relatives of hirsute women than nonhirsute women or controls. The frequency of various abnormalities was sought and analyzed in relatives based on the phenotype of the affected case. Also noted in this study (on the basis of a questionnaire) was an increased incidence of premature balding among male relatives of a subgroup of 136 hirsute female patients (see Table III). Patients and affected famJanuary 16,

TABLE I Interview Results of Relatives of Patients With Histologically Proven PCOS Maternal

Paternal

Slbships Affectedmembers

28 39

15 41

Total

93

41

Percentage affected

41%

87%

Maternal and Paternal

5 9 10 90%

!

I

A total of 28 slbshrps had only maternal relatives affected; 15 sibshlps had only paternal aunts affected; and 5 had both paternal and maternal aunts affected. PCOS = polycystlc ovary syndrome. Adapted from [91.

ily members were not systematically characterized either alinically or endocrinologically. The authors concluded that the mode of inheritance was a “modified dominant form(s) of inheritance.” More recent studies from England have focused on polycystic ovaries identified by ultrasound to characterize PCOS. Hague et al [13] utilized highresolution ultrasonography to identify polycystic ovaries in women presenting to a reproductive endocrinology clinic complaining of menstrual disturbances, hyperandrogenic phenomena, obesity, and infertility. Female first-degree relatives were then subjected to ovarian ultrasound examination. Males were not studied and ethnicity was not stated. Ovarian morphology of polycystic ovaries was according to the criteria of Adams et al [141, (~10 cysts/follicles 58 mm diameter around the periphery of the ovary or through the stroma, combined with increased stroma). The ultrasound appearance of the ovaries was felt to be a more sensitive diagnostic marker than either symptoms or biochemical markers. An initial cohort showed that 28 of 30 sisters had polycystic ovaries, so more families were recruited. Eventually 61 families were studied and 56 had more than one family member affected (92%). Overall, 45 of 52 siblings (87%) or probands were affected. Segregation ratios were in excess of those expected in an autosomal mendelian inheritance. Lunde et al [15] studied a group of 132 Norwegian women who had been identified on the basis of an ovarian wedge resection, and compared them with a control group of 71 women. Criteria for inclusion as a proband included “multicystic ovaries” and two or more of the following symptoms: menstrual irregularities, hirsutism, infertility, and/or obesity. Only 10 had a severe form of PCOS, defined as polycystic ovaries twice the normal size, amenorrhea, hirsutism, infertility, and obesity. Information was obtained about family members by questionnaire. 1995 The American Journal of Medicine

Volume 98 (suppl 1A)

IA-11s

SYMPOSIUMONANDROGENSAND WOMEN’SHEALTH/ LEGRO TABLE II Prevalance of PCOS Symptoms Among First-Degree Relatives of Probands Number with Female First-Degrae Relative Suffering From:

Probands

HlfSUte

Ovarian Size

Total Number of Patients

Hirsulism n 1%)

Enlarged

188

38 1201

30(16)

96

13(14)

15(161

Normal Nonhirsute

45 45 179

Enlarged Normal

Controls

ll2) 112) 7(41

8(181 7(161 8 141

19ilO) 10(101 5(H) 3(71 a 14)

japted from 1121. :OS = polycystic ovary syndrome

TABLEIll Prevalanceof PrematureBaldingAmongMale First-DegreeRelativesof Probands NlEllbNWRh Probands Body Hair Growth

Ovarian Size

Hirsute

Enlarged

Total Number of Patients

44 32 60 34

Normal Undetermined Nonhusute

Enlarged

Controls

178

Premature BaMg Among Male Fti. Oe&!reeRefafives (74)

8(18) a(251 17128) 319) 14(a)

iapted from [121.

Findings were consistent with the earlier study by Ferriman and Purdie [XL?]. Female first-degree relatives of PCOS patients had a significantly higher percentage of PCOS-related symptoms (hirsutism, menstrual irregularities, and infertility) compared with controls (31.4% vs 3.2%); 19.7% of male firstdegree relatives of PCOS patients were reported to have early baldness or “excessive hairiness,” compared with 6.5% of controls. No clear mode of inheritance was ascertained, although the authors felt that the findings were consistent with an autosomal dominant mode of inheritance for a large number of the families. Recently a report from Carey et al [16] in which affected probands and family members were more fully characterized suggested a single gene with an autosomal dominant pattern of transmission as the cause of polycystic ovaries. Probands were identified on the basis of ovarian morphology on ultrasound.as per the criteria reviewed above. However, probands and family members, including some males, underwent a more extensive evaluation, consisting of history; measurement of physical indices and hirsutism; measurement of serum androgens and other steroids, including 17a-hydroxyprogesterone, gonadotropins, and prolactin; assessment of insulin resistance by oral glucose tolerance lA-12s

January

16,

1995

The American

Journal

of Medicine

Volume

test in obese patients, and ultrasound visualization of polycystic ovaries in women. A total of 14 families were identified, although information was only available on 10 families to perform classic segregation analysis. Eight of these families were Caucasian, one was Iranian, and one was Asian. None of the families had CAH as determined by 17ahydroxyprogesterone or type A IR. Although the authors found a highly significant association of ovarian morphology with a positive history of PCOS (p <0.0034), only 43% had an abnormal testosterone value (Figure 2). A combination of symptoms with biochemical markers was able to identify 92% of women with ultrasound evidence of polycystic ovaries. Affected status was assigned in firstdegree relatives on the basis of ultrasound findings consistent with polycystic ovaries, and in the extended family on the basis of a positive history. Female first-degree relatives were found to have a 51% chance of being affected. Premature balding was found to be an accurate phenotype for male carriers (8 of 22 males with premature male pattern baldness). No endocrinologic male marker of affected status was noted. If male pattern baldness is accepted as the male phenotype, then the segregation is consistent with autosomal dominant inheritance. 98 (suppl

1A)

SYMPOSIUM ON ANDROGENSAND WOMEN’S HEALTH / LEGRO

These studies point up some of the difficulties in studying the genetics of PCOS. Many of these studies are subject to an ascertainment bias, as only familial aggregates were studied. Thus, strictly speaking they involved the entity of familial PCOS. This may not be representative of isolated cases of PCOS any more than familial aggregates of breast and ovarian cancer may be representative of the more common nonfamilial cases of these cancers. The uncertain phenotypic criteria make assignment of affected status difficult, and different authors have utilized different criteria (Table IV). Many of the studies rely mainly on historical criteria to do this. PCOS remains a diagnosis of exclusion, and many studies have failed to exclude CAH, specifically 21-hydroxylase deficiency. None of the studies has fully characterized the endocrinologic and metabolic sequelae of an entire pedigree. The male phenotype has also been incompletely studied and remains uncertain (see Table V), although the common thread seems to be some disorder of androgen metabolism. In many of the studies, abnormalities of male hair distribution were diagnosed by vague criteria via questionnaire. Despite these shortcomings, the study of familial aggregates has consistently suggested that the mode of inheritance appears to be dominant. This fact, in and of itself, would tend to exclude many of the other etiologies of hyperandrogenism, including steroidogenic enzyme deficiencies, which are autosomal recessive.

GENETIC INVESTIGATIONSOF PCOS

among 19 women with PCOS compared with 46 controls, Multiple genetic causes of adult-onset hyperandrogenism and chronic anovulation have been identified. The prevalence of many of these mutations among hyperandrogenic women is still being established. As other authors in this symposia have reported, mutations that can cause CAH in the 21hydroxylase gene and the 3P-hydroxysteroid reductase gene have been identified. Mutations in the insulin receptor have also been identified in women with frank diabetes mellitus and significant IR with accompanying hyperandrogenism to the point of virilization [25]. However, these have not yet been documented in PCOS. Complete sequencing of all 22 exons of the insulin receptor in two women with PCOS has not yielded any mutations [26]. Recently, a patient with polycystic ovaries, sexual infantilism, and ambiguous external genitalia was reported to have aromatase deficiency, due to point mutations in both aromatase genes [27]. We have explored the relationship between several potential candidate genes and women affected by HCA. A group of >200 female Hispanic patients who presented to the Endocrine and Infertility Clinic at Women’s Hospital in Los Angeles were characterized on the basis of menstrual and reproductive history as well as serum hormonal assays. We limited our analysis to Hispanic women to control partially for ethnic differences in allele frequencies [28]. We chose candidate genes that had polymorphic regions or established restricted fragment length polymorphisms. These include to date the androgen receptor and two dopamine receptor subtypes, the Dz dopamine receptor and the Da dopamine receptor.

Genetic investigation of PCOS (specifically chromosomal and molecular genetic studies) was first undertaken with chromosome studies. There have been isolated case reports or small series reporting 40 I polyploidies [17] and aneuploidies, specifically X chromosomal aneuploidies. These include XX/XXX and XX/X0 mosaics [l&191. Larger cytogenetic 30 series of PCOS patients, however, have found norNo. mal karyotypes. Stenchever et al [20] reported norof mal karyotypes in 41 patients and Knorr et aZ 1211 patients 2. also reported the same in 16 patients. Few molecular genetic studies of PCOS have been undertaken to date. Human leukocyte antigen (HLA) association studies of familial groupings of PCOS have shown conflicting results. Mandel et al [22] studied four families with ~2 affected siblings Bx Sx+Bx Sx LH Test and found no linkage to the HLA types studied. Figure 2. Ultrasound screening for women with polycysbc ovaries. Values In Hague et al [23] reported an association with parentheses are percentages of these women with clrnical and/or biochemical DRW6 in 75 patients with polycystic ovaries com- features suggestive of PCOS. Bx = biochemical abnormality; LH = abnormal pared with 110 control women. However, when 16 luteinizing hormone value (36%); Sx = positive symptoms (84%); families with PCOS were studied, no HLA linkage Sx+Bx = positrve symptoms and/or biochemical abnormality (92%); Test was noted. In a similar but smaller study, Ober et al = abnormal testosterone value (43%); Bx, LH, and/or testosterone abnormal[24] reported an association with DQAl”0501 ity (54%). From [161. January16, 1995 The American Journal of Medicine Volume98

kuppl

1A)

lA-13S

SYMPOSIUM ON ANDROGENS AND WOMEN’S HEALTH/ LEGRO

TABLEIV FamilialStudiesof PCOSand PCOSDiagnosticCriteria and ProposedModeof inheritance PCOS Criteria

Author

Mode of Inheritance

Cooper et al(1968) [71

Ollgomenorrhea, hlrstiism, polycysbc ovaries (by culdoscopy)

Autosomal dominant with reduced penetrance

Givens eta1(1971,1975, 19881[8-101

Hirsutism and polycystic ovaries (exam and surgery)

MInked dominant

fernman and Purdle (19791 [I21

Hirsutism, oligomenorrhea (by aircontrast gynecography)

Modified dominant

Hagueetal~1988)l131

Polycystic ovaries by uitrasound and cltnical symptoms of PCOS

Segregabon rattos exceeded autosomal dominant pattern

Multicystic ovaries on wedge resectron and clinical symptoms of PCOS

Unclear, but most consistent with autosomal dominant

Polycystic ovaries (by transabdomlnal ultrasound)

Autosomal dominant with 90% penetrance

tunde

etal(1989) [RI

Carey eta1119931

[161

PCOS = polycystic ovary syndrome.

DNA was obtained from peripheral leukocytes for genetic analysis. We found no association between PCOS patients and either androgen receptor alleles [29] or dopamine Dz receptor alleles [30]. However, we found a significant clustering of women with PCOS among one of the genotypes of the dopamine D3 receptor (unpublished observation). These must, however, be considered preliminary results and further studies are indicated. Choosing candidate genes for PCOS can open up a Pandora’s box. Because the exact biochemical cause of the metabolic abnormalities is uncertain (as well as what abnormalities characterize the syndrome), there are many potential candidates. With ~5% of human genes identified, a recent survey of cloned potential genes, which includes such genes as enzymes, receptors, and transcription factors, identified >250 candidate genes. As more potential candidate genes are identified, the search for association could become a wild goose chase.

TABLEV ProposedMale Affected or Carrier Phenotypein Familial Kindredsof PCOS Author

Male Phenolylpe

Cooper et al119681 171 Glvensetai11971,1975,1988)[8-18 Ferrlman and Purdie (19791 WI Hague eta1~198811131 tunde etall1989l1151 Careyeta1(19931[161

Increased “plosity” Oligospermra, lack of facial hair Premature balding Not lnvestlgated Early baldness or excessive hairiness Premature balding

nc - nn,t,r,,ct,r ,,l,Q”l, nmrlmmo

lA-14s

January

16,

1995

The American

Journal

of Medicine

Volume

GENETICMODELS FOR INVESTIGATINGPCOS A model for the systematic investigation of the genetics of PCOS is needed. Association studies such as the ones discussed above are useful as preliminary tests for an association between candidate genes and the disease phenotype. Association studies also offer the advantage of studying diseases whose mode of inheritance is uncertain and whose presentation is poorly defined or subject to late onset or variable penetrance. All of these criteria are applicable to the study of PCOS. Only the affected (and when relevant, nonaffected controls) need be studied. This would eliminate some of the difficulty in classifying premenarchal and postmenopausal women as well as men. However, association studies may be weakened by the heterogeneity, both genetic and nongenetic, of the syndrome. PCOS is not a single untamed beast, but a veritable menagerie of untamed beasts. Clumping together common phenotypes due to diverse etiologies will weaken any attempted genetic analysis. Certainly the stigma that gave this syndrome its name, polycystic ovaries, is the prime example of this. Any condition that causes hyperandrogenemia can result in polycystic ovaries, including Cushing’s disease and CAH. Hirsutism is another example of a phenotypic characteristic with diverse etiologies, including increased precursor androgens as well as increased peripheral activity of androgenic enzymes. Such phenotypes, in addition to involving separate genes or combinations of genes, may also have different modes of inheritance. Large family clusterings of PCOS offer the best opportunity for identifying unique strains of PCOS. 98 (suppl

1A)

SYMPOSIUM ON ANDROGENSAND WOMEN’S HEALTH / LEGRO

Familial clusterings of PCOS may represent a homogenous etiology of the syndrome, despite significant phenotypic heterogeneity within a given pedigree. Careful phenotyping of the complete pedigree, including all available family members for the full spectrum of reproductive and metabolic abnormalities will allow assignment of affected status within that particular pedigree. This phenotyping may also serve to allow assignment of affected status to other pedigree members, such as men, prepubertal girls and boys, and postmenopausal women. There will be no a priori complex of symptoms or traits that will comprise the syndrome, other than the hyperandrogenemia and ovulatory disturbance that brought the pedigree to attention in the first place. Traits and quantifiable abnormalities will be examined both individually and in various combinations to identify genetically distinct groups. These may vary from kindred to kindred. Such combinations may involve hyperandrogenemia and ultrasound visualization of polycystic ovaries, or gonadotropin abnormalities combined with upper body obesity, or exaggerated adrenal response to adrenocorticotropic hormone stimulation testing combined with defective insulin receptor phosphorylation. Linkage analysis will be performed between polymorphic markers spaced at regular genetic intervals and these familial traits. Choosing candidate genes for beginning linkage analysis would probably prove unproductive. Ott [31] reports that these are frequently the first sites chosen for linkage analysis, but “only rarely, however, has this approach uncovered a disease linkage.” Large multigeneration pedigrees that are well characterized, such as those identified by Givens [S] or by Dunaif (Figure 3) may provide enough statistical power

Figure 3. Large familial kindred of polycystic ovary syndrome. NIDDM = non-insulindependent diabetes mellitus; PCOS = polycystlc ovary syndrome. (Data courtesy of Andrea Dunaif, M.D.)

Key:

for linkage analysis. The linkage of Huntington’s disease to a gene on the short arm of chromosome 4 was based largely on an extensive Venezuelan pedigree. There are multiple models of linkage analysis with which to examine the resulting data. Given the uncertain mode of inheritance, linkage analysis could be performed with expectations of a variety of modes of inheritance, hoping that a strong relationship would appear under one of the models. Another method in such cases is to examine the risk in relatives of probands compared with the risk of relatives from the control population, a ratio known as A. This ratio decreases with the degree of relationship between proband and relatives. This has been used to estimate the number of genes involved in a disease etiology [32]. Several models for quantitative linkage analysis used to analyze complex traits also exist [33-361. The power to detect linkage when affected pairs of relatives are used is greatest for grandparentgrandchild pairs, followed by cousins, and then siblings [37]. Due to the difficulty in assigning affected status in PCOS grandparents, cousin or sibling pairs may be a more realistic alternative. Utilizing sibling pairs (when there is no recombination between marker and disease gene, and a power of 80% is desired) a sample size >200 is required to detect linkage with a A = 2. With a higher risk ratio (h = 5) only 60 pairs are required. Adding unaffected relatives to the analysis in addition to the affected relative pairs will increase the informativeness of the linkage analysis. Available studies suggest there is a strong familial component to PCOS. Our expanding knowledge of the human genome fostered by the Human Genome Project and increasingly sophisticated mathe-

fF3 Untested;

January 16, 1995

0 PCOS;

0 NIDDM;

The American Journal of Medicine

(3 PCOS + NIDDM

Volume 98 (suppI 1A)

lA-15s

matical tools for performing linkage studies may allow us to map the disease locus for PCOS. There is no guarantee that such an analysis would result in identifying a candidate locus for PCOS. However, even a negative analysis would prove useful in eliminating suspicions that there is a genetic basis to PCOS. This would allow channeling of resources for detecting other epigenetic etiologies of the syndrome.

REFERENCES 1. Dunaif A, Givens JR, Haseltine FP, Merriam GR, eds. Current Issues in Endocrinology and Metabolism: Polycystic Ovary Syndrome. Boston: Blackwell Scientific Publications, 1992. 2. Givens JR. Polycystic ovaries: a sign, not a diagnosis. Sem Reprod Endocrinol 1984; 2: 271-80. 3. Polson DW, Adams J, Wadsworth J, Franks S. Polycystlc ovaries-a common finding in normal women. Lancet 1988; i: 870-2. 4. Aono T, Miyazaki M, Miyake A, Kinugasa T, Kurachi K, Matsumoto K. Responses of serum gonadotropins to LH-releasing hormone and estrogens in Japanese women with polycystic ovaries. Acta Endocrinologica 85: 840-8. 5. Carmina E, Koyama T, Chang L, Stanczyk FZ, Lobo RA. Does ethnicity influence the prevalance of adrenal hyperandrogenism and insulin resistance tn polycystic ovary syndrome? Am J Obstet Gynecol 1992; 167: 1807-12. 6. Dunaif A, Sorbara L, Delson R, Green G. Ethnicity and polycystic ovary syndrome are associated with independent and additive decreases in insulin action in Carib bean-Hispanic women. Diabetes 1993; 42: 1462-8. 7. Cooper HE, Spellacy WN, Prem KA, Cohen WD. Hereditary factors in Stein-Leventhal syndrome. Am J Obstet Gynecol 1968; 100: 371-87. 8. Givens JR. Familial ovarian hyperthecosis: a study of two families. Am J Obstet Gynecol 1971; 11: 959-72. 9. Wilroy RS Jr, Givens JR, Wiser WL, Coleman SA, Andersen RN, Fish SA. Hyperthecosis: an Inheritable form of polycystic ovarian disease. Birth Defects 1975; 11: 81-5. 10. Givens JR. Familial polycystic ovarian disease. Endocrinol Metab Clin North Am 1988; 17: 1-17. 11. Cohen PN, Givens JR, Wiser WL, et al. Polycystic ovarian disease, maturation arrest of spermatogenesis and Klinefelter’s syndrome in siblings of a family with familial hirsutism. Fertil Steril 1975; 26: 1228-38. 12. Ferriman D, Purdie AW. The inheritance of polycystic ovarian disease and possible relationship to premature balding. Clin Endocrinol 1979; 1: 291-9. 13. Hague WM, Adams J, Reeders ST, Peto TEA, Jacobs HS. Familial polycystlc ovaries: a genetic disease? Clin Endocrinol 1988; 29: 593-605. 14. Adams J, Polson DW, Frank S, et al. Multifollicular ovaries: clinical and endocrine features and response to pulsatile gonadotropin releasing hormone. Lancet 1985; ii: 1375-9. 15. Lunde 0, Magnus P, Sandvik L, Hoglo S. Familial clusterrng in the polycystic ovarian syndrome. Gynecol Obstet Invest 1989; 28: 23-30. 16. Carey AH, Chan KL, Short F, White D, Williamson R, Franks S. Evidence for a

lA-16s

January

16, 1995 The American Journal of Medicine

single gene effect causing polycystlc ovaries and male pattern baldness. Clan Endocrinol 1993; 38: 653-9. 17. Rojanasakul A, Gustavson KH, Lithell H, Nillius SJ. Tetraploidy in two sisters with the polycystic ovary syndrome. Clin Genet 1985; 27: 167-74. 18. Netter A, Block-Michel H, Salomon Y, Thervet F, Degrouchy J, Lamy M. Etude du caryotype dans la maladie de Stein-Leventhal. Ann Endocnnol (Pans) 1961; 22: 8419. 19. Bishun NP, Morton WRM. Chromosome mosaicism in Stein-Leventhal syndrome (letter). Br Med J 1964; 1200. 20. Stenchever MA, Macintyre MN, Jarvls JA, Hempel JM. Cytogenetic evaluation of 41 patients with Stein-Leventhal syndrome. Obstet Gynecol 1968; 32: 794-801. 21. Knorr K, Knorr G, Gartner H, Uebele-Kallhardt B. Chromosomenanalytlsche Befunde bei Stein Leventhal Syndrome. Endokrinolgle 1969; 54: 364-73. 22. Mandel FP, Chang RJ, DuPont B, et al. HLA genotyping In family members and patients with familial polycystlc ovarian dtsease. J Clin Endocrinol Metab 1983; 56: 862-4. 23. Hague WM, Adams J, Algar V, et al. HLA associabons in pabents with polycystic ovaries and in patients with congenital adrenal hyperplasia caused by Pl-hydroxylase deficiency. Clin Endocrlnol (Oxf) 1990; 32: 407-15. 24. Ober C, Weil S, Steck T, Billstrand C, Levrant S, Barnes R. Increased risk for polycystic ovary syndrome associated with human leukocyte antigen DQAl’O501. Am J Obstet Gynecol 1992; 167: 1803-6. 25. Taylor SI, Cama A, Accill D, et al. Mutations in the Insulin receptor gene. Endocr Rev 1992; 13: 566-95. 26. Dunalf A. Insulin resistance and ovarian hyperandrogenism. Endocrinologist 1992; 2: 248-50. 27. Ito Y, Fisher CR, Conte FA, Grumback MM, Simpson ER. Molecular basis of aromatase deficiency in an adult female with sexual infantilism and polycystic ovaries. Proc Natl Acad Sci USA 1993; 90: 11673-7. 28. Cooper DN, Clayton JF. DNA polymorphism and the study of disease associations. Hum Genet 1988; 78: 299-312. 29. Legro RS, Shahbahrami B, Lobo RA, Kovacs B. Size polymorphism of the androgen receptor among female Hispanics and correlation with androgenlsm. Obstet Gynecol 1994; 83: 701-6. 30. Legro RS, Dietz G, Cornings D, Lobo RA, Kovacs B. DZ receptor gene haplotypes in female Hispanics are significantly associated with gonadotropins and fecundity but not ovulatory status. Hum Reprod. 1994; 9: 1271-5. 31. Ott J. Analysis of Human Genetic Linkage. Baltimore: The Johns Hopkins llnlversity Press, 1992. 32. Risch N. Linkage strategies for genetlcally complex traits. I. Multilocus models. Am J Hum Genet 1990; 46: 222-8. 33. Risch N. Linkage strategies for genetically complex traits. II. The power of affected relative pairs. Am J Hum Genet 1990; 46: 242-53. 34. Risch N. Linkage strategies for genetically complex traits. III. The effect of markers of polymorphism on analysis of affected relative pairs. Am J Hum Genet 1990; 46: 222-8. 35. Bishop DR, Williamson JA. The power of identlty-by-state methods for linkage analysis. Am J Hum Genet 1990; 46: 254-65. 36. Fishman PM, Suarez B, Hedge SE, Reich T. A robust method for the detection of linkage in familial diseases. Am J Hum Genet 1978; 30: 308-21. 37. Ott J. Cutting a Gordian knot in linkage analysis of complex human traits. Am J Hum Genet 1990; 46: 219-21.

Volume 98 (suppl IA)