Yvonne V. Louwers

Causal mechanisms and balancing selection inferred from genetic associations with polycystic ovary syndrome

Polycystic ovary syndrome (PCOS) is the most common reproductive disorder in women, yet there is little consensus regarding its aetiology. Here we perform a genome-wide association study of PCOS in up to 5,184 self-reported cases of White European ancestry and 82,759 controls, with follow-up in a further ∼2,000 clinically validated cases and ∼100,000 controls. We identify six signals for PCOS at genome-wide statistical significance (P<5 × 10−8), in/near genes ERBB4/HER4, YAP1, THADA, FSHB, RAD50 and KRR1. Variants in/near three of the four epidermal growth factor receptor genes (ERBB2/HER2, ERBB3/HER3 and ERBB4/HER4) are associated with PCOS at or near genome-wide significance. Mendelian randomization analyses indicate causal roles in PCOS aetiology for higher BMI (P=2.5 × 10−9), higher insulin resistance (P=6 × 10−4) and lower serum sex hormone binding globulin concentrations (P=5 × 10−4). Furthermore, genetic susceptibility to later menopause is associated with higher PCOS risk (P=1.6 × 10−8) and PCOS-susceptibility alleles are associated with higher serum anti-Müllerian hormone concentrations in girls (P=8.9 × 10−5). This large-scale study implicates an aetiological role of the epidermal growth factor receptors, infers causal mechanisms relevant to clinical management and prevention, and suggests balancing selection mechanisms involved in PCOS risk.

Cardiovascular and metabolic profiles amongst different polycystic ovary syndrome phenotypes: who is really at risk?

OBJECTIVE To study the cardiometabolic profile characteristics and compare the prevalence of cardiovascular (CV) risk factors between women with different polycystic ovary syndrome (PCOS) phenotypes. DESIGN A cross-sectional multicenter study analyzing 2,288 well phenotyped women with PCOS. SETTING Specialized reproductive outpatient clinic. PATIENT(S) Women of reproductive age (18-45 years) diagnosed with PCOS. INTERVENTION(S) Women suspected of oligo- or anovulation underwent a standardized screening consisting of a systematic medical and reproductive history taking, anthropometric measurements, and transvaginal ultrasonography followed by an extensive endocrinologic/metabolic evaluation. MAIN OUTCOME MEASURE(S) Differences in cardiometabolic profile characteristics and CV risk factor prevalence between women with different PCOS phenotypes, i.e., obesity/overweight, hypertension, insulin resistance, dyslipidemia, and metabolic syndrome. RESULT(S) Women with hyperandrogenic PCOS (n=1,219; 53.3% of total) presented with a worse cardiometabolic profile and a higher prevalence of CV risk factors, such as obesity and overweight, insulin resistance, and metabolic syndrome, compared with women with nonhyperandrogenic PCOS. In women with nonhyperandrogenic PCOS overweight/obesity (28.5%) and dyslipidemia (low-density lipoprotein cholesterol≥3.0 mmol/L; 52.2%) were highly prevalent. CONCLUSION(S) Women with hyperandrogenic PCOS have a worse cardiometabolic profile and higher prevalence of CV risk factors compared with women with nonhyperandrogenic PCOS. However, all women with PCOS should be screened for the presence of CV risk factors, since the frequently found derangements at a young age imply an elevated risk for the development of CV disease later in life.

The phenotype of polycystic ovary syndrome ameliorates with aging

OBJECTIVE To assess the effects of aging on the features of polycystic ovary syndrome (PCOS). DESIGN Retrospective longitudinal follow-up study. SETTING Tertiary care center. PATIENT(S) Patients with PCOS, diagnosed according to the 2003 Rotterdam criteria, who visited the outpatient clinic on consecutive occasions with a minimum interval of 6 months. INTERVENTION(S) Comparisons were made between the first visit and the consecutive visit grouped by intervals. MAIN OUTCOME MEASURE(S) Changes in clinical and endocrine characteristics. RESULT(S) A total of 254 women visited the outpatient clinic on 2 occasions each. Consecutive visits were grouped into 0.5 to 3.9 years (n = 172; mean follow-up, 2.6 years) and 4.0 to 7.0 years (n = 82; mean follow-up, 5.5 years). At their second visit, significantly more women had regained a regular cycle. The total antral follicle count was similar. Serum levels of testosterone, androstenedione, and dehydroepiandrosterone sulfate had decreased significantly. Plasma glucose levels had increased, whereas serum insulin levels and homeostasis model assessment score had significantly decreased. CONCLUSION(S) The PCOS phenotype changed with aging, suggesting an amelioration of the phenotype and ovarian dysfunction as indicated by the increase in number of regular menstrual cycles, decrease in serum androgen levels, and decrease in insulin resistance.

The role of vitamin D in metabolic disturbances in polycystic ovary syndrome: a systematic review

CONTEXT Metabolic disturbances, in particular, insulin resistance (IR) and dyslipidemia, are common in women suffering from polycystic ovary syndrome (PCOS). Evidence is accumulating that vitamin D status may contribute to the development of metabolic disturbances in PCOS. OBJECTIVE The aim of this study was to carry out a systematic review addressing the association between vitamin D status, vitamin D receptor polymorphisms, and/or polymorphisms related to vitamin D metabolism and metabolic disturbances in women with PCOS. DESIGN AND METHODS A systematic search of electronic databases was carried out up to January 2013 for observational studies and clinical trials in women suffering from PCOS with outcome measures that were related to vitamin D status. We conducted univariate and multivariate regression analyses of the weighted means to gain insights into the association between vitamin D, BMI, and IR based on existing literature. RESULTS We found 29 eligible trials with inconsistency in their results. One well-designed randomized controlled trial has been carried out until now. Univariate regression analyses of the weighted means revealed vitamin D to be a significant and independent predictor of IR in both PCOS and control women. The significance disappeared after adjustment for BMI in PCOS women. CONCLUSIONS Current evidence suggests an inverse association between vitamin D status and metabolic disturbances in PCOS. Owing to the heterogeneity of the studies, it is hard to draw a definite conclusion. The causal relationship between vitamin D status and metabolic disturbances in PCOS remains to be determined in well-designed placebo-controlled randomized clinical trials.

Replication of association of a novel insulin receptor gene polymorphism with polycystic ovary syndrome

OBJECTIVE To evaluate association with polycystic ovary syndrome (PCOS) of 295 variants in 39 genes central to metabolic insulin signaling and glycogen synthase kinase 3β (GSK-3β) regulation, followed by replication efforts. DESIGN Case-control association study, with discovery and replication cohorts. SETTING Subjects were recruited from reproductive endocrinology clinics, and controls were recruited from communities surrounding the University of Alabama at Birmingham and Erasmus Medical Center, Rotterdam. PATIENT(S) A total of 273 cases with PCOS and 173 control subjects in the discovery cohort; and 526 cases and 3,585 control subjects in the replication cohort. All subjects were caucasian. INTERVENTION(S) Phenotypic and genotypic assessment. MAIN OUTCOME MEASURE(S) Detection of 295 single-nucleotide polymorphisms (SNPs), PCOS status. RESULT(S) Several SNPs were associated with PCOS in the discovery cohort. Four insulin receptor (INSR) SNPs and three insulin receptor substrate 2 (IRS2) SNPs associated with PCOS were genotyped in the replication cohort. One INSR SNP (rs2252673) replicated association with PCOS. The minor allele conferred increased odds of PCOS in both cohorts, independent of body mass index. CONCLUSION(S) A pathway-based tagging SNP approach allowed us to identify novel INSR SNPs associated with PCOS, one of which confirmed association in a large replication cohort.

Oligoovulatory and anovulatory cycles in women with polycystic ovary syndrome (PCOS): what's the difference?

CONTEXT Polycystic ovary syndrome (PCOS) is a heterogeneous disorder. The phenotype may differ between patients who exhibit signs of recent ovulation and anovulatory PCOS patients. OBJECTIVE Our objective was to study differences in clinical and endocrine characteristics and response to ovulation induction (OI) treatment comparing oligoovulatory and anovulatory PCOS patients. DESIGN AND SETTING We conducted a retrospective cohort study at a tertiary hospital. PATIENTS PCOS patients (n=1750) presenting with oligo- or amenorrhea were diagnosed according to the Rotterdam 2003 consensus criteria. Arbitrarily, oligoovulatory PCOS was defined by a single random serum progesterone level of 10 nmol/liter or higher. MAIN OUTCOME MEASURES We evaluated the incidence of oligo- or amenorrhea, menstrual cycle length, serum androgen levels, follicle count, and OI outcome parameters. RESULTS Anovulatory women (n=1541 of 1750, 88.1%) were more often amenorrheic (P<0.001) and presented with a longer cycle duration (P<0.001) compared with oligoovulatory women (n=209 of 1750, 11.9%). Serum levels of testosterone (P<0.001), the free androgen index (P<0.001), and total follicle count (P<0.005) were higher in anovulatory compared with oligoovulatory patients. During clomiphene citrate OI, more oligoovulatory women gained regular menstrual cycles (P<0.05), whereas after second-line treatment with recombinant FSH, more anovulatory women became pregnant (P<0.05). CONCLUSIONS Oligoovulatory women with PCOS exhibit a milder phenotype of ovarian dysfunction and have a more favorable response to OI treatment using clomiphene citrate compared with anovulatory PCOS patients. However, during second-line treatment with recombinant FSH, anovulatory PCOS patients presented with a higher chance of pregnancy compared with oligoovulatory patients.

Influence of oral contraceptives on anthropomorphometric, endocrine, and metabolic profiles of anovulatory polycystic ovary syndrome patients.

OBJECTIVE To evaluate the influence of oral contraceptive pills (OCPs) on anthromorphometric, endocrine, and metabolic parameters in women with polycystic ovary syndrome (PCOS). DESIGN Retrospective cross-sectional cohort study for the period 1993-2011. SETTING Tertiary university hospital. PATIENT(S) PCOS patients, who never, ever, or at time of screening were using OCPs were included. A total of 1,297 patients, of whom 827 were white, were included. All PCOS patients diagnosed according to the Rotterdam 2003 consensus criteria were divided into three groups: current users, (n = 76; 6% of total), ever users (n = 1,018; 78%), and never users (n = 203; 16%). Ever users were subdivided based on the OCP-free interval. INTERVENTION(S) None. MAIN OUTCOME MEASURE(S) Anthromorphometric (blood pressure, cycle duration) and ultrasound (follicle count, mean ovarian volume) parameters, endocrine (SHBG, testosterone, free androgen index, antimüllerian hormone [AMH]) and lipid profiles. RESULT(S) Current users and ever users were compared with never users. In current users, SHBG was increased and androgen levels decreased. Patients with an OCP-free interval of <1 year had a higher mean follicle count, higher AMH level, and increased serum androgen level compared with never users. SHBG levels remained increased until 5-10 years after cessation of OCP use. CONCLUSION(S) OCP use causes a milder phenotypic presentation of PCOS regarding hyperandrogenism. However, it does not alter parameters associated with increased health risks.

Variants in SULT2A1 Affect the DHEA Sulphate to DHEA Ratio in Patients With Polycystic Ovary Syndrome But Not the Hyperandrogenic Phenotype

CONTEXT Because of the elevated dehydroepiandrosterone sulfate (DHEAS) levels in polycystic ovary syndrome (PCOS) and the heritability of DHEAS serum levels, genes encoding the enzymes that control the sulfation of dehydroepiandrosterone (DHEA) to DHEAS and vice versa are obvious candidate genes to explain part of the heritability of PCOS. OBJECTIVE The objective of the study was to determine the role of genetic variants in sulfotransferase (SULT2A1), 3-phosphoadenosine 5-phosphosulfate synthase isoform 2 (PAPSS2), and steroid sulfatase (STS) in PCOS and in hormone levels related to the hyperandrogenic phenotype of PCOS. DESIGN This was a candidate-gene study. PATIENTS The discovery set consisted of 582 patients and 2017 controls. MAIN OUTCOME MEASURES A pruned subset of 28 single-nucleotide polymorphisms (SNPs) in SULT2A1, PAPSS2, and STS was generated based on pairwise genotypic correlation. Association with PCOS was tested, and we studied whether the SNPs modulate DHEAS levels, DHEA levels, and their ratio in PCOS. Significant SNPs were replicated in an independent sample of patients. RESULTS None of the SNPs in SULT2A1, PAPSS2, and STS constituted risk alleles for PCOS. SNP rs2910397 in SULT2A1 decreased the DHEAS to DHEA ratio in PCOS by 5% in the discovery sample. Meta-analysis of discovery and replication sample resulted in a combined effect of -0.095 (P = .027). However, carrying the minor T allele did not contribute to differences in the hyperandrogenic phenotype, including the levels of T and androstenedione, of PCOS patients. CONCLUSIONS Genetic variants in SULT2A1, PAPSS2, and STS do not predispose to PCOS. Although a variant in SULT2A1 decreased the DHEAS to DHEA ratio, no changes in other androgenic hormone levels were observed.

Bone morphogenetic proteins and the polycystic ovary syndrome

BackgroundPolycystic Ovary Syndrome (PCOS) is defined by two out of the following three criteria being met: oligo- or anovulation, hyperandrogenism, and polycystic ovaries. Affected women are often obese and insulin resistant. Although the etiology is still unknown, members of the Transforming Growth Factor β (TGFβ) family, including Bone Morphogenetic Proteins (BMPs) and anti-Müllerian hormone (AMH), have been implicated to play a role. In this pilot study we aimed to measure serum BMP levels in PCOS patients.MethodsTwenty patients, fulfilling the definition of PCOS according to the Rotterdam Criteria, were randomly selected. Serum BMP2, -4, -6 and −7 levels were measured using commercially available BMP2, BMP4, BMP6 and BMP7 immunoassays.ResultsSerum BMP2, serum BMP4 and serum BMP6 levels were undetectable. Three patients had detectable serum BMP7 levels, albeit at the lower limit of the standard curve.ConclusionsBMP levels were undetectable in almost all patients. This suggests that with the current sensitivity of the BMP assays, measurement of serum BMP levels is not suitable as a diagnostic tool for PCOS.

The Impact of Self-Reported Ethnicity Versus Genetic Ancestry on Phenotypic Characteristics of Polycystic Ovary Syndrome (PCOS)

CONTEXT It is well established that ethnicity is associated with the phenotype of polycystic ovary syndrome (PCOS). Self-reported ethnicity was shown to be an inaccurate proxy for ethnic origin in other disease traits, and it remains unclear how in PCOS patients self-reported ethnicity compares with a biological proxy such as genetic ancestry. OBJECTIVE We compared the impact of self-reported ethnicity versus genetic ancestry on PCOS and tested which of these 2 classifications better predicts the variability in phenotypic characteristics of PCOS. PATIENTS A total of 1499 PCOS patients from The Netherlands, comprising 11 self-reported ethnic groups of European, African, American, and Asian descent were genotyped with the Illumina 610K Quad BeadChip and merged with the data genotyped with the Illumina HumanHap650K available for the reference panel collected by the Human Genome Diversity Project (HGDP), in a collaboration with the Centre Etude Polymorphism Humain (CEPH), including 53 populations for ancestry reference. MAIN OUTCOME MEASURES Algorithms for inferring genetic relationships among individuals, including multidimensional scaling and ADMIXTURE, were applied to recover genetic ancestry for each individual. Regression analysis was used to determine the best predictor for the variability in PCOS characteristics. RESULTS The association between self-reported ethnicity and genetic ancestry was moderate. For amenorrhea, total follicle count, body mass index, SHBG, dehydroepiandrosterone sulfate, and insulin, mainly genetic ancestry clusters ended up in the final models (P values < .004), indicating that they explain a larger proportion of variability of these PCOS characteristics compared with self-reported ethnicity. Especially variability of insulin levels seems predominantly explained by genetic ancestry. CONCLUSIONS Self-reported ancestry is not a perfect proxy for genetic ancestry in patients with PCOS, emphasizing that by using genetic ancestry data instead of self-reported ethnicity, PCOS-relevant misclassification can be avoided. Moreover, because genetic ancestry explained a larger proportion of phenotypic variability associated with PCOS than self-reported ethnicity, future studies should focus on genetic ancestry verification of PCOS patients for research questions and treatment as well as preventive strategies in these women.