How Have Genetic Analyses Improved Our Knowledge of PCOS?
A Q&A with Andrea Dunaif, MD
Modern genetic approaches have allowed for marked advances to be made in current understanding of the etiology of polycystic ovary syndrome (PCOS). Andrea Dunaif, MD, described these advances in a review article in the March The Journal of Clinical Endocrinology and Metabolism.1
EndocrineWeb spoke with Dr. Dunaif to discuss the major advances made in this field, and how these findings may lead to improved diagnosis and treatment of PCOS. Dr. Dunaif is Charles F. Kettering Professor of Endocrinology & Metabolism and Vice Chair for Research in the Department of Medicine at Northwestern University Feinberg School of Medicine in Chicago, IL. In addition, Dr. Dunaif was the senior author of the first genome-wide association (GWAS) study that mapped PCOS susceptibility loci in European ancestry PCOS cases and control women.
Q: How has genetic susceptibility research advanced our understanding of the etiology of PCOS?
Dr. Dunaif: Before modern genetic approaches, PCOS was described as a vicious cycle of abnormalities with no knowledge of the initiating insult. Fortunately, genetic research has markedly advanced our understanding of PCOS. For the first time, we are learning about which biological pathways may initiate the vicious cycle of PCOS. Further, new biologic pathways contributing to PCOS are being discovered.
Research has shown that gonadotropin secretory abnormalities found in PCOS are caused by increased frequency of pulsatile GnRH secretion that selectively increases luteinizing hormone (LH) secretion while slightly suppressing follicle stimulating hormone (FSH) release. The increased LH drives ovarian theca cell testosterone production. In the face of relative FSH deficiency, those follicles are arrested in development. Accordingly, the theca cell-derived testosterone isn’t aromatized as efficiently into estradiol because of a lack of FSH-stimulated granulosa cell aromatase capacity.
We also know that the ovarian theca cells of women with PCOS secrete more androgens than those of reproductively normal women. The increased LH drive, increased theca cell androgen secretion, and decreased granulosa aromatase activity result in increased ovarian androgen production. Increased circulating testosterone causes the hyperandrogenic signs of PCOS, primarily hirsutism but also acne and androgenic alopecia. In addition, testosterone feeds back on the hypothalamic-pituitary axis, decreasing sensitivity of that axis to the normal effect of estradiol and progesterone to slow the GnRH pulse generator, creating a self-sustaining cycle. This cycle can be initiated experimentally by administering either testosterone or high-frequency GnRH pulses.
For the past 20 years, researchers have been focused on testosterone as a cause of PCOS because all the features of the syndrome can be recreated in animal models by androgen administration during gestation, neonatally, or peripubertally.
However, GWAS have implicated genes regulating gonadotropin action and secretion in the pathogenesis of PCOS. Han Chinese GWAS studies identified two loci that contain genes for luteinizing hormone (LH)/human chorionic gonadotropin receptor (LHCGR) and FSH receptor (FSHR) that mediate ovarian gonadotropin action.2,3 The European GWAS studies identified a locus in the region of the FSH B polypeptide (FSHB) gene. Further, the same genetic signal is associated with circulating LH levels.4,5 These genes are all high-priority PCOS candidate genes given the central role of gonadotropins in the pathogenesis of ovarian androgen excess.
Currently, a total 19 PCOS susceptibility loci have been mapped in European and Han Chinese PCOS cohorts. These loci include known susceptibility genes for type 2 diabetes, THADA and insulin-receptor gene (INSR), and a gene associated with age of menopause, RAD50. Another susceptibility locus contained the DENND1A gene encoding DENND1A protein, which has recently been shown to be involved in testosterone production and the PCOS theca cell phenotype.6 However, the DENND1A PCOS susceptibility variants are intronic and do not alter the amino acid sequence of the DENND1A protein. Therefore, it is not known how these GWAS variants contribute to the pathogenesis of PCOS.
The remaining PCOS GWAS susceptibility loci contain genes that are not obvious PCOS candidate genes. However, arguably, the most important strength of GWAS is that they allow an unbiased interrogation of the entire genome for disease-associated susceptibility loci. Accordingly, they provide insight into novel biologic pathways contributing to disease pathogenesis. Thus, it is likely that further analyses of the PCOS GWAS loci will substantially advance our understanding of the etiology of PCOS.
Q: How much do the currently identified susceptibility loci contribute to the risk for PCOS?
Dr. Dunaif: The frustrating thing about GWAS is that very few of the genes found have large biologic effects. Most GWAS loci are associated with small increases in disease risk. Even genetic risk scores that sum the contributions of multiple disease-associated GWAS loci are inferior for disease prediction to algorithms based on clinical features and biomarkers. The relatively small contribution of GWAS loci to disease risk is not surprising since GWAS by design detect common allelic variants with minor allele frequencies of ≥5%. Common variants would be expected to have a modest effect on phenotype because they have not been subjected to strong selective pressure.
The heritability of PCOS has been estimated to be about 80%. The GWAS loci so far identified account for less than 5% of this heritability. This so-called “missing heritability” is found in most complex diseases/traits (ie, those that involve an interaction of susceptibility genes and environmental factors) including type 2 diabetes and obesity, where the GWAS susceptibility variants account for only 5% to 10% of the heritability of these conditions. This missing heritability may be the result of rare genetic variants with much larger biological effects that are not detected by GWAS. It is also possible that epigenetic changes are much more important than previously appreciated in disease pathogenesis.
Q: In your review, you note that several of the Han Chinese GWAS loci (DENND1A, THADA, YAP1, LHCGR, and FSHR) have been replicated in European cohorts diagnosed with PCOS by NIH or Rotterdam criteria. What are the clinical implications of the shared genetic susceptibility found across different PCOS phenotypes and ethnicities?
Dr. Dunaif: The observation that several genetic loci are shared between Han Chinese and European PCOS has evolutionary implications suggesting that these shared loci are conserved because the ancestral Eurasian population migrated out of Africa approximately 100,000 years ago, and these populations diverged about 40,000 years ago. The fact that the susceptibility genes do not differ according to NIH or Rotterdam diagnostic criteria or by self-report, as in the 23andMe study,7 suggests that we don’t need to fight over which diagnostic criteria to use.
Understanding that PCOS is easy to diagnose is important because women with PCOS are underserved by the medical community, and often have to see an average of four physicians before receiving an accurate diagnosis. I hope that these findings will increase recognition of PCOS in the primary care and general internal medicine communities, where it is currently considered to be an obscure, mysterious disease. PCOS is important to diagnose, particularly as many women with the condition are at high risk for diabetes and other cardiometabolic outcomes, including coronary artery disease.
Q: What advances in the PCOS do you think (or hope) will be made in the near future?
Dr. Dunaif: The wide availability of next generation sequencing and new methods to perform genome-wide assessments of variant function should permit a much more rapid elucidation of the biology of GWAS loci than has been possible in the past. This information will not only elucidate the pathogenesis of PCOS, but also will also provide new drug targets.
Q: What else would you like to tell readers of EndocrineWeb?
Dr. Dunaif: It is time to stop thinking of PCOS as a heterogeneous disorder that is difficult to diagnose. GWAS have already shown that there are a number of shared PCOS susceptibility loci across phenotypes defined by the different diagnostic criteria as well as across disparate ethnicities. The focus should now be on the unifying features of PCOS and the ease of diagnosis. This change in emphasis will be critical for broader engagement of healthcare professionals in the care of women with PCOS.
April 14, 2016