There may be at least two different biologically relevant subtypes of polycystic ovary syndrome (PCOS).
By comparing the genomes of 893 women of European ancestry diagnosed with PCOS, researchers found that these women could be grouped into at least two distinct subsets, which are associated with novel gene regions and which have different biological responses.
'We're starting to make headway on what causes PCOS. It's very frustrating for patients because it's poorly understood and patients often see several physicians before PCOS is diagnosed,' said senior author Dr Andrea Dunaif. 'Through genetics, we're beginning to understand the condition and may have specific targeted therapies in the not-too-distant future.'
PCOS is an infertility disorder that affects at least 15 percent of women of reproductive age. The cause of PCOS is unknown, but it does appear to run in families, suggesting a genetic basis to the disorder. PCOS is currently diagnosed based upon physical features and symptoms, including irregular or missing periods, raised levels of male sex hormones, hirsutism and small cysts on the surface of the ovaries.
There is a variation of symptoms among PCOS patients and this study, conducted by scientists at the Mount Sinai Health System in New York, aimed to discover whether there were any genetic, clinical or biochemical differences that could explain this.
Dr Dunaif and her team identified PCOS subtypes by classifying cases into relative groups called clusters from the results of their previously published PCOS genome-wide association study (GWAS). The clusters were then repeated in an independent group of 263 PCOS cases and the researchers identified two distinct PCOS subtypes: a 'reproductive' group and a 'metabolic' group.
Women in the 'reproductive' group (approximately 23 percent of cases) had raised levels of luteinising hormone (LH), which triggers ovulation, and higher sex hormone-binding globulin (SHBG) levels, a protein that regulates the ability of testosterone to enter its target tissues. In addition, this group had lower insulin levels and a lower body mass index (BMI).
Women in the 'metabolic' group (approximately 37 percent of cases) had raised levels of glucose and insulin and a higher BMI. They also had lower LH and SHBG levels.
The remaining women (approximately 40 percent of cases) had no distinguishable characteristics, with traits of both the metabolic and reproductive subgroups.
The researchers discovered that the subtypes tended to cluster in families and that carriers of rare genetic variants in DENND1A, a gene involved in male hormone production, were more likely to have the reproductive subtype of PCOS.
Rather than a catch-all approach for PCOS patients, the researchers suggest that these distinct forms may benefit from different treatment approaches, as they are underpinned by different biological mechanisms, and this could improve long-term outcomes for patients.
'In contrast to classifying disorders based on expert opinion, this is a very powerful objective approach to categorising syndromes like PCOS into distinct subtypes with different causes, treatment and clinical outcomes, said Dr Dunaif.
This study was published in PLOS Medicine.