Metabolic heterogeneity of polycystic ovary syndrome: from pathophysiology to phenotypic stratification

Abstract

Introduction. Polycystic ovary syndrome (PCOS) is one of the most prevalent endocrine disorders in women of reproductive age; however, its clinical heterogeneity complicates timely stratification of long-term cardiometabolic risk. The lack of a generally accepted algorithm for a differentiated approach at the initial diagnostic stage highlights the need for developing additional assessment criteria to enable personalized prognosis and patient management strategies. Aim. To analyze current concepts of the pathophysiology of PCOS and to substantiate the necessity for metabolic phenotyping as the first step after diagnosis. Materials and methods. A narrative review of the literature presented in PubMed and eLibrary.ru databases for the period 2016–2026 was conducted. The diagnosis of PCOS in the included studies was based on the Rotterdam criteria (2003). The search was performed using the keywords: “polycystic ovary syndrome”, “insulin resistance”, “phenotyping”, “cardiovascular outcomes”, “hyperandrogenism”, and “cardiometabolic risk”. Results. PCOS is characterized by significant metabolic heterogeneity. The prevalence of insulin resistance ranges from 50 to 70%, including in some patients with normal body weight. The risk of developing type 2 diabetes mellitus (T2DM) in women with PCOS is four times higher than the rate in the general population. Depending on the leading pathogenetic mechanism, two main phenotypes have been identified: 1) the insulin-resistant phenotype (IR-phenotype), characterized by visceral obesity, hyperinsulinemia, dyslipidemia, and a high risk of T2DM, metabolic dysfunction-associated steatotic liver disease, and cardiovascular events; 2) the reproductive phenotype (R-phenotype), with normal body weight, isolated hyperandrogenism in the absence of systemic metabolic disorders, and a population-level cardiovascular risk. Phenotype differentiation determines the follow-up strategy: the IR phenotype requires extended metabolic screening, whereas monitoring of the R-phenotype is focused on the reproductive system. Conclusion. Verification of the metabolic phenotype helps avoiding both overdiagnosis in patients with a favorable prognosis and underestimation of high cardiometabolic risk in women with the IR-phenotype.