When one size does not fit all: Reconsidering PCOS etiology, diagnosis, clinical subgroups, and subgroup-specific treatments

Polycystic Ovary Syndrome (PCOS) is a complex endocrine disorder that affects a large proportion of women. Due to its heterogeneity, the best diagnostic strategy has been a matter of contention. Since 1990 scientific societies in the field of human reproduction have tried to define the pivotal criteria for the diagnosis of PCOS. The consensus Rotterdam diagnostic criteria included the presence of hyperandrogenism, oligo/anovulation, and polycystic ovarian morphology (PCOM), and have now been updated to evidence based diagnostic criteria in the 2018 and 2023 International Guideline diagnostic criteria endorsed by 39 societies internationally. Within the Rotterdam Criteria, at least two out of three of the above-mentioned features are required to be present to diagnose PCOS, resulting in four phenotypes being identified: phenotype A, characterized by the presence of all the features, phenotype B, exhibiting hyperandrogenism and oligo-anovulation, phenotype C, presenting as hyperandrogenism and PCOM and finally the phenotype D that is characterized by oligo-anovulation and PCOM, lacking the hyperandrogenic component. However, it is the hypothesis of the EGOI group that the Rotterdam phenotypes A, B, and C have a different underlying causality to phenotype D. Recent studies have highlighted the strong correlation between insulin resistance and hyperandrogenism, and the pivotal role of these factors in driving ovarian alterations, such as oligo-anovulation and follicular functional cyst formation. This new understanding of PCOS pathogenesis has led the authors to hypothesis that phenotypes A, B, and C are endocrine-metabolic syndromes with a metabolic clinical onset. Conversely, the absence of hyperandrogenism and metabolic disturbances in phenotype D suggests a different origin of this condition, and point towards novel pathophysiological mechanisms; however, these are still not fully understood. Further questions have been raised regarding the suitability of the “phenotypes” described by the Rotterdam Criteria by the publication by recent GWAS studies, which demonstrated that these phenotypes should be considered clinical subtypes as they are not reflected in the genetic picture. Hence, by capturing the heterogeneity of this complex disorder, current diagnostic criteria may benefit from a reassessment and the evaluation of additional parameters such as insulin resistance and endometrial thickness, with the purpose of not only improving their diagnostic accuracy but also of assigning an appropriate and personalized treatment. In this framework, the present overview aims to analyze the diagnostic criteria currently recognized by the scientific community and assess the suitability of their application in clinical practice in light of the newly emerging evidence.