Once treated primarily as a fertility disorder, PCOS is being reframed as a whole-body endocrine and metabolic condition, opening the door to earlier diagnosis, sharper risk profiling, and more personalized care.
Pathophysiology of PCOS/PMOS. PCOS/PMOS is a complex syndrome, involving multiple feedback loops in the anterior pituitary, liver, pancreas, and ovaries, ultimately contributing to a hyperandrogenic state. In the pituitary gland, LH is released in higher amounts than FSH, simulating the theca cells of the ovary to increase androgen production. Androgens, in turn, stimulate the pituitary gland to release LH and FSH. Elevated androgen also acts on the liver, decreasing SHBG and increasing circulating free androgens. Finally, at the level of the pancreas, androgens lead to hyperinsulinemia, which in turn decreases SHBG. Insulin also acts on theca cells to promote the release of androgens. All these factors lead to HA, which leads to anovulation and PCOM in the ovary. LH, luteinizing hormone; FSH, follicle-stimulating hormone. SHBG, sex hormone binding globulin. AI-assisted redesign of Figure 1 by Chan JL, Masini I, and Pisarska MD, “Polyendocrine metabolic ovarian syndrome (PMOS)/polycystic ovary syndrome (PCOS): current and future trends,” Journal of Clinical Investigation, 2026;136(12). . Original article published under the Creative Commons Attribution 4.0 International License.
In a recent comprehensive review published in Journal of Clinical Investigationresearchers compiled decades of research on polycystic ovary syndrome (PCOS) to clarify the changing diagnostic paradigms surrounding the disease. It describes the interplay between genetic, epigenetic, environmental, metabolic, and neuroendocrine factors that may together contribute to disease development, phenotypic variation, and long-term complications, as well as emerging targeted approaches to improve patient outcomes.
The review findings highlight how the management of PCOS, recently proposed to be renamed polyendocrine metabolic ovarian syndrome (PMOS), has long been hampered by diagnostic inconsistency and symptom-focused treatments. It concludes by suggesting that PMOS treatment is moving from its one-size-fits-all roots toward personalized, multisystemic medicine that could support more targeted future care for this highly prevalent but poorly understood endocrinologic disorder.
Background
Decades of research and clinical reports have identified polycystic ovary syndrome (PCOS) as the most common endocrine disorder among women of reproductive age worldwide. Currently, studies estimate that the condition affects between 5% and 20% of the global female population, depending on the diagnostic criteria used.
Unfortunately, despite its widespread prevalence, the exact biological mechanisms driving the condition remain poorly understood. Consequently, researchers believe that current prevalence estimates may underrepresent true prevalence, potentially leaving millions of patients misdiagnosed or undertreated.
PCOS has been culturally and historically treated primarily as a fertility issue due to its association with irregular menstrual cycles and polycystic ovary morphology. However, modern clinical knowledge reveals a much more systemic reality. Recent research has demonstrated that the disorder is underpinned by hormonal and metabolic disturbances, often involving elevated androgens and/or insulin resistance.
Scientists now know that this combination of reproductive, endocrine and metabolic dysfunction is associated with a patient’s risk for lifetime metabolic complications, including insulin resistance, type 2 diabetes.T2DM), and cardiovascular diseases (CVD), although a definitive causal relationship between PCOS itself and CVD has not been established. Emerging research further links the condition to clinically significant mental health struggles such as depression and anxiety.
To better reflect this multisystem metabolic nature, an international consortium recently advocated renaming the condition Polyendocrine Metabolic Ovarian Syndrome (PMOS).
About the review
The present comprehensive review aimed to elucidate the complex architecture of PMOS by synthesizing findings from clinical studies, genetic analyses, mechanistic research, animal models, and emerging in vitro platforms.
It specifically assessed the implementation of the updated 2023 international evidence-based guidelines, which constitute the Rotterdam criteria for adults. These criteria require individuals to meet two of the three main diagnostic features: ovulatory dysfunction (O.D), biochemical or clinical hyperandrogenism (HA), and polycystic ovary morphology (PCOM), with the exception of other hormonal disorders.
In the field of diagnostic data, the review details advanced biochemical monitoring, such as the use of liquid chromatography-tandem mass spectrometry (LC-MS/MS) to measure total testosterone and pelvic transvaginal ultrasound with high-bandwidth transducers (8 MHz or higher) to map the number of antral follicles. In addition, the paper assesses the validity of the anti-Müllerian hormone (AMH) levels as an objective, non-invasive proxy for ultrasound mapping, while emphasizing that AMH should not be used as a stand-alone diagnostic test.
The review further explores advanced methodologies driving contemporary PMOS research, such as genome-wide association studies (GWAS) the investigation of risk alleles in various human lineages and advanced microfluidic platforms (eg, the “LATTICE” organ-on-a-chip framework) that allow researchers to study 3D cellular interactions and accelerate high-throughput automated drug testing at PMOS-like hormone levels.
Review of findings
The review revealed that, diagnostically, research using an automated AMH assay found that, in women aged 23 to 35 years, a serum AMH level above 3.2 ng/mL could serve as a potential surrogate for PCOM assessment, with a sensitivity of 88.6% and a specificity of 80.3%. In diagnostic bioinformatics research, machine learning models using specific hormone values, especially luteinizing hormone (LH), follicle-stimulating hormone (FSH), estradiol and sex hormone-binding globulin (SHBG), successfully predicted the development of PMOS before clinical onset with an area under the curve (AUC) to 0.85 in an out-of-sample test.
The review further identified a strong hereditary component, noting that 60% to 70% of girls born to mothers with PMOS develop the condition themselves. It also highlighted how distinct reproductive and metabolic phenotypes may carry different risk profiles, reinforcing the need for more accurate classification.
On the therapeutic front, clinical updates show that the aromatase inhibitor letrozole achieves significantly higher live birth rates for ovulation induction compared to the traditional choice, clomiphene citrate (27.5% vs 19.1%). Drug repurposing studies also highlighted that the antimalarial agent artemether alleviated hyperandrogenism, irregular estrous cycles, PCOM, and infertility in PCOS-like rodent models by targeting ovarian enzymes to inhibit excess androgen synthesis. In a small pilot human trial, the related compound dihydroartemisinin was associated with reduced hyperandrogenism, lower AMH levels, reduced PCOM, and normalized menstrual cycles.
conclusions
This review highlights that PMOS is a multifaceted, highly heritable condition that requires precision individualized care rather than a one-size-fits-all approach. While current treatments mainly manage symptoms such as abnormal bleeding or metabolic dysfunction through combined hormonal contraceptives (CHC), lifestyle intervention and metformin, the review findings suggest that the future of PMOS treatments may depend on identifying common biological pathways and clinically relevant subtypes.
Encouragingly, new treatments such as neurokinin 3 (NK3) receptor antagonists and glucagon-like peptide-1 (GLP-1) receptor agonists or poly-agonists show early or specific clinical potential in reducing hyperandrogenism or improving metabolic and reproductive traits, although several approaches remain preclinical and GLP-1 therapies require caution regarding conception and pregnancy. Addressing current diagnostic gaps through machine learning and validating distinct reproductive-metabolic subphenotypes could allow future clinicians to intervene early, thereby changing the long-term health course for millions of women.
