Metabolic-inflammatory heterogeneity in PCOS assisted reproduction: a compartmental susceptibility model

Polycystic ovary syndrome (PCOS) is a heterogeneous endocrine-metabolic disorder rather than a uniform reproductive diagnosis. Women entering assisted reproductive technology (ART) differ in adiposity, insulin sensitivity, lipid metabolism, androgen excess, inflammatory activity, and oxidative stress, but these systemic traits do not map uniformly onto the follicular or endometrial compartments. We propose a testable compartmental susceptibility model in which systemic metabolic-inflammatory burden acts as a background exposure and clinically relevant effects emerge when stage-specific follicular-oocyte or endometrial-maternal vulnerability is also present. Evidence was interpreted with a pragmatic hierarchy that separates guideline/RCT evidence, registry and externally validated prognostic evidence, adjusted observational associations, mechanistic discovery studies, and protocol-only evidence. Current data support metabolic assessment, prevention of excessive ovarian response, and maternal risk surveillance, but they do not validate a metabolic-inflammatory subtype for selecting stimulation, trigger, frozen-embryo-transfer preparation, immune adjuncts, or other phenotype-specific treatment. Local immune, extracellular-vesicle, and multi-omics signals remain discovery tools rather than clinical tests. Precision ART in PCOS will require paired systemic-local phenotyping, external validation, and prospective phenotype-by-treatment interaction studies using cumulative live birth and maternal-neonatal safety as primary endpoints.