FABP4 is a small cytoplasmic protein expressed abundantly in adipocytes and macrophages, where it buffers and traffics fatty acids, coordinating lipid storage, lipolysis, and inflammatory signaling. Over the past two decades it has attracted growing attention as an adipokine—a hormone-like molecule released from fat tissue—whose circulating levels track closely with obesity, insulin resistance, and cardiovascular risk. In adults with PCOS, several studies have reported elevated FABP4 concentrations, but the adolescent population has remained largely unexplored. That gap matters, because adolescence is precisely the window in which PCOS first declares itself and in which early intervention could alter long-term metabolic trajectories. The new research, led by Bahadır Codal and colleagues across hospitals and universities in Türkiye, set out to determine whether FABP4 could serve as both a metabolic correlate and a diagnostic marker in this younger cohort.

The study was designed as a prospective observational investigation involving 88 adolescents aged 10 to 19 years, evenly divided between 44 girls diagnosed with PCOS and 44 healthy controls. Diagnosis was made according to the 2023 international ESHRE/ASRM criteria, the most recent consensus framework for identifying PCOS across the lifespan. Each participant underwent a comprehensive clinical, anthropometric, and biochemical workup, including measurements of body mass index, androgen levels, fasting glucose and insulin, and derived indices of insulin resistance such as HOMA-IR. Serum FABP4 concentrations were quantified using an enzyme-linked immunosorbent assay, the standard immunoassay technique that relies on antibody-based capture and detection to produce a quantitative readout of protein concentration in blood.

The headline finding was unambiguous: median circulating FABP4 levels were significantly higher in adolescents with PCOS than in controls, at 5.74 nanograms per milliliter versus 3.59 nanograms per milliliter, a difference that achieved strong statistical significance with a p-value below 0.001. In other words, girls with PCOS carried roughly 60 percent more of this fat-derived protein in their bloodstream than their peers. But the researchers went further, asking whether this elevation simply reflected differences in body weight or whether it carried information independent of adiposity. Using multivariable linear regression, a statistical technique that estimates the independent contribution of each variable while holding the others constant, they found that PCOS status itself, along with body mass index and insulin resistance as measured by HOMA-IR, each contributed to higher FABP4 concentrations.

This triad of associations—PCOS status, BMI, and insulin resistance—fits neatly with the established biology of FABP4. The protein is secreted by hypertrophied adipocytes and by macrophages infiltrating inflamed fat depots, and its circulating levels rise as fat tissue expands and as insulin signaling deteriorates. In PCOS, insulin resistance is a central pathophysiologic thread, driving compensatory hyperinsulinemia that in turn stimulates ovarian androgen production. The observation that FABP4 tracks with HOMA-IR in adolescents suggests that the protein may serve as a readout of the metabolic disturbance that underlies the reproductive phenotype, capturing a dimension of the disease that androgen measurements alone do not fully represent.

The diagnostic question was addressed with a more sophisticated statistical armamentarium. Because the outcome of interest—PCOS diagnosis—is binary, the team employed Firth’s penalized logistic regression, a method designed to produce stable estimates in small samples where conventional maximum-likelihood logistic regression can fail or produce inflated coefficients. This analysis identified two variables independently associated with PCOS: the free androgen index, or FAI, a calculated measure of bioavailable testosterone that reflects hyperandrogenism, and FABP4 itself. That a fat-derived adipokine stood alongside the classic hormonal marker of the disease is the finding most likely to capture the attention of clinicians and researchers alike.

To translate these associations into clinically meaningful performance metrics, the researchers constructed receiver operating characteristic curves, the standard framework for evaluating how well a continuous biomarker discriminates between diseased and healthy individuals. A FABP4 threshold of 4.08 nanograms per milliliter yielded an area under the curve of 0.849, with 81.8 percent sensitivity and 75.0 percent specificity. In practical terms, an AUC of 0.849 indicates good—not perfect—discrimination, meaning the test correctly ranks a randomly chosen PCOS patient above a randomly chosen control in about 85 percent of paired comparisons. The sensitivity figure suggests the marker would miss roughly one in five true cases at this cutoff, while the specificity figure implies a meaningful rate of false positives if used alone.

Perhaps the most clinically relevant result emerged when FABP4 was combined with the free androgen index. Adding FABP4 to FAI improved the model’s discrimination substantially, with a change in AUC of 0.133 that was statistically significant at p equals 0.001. Crucially, this improvement persisted even after adjustment for body mass index, indicating that FABP4 contributes metabolic information that is not merely a proxy for body fatness. The authors interpret this as evidence that FABP4 provides complementary metabolic context when read alongside androgen markers—a framing that is deliberately cautious and appropriate for a biomarker still in its validation infancy.

Indeed, the study’s limitations are clearly acknowledged and worth emphasizing. The cohort of 88 participants is modest, and the case-control design, while efficient, cannot establish whether elevated FABP4 precedes the development of PCOS or is a consequence of it. The authors themselves note that BMI-matched studies, investigations that characterize body composition in detail rather than relying on BMI alone, and external validation in independent cohorts are all needed before FABP4 can be considered a clinically generalizable adjunctive biomarker in adolescent PCOS. Prior literature in adults has been mixed on this point, with some studies finding FABP4 associations that dissolve once obesity is properly controlled, and the new adolescent data, while encouraging, do not yet resolve that uncertainty.

Nevertheless, the study represents a meaningful step forward in a field where objective tools are badly needed. Diagnosing PCOS in adolescents has real consequences: delayed diagnosis allows metabolic dysfunction—insulin resistance, dyslipidemia, and long-term cardiovascular risk—to progress unchecked, while premature diagnosis can lead to unnecessary treatment and anxiety during a life stage when physiology is still settling. A biomarker that captures the metabolic dimension of the syndrome, complements hormonal testing, and performs well in a teenage population could eventually help clinicians navigate this diagnostic tightrope with greater confidence. For now, FABP4 should be viewed as a promising signal rather than a ready-made test, but the convergence of adipose biology, insulin resistance, and ovarian function that it represents may prove to be one of the more illuminating threads in the ongoing effort to understand PCOS where it begins—in adolescence.

Subject of Research: Circulating FABP4 as a metabolic biomarker for adolescent polycystic ovary syndrome

Article Title: Circulating Fatty Acid-Binding Protein 4 (FABP4) in Adolescents with Polycystic Ovary Syndrome: Metabolic Correlates and Diagnostic Performance

Article References: Codal, B., Kurt, A., Kaplan, H., Çakır, A., Hızlı, B., & Özler, M. R. (2026). Circulating Fatty Acid-Binding Protein 4 (FABP4) in Adolescents with Polycystic Ovary Syndrome: Metabolic Correlates and Diagnostic Performance. Reproductive Sciences. https://doi.org/10.1007/s43032-026-02189-2

Image Credits: AI Generated

DOI: 10.1007/s43032-026-02189-2

Keywords: PCOS, FABP4, adolescents, biomarkers, insulin resistance, hyperandrogenism, adipokines, free androgen index, ROC analysis, metabolic syndrome, reproductive endocrinology, HOMA-IR