Polycystic ovary syndrome (PCOS) is the most common endocrine disorder affecting 5-15 % of women of reproductive age worldwide, depending upon diagnostic criteria (Bozdag et al., 2016[7]; Salari et al., 2024[39]). It is characterized by a range of clinical features, including irregular menstrual cycles, anovulatory infertility, hyperandrogenism and hirsutism (Ehrmann, 2005[13]; Spritzer et al., 2022[44]). Beyond reproductive dysfunction, PCOS is also associated with an increased risk of metabolic comorbidities such as insulin resistance (IR), obesity, type 2 diabetes (T2D), metabolic syndrome and other cardiovascular disorders (Ehrmann, 2005[13]; Goodarzi et al., 2011[17]; Moran et al., 2015[29]; Zhu et al., 2019[55]).

A well-established connection in PCOS pathophysiology is its strong association with both IR and obesity. Studies indicate that 38-88 % of women with PCOS are classified as overweight or obese, and the majority (50-90 %) exhibit IR (Barber et al., 2016[4], 2019[5]). Given that IR is exacerbated by obesity, it is proposed that increased adipose tissue dysfunction in obesity leads to chronic inflammation, altered adipokine secretion and lipotoxicity, thus further aggravating IR though, of note, this pathway is largely mediated via pro-inflammatory cytokines such as tumor necrosis factor alpha (TNFα) and interleukin 6 (IL6) (Chait and Den Hartigh, 2020[9]; Hammarstedt et al., 2018[18]; Kahn and Flier, 2000[21]). In turn, IR plays a critical role in PCOS pathophysiology. Hyperinsulinemia, a compensatory mechanism of IR, activates the insulin signaling MAPK (Mitogen-Activated Protein Kinase) pathway, which regulates cellular growth, differentiation and steroidogenesis (Santarpia et al., 2012[40]). Specifically, hyperinsulinemia stimulates ovarian theca cells to produce excess androgens, disrupting follicular maturation and contributing to anovulation and menstrual irregularities, hallmark clinical features of PCOS (Barber et al., 2019[5]; White et al., 1995[51]). The frequent co-occurrence of obesity and PCOS, alongside their shared link to IR, underscores the complex interplay of metabolic and endocrine dysfunction in the syndrome.

However, while obesity and IR are central to PCOS pathology, a significant subset of women with PCOS do not present with obesity yet still experience menstrual irregularities, hyperandrogenism, hyperinsulinemia and metabolic disturbances (Carmina and Lobo, 2022[8]; Toprak et al., 2001[48]; Zhu et al., 2019[55]) This suggests that mechanisms beyond obesity-driven IR contribute to the syndrome. One potential target for further investigation is the insulin signaling MAPK pathway, specifically the Rat sarcoma (Ras) proteins, which serve as key regulators of cellular growth, differentiation and apoptosis through growth factor-mediated signaling (Simanshu et al., 2017[42]).

Previous research has provided evidence that circulating growth factors acting through Ras-mediated signaling pathways differ between individuals with PCOS and healthy controls at a BMI up to 29kg/m² (Niinuma et al., 2025[31]) with epidermal growth factor receptor (EGFR), fibroblast growth factor 8 (FGF8), fibroblast growth factor receptor 1 (FGFR1), vascular endothelial growth factor D (VEGFD) insulin like growth factor 1 (IGF1) and its receptor (IGF1sR) differing, but when subsequently stratifying for HOMA-IR, only FGFR1, VEGFD, IGF1 and IGF1sR differed between groups. These growth factors interact with specific cell surface receptors to activate Ras signaling cascades, influencing key processes such as follicular development, insulin signaling and metabolic regulation (Apte et al., 2019[3]; Kobayashi et al., 2024[22]; Orian-Rousseau et al., 2007[32]; Ornitz and Itoh, 2015[33]; Pandey et al., 2023[35]; Reichardt, 2006[36]; Surve et al., 2021[47]; Werner, 2023[50]). Upon binding to transmembrane receptors, these circulating growth factors initiate one of two major intracellular signaling pathways, which may contribute to the underlying mechanisms of PCOS, as illustrated in Figure 2(Fig. 2). Notably, endothelial dysfunction proteins differ in obese, weight-matched women with PCOS and persisted after BMI was stratified (Niinuma et al., 2025[31]), highlighting the need to identify PCOS-related features that are independent of obesity in matched cohorts.

To address this gap, our study focused on non-obese (BMI 20-25 kg/m²), weight-matched women with and without PCOS, to investigate the role of Ras activation in the absence of obesity-driven metabolic dysfunction. We hypothesized that non-obese, insulin-sensitive and without overt systemic inflammation (normal CRP) weight-matched women with and without PCOS would not differ and would not show a difference in plasma growth factors that activate Ras.

Demographic, hormonal and metabolic data for the 44 PCOS subjects and 78 controls are shown in Table 1(Tab. 1). The two cohorts were weight and age-matched and had neither IR (as adjudged by HOMA-IR) nor evidence of increased systemic inflammation, as CRP (mg L^-1) was not elevated (normal range 0-3 mg L^-1) and did not differ between the groups; subjects with PCOS did have hyperandrogenemia and elevated AMH (both p<0.0001 versus controls).

The non-obese cohort of PCOS patients recruited for this study was matched to controls for BMI, IR and CRP. We hypothesized that, with these parameters being normal, there would be no difference in growth factor-related protein markers signaling through Ras proteins. The results here showed that VEGFA and EGFR were increased, whilst EGFR1 was decreased in PCOS after correcting for multiple comparisons. EGFR, EGFR1 and VEGFA are tightly connected in endothelial, metabolic and ovarian signaling pathways (Lee et al., 2025[25]). The elevation in soluble EGFR may be explained by increased ectodomain shedding mediated by ADAM10/17 metalloproteinases, which are strongly activated by inflammatory cytokines, hyperinsulinemia and oxidative stress found in PCOS, obesity and insulin resistant states (Blobel, 2005[6]; Malaguarnera et al., 2025[28]). EGFR shedding may raise soluble EGFR levels while reducing membrane-bound EGFR1, that may be the explanation for the findings reported here. Similar dissociation between circulating EGFR and tissue EGFR1 has been reported in metabolic syndrome, early diabetes and endothelial injury models (Shraim et al., 2021[41]).

In comparison with our previous study, where BMI was stratified up to 29 kg/m² (Niinuma et al., 2025[31]), EGFR differed in both studies; fibroblast growth factor 8 (FGF8), FGFR1, IGF1 differed in the previous study and here they were either significantly different (IGF1, p=0.05) or showed a trend (FGF8, FGFR1) towards significance (0.06 and 0.07, respectively), but none remained significant after FDR testing. VEGFD, and IGF1sR, that differed previously (18), could not be compared in this study as the SOMA panel used for the samples in this study did not contain either. The major difference between the studies was that here the BMI was 20-25 kg/m² in comparison to a BMI of up to 29 kg/m² (18), and the differences seen are likely due to BMI. In addition, in the previously reported study, all the PCOS patients were phenotype A (irregular menses with hyperandrogenism and polycystic ovaries on ultrasound). In this study, 15 of the PCOS patients were phenotype A, whilst the remainder were a mixture of phenotype B or C: phenotype B (irregular menses with hyperandrogenism); phenotype C (irregular menses and polycystic ovaries on transvaginal scanning). There were too few subjects, and less than the power analysis would allow, to perform a subgroup analysis.

VEGFA was elevated in PCOS patients in comparison to controls as has previously been reported (Agrawal et al., 1998[2], 2002[1]; Dambala et al., 2017[12], 2019[11]). VEGF is a pro-angiogenic factor whose increase can contribute to hyperplasia, hypervascularity and the gradual development of endothelial dysfunction, key features of PCOS (Ferrara et al., 2003[14]). However, age and PCOS phenotype may affect these growth factors; a study investigating adolescent PCOS patients found no significant differences in VEGF levels between cases and controls, suggesting that VEGF expression may vary depending on the development of PCOS and its subpopulations (Skrzynska et al., 2024[43]). VEGFA is expressed in granulosa and theca cells from early stages and rises as follicles grow, promoting angiogenesis in the thecal vasculature for the enlarging follicle, and further increases as a dominant follicle develops (Fraser, 2006[15]; Jankowska-Ziemak et al., 2025[19]). In PCOS, VEGFA rises not only as an endothelial response but also due to ovarian overproduction: hyperinsulinemia, elevated LH, and local oxidative stress in granulosa and theca cells stimulate VEGFA transcription, contributing to stromal hypervascularity and abnormal folliculogenesis (Li et al., 2020[27]). Clinically, VEFGA is thought to contribute to the development of ovarian hyperstimulation syndrome during controlled ovarian stimulation (Ferrara et al., 2003[14]). Ovarian cells express vascular endothelial growth factor receptor 1 (VEGFR1/FLT1) and vascular endothelial growth factor receptor 2 (VEGFR2/KDR) (Ortega et al., 2015[34]), with VEGFR2 meditating most of the mitogenic/permeability effects of VEFGA. In this study, the increase of VEFGA was highly significant but in real terms there was only a 1.2-fold increase that, on its own, is unlikely to be clinically significant but, in combination with other factors, may have clinical relevance.

The present study showed significantly increased EGFR and reduced EGFR1 levels in women with PCOS. The EGFR axis is essential for normal folliculogenesis, granulosa cell proliferation, and the coordination of LH-mediated oocyte maturation. Altered expression of EGFR family signaling components has been associated with defective ovulatory responses in both human and experimental models (Zhang et al., 2022[54]).

In healthy follicles, activation of EGF-family ligands triggers the EGFR-extracellular signal-related protein kinases 1 and 2 (ERK1/2) signaling cascade required for cumulus-oocyte complex expansion and luteinizing hormone (LH)-induced resumption of meiosis. Decreased EGFR or reduced phosphorylation of EGFR disrupts these downstream pathways, leading to insufficient mitogen-activated protein kinase (MAPK) activation and impaired oocyte maturation (Richani and Gilchrist, 2018[37]). A reduction in EGFR signaling may reflect in a functional deficiency in peri-ovulatory signaling events necessary for follicular progression and ovulation (Sugimura et al., 2018[45]).

Impaired EGFR function has also been linked to abnormal follicular angiogenesis. EGFR-mediated pathways normally cross-regulate angiogenic factors such as VEGFA; when EGFR signaling is reduced, compensatory upregulation of VEGFA may occur, that may be the mechanism seen here, and may suggest a decoupling of angiogenic and growth-factors within the PCOS ovary. Overall, the increased EGFR and decreased EGFR1 in PCOS may indicate an increase in angiogenic drive while EGF-mediated follicular maturation signaling declines, resulting in suboptimal oocyte development (Yang et al., 2016[53]).

These findings suggest that growth factor-related proteins involved in Ras signaling differ among non-obese women with PCOS. This suggests that differences in Ras-mediated signaling specifically for VEGFA, EGF and EGFR may be inherent in PCOS, whilst the other Ras-mediated signaling growth factors observed in prior studies may be primarily driven by obesity, chronic inflammation or IR, or a combination of these parameters, rather than PCOS itself. Hypothetically, even a modest increase in weight, above what is considered normal, could lead to more pronounced Ras activation in women with PCOS compared to their non-PCOS counterparts at the same weight.

From a translational perspective, EGFR-pathway dysregulation may represent a potential therapeutic target. Restoration of EGF-family signaling has been suggested as a potential strategy to improve oocyte competence in PCOS undergoing assisted reproduction (Vitale and Dolmans, 2024[49]).

Strengths of this study include that the cohort was a well-defined non-obese non-IR Caucasian PCOS population closely matched to controls, allowing comparison of a panel of growth factor-related protein markers signaling through Ras, without the bias of BMI, IR or chronic inflammation. Limitations include the small population size; however, the power analysis (though based upon FGFR1 changes between PCOS and controls) suggested that the study was adequately powered to reveal differences, but the findings should be treated with caution as exploratory, with further validation studies in larger cohorts being necessary. That the study was performed on an exclusively Caucasian population of non-obese women many of who were attending an IVF clinic, and who were thus potentially subfertile, may also be seen as a limitation and would need to be repeated in other ethnic groups, other BMI strata, and in women known not to be subfertile, are needed to confirm these findings. The use of CRP as the only measure of systemic inflammation is a limitation, as other more sensitive inflammatory markers may have shown elevations despite a CRP in the normal range. The use of HOMA-IR as the only measure of insulin resistance is a limitation, as this is an indirect measure and lacks the sensitivity of a euglycemic clamp, though multiple studies have shown a strong, statistically significant correlation between HOMA and clamp measurements (Lansang et al., 2001[24]). The cross-sectional study design can be viewed as a limitation, as temporal relationships between VEGFA, EGF and EGFR cannot be determined, nor can changes in these proteins be a cause or effect of the PCOS condition.

The EGFR and VEGF pathways interact closely and when EGFR signaling decreases, VEGFA may increase to maintain angiogenic balance, suggesting that in non-obese PCOS there may be a signal for compensatory angiogenesis in a dysfunctional endothelial environment.

Sara Anjum Niinuma, Haniya Habib, and Ashleigh Suzu-Nishio Takemoto contributed equally as first author.