Abstract #1398611: A High Chemerin: Adiponectin Ratio is Associated with Dysglycemia in Patients with the Polycystic Ovarian Syndrome (PCOS)

Abstract

PCOS is a multifactorial disease, which in turn increases the chance of progression to Type 2 diabetes (T2D) and metabolic syndrome. Since PCOS is now recognized as the ovarian manifestation of metabolic syndrome, identifying potential biomarkers predictive of glucose intolerance in a subset of PCOS could have important implications for the prevention of T2D in this young population. We observed impaired glucose tolerance (IGT) in PCOS to precede impaired fasting glucose (IFG), indicating impaired insulin-mediated peripheral glucose disposal. Two-hour post-glucose plasma glucose (2h-PGPG), the gold standard protocol to detect IGT, has its limitations, i.e., expensive, labor-intensive, and variability issues. Identifying a biomarker is therefore an unmet need for community studies. Chemerin and Adiponectin are two important organokines that are inversely associated with each other in people with metabolic syndrome (MetS). We looked at Chemerin: Adiponectin ratio as a possible biomarker for glucose intolerance in PCOS. Both fasting plasma glucose and 2h-PGPG were measured biochemically by the GOD-POD method. HbA1c level was measured by HPLC. Serum chemerin and adiponectin levels were measured by ELISA using commercial kits from Cusa-Bio. Serum c-peptide level was used to calculate HOMA2IR and HOMA2β, using ELISA kits from G-Biosciences. Subjects with PCOS with glucose intolerance (PCOS-IGT); median (interquartile range), 26.71 (10.19 - 45.09) [n = 44] showed higher serum chemerin: adiponectin ratio (CAR) compared to controls (PCOS-NGT); median (interquartile range), 6.977 (5.470 - 14.06). CAR correlated positively with 2h-PGPG (r = 0.5347, p < 0.01) and negatively with HOMA-2β (r = -0.5029, p < 0.05). A CAR cut-off value of 14.15 on the ROC curve provided the optimum trade-off between sensitivity and specificity (70.59% and 78.95% respectively). The same cut-off value showed sensitivity and specificity of 76.92%, with an AUC of 0.8964 and likelihood ratio of 3.333, in the case of comparison between PCOS-NGT and PCOS subjects having 2h-PGPG in the diabetic range (PCOS-T2D). With a power (1 – β) of 0.9258, CAR could identify 93% of the subjects having peak plasma glucose levels on OGTT ≥ 140 mg/dL. Moreover, CAR correlated strongly with TG: HDLc ratio (r = 0.62, p < 0.0001). These results suggest that CAR may be utilized as a biomarker for dysglycaemia in the study population. CAR could be a surrogate marker of dysglycaemia in PCOS. The strong correlation between CAR and TG: HDLc ratio is consistent with the premise that lipid abnormalities might precede glucose abnormalities in PCOS, like the situation in metabolic syndrome.