OR07-3 Validation of Surrogate Models to Assess Tissue and Whole-Body Insulin Resistance Among High-Risk Adolescent Girls

Abstract

Polycystic Ovarian Syndrome (PCOS) is a common endocrine disorder which develops in adolescence and affects 6-10% of women. PCOS is associated with insulin resistance (IR), which can occur in multiple tissues, in particular skeletal muscle and liver. Reliable assessment of tissue-specific IR in youth with PCOS is important for development of new therapies. The gold-standard method to assess tissue-IR is the hyper-insulinemic euglycemic clamp, a method which is too intensive for use in routine research. We aimed to validate surrogate indices (the oral minimal model (OMM)-derived whole-body index and the Abdul-Ghani model-derived muscle index and liver index) against their respective clamp measurements in a population of high-IR-risk adolescent girls. 45 adolescent girls (14.6 ± 1.7 years; BMI %ile 23.3%-98.2%) underwent a standard 2-hour oral glucose tolerance test (OGTT) (75g glucose) and a multi-phase hyper-insulinemic euglycemic clamp (10,16 and 80 mU/m2/min) with a glucose isotope tracer. This study was a secondary analysis. OMM total Si was computed using SAAM II software, using 0, 15, 30, 60, 90, 120 min time points and the assumption that glucose rate of appearance (Ra) decays exponentially after 120 min. Abdul-Ghani muscle and liver insulin resistance indices (IRIs) were calculated: liver IRI= AUC0-30min Glucose (mmol L-1 min) x AUC0-30min Insulin (pmol L-1 min); muscle IRI= mean insulin0-120min. Correlation analyses were performed using Spearman’s or Pearson’s correlation, as appropriate. OMM total Si correlated with clamp-measured insulin sensitivity (glucose infusion rate, r=0.65; p<0.0001), whereas muscle IRI did not (p=0.45). Liver IRI correlated moderately with clamp-derived hepatic insulin sensitivity (insulin concentration required to suppress 50% of basal endogenous glucose production; r=0.35; p=0.03). In adolescent girls at high risk of IR, OMM provided a strong surrogate characterization of whole-body IR when evaluated against the clamp. The Abdul-Ghani model indices are simpler to calculate compared to OMM; however, these results suggest that the Abdul-Ghani indices may not be adequate for describing and distinguishing among individuals within a narrower IR range, such as that found in our study population. By contrast, OMM offers an effective, OGTT-based methodology to be used in research studies for characterizing whole-body IR that is less resource-intensive compared to the clamp. Future work is needed to determine if the sensitivity of this model can be further improved with a longer-duration OGTT as obese youth can have an exaggerated and prolonged response to an OGTT. Funding: Thrasher Research Foundation, AHA CRP, Endocrine Society Fellowship in Women's Health, NIDDK K23