SAT-247 Two-Year Treatment with Metformin during Puberty Is Associated with Improvements in Body Composition, but Not in Insulin Sensitivity or β-Cell Function in Obese Youth

Background. The incidence of youth-onset type 2 diabetes in obese youth is tightly linked with the physiologic insulin resistance of puberty. The Health Influences of Puberty (HIP) Study was a placebo-controlled trial of metformin to test whether metformin mitigates the decrease in insulin sensitivity and β-cell function seen in obese youth during puberty. Methods. Obese (BMI ≥95%ile) non-diabetic youth (n=23) were randomized to metformin (n=11) or placebo (n=12) early in puberty (Tanner 2-3 [T2-3]) and treated until the end of puberty, Tanner 5 (T5) (mean treatment time 2.3+ 0.6 yrs). Exclusion criteria: pre-diabetes or diabetes, measured by oral glucose tolerance testing, obesity co-morbidities requiring medications, and medications affecting glucose metabolism. Study visits at baseline and treatment completion were performed following a 3-day energy-balanced diet and activity restriction and included intravenous glucose tolerance testing, fasting laboratory measures, anthropometrics, and body composition (DXA). Bergman’s minimal model was used to estimate insulin sensitivity (S(i)), insulin secretion (acute insulin response to glucose [AIR(g)]), and disposition index (DI). T-tests/chi-square were used to assess group differences in change (T5-T2/3) over time; linear regression models were used to adjust for sex, race/ethnicity, and baseline Tanner stage. Results: Baseline participant characteristics were: 65% female, 83% Hispanic, 70% T2, age 10.8+1.2 years, BMI z-score 2.3±0.35. There were no significant treatment group differences in baseline characteristics. Metformin treatment was associated with significant improvements in %body fat (-3.5±4.4 vs. 0.33±3.5%, p=0.04), and leptin (-4.1±13.7 vs. 18.2±17.7 ng/mL, p=0.003) and a smaller increase in waist circumference (2.21±10.32 vs. 14.1±7.87 cm, p=0.006). Metformin-associated improvement in BMI z-score was not statistically significant (-0.08±0.31 v. 0.15±0.13, p=0.20). There were no significant treatment group differences in insulin sensitivity or secretion; metformin vs. placebo: S(i) (0.26±3.2 v. -1.1±3.9x10(-4)/min(-1)/μIU/mL, p=0.37), AIR(g) (-397±1217 vs. -240±1151 μIU/mL, p=0.54), or DI (-1012±2326 vs. -1701±3261x10(-4)/min(-1), p=0.56). Conclusions: Two years of metformin treatment during puberty in obese youth was associated with improvements in body fat and leptin. However, these improvements were not associated with improved insulin sensitivity or β-cell function. Larger studies are needed to better understand potential metabolic benefits of metformin treatment in obese youth during puberty, a critical window of metabolic change.