Metformin used as adjuvant therapy in youth with T1D and elevated BMI decreased total daily insulin dose (TDID) and insulin resistance, although hemoglobin A1c (A1c) did not change. Yet, the factors that are associated with greater benefit from metformin are not well known. Here, we tested whether baseline measures predict response to metformin in youth with T1D and increased BMI-z score.

We analyzed publicly available data from the metformin treatment arm of the T1D Exchange Metformin Study (NCT01881828) , with 61 adolescents (median age 15.3 yrs, range 12-18.9; 36% males; 72% non-Hispanic Whites, 98% overweight or obese; median diabetes duration 7.4 yrs, range 1.7-15) . We assessed baseline age, sex, diabetes duration, race/ethnicity, BMI z-score, waist circumference, body fat percentage (BF%) measured by Dual-energy X-ray Absorptiometry (DXA) scan and serum adipokines. Outcome measures included changes in A1c, TDID and BF% at 13 and 26 weeks (wks) . After adjusting for baseline values, multiple linear regression models were used to identify predictors for the outcomes. Decreases in TDID at 13 and 26 wks showed a significant negative association with baseline leptin (p=0.0008, p=0.004, respectively) . Decreases in BF% at 26 wks were positively associated with baseline BMI z-score (p=0.03) and negatively associated with baseline BF% (p=0.0002) . Increase in A1c at 26 wks trended towards a positive association with baseline BF% (p=0.07) and a negative association with baseline adiponectin (p=0.06) . Compared to females, males showed a significant reduction in BF% at 26 wks (p<0.0001) and a trend towards a reduction in TDID at 13 wks (p=0.06) .In sum, in adolescents with T1D and elevated BMI treated with metformin as adjuvant to insulin, baseline leptin was the strongest predictor of decrease in TDID at 13 and 26 wks. Decrease in BF% was predicted by higher BMI z-score and lower BF% at baseline. These findings may aid a precision medicine approach in this population.


J.Wang: None. S.Barua: None. M.Tosur: Advisory Panel; Provention Bio, Inc. M.J.Redondo: Advisory Panel; Provention Bio, Inc. H.M.Ismail: n/a.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at