Study was performed with Baylor College of Medicine Institutional Review Board approval, and informed consent was obtained in accordance with federal/institutional guidelines.

Subjects with type 1 diabetes using multiple daily injections or insulin pump, aged 13–22 years, with diabetes for ≥1 year, BMI <90th percentile for age, hemoglobin >12 g/dl, and hemoglobin A1C <8.5% were recruited. One subject had treated hypothyroidism with no other chronic conditions. Subjects were not on any concomitant medications, which affected blood glucose concentrations. Pregnant/lactating females were excluded. Two subjects failed screening. One subject had hypoglycemia needing rescue glucose boluses with insulin alone and did not participate in further studies.

Eight subjects completed the three-part study. Followed by a baseline study with insulin alone, subjects were randomized to two different doses of exenatide (1.25 and 2.5 μg), administered in a double-blinded randomized controlled manner. Studies were at least 3 weeks apart. All subjects had type 1 diabetes (antibody positive), with minimal or no endogenous C-peptide response to meals. (A pilot study done previously suggested that higher doses [7–10 μg] of exenatide used in normal BMI subjects with type 1 diabetes resulted in hypoglycemia, and hence the lowered doses were examined.)

Subjects were 17 ± 1 years (range 15.8–18), had diabetes for 5 ± 3.3 years, weighed 67 ± 8.7 kg, had a BMI of 23.8 ± 2.1 kg/m², had an A1C of 7.4 ± 0.7%, and had a total daily insulin dose of 0.9 ± 0.2 units/kg/day.

Postprandial hyperglycemia was reduced with 1.25 and 2.5 μg adjunctive exenatide versus insulin monotherapy (P < 0.0001) (Fig. 1). Delta plasma glucose area under the curve (AUC_0–120) was reduced in the early postprandial period in studies with 1.25 μg (49 ± 156 mmol/l per min) (P < 0.008) and 2.5 μg (44 ± 281 mmol/l per min) exenatide versus insulin alone (379 ± 259 mmol/l per min) (P < 0.007).

Gastric emptying as measured by ¹³CO₂ in breath was significantly delayed with 1.25 μg exenatide versus insulin alone (P < 0.004) and 2.5 μg exenatide when compared with insulin monotherapy (P < 0.0001).

Glucagon and C-peptide concentrations were not statistically different between studies using exenatide versus insulin alone (P < 0.1 and P < 0.06, respectively). GLP-1 was lower with 2.5 μg exenatide compared with insulin (P < 0.0001) but not with 1.25 μg exenatide (P < 0.2).

Insulin levels were lower between exenatide groups versus insulin alone (P < 0.0001), as expected with a 20% reduction in insulin.

Our study demonstrates reduction in postprandial glucose after exenatide injection in adolescents with type 1 diabetes. These effects were associated with delayed gastric emptying, and the results are consistent with published studies in adults (¹³,¹⁴). However, as opposed to previous reports, glucagon suppression was not noted with exenatide, which could be because of the small sample size of our study.

Both exenatide doses were comparable in reducing postprandial glucose excursions, and hence the lower dose could be tested as an initial dose and titrated to response and tolerability in a larger cohort of subjects. Pharmacokinetic data are included in the online appendix (available at http://care.diabetesjournals.org/cgi/content/full/dc09-1959/DC1), and data are comparable to adults. Caution must be exercised if blood glucose is low to omit exenatide or use lowered insulin and/or exenatide dose.

In conclusion, adjunctive exenatide therapy has therapeutic potential in adolescents with type 1 diabetes. Further studies are ongoing using exenatide for a 4-month period.

Clinical trial reg. no. NCT00456300, clinicaltrials.gov.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

R.A.H. received a grant from the National Institutes of Health (grant R01DK077166-01). R.A.H. is a McNair scholar and was supported by The Robert and Janice McNair Foundation, McNair Scholars Program at Texas Children's Hospital. L.M.R. was supported by grant K23 DK075931. This study was also supported by General Clinical Research Center (GCRC) grant RR00188.

No potential conflicts of interest relevant to this article were reported.

This work was presented as an oral presentation at the 2009 American Diabetes Association Annual Meeting, New Orleans, Louisiana, 5–9 June 2009.

We would like to thank Thanh Nguyen, MD, and staff of the GCRC and Texas Children's Hospital's investigational pharmacy for their assistance in conducting this study.