• beta-cell function;
  • clinical trial;
  • intensive insulin therapy

Aim: Studies evaluating the effects of oral antidiabetic drugs (OADs) on beta-cell function in type 2 diabetes mellitus (T2DM) are confounded by an inability to establish the actual baseline degree of beta-cell dysfunction, independent of the deleterious effects of hyperglycaemia (glucotoxicity). Because intensive insulin therapy (IIT) can induce normoglycaemia, we reasoned that short-term IIT could enable evaluation of the beta-cell protective capacity of OADs, free from confounding hyperglycaemia. We applied this strategy to assess the effect of sitagliptin on beta-cell function.

Methods: In this pilot study, 37 patients with T2DM of 6.0 + 6.4 years duration and A1c 7.0 + 0.8% on 0–2 OADs were switched to 4–8 weeks of IIT consisting of basal detemir and premeal insulin aspart. Subjects achieving fasting glucose <7.0 mmol/l 1 day after completing IIT (n = 21) were then randomized to metformin with either sitagliptin (n = 10) or placebo (n = 11). Subjects were followed for 48 weeks, with serial assessment of beta-cell function [ratio of AUCCpep to AUCgluc over Homeostasis Model Assessment of Insulin Resistance (HOMA-IR) (AUCCpep/gluc/HOMA-IR)] on 4-h meal tests.

Results: During the study, fasting glucagon-like-peptide-1 was higher (p = 0.003) and A1c lower in the sitagliptin arm (p = 0.016). Nevertheless, although beta-cell function improved during the IIT phase, it declined similarly in both arms over time (p = 0.61). By study end, AUCCpep/gluc/HOMA-IR was not significantly different between the placebo and sitagliptin arms (median 71.2 vs 80.4; p = 0.36).

Conclusions: Pretreatment IIT can provide a useful strategy for evaluating the beta-cell protective capacity of diabetes interventions. In this pilot study, improved A1c with sitagliptin could not be attributed to a significant effect on preservation of beta-cell function.