XENical in the prevention of diabetes in obese subjects (XENDOS) study: a randomized study of orlistat as an adjunct to lifestyle changes for the prevention of type 2 diabetes in obese patients.

OBJECTIVE It is well established that the risk of developing type 2 diabetes is closely linked to the presence and duration of overweight and obesity. A reduction in the incidence of type 2 diabetes with lifestyle changes has previously been demonstrated. We hypothesized that adding a weight-reducing agent to lifestyle changes may lead to an even greater decrease in body weight, and thus the incidence of type 2 diabetes, in obese patients. RESEARCH DESIGN AND METHODS In a 4-year, double-blind, prospective study, we randomized 3,305 patients to lifestyle changes plus either orlistat 120 mg or placebo, three times daily. Participants had a BMI >/=30 kg/m2 and normal (79%) or impaired (21%) glucose tolerance (IGT). Primary endpoints were time to onset of type 2 diabetes and change in body weight. Analyses were by intention to treat. RESULTS Of orlistat-treated patients, 52% completed treatment compared with 34% of placebo recipients (P < 0.0001). After 4 years' treatment, the cumulative incidence of diabetes was 9.0% with placebo and 6.2% with orlistat, corresponding to a risk reduction of 37.3% (P = 0.0032). Exploratory analyses indicated that the preventive effect was explained by the difference in subjects with IGT. Mean weight loss after 4 years was significantly greater with orlistat (5.8 vs. 3.0 kg with placebo; P < 0.001) and similar between orlistat recipients with impaired (5.7 kg) or normal glucose tolerance (NGT) (5.8 kg) at baseline. A second analysis in which the baseline weights of subjects who dropped out of the study was carried forward also demonstrated greater weight loss in the orlistat group (3.6 vs. 1.4 kg; P < 0.001). CONCLUSIONS Compared with lifestyle changes alone, orlistat plus lifestyle changes resulted in a greater reduction in the incidence of type 2 diabetes over 4 years and produced greater weight loss in a clinically representative obese population. Difference in diabetes incidence was detectable only in the IGT subgroup; weight loss was similar in subjects with IGT or NGT [correction].