Sitagliptin: new drug. Type 2 diabetes: limited efficacy, too many unknown risks.

(1) Many drugs are available for the treatment of type 2 diabetes, but only metformin and glibenclamide have a proven impact on morbidity and mortality outcomes (only morbidity in the case of glibenclamide). If monotherapy with one of these drugs is inadequately effective, there is a choice of abandoning strict glycaemic control, combining the two drugs, or adding insulin. (2) Sitagliptin, a glucose-lowering inhibitor of DPP-4 (dipeptidyl dipeptidase 4), the enzyme responsible for catabolising physiological incretins, is the latest addition to the list of oral glucose-lowering drugs. (3) Sitagliptin has not been tested for its effect on morbidity or mortality endpoints. (4) Five placebo-controlled trials lasting from 18 to 24 weeks have evaluated sitagliptin monotherapy (3 trials), sitagliptin combined with metformin, or sitagliptin combined with pioglitazone. These trials showed that sitagliptin induced a limited reduction in glycated haemoglobin levels, which usually remained above the cutoff point (7%) generally used to define proper glycaemic controlled. (5) A trial comparing sitagliptin + metformin versus glipizide + metformin, and a direct comparison of sitagliptin versus glipizide, provided a too low level of evidence to convincingly demonstrate the non-inferiority of the sitagliptin combination. (6) In one trial the sitagliptin + metformin combination was significantly more effective on glycated haemoglobin levels than either drug used alone. (7) In the short term, the main adverse effects of sitagliptin are nausea and constipation. In the long term, there is a risk of infections, especially upper respiratory tract infections. Cases of depression have also been reported. Sitagliptin sometimes increases creatinine levels. Pharmacological data suggest there might be an increased risk of cancer and muscular and neurological disorders. (8) In summary, whether used alone or in combination, the antidiabetic effects of sitagliptin, so far studied on surrogate endpoints, are too modest, given the outstanding safety issues, to recommend its use in patients with type 2 diabetes.