Saxagliptin, the active substance in Onglyza, has been shown to be more effective than placebo (a dummy treatment) at controlling blood glucose in 8 main studies in over 3,900 patients. In these studies saxagliptin was used as an ‘add-on’ to other diabetes medicines in patients in whom previous treatment had failed. The main measure of effectiveness was the reduction in blood levels of a substance called glycosylated haemoglobin (HbA1c) after 24 weeks of treatment. This gives an indication of how well the blood glucose is controlled. Results showed that:

• In 743 patients not satisfactorily controlled on metformin alone, HbA1c levels fell by around 0.7 percentage points (from around 8.1% to around 7.4%) in patients adding saxagliptin compared with a rise of around 0.1 percentage points in patients adding placebo.
• In 768 patients not satisfactorily controlled on a sulphonylurea, HbA1c levels fell by around 0.6 percentage points in patients adding saxagliptin compared with a rise of around 0.1 percentage points in patients who added placebo.
• In 565 patients not satisfactorily controlled on a thiazolidinedione (diabetes medicines such as pioglitazone and rosiglitazone), HbA1c levels fell by around 0.9 percentage points in patients adding saxagliptin compared with a fall of around 0.3 percentage points in patients who added placebo.
• In 457 patients not satisfactorily controlled on insulin (with or without metformin), HbA1c levels fell by around 0.7 percentage points in patients adding saxagliptin compared with a fall of around 0.3 percentage points in patients adding placebo.
• In 257 patients who took saxagliptin with metformin and a sulphonylurea, HbA1c levels fell by around 0.7 percentage points, compared with a fall of 0.1 percentage points in patients who were given placebo in place of saxagliptin.
• In 534 patients not satisfactorily controlled with metformin alone, adding saxagliptin reduced HbA1c levels by around 0.9 percentage points and adding saxagliptin and dapagliflozin reduced HbA1c levels by 1.5 percentage points. Adding dapagliflozin to metformin reduced HbA1c levels by 1.2 percentage points. HbA1c levels were on average around 9% at the start of the study.
• In 315 patients not satisfactorily controlled with metformin and dapagliflozin, adding saxagliptin to treatment with dapagliflozin and metformin reduced HbA1c levels by 0.5 percentage points, compared with a reduction of 0.2 percentage points when placebo was added to dapagliflozin and metformin. HbA1c levels were around 8% at the start of the study.
• In 320 patients not controlled with metformin and saxagliptin, adding dapagliflozin to saxagliptin and metformin reduced HbA1c levels by 0.8 percentage points, compared with a reduction of 0.1 percentage points when placebo was added to saxagliptin and metformin.

Saxagliptin given on its own has also been shown to be more effective than placebo at controlling blood glucose in four studies: in patients given saxagliptin HbA1c levels fell by around 0.5 percentage points more than in patients given placebo.

A further study compared saxagliptin with metformin in patients who had not previously received substantial treatment with diabetes medicines. However, the results were not considered to be clinically relevant and the company withdrew its application for the use of saxagliptin as an initial combination medicine in previously untreated patients.