The coadministration of saxagliptin and metformin immediate-release tablets has been studied in adults with type 2 diabetes inadequately controlled on metformin alone and in treatment-naive patients inadequately controlled on diet and exercise alone. In these two trials, treatment with saxagliptin dosed in the morning plus metformin immediate-release tablets at all doses produced clinically relevant and statistically significant improvements in A1C, fasting plasma glucose (FPG), and 2-hour postprandial glucose (PPG) following a standard oral glucose tolerance test (OGTT), compared to control. Reductions in A1C were seen across subgroups including gender, age, race, and baseline BMI.
In these two trials, decrease in body weight in the treatment groups given saxagliptin in combination with metformin immediate-release was similar to that in the groups given metformin immediate-release alone. Saxagliptin plus metformin immediate-release was not associated with significant changes from baseline in fasting serum lipids compared to metformin alone.
The coadministration of saxagliptin and metformin immediate-release tablets has also been evaluated in an active-controlled trial comparing add-on therapy with saxagliptin to glipizide in 858 patients inadequately controlled on metformin alone, in a placebo-controlled trial where a subgroup of 314 patients inadequately controlled on insulin plus metformin received add-on therapy with saxagliptin or placebo, a trial comparing saxagliptin to placebo in 257 patients inadequately controlled on metformin plus a sulfonylurea, and a trial comparing saxagliptin to placebo in 315 patients inadequately controlled on dapagliflozin and metformin.
In a 24-week, double-blind, randomized trial, patients treated with metformin immediate-release 500 mg twice daily for at least 8 weeks were randomized to continued treatment with metformin immediate-release 500 mg twice daily or to metformin extended-release either 1,000 mg once daily or 1,500 mg once daily. The mean change in A1C from baseline to Week 24 was 0.1% (95% confidence interval 0%, 0.3%) for the metformin immediate-release treatment arm, 0.3% (95% confidence interval 0.1%, 0.4%) for the 1,000 mg metformin extended-release treatment arm, and 0.1% (95% confidence interval 0%, 0.3%) for the 1,500 mg metformin extended-release treatment arm. Results of this trial suggest that patients receiving metformin immediate-release treatment may be safely switched to metformin extended-release once daily at the same total daily dose, up to 2,000 mg once daily. Following a switch from metformin immediate-release to metformin extended-release, glycemic control should be closely monitored and dosage adjustments made accordingly.
Saxagliptin Morning and Evening Dosing
A 24-week monotherapy trial was conducted to assess a range of dosing regimens for saxagliptin. Treatment-naive patients with inadequately controlled diabetes (A1C ≥7% to ≤10%) underwent a 2-week, single-blind diet, exercise, and placebo lead-in period. A total of 365 patients were randomized to 2.5 mg every morning, 5 mg every morning, 2.5 mg with possible titration to 5 mg every morning, or 5 mg every evening of saxagliptin, or placebo. Patients who failed to meet specific glycemic goals during the study were treated with metformin rescue therapy added on to placebo or saxagliptin; the number of patients randomized per treatment group ranged from 71 to 74.
Treatment with either saxagliptin 5 mg every morning or 5 mg every evening provided significant improvements in A1C versus placebo (mean placebo-corrected reductions of −0.4% and −0.3%, respectively).
Coadministration of Saxagliptin with Metformin Immediate-Release in Treatment-Naive Patients
A total of 1,306 treatment-naive patients with type 2 diabetes mellitus participated in this 24-week, randomized, double-blind, active-controlled trial to evaluate the efficacy and safety of saxagliptin coadministered with metformin immediate-release in patients with inadequate glycemic control (A1C ≥8% to ≤12%) on diet and exercise alone. Patients were required to be treatment-naive to be enrolled in this study.
Patients who met eligibility criteria were enrolled in a single-blind, 1-week, dietary and exercise placebo lead-in period. Patients were randomized to one of four treatment arms: saxagliptin 5 mg + metformin immediate-release 500 mg, saxagliptin 10 mg + metformin immediate-release 500 mg, saxagliptin 10 mg + placebo, or metformin immediate-release 500 mg + placebo (the maximum recommended approved saxagliptin dose is 5 mg daily; the 10 mg daily dose of saxagliptin does not provide greater efficacy than the 5 mg daily dose and the 10 mg saxagliptin dosage is not an approved dosage). Saxagliptin was dosed once daily. In the 3 treatment groups using metformin immediate-release, the metformin dose was up-titrated weekly in 500 mg per day increments, as tolerated, to a maximum of 2,000 mg per day based on FPG. Patients who failed to meet specific glycemic goals during this study were treated with pioglitazone rescue as add-on therapy.
Coadministration of saxagliptin 5 mg plus metformin immediate-release provided significant improvements in A1C, FPG, and PPG compared with placebo plus metformin immediate-release (Table 7).
Table 7: Glycemic Parameters at Week 24 in a Placebo-Controlled Trial of Saxagliptin Coadministration with Metformin Immediate-Release in Treatment-Naive Patients*
| Efficacy Parameter | Saxagliptin 5 mg + Metformin N=320 | Placebo + Metformin N=328 |
| Hemoglobin A1C (%) | N=306 | N=313 |
| Baseline (mean) | 9.4 | 9.4 |
| Change from baseline (adjusted mean†) | −2.5 | −2.0 |
| Difference from placebo + metformin (adjusted mean†) | −0.5‡ | |
| 95% Confidence Interval | (−0.7, −0.4) | |
| Percent of patients achieving A1C < 7% | 60%§ (185/307) | 41% (129/314) |
| Fasting Plasma Glucose (mg/dL) | N=315 | N=320 |
| Baseline (mean) | 199 | 199 |
| Change from baseline (adjusted mean†) | −60 | −47 |
| Difference from placebo + metformin (adjusted mean†) | −13§ | |
| 95% Confidence Interval | (−19, −6) | |
| 2-hour Postprandial Glucose (mg/dL) | N=146 | N=141 |
| Baseline (mean) | 340 | 355 |
| Change from baseline (adjusted mean†) | −138 | −97 |
| Difference from placebo + metformin (adjusted mean†) | −41§ | |
| 95% Confidence Interval | (−57, −25) | |
* Intent-to-treat population using last observation on study or last observation prior to pioglitazone rescue therapy for patients needing rescue.
† Least squares mean adjusted for baseline value.
‡ p-value <0.0001 compared to placebo + metformin
§p-value <0.05 compared to placebo + metformin
Addition of Saxagliptin to Metformin Immediate-Release
A total of 743 patients with type 2 diabetes participated in this 24-week, randomized, double-blind, placebo-controlled trial to evaluate the efficacy and safety of saxagliptin in combination with metformin immediate-release in patients with inadequate glycemic control (A1C ≥7% and ≤10%) on metformin alone. To qualify for enrollment, patients were required to be on a stable dose of metformin (1,500 to 2,550 mg daily) for at least 8 weeks.
Patients who met eligibility criteria were enrolled in a single-blind, 2-week, dietary and exercise placebo lead-in period during which patients received metformin immediate-release at their pre-study dose, up to 2,500 mg daily, for the duration of the study. Following the lead-in period, eligible patients were randomized to 2.5 mg, 5 mg, or 10 mg of saxagliptin or placebo in addition to their current dose of open-label metformin immediate-release (the maximum recommended approved saxagliptin dose is 5 mg daily; the 10 mg daily dose of saxagliptin does not provide greater efficacy than the 5 mg daily dose and the 10 mg dosage is not an approved dosage). Patients who failed to meet specific glycemic goals during the study were treated with pioglitazone rescue therapy, added on to existing study medications. Dose titrations of saxagliptin and metformin immediate-release were not permitted.
Saxagliptin 2.5 mg and 5 mg add-on to metformin immediate-release provided significant improvements in A1C, FPG, and PPG compared with placebo add-on to metformin immediate-release (Table 8). Mean changes from baseline for A1C over time and at endpoint are shown in Figure 1. The proportion of patients who discontinued for lack of glycemic control or who were rescued for meeting prespecified glycemic criteria was 15% in the saxagliptin 2.5 mg add-on to metformin immediate-release group, 13% in the saxagliptin 5 mg add-on to metformin immediate-release group, and 27% in the placebo add-on to metformin immediate-release group.
Table 8: Glycemic Parameters at Week 24 in a Placebo-Controlled Study of Saxagliptin as Add-On Combination Therapy with Metformin Immediate-Release*
| Efficacy Parameter | Saxagliptin 2.5 mg + Metformin N=192 | Saxagliptin5 mg + Metformin N=191 | Placebo + Metformin N=179 |
| Hemoglobin A1C (%) | N=186 | N=186 | N=175 |
| Baseline (mean) | 8.1 | 8.1 | 8.1 |
| Change from baseline (adjusted mean†) | −0.6 | −0.7 | +0.1 |
| Difference from placebo (adjusted mean†) | −0.7‡ | −0.8‡ | |
| 95% Confidence Interval | (−0.9, −0.5) | (−1.0, −0.6) | |
| Percent of patients achieving A1C <7% | 37%§ (69/186) | 44%§ (81/186) | 17% (29/175) |
| Fasting Plasma Glucose (mg/dL) | N=188 | N=187 | N=176 |
| Baseline (mean) | 174 | 179 | 175 |
| Change from baseline (adjusted mean†) | −14 | −22 | +1 |
| Difference from placebo (adjusted mean†) | −16§ | −23§ | |
| 95% Confidence Interval | (−23, −9) | (−30, −16) | |
| 2-hour Postprandial Glucose (mg/dL) | N=155 | N=155 | N=135 |
| Baseline (mean) | 294 | 296 | 295 |
| Change from baseline (adjusted mean†) | −62 | −58 | −18 |
| Difference from placebo (adjusted mean†) | −44§ | −40§ | |
| 95% Confidence Interval | (−60, −27) | (−56, −24) | |
* Intent-to-treat population using last observation on study or last observation prior to pioglitazone rescue therapy for patients needing rescue.
† Least squares mean adjusted for baseline value.
‡ p-value < 0.0001 compared to placebo + metformin.
§ p-value < 0.05 compared to placebo + metformin.
Figure 1: Mean Change from Baseline in A1C in a Placebo-Controlled Trial of Saxagliptin as Add-On Combination Therapy with Metformin Immediate-Release*
*Includes patients with a baseline and week 24 value.
Week 24 (LOCF) includes intent-to-treat population using last observation on study prior to pioglitazone rescue therapy for patients needing rescue. Mean change from baseline is adjusted for baseline value.
Saxagliptin Add-On Combination Therapy with Metformin Immediate-Release versus Glipizide Add-On Combination Therapy with Metformin Immediate-Release
In this 52-week, active-controlled trial, a total of 858 patients with type 2 diabetes and inadequate glycemic control (A1C >6.5% and ≤10%) on metformin immediate-release alone were randomized to double-blind add-on therapy with saxagliptin or glipizide. Patients were required to be on a stable dose of metformin immediate-release (at least 1,500 mg daily) for at least 8 weeks prior to enrollment.
Patients who met eligibility criteria were enrolled in a single-blind, 2-week, dietary and exercise placebo lead-in period during which patients received metformin immediate-release (1,500 to 3,000 mg based on their prestudy dose). Following the lead-in period, eligible patients were randomized to 5 mg of saxagliptin or 5 mg of glipizide in addition to their current dose of open-label metformin immediate-release. Patients in the glipizide plus metformin immediate-release group underwent blinded titration of the glipizide dose during the first 18 weeks of the trial up to a maximum glipizide dose of 20 mg per day. Titration was based on a goal FPG ≤110 mg/dL or the highest tolerable glipizide dose. Fifty percent (50%) of the glipizide-treated patients were titrated to the 20-mg daily dose; 21% of the glipizide-treated patients had a final daily glipizide dose of 5 mg or less. The mean final daily dose of glipizide was 15 mg.
After 52 weeks of treatment, saxagliptin and glipizide resulted in similar mean reductions from baseline in A1C when added to metformin immediate-release therapy (Table 9). This conclusion may be limited to patients with baseline A1C comparable to those in the trial (91% of patients had baseline A1C <9%).
From a baseline mean body weight of 89 kg, there was a statistically significant mean reduction of 1.1 kg in patients treated with saxagliptin compared to a mean weight gain of 1.1 kg in patients treated with glipizide (p<0.0001).
Table 9: Glycemic Parameters at Week 52 in an Active-Controlled Trial of Saxagliptin versus Glipizide in Combination with Metformin Immediate-Release*
| Efficacy Parameter | Saxagliptin 5 mg + Metformin N=428 | Titrated Glipizide + Metformin N=430 |
| Hemoglobin A1C (%) | N=423 | N=423 |
| Baseline (mean) | 7.7 | 7.6 |
| Change from baseline (adjusted mean†) | −0.6 | −0.7 |
| Difference from glipizide + metformin (adjusted mean†) | 0.1 | |
| 95% Confidence Interval | (−0.02, 0.2)‡ | |
| Fasting Plasma Glucose (mg/dL) | N=420 | N=420 |
| Baseline (mean) | 162 | 161 |
| Change from baseline (adjusted mean†) | −9 | −16 |
| Difference from glipizide + metformin (adjusted mean†) | 6 | |
| 95% Confidence Interval | (2, 11)§ | |
* Intent-to-treat population using last observation on study.
† Least squares mean adjusted for baseline value.
‡ Saxagliptin + metformin is considered non-inferior to glipizide + metformin because the upper limit of this confidence interval is less than the prespecified non-inferiority margin of 0.35%.
§ Significance not tested.
Saxagliptin Add-On Combination Therapy with Insulin (with or without Metformin Immediate-Release)
A total of 455 patients with type 2 diabetes participated in this 24-week, randomized, double-blind, placebo-controlled trial to evaluate the efficacy and safety of saxagliptin in combination with insulin in patients with inadequate glycemic control (A1C ≥7.5% and ≤11%) on insulin alone (N=141) or on insulin in combination with a stable dose of metformin immediate-release (N=314). Patients were required to be on a stable dose of insulin (≥30 units to ≤150 units daily) with ≤20% variation in total daily dose for ≥8 weeks prior to screening. Patients entered the trial on intermediate- or long-acting (basal) insulin or premixed insulin. Patients using short-acting insulins were excluded unless the short-acting insulin was administered as part of a premixed insulin.
Patients who met eligibility criteria were enrolled in a single-blind, four-week, dietary and exercise placebo lead-in period during which patients received insulin (and metformin immediate-release if applicable) at their pretrial dose(s). Following the lead-in period, eligible patients were randomized to add-on therapy with either saxagliptin 5 mg or placebo. Doses of the antidiabetic therapies were to remain stable but patients were rescued and allowed to adjust the insulin regimen if specific glycemic goals were not met or if the investigator learned that the patient had self-increased the insulin dose by >20%. Data after rescue were excluded from the primary efficacy analyses.
Add-on therapy with saxagliptin 5 mg provided significant improvements from baseline to Week 24 in A1C and PPG compared with add-on placebo (Table 10). Similar mean reductions in A1C versus placebo were observed for patients using saxagliptin 5 mg add-on to insulin alone and saxagliptin 5 mg add-on to insulin in combination with metformin immediate-release (−0.4% and −0.4%, respectively). The percentage of patients who discontinued for lack of glycemic control or who were rescued was 23% in the saxagliptin group and 32% in the placebo group.
The mean daily insulin dose at baseline was 53 units in patients treated with saxagliptin 5 mg and 55 units in patients treated with placebo. The mean change from baseline in daily dose of insulin was 2 units for the saxagliptin 5 mg group and 5 units for the placebo group.
Table 10: Glycemic Parameters at Week 24 in a Placebo-Controlled Trial of Saxagliptin as Add-On Combination Therapy with Insulin*
| Efficacy Parameter | Saxagliptin 5 mg + Insulin (+/− Metformin) N=304 | Placebo + Insulin (+/− Metformin) N=151 |
| Hemoglobin A1C (%) | N=300 | N=149 |
| Baseline (mean) | 8.7 | 8.7 |
| Change from baseline (adjusted mean†) | −0.7 | −0.3 |
| Difference from placebo (adjusted mean†) | −0.4‡ | |
| 95% Confidence Interval | (−0.6, −0.2) | |
| 2-hour Postprandial Glucose (mg/dL) | N=262 | N=129 |
| Baseline (mean) | 251 | 255 |
| Change from baseline (adjusted mean†) | −27 | −4 |
| Difference from placebo (adjusted mean†) | −23§ | |
| 95% Confidence Interval | (−37, −9) | |
* Intent-to-treat population using last observation on study or last observation prior to insulin rescue therapy for patients needing rescue.
† Least squares mean adjusted for baseline value and metformin use at baseline.
‡ p-value <0.0001 compared to placebo + insulin.
§ p-value <0.05 compared to placebo + insulin.
The change in fasting plasma glucose from baseline to Week 24 was also tested, but was not statistically significant. The percent of patients achieving an A1C <7% was 17% (52/300) with saxagliptin in combination with insulin compared to 7% (10/149) with placebo. Significance was not tested.
Saxagliptin Add-On Combination Therapy with Metformin plus Sulfonylurea
A total of 257 subjects with type 2 diabetes participated in this 24-week, randomized, double-blind, placebo-controlled trial to evaluate the efficacy and safety of saxagliptin in combination with metformin plus a sulfonylurea in patients with inadequate glycemic control (A1C ≥7% and ≤10%). Patients were to be on a stable combined dose of metformin extended-release or immediate-release (at maximum tolerated dose, with minimum dose for enrollment being 1,500 mg) and a sulfonylurea (at maximum tolerated dose, with minimum dose for enrollment being ≥50% of the maximum recommended dose) for ≥8 weeks prior to enrollment.
Patients who met eligibility criteria were entered in a 2-week enrollment period to allow assessment of inclusion/exclusion criteria. Following the 2-week enrollment period, eligible patients were randomized to either double-blind saxagliptin (5 mg once daily) or double-blind matching placebo for 24 weeks. During the 24-week double-blind treatment period, patients were to receive metformin and a sulfonylurea at the same constant dose ascertained during enrollment. Sulfonylurea dose could be down titrated once in the case of a major hypoglycemic event or recurring minor hypoglycemic events. In the absence of hypoglycemia, titration (up or down) of study medication during the treatment period was prohibited.
Saxagliptin in combination with metformin plus a sulfonylurea provided significant improvements in A1C and PPG compared with placebo in combination with metformin plus a sulfonylurea (Table 11). The percentage of patients who discontinued for lack of glycemic control was 6% in the saxagliptin group and 5% in the placebo group.
Table 11: Glycemic Parameters at Week 24 in a Placebo-Controlled Trial of Saxagliptin as Add-On Combination Therapy with Metformin plus Sulfonylurea*
| Efficacy Parameter | Saxagliptin 5 mg + Metformin plus Sulfonylurea N=129 | Placebo + Metformin plus Sulfonylurea N=128 |
| Hemoglobin A1C (%) | N=127 | N=127 |
| Baseline (mean) | 8.4 | 8.2 |
| Change from baseline (adjusted mean†) | −0.7 | −0.1 |
| Difference from placebo (adjusted mean†) | −0.7‡ | |
| 95% Confidence Interval | (−0.9, −0.5) | |
| 2-hour Postprandial Glucose (mg/dL) | N=115 | N=113 |
| Baseline (mean) | 268 | 262 |
| Change from baseline (adjusted mean†) | −12 | 5 |
| Difference from placebo (adjusted mean†) | −17§ | |
| 95% Confidence Interval | (−32, −2) | |
* Intent-to-treat population using last observation prior to discontinuation.
† Least squares mean adjusted for baseline value.
‡ p-value <0.0001 compared to placebo + metformin plus sulfonylurea
§p-value <0.05 compared to placebo + metformin plus sulfonylurea.
The change in fasting plasma glucose from baseline to Week 24 was also tested, but was not statistically significant. The percent of patients achieving an A1C <7% was 31% (39/127) with saxagliptin in combination with metformin plus a sulfonylurea compared to 9% (12/127) with placebo. Significance was not tested.
Saxagliptin Add-on Combination Therapy with Metformin plus an SGLT2 Inhibitor
A total of 315 patients with type 2 diabetes participated in this 24-week randomized, double-blind, placebo-controlled trial to evaluate the efficacy and safety of saxagliptin added to dapagliflozin (an SGLT2 inhibitor) and metformin in patients with a baseline of HbA1c ≥7% to ≤10.5%. The mean age of these subjects was 54.6 years, 1.6% were 75 years or older and 52.7% were female. The population was 87.9% White, 6.3% Black or African American, 4.1% Asian, and 1.6% Other race. At baseline the population had diabetes for an average of 7.7 years and a mean HbA1c of 7.9%. The mean eGFR at baseline was 93.4 mL/min/1.73 m2. Patients were required to be on a stable dose of metformin (≥1,500 mg per day) for at least 8 weeks prior to enrollment. Eligible subjects who completed the screening period entered the lead in treatment period, which included 16 weeks of open-label metformin and 10 mg dapagliflozin treatment. Following the lead-in period, eligible patients were randomized to saxagliptin 5 mg (N=153) or placebo (N=162).
The group treated with add-on saxagliptin had statistically significant greater reductions in HbA1c from baseline versus the group treated with placebo (see Table 12).
Table 12: HbA1c Change from Baseline at Week 24 in a Placebo-Controlled Trial of Saxagliptin as Add-On to Dapagliflozin and Metformin§
| | Saxagliptin 5 mg (N=153)† | Placebo (N=162)† |
| | In combination with Dapagliflozin and Metformin |
| Hemoglobin A1C (%)* | | |
| Baseline (mean) | 8.0 | 7.9 |
Change from baseline (adjusted mean‡) 95% Confidence Interval | -0.5 (-0.6, -0.4) | -0.2 (-0.3, -0.1) |
Difference from placebo (adjusted mean) 95% Confidence Interval | -0.4¶ (-0.5, -0.2) | |
* Analysis of Covariance including all post-baseline data regardless of rescue or treatment discontinuation. Model estimates calculated using multiple imputation to model washout of the treatment effect using placebo data for all subjects having missing week 24 data.
† Number of randomized and treated patients.
‡ Least squares mean adjusted for baseline value.
§There were 6.5% (n=10) of randomized subjects in the saxagliptin arm and 3.1% (n=5) in the placebo arm for whom change from baseline HbA1c data was missing at week 24. Of the subjects who discontinued study medication early, 9.1% (1 of 11) in the saxagliptin arm and 16.7% (1 of 6) in the placebo arm had HbA1c measured at week 24.
¶ p-value <0.0001
The known proportion of patients achieving HbA1c <7% at Week 24 was 35.3% in the saxagliptin treated group compared to 23.1% in the placebo treated group.