Byetta results
In DURATION-1, mean weight change was -3.7 kg (~3.6%) at 30 wks at 2 mg weekly (Bydureon); in DURATION-1, A1c fell -1.9% at 30 wks. Below is what Byetta's own FDA label reports. Source.
Section 14, word for word, from the FDA label on DailyMed (Astrazeneca Pharmaceuticals LP), revised , version 26. Read every day: a new FDA version updates this page by itself.
What Byetta's clinical studies found
BYETTA has been studied as monotherapy and in combination with metformin, a sulfonylurea, a thiazolidinedione, a combination of metformin and a sulfonylurea, a combination of metformin and a thiazolidinedione, or in combination with insulin glargine with or without metformin and/or thiazolidinedione.
Monotherapy
In a randomized, double-blind, placebo-controlled trial of 24 weeks duration, BYETTA 5 mcg BID (n=77), BYETTA 10 mcg BID (n=78), or placebo BID (n=77) was used as monotherapy in patients with entry HbA1c ranging from 6.5% to 10%. All patients assigned to BYETTA initially received 5 mcg BID for 4 weeks. After 4 weeks, those patients either continued to receive BYETTA 5 mcg BID or had their dose increased to 10 mcg BID. Patients assigned to placebo received placebo BID throughout the trial. BYETTA or placebo was injected subcutaneously before the morning and evening meals. The majority of patients (68%) were White, 26% were West Asian, 3% were Hispanic or Latino ethnicity, 3% were Black or African American, and 0.4% were of East Asian ethnicity.
The primary endpoint was the change in HbA 1c from baseline to Week 24 (or the last value at time of early discontinuation). Compared to placebo, BYETTA 5 mcg BID and 10 mcg BID resulted in statistically significant reductions in HbA 1c from baseline at Week 24 (Table 7).
| Placebo BID | BYETTA 5 mcg BID | BYETTA 10 mcg BYETTA 5 mcg twice daily (BID) for 1 month followed by 10 mcg BID for 5 months before the morning and evening meals. BID | |
|---|---|---|---|
| Intent-to-Treat Population (N) | 77 | 77 | 78 |
| HbA1c (%), Mean | HbA1c (%), Mean | HbA1c (%), Mean | HbA1c (%), Mean |
| Baseline | 7.8 | 7.9 | 7.8 |
| Change at Week 24 Least squares means are adjusted for screening HbA1c strata and baseline value of the dependent variable. | −0.2 | −0.7 | −0.9 |
| Difference from placebo (95% CI) | −0.5 [−0.9, −0.2] p <0.01, treatment vs. placebo. | −0.7 [−1.0, −0.3] | |
| Proportion Achieving HbA1c <7% | 38% | 48% | 53% |
| Body Weight (kg), Mean | Body Weight (kg), Mean | Body Weight (kg), Mean | Body Weight (kg), Mean |
| Baseline | 86.1 | 85.1 | 86.2 |
| Change at Week 24 | −1.5 | −2.7 | −2.9 |
| Difference from placebo (95% CI) | −1.3 [−2.3, −0.2] | −1.5 [−2.5, −0.4] | |
| Fasting Serum Glucose Measured using the hexokinase-based glucose method. (mg/dL), Mean | Fasting Serum Glucose Measured using the hexokinase-based glucose method. (mg/dL), Mean | Fasting Serum Glucose Measured using the hexokinase-based glucose method. (mg/dL), Mean | Fasting Serum Glucose Measured using the hexokinase-based glucose method. (mg/dL), Mean |
| Baseline | 159 | 166 | 155 |
| Change at Week 24 | −5 | −17 | −19 |
| Difference from placebo (95% CI) | −12 [−23.2, −1.3] | −14 [−24.5, −2.5] | |
| BID = twice daily. | BID = twice daily. | BID = twice daily. | BID = twice daily. |
On average, there were no adverse effects of exenatide on blood pressure or lipids.
Combination Therapy with Oral Antihyperglycemic Medicines
Three 30-week, double-blind, placebo-controlled trials were conducted to evaluate the safety and efficacy of BYETTA in patients with type 2 diabetes whose glycemic control was inadequate with metformin alone, a sulfonylurea alone, or metformin in combination with a sulfonylurea. In addition, a 16-week, placebo-controlled trial was conducted where BYETTA was added to existing thiazolidinedione (pioglitazone or rosiglitazone) treatment, with or without metformin, in patients with type 2 diabetes with inadequate glycemic control.
In the 30-week trials, after a 4-week placebo lead-in period, patients were randomly assigned to receive BYETTA 5 mcg BID, BYETTA 10 mcg BID, or placebo BID before the morning and evening meals, in addition to their existing oral antidiabetic agent. All patients assigned to BYETTA initially received 5 mcg BID for 4 weeks. After 4 weeks, those patients either continued to receive BYETTA 5 mcg BID or had their dose increased to 10 mcg BID. Patients assigned to placebo received placebo BID throughout the study. A total of 1446 patients were randomized in the three 30-week trials: 991 (69%) were White, 224 (16%) were Hispanic or Latino and 174 (12%) were of Black of African American ethnicity. Mean HbA1c values at baseline for the trials ranged from 8.2% to 8.7%.
In the placebo-controlled trial of 16 weeks duration, BYETTA (n=121) or placebo (n=112) was added to existing thiazolidinedione (pioglitazone or rosiglitazone) treatment, with or without metformin. Randomization to BYETTA or placebo was stratified based on whether the patients were receiving metformin. BYETTA treatment was initiated at a dose of 5 mcg BID for 4 weeks then increased to 10 mcg BID for 12 more weeks. Patients assigned to placebo received placebo BID throughout the study. BYETTA or placebo was injected subcutaneously before the morning and evening meals. In this trial, 79% of patients were taking a thiazolidinedione and metformin and 21% were taking a thiazolidinedione alone. The majority of patients (84%) were White, 8% were Hispanic or Latino ethnicity, and 3% were of Black or African American ethnicity. The mean baseline HbA1c values were 7.9% for BYETTA and placebo.
The primary endpoint in each study was the mean change in HbA 1c from baseline to study end (or early discontinuation). Table 8 summarizes the study results for the 30- and 16-week clinical trials.
| Placebo BID | BYETTA 5 mcg BID | BYETTA 10 mcg BYETTA 5 mcg twice daily for 1 month followed by 10 mcg BID for 6 months for the 30-week trials or 10 mcg BID for 3 months in the 16-week trial before the morning and evening meals. BID | |
|---|---|---|---|
| In Combination with Metformin (30 Weeks) | In Combination with Metformin (30 Weeks) | In Combination with Metformin (30 Weeks) | |
| Intent-to-Treat Population (N) | 113 | 110 | 113 |
| HbA1c (%), Mean | HbA1c (%), Mean | HbA1c (%), Mean | HbA1c (%), Mean |
| Baseline | 8.2 | 8.3 | 8.2 |
| Change at Week 30 Least squares means are adjusted for baseline HbA1c strata or value, investigator site, baseline value of the dependent variable (if applicable), and background antihyperglycemic therapy (if applicable). | −0.0 | −0.5 | −0.9 |
| Difference from placebo (95% CI) | −0.5 [−0.7, −0.2] p <0.01, treatment vs. placebo. | −0.9 [−1.1, −0.6] | |
| Proportion Achieving HbA1c <7% | 12% | 32% | 40% |
| Body Weight (kg), Mean | Body Weight (kg), Mean | Body Weight (kg), Mean | Body Weight (kg), Mean |
| Baseline | 99.9 | 100.0 | 100.9 |
| Change at Week 30 | −0.2 | −1.3 | −2.6 |
| Difference from placebo (95% CI) | −1.1 [−2.2, −0.0] | −2.4 [−3.5, −1.3] | |
| Fasting Plasma Glucose Measured using the hexokinase-based glucose method. (mg/dL), Mean | Fasting Plasma Glucose Measured using the hexokinase-based glucose method. (mg/dL), Mean | Fasting Plasma Glucose Measured using the hexokinase-based glucose method. (mg/dL), Mean | Fasting Plasma Glucose Measured using the hexokinase-based glucose method. (mg/dL), Mean |
| Baseline | 169 | 176 | 168 |
| Change at Week 30 | +14 | −5 | −10 |
| Difference from placebo (95% CI) | −20 [−32, −7] | −24 [−37, −12] | |
| In Combination with a Sulfonylurea (30 Weeks) | In Combination with a Sulfonylurea (30 Weeks) | In Combination with a Sulfonylurea (30 Weeks) | |
| Intent-to-Treat Population (N) | 123 | 125 | 129 |
| HbA1c (%), Mean | HbA1c (%), Mean | HbA1c (%), Mean | HbA1c (%), Mean |
| Baseline | 8.7 | 8.5 | 8.6 |
| Change at Week 30 | +0.1 | −0.5 | −0.9 |
| Difference from placebo (95% CI) | −0.6 [−0.9, −0.3] | −1.0 [−1.3, −0.7] | |
| Proportion Achieving HbA1c <7% | 10% | 25% | 36% |
| Body Weight (kg), Mean | Body Weight (kg), Mean | Body Weight (kg), Mean | Body Weight (kg), Mean |
| Baseline | 99.1 | 94.9 | 95.2 |
| Change at Week 30 | −0.8 | −1.1 | −1.6 |
| Difference from placebo (95% CI) | −0.3 [−1.1, 0.6] | −0.9 [−1.7, −0.0] | |
| Fasting Plasma Glucose (mg/dL), Mean | Fasting Plasma Glucose (mg/dL), Mean | Fasting Plasma Glucose (mg/dL), Mean | Fasting Plasma Glucose (mg/dL), Mean |
| Baseline | 194 | 180 | 178 |
| Change at Week 30 | +6 | −5 | −11 |
| Difference from placebo (95% CI) | −11 [−25, 3] | −17 [−30, −3] | |
| In Combination with Metformin and a Sulfonylurea (30 Weeks) | In Combination with Metformin and a Sulfonylurea (30 Weeks) | In Combination with Metformin and a Sulfonylurea (30 Weeks) | |
| Intent-to-Treat Population (N) | 247 | 245 | 241 |
| HbA1c (%), Mean | |||
| Baseline | 8.5 | 8.5 | 8.5 |
| Change at Week 30 | +0.1 | −0.7 | −0.9 |
| Difference from placebo (95% CI) | −0.8 [−1.0, −0.6] | −1.0 [−1.2, −0.8] | |
| Proportion Achieving HbA1c <7% | 8% | 25% | 31% |
| Body Weight (kg), Mean | |||
| Baseline | 99.1 | 96.9 | 98.4 |
| Change at Week 30 | −0.9 | −1.6 | −1.6 |
| Difference from placebo (95% CI) | −0.7 [−1.2, −0.2] | −0.7 [−1.3, −0.2] | |
| Fasting Plasma Glucose (mg/dL), Mean | Fasting Plasma Glucose (mg/dL), Mean | Fasting Plasma Glucose (mg/dL), Mean | Fasting Plasma Glucose (mg/dL), Mean |
| Baseline | 181 | 182 | 178 |
| Change at Week 30 | +13 | −11 | −12 |
| Difference from placebo (95% CI) | −24 [−33, −15] | −25 [−34, −16] | |
| In Combination with a Thiazolidinedione or a Thiazolidinedione plus Metformin (16 Weeks) | In Combination with a Thiazolidinedione or a Thiazolidinedione plus Metformin (16 Weeks) | In Combination with a Thiazolidinedione or a Thiazolidinedione plus Metformin (16 Weeks) | |
| Intent-to-Treat Population (N) | 112 | Dose not studied | 121 |
| HbA1c (%), Mean | HbA1c (%), Mean | HbA1c (%), Mean | HbA1c (%), Mean |
| Baseline | 7.9 | Dose not studied | 7.9 |
| Change at Week 16 | +0.1 | Dose not studied | −0.7 |
| Difference from placebo (95% CI) | Dose not studied | −0.9 [−1.1, −0.7] | |
| Proportion Achieving HbA1c <7% | 15% | Dose not studied | 51% |
| Body Weight (kg), Mean | Body Weight (kg), Mean | Body Weight (kg), Mean | Body Weight (kg), Mean |
| Baseline | 96.8 | Dose not studied | 97.5 |
| Change at Week 16 | −0.0 | Dose not studied | −1.5 |
| Difference from placebo (95% CI) | Dose not studied | −1.5 [−2.2, −0.7] | |
| Fasting Serum Glucose (mg/dL), Mean | Fasting Serum Glucose (mg/dL), Mean | Fasting Serum Glucose (mg/dL), Mean | Fasting Serum Glucose (mg/dL), Mean |
| Baseline | 159 | Dose not studied | 164 |
| Change at Week 16 | +4 | Dose not studied | −21 |
| Difference from placebo (95% CI) | Dose not studied | −25 [−33, −16] | |
| BID = twice daily. | BID = twice daily. | BID = twice daily. | BID = twice daily. |
HbA1c
The addition of BYETTA to a regimen of metformin, a sulfonylurea, or both, resulted in statistically significant reductions from baseline in HbA 1c compared with patients receiving placebo added to these agents in the three controlled trials (Table 8).
In the 16-week trial of BYETTA add-on to thiazolidinediones, with or without metformin, BYETTA resulted in statistically significant reductions from baseline in HbA 1c compared with patients receiving placebo (Table 8).
Postprandial Glucose
Postprandial glucose was measured after a mixed meal tolerance test in 9.5% of patients participating in the 30-week add-on to metformin, add-on to sulfonylurea, and add-on to metformin in combination with sulfonylurea clinical trials. In this pooled subset of patients, BYETTA reduced postprandial plasma glucose concentrations in a dose-dependent manner. The mean (SD) change in 2-hour postprandial glucose concentration following administration of BYETTA at Week 30 relative to baseline was − 63 (65) mg/dL for 5 mcg BID (n=42), − 71 (73) mg/dL for 10 mcg BID (n=52), and +11 (69) mg/dL for placebo BID (n=44).
Combination with Insulin Glargine
-Week Placebo-Controlled Trial
A 30-week, double-blind, placebo-controlled trial was conducted to evaluate the efficacy and safety of BYETTA (n=137) versus placebo (n=122) when added to titrated insulin glargine, with or without metformin and/or thiazolidinedione, in patients with type 2 diabetes with inadequate glycemic control.
All patients assigned to BYETTA initially received 5 mcg BID for 4 weeks. After 4 weeks, those patients assigned to BYETTA had their dose increased to 10 mcg BID. Patients assigned to placebo received placebo BID throughout the trial. BYETTA or placebo was injected subcutaneously before the morning and evening meals. Patients with an HbA1c ≤8.0% decreased their prestudy dose of insulin glargine by 20% and patients with an HbA1c ≥8.1% maintained their current dose of insulin glargine. Five weeks after initiating randomized treatment, insulin doses were titrated with guidance from the investigator toward predefined fasting glucose targets according to the dose titration algorithm provided in Table 9. The majority of patients (78%) were White, 10% were American Indian or Alaska Native, 9% were Black or African American, 3% were Asian, and 0.8% were of multiple origins.
The primary endpoint was the change in HbA1c from baseline to Week 30. Compared to placebo, BYETTA 10 mcg BID resulted in statistically significant reductions in HbA1c from baseline at Week 30 (Table 9) in patients receiving titrated insulin glargine.
| Placebo BID + Titrated Insulin Glargine | BYETTA 10 mcg BYETTA 5 mcg twice daily for 1 month followed by 10 mcg BID for 5 months for the 30-week trial. BID + Titrated Insulin Glargine | |
|---|---|---|
| Intent-to-Treat Population (N) | 122 | 137 |
| HbA1c (%), Mean | HbA1c (%), Mean | HbA1c (%), Mean |
| Baseline | 8.5 | 8.3 |
| Change at Week 30 Least squares means are based on a mixed model adjusting for treatment, pooled investigator, visit, baseline HbA1c value, and treatment by visit, where subject is treated as a random effect. | −1.0 | −1.7 |
| Difference from placebo (95% CI) | −0.7 [−1.0, −0.5] p <0.01, treatment vs. placebo. | |
| Proportion Achieving HbA1c <7% | 29% | 56% |
| Body Weight (kg), Mean | Body Weight (kg), Mean | Body Weight (kg), Mean |
| Baseline | 93.8 | 95.4 |
| Change at Week 30 Least squares means are based on a mixed model adjusting for treatment, pooled investigator, visit, baseline HbA1c stratum, baseline value of the dependent variable (where applicable), and treatment by visit, where subject is treated as a random effect. | 1.0 | −1.8 |
| Difference from placebo (95% CI) | −2.7 [−3.7, −1.7] | |
| Fasting Serum Glucose Patients in both groups titrated insulin glargine dose to achieve optimal fasting glucose concentrations. (mg/dL), Mean | Fasting Serum Glucose Patients in both groups titrated insulin glargine dose to achieve optimal fasting glucose concentrations. (mg/dL), Mean | Fasting Serum Glucose Patients in both groups titrated insulin glargine dose to achieve optimal fasting glucose concentrations. (mg/dL), Mean |
| Baseline | 133 | 132 |
| Change at Week 30 | −16 | −23 |
| Difference from placebo (95% CI) | −7 [−18, 3] | |
| BID = twice daily. | BID = twice daily. | BID = twice daily. |
| Fasting Plasma Glucose Values (mg/dL) | Dose Change (U) |
|---|---|
| <56 Value for at least 1 fasting plasma glucose measurement since the last assessment. | − 4 |
| 56 to 72 | − 2 |
| 73 to 99 Based on the average of fasting plasma glucose measurements taken over the prior 3 to 7 days. The increase in the total daily dose should not have exceeded more than 10 units per day or 10% of the current total daily dose, whichever was greater. | 0 |
| 100 to 119 | +2 |
| 120 to 139 | +4 |
| 140 to 179 | +6 |
| ≥180 | +8 |
| Abbreviations: U = units. | Abbreviations: U = units. |
-Week Comparator-Controlled Noninferiority Trial
A 30 week, open-label, active comparator-controlled, noninferiority study was conducted to evaluate the safety and efficacy of BYETTA (n=315) versus titrated insulin lispro (n=312) on a background of optimized basal insulin glargine and metformin in patients with type 2 diabetes with inadequate glycemic control.
Following a 12-week basal insulin optimization (BIO) phase, subjects with an HbA 1c >7.0% entered a 30-week intervention phase and were randomized to add either BYETTA or insulin lispro to their existing regimen of insulin glargine and metformin. Insulin glargine was titrated to a target fasting plasma glucose of 72 to 100 mg/dL.
All patients assigned to BYETTA initially received 5 mcg BID for four weeks. After four weeks, their dose was increased to 10 mcg BID. Patients in the BYETTA-treated arm with an HbA 1c ≤8.0% at the end of the BIO phase decreased their insulin glargine dose by at least 10%.
All patients assigned to insulin lispro three times daily (TID) maintained their prior total daily insulin dose at baseline; however, the initial insulin lispro dose was ⅓ to ½ of the total daily insulin dose with the insulin glargine dose reduced accordingly. The insulin lispro dose was titrated based on preprandial glucose values.
The majority of patients (87%) were White, 7% were American Indian or Alaska Native, 5% were Asian, and <1% were of Black or African American ethnicity.
The primary endpoint was the change in HbA 1c from baseline to Week 30. Both BYETTA 10 mcg BID and titrated lispro provided a mean reduction in HbA 1c at Week 30 that met the pre-specified non-inferiority margin of 0.4%.
| Titrated Insulin Lispro TID + Titrated Insulin Glargine | BYETTA 10 mcgBYETTA 5 mcg BID for 1 month followed by 10 mcg BID for 5 months for the 30-week trial. BID + Titrated Insulin Glargine | |
| Intent-to-Treat Population (N) | 312 | 315 |
| HbA1c (%), Mean | HbA1c (%), Mean | HbA1c (%), Mean |
| Baseline | 8.2 | 8.3 |
| Change at Week 30 Least squares means are based on a mixed model adjusting for treatment, country, prior use of sulfonylurea (yes/no), visit, corresponding baseline, and treatment by visit interaction, where subject is treated as a random effect. Data at 30 weeks are available from 88% and 84% of the intent-to-treat subjects in the Lispro and BYETTA groups, respectively. | −1.1 | −1.1 |
| Difference from Insulin Lispro (95% CI) | −0.0 [−0.2, 0.1] | |
| Body Weight (kg), Mean | Body Weight (kg), Mean | Body Weight (kg), Mean |
| Baseline | 89.3 | 89.9 |
| Change at Week 30 | 1.9 | −2.6 |
| Difference from Insulin Lispro (95% CI) | −4.5 [−5.2, −3.9] | |
| Fasting Serum Glucose Patients titrated insulin glargine or insulin lispro dose to achieve prespecified target fasting and preprandial glucose concentrations. (mg/dL), Mean | Fasting Serum Glucose Patients titrated insulin glargine or insulin lispro dose to achieve prespecified target fasting and preprandial glucose concentrations. (mg/dL), Mean | Fasting Serum Glucose Patients titrated insulin glargine or insulin lispro dose to achieve prespecified target fasting and preprandial glucose concentrations. (mg/dL), Mean |
| Baseline | 126 | 130 |
| Change at Week 30 | 5 | −7 |
| Difference from Insulin Lispro (95% CI) | −12 [−19, −4] | |
| BID = twice daily. TID = three times daily. | BID = twice daily. TID = three times daily. | BID = twice daily. TID = three times daily. |
Source: Byetta prescribing information (FDA label on DailyMed), revised 2025-09-02
Byetta results: common questions
How much weight can you lose on Byetta?
In DURATION-1, people on 2 mg weekly (Bydureon) had a mean weight change of -3.7 kg (~3.6%) at 30 wks. That is an average across a trial, not a prediction for one person, and it depends on the dose reached and how long treatment continued.
How long does Byetta stay in your system?
Byetta's FDA label gives a half-life of 2.4 hours and no figure for how long it stays. By the usual rule of five half-lives, that is about 12 hours, which is our arithmetic and not the label's.
Can Byetta stop working?
Weight loss on these medicines levels off rather than continuing for good. The clearest measure comes from tirzepatide's trials: among adherent people who lost at least 5%, half had plateaued by about 24 to 36 weeks, depending on starting BMI. Talk to your prescriber if weight comes back while you are still taking it. When weight loss reaches a plateau
Trial averages do not predict what one person will experience.