Exenatide Before Prandial Insulin? – Not Just Yet!

Written By

Dave L. Dixon, Pharm.D., BCPS, CDE

Reviewed By

Dawn Fuke, Pharm.D., BCPS
Stuart T. Haines, Pharm.D., BCPS, BCACP


Buse JB, Bergenstal RM, Glass LC, et al. Use of Twice-Daily Exenatide in Basal Insulin-Treated Patients With Type 2 Diabetes. Ann Intern Med 2011; 154: 103-112.

For diabetes patients uncontrolled on basal insulin therapy, is adding exenatide a good choice? Current treatment algorithms advocate for the addition of prandial insulin in patients who have not reached their glycemic targets with basal insulin.1 Exenatide is not FDA-approved for use in combination with basal insulin because current studies evaluating the combination of exenatide and basal insulin lack sufficient scientific vigor to provide reliable conclusions.2 This study sought to address this issue using a parallel, randomized, placebo-controlled trial design.3

The purpose of this study was to assess whether exenatide would reduce A1c levels more than placebo in patients on insulin glargine.   Participants included in this study had type 2 diabetes and an A1c between 7.1% and 10.5%, while receiving insulin glargine either as monotherapy or in combination with metformin, pioglitazone, or both. Patients were randomized to exenatide 10 mcg twice daily or placebo for 30 weeks. Those with baseline HbA1c levels > 8.0% remained on their current insulin dose, while participants with HbA1c levels ≤ 8.0% decreased their glargine dose by 20% to minimize the risk of hypoglycemia.  Glucometer readings were taken seven times a day by participants and insulin dose adjustments were made by the investigators using the Treat-to-Target Trial algorithm4 at least weekly from week 5 to 10, and bi-weekly thereafter. The primary outcome was change in A1c.

Of the 261 patients enrolled, 213(exenatide group, n=112; placebo group, n=101) patients completed the study. The mean age of the participants was 59 years, mean duration of diabetes was 12 years, and baseline A1c was 8.4%. Other baseline characteristics were similar with exception of gender and oral hypoglycemic use. More males were in the placebo group, 64% (vs. 51% in the exenatide group) and the use of metformin plus pioglitazone was higher in the exenatide group, 17% (vs. 7% in the exenatide group). The authors did, however, note that their results were independent of race, oral anti-hyperglycemic agent use, gender, or age.

Key findings from this study were:


Exenatide (n=137)

Placebo (n=122)



Change in A1c from baseline





A1c ≤7.0%





A1c ≤6.5%





Body weight

- 1.78 kg

0.96 kg

< 0.001


Compared to placebo, the exenatide group exhibited a greater decrease in A1c and had greater achievement of A1c goals of ≤7.0% and ≤6.5%.  Not surprising, weight decreased significantly in the exenatide group and increased in the placebo group. The number of hypoglycemic events per participant did not differ between the exenatide and placebo group, 1.4 vs. 1.2, respectively. Major nocturnal hypoglycemia was an issue in only one patient in the placebo group. Tolerability was a concern, however, as the drop out rate due to adverse events was 9% in the exenatide group vs. 1% in the placebo group. The most common adverse effects were nausea, diarrhea, vomiting, headache, and constipation.

Overall, the study results were not unexpected considering exenatide’s known effects on blood glucose and propensity to decrease weight.  What cannot be concluded from this study is whether adding exenatide is “better” than adding prandial insulin, because this was not an active comparator study.  Patient’s entering this study were not at goal on basal insulin therapy and the mean glargine dose was approximately 50 units per day.  What we can conclude from this study is that adding exenatide is more efficacious than continued titration of basal insulin. Given the dose of insulin glargine these patients were receiving and the mean duration of diabetes in the study participants, it is not surprising that up titration of glargine (which primarily addresses fasting blood glucose) improved glycemic control less than the addition of exenatide (which primarily addresses post-prandial blood glucose).

Many clinicians and patients have legitimate concerns about hypoglycemia when more intensive insulin therapy is needed.  This is one area where exenatide seems to have a notable advantage over prandial insulin.  There are, however, two major factors limiting the use of exenatide...affordability and tolerability.  Exenatide costs hundreds of dollars per month and requires a higher co-payment. In this study, one out of every 13 patients treated with exenatide discontinued therapy due to adverse effects. GLP-1 agonists are also associated with pancreatitis, renal impairment, and gastroparesis.

The big question - should clinicians recommend exenatide over prandial insulin in patients not at goal on basal insulin therapy? The evidence here shows that exenatide use does result in glycemic improvement and modest weight loss, without increasing the risk of hypoglycemia. However, when tolerability and cost are taken into consideration, prandial insulin remains the best overall choice in the majority of patients not yet at goal on basal insulin. Exenatide should be considered a second-line agent in patients who are poor candidates for prandial insulin, such as those at highest risk of hypoglycemia or with poor hypoglycemic awareness.

Have you started using exenatide with basal insulins instead of prandial insulin? Should the next diabetes treatment algorithm include exenatide as an alternative to prandial insulin? What are your thoughts?