Comparing New Diabetes Drugs

Dean L.

Publication Details

More than 20 million Americans have diabetes. Type 1 diabetes is treated with insulin whereas a variety of oral medications may be used to lower the blood sugar levels in people with type 2 diabetes. Oral medications include sulfonylureas and thiazolidiones. About 70% of people with type 2 diabetes will eventually require more than one oral agent, with or without insulin.

Three new medications which target different receptors and hormones to those of "traditional" oral agents have recently been approved. Exenatide, sitagliptin, and pramlintide have been approved to treat type 2 diabetes, and pramlintide has also been approved as an add-on therapy to insulin to treat type 1 diabetes.

The "Drug Class Review on Newer Drugs for the Treatment of Diabetes Mellitus" compares the safety and effectiveness of three drugs. A summary of the findings is below.

In type 1 diabetes, how does adding pramlintide to insulin therapy compare to using insulin alone?

A marker of long-term blood glucose control is HbA1c (glycated hemoglobin). A high level of HbA1c indicates poor control of diabetes over the previous 2 to 3 months.

In adults using pramlintide (30 or 60 mcg per meal) added on to flexibly dosed insulin, one trial found little improvement in HbA1c values between treatment groups at 29 weeks. But at 52 weeks, a second trial showed pramlintide lowering HbA1c by 0.27%.

Compared to adults using fixed-dose insulin therapy plus placebo, adults using pramlintide add-on therapy (60 mcg 3 or 4 times a day) had a similarly small improvement in HbA1c (between group differences 0.2 to 0.3%). Weight loss was more evident in pramlintide-treated adults (range of mean change across trials -0.4 to -1.3 kg) than placebo in this group (+0.8 to +1.2 kg). [full review]

In type 1 diabetes, how does adding pramlintide to insulin therapy compare to insulin alone in adverse effects?

Adverse effects such as nausea, vomiting, and anorexia were more common in adults using pramlintide plus insulin than in those using placebo plus insulin. Severe hypoglycemia was more common in the first four weeks in adults using pramlintide than those without and rates of hypoglycemia remained slightly higher at 1-year follow up. No data are available for children. [full review]

Evidence is lacking on the long-term benefits and harms of adding pramlintide to treatment. [full review]

In type 2 diabetes, how does pramlintide compare to insulin therapy?

Pramlintide added on to flexibly dosed glargine regimens resulted in a 0.35% reduction in HbA1c and a weight loss of 2.3 kg at 16 weeks, compared with placebo.

Adults using fixed-dose insulin plus pramlintide (90 mcg or 120 mcg) for one year had improved HbA1c levels (between-group difference: -0.13 and -0.4%) and weight loss (between-group difference: -1.1 to -1.85 kg) compared with placebo. [full review]

In type 2 diabetes, how does exenatide compare to oral agents and insulin therapy?

In adults taking various oral agents, the addition of exenatide improves HbA1c more than placebo does. The 5 mcg twice daily dose of exenatide lowered HbA1c by 0.6% (95% CI -0.8 to -0.4%) while the 10 mcg twice daily dose did so by 1.0% (95% CI -1.2 to -0.8%).

When exenatide (10 mcg dose) was compared with insulin in adults taking oral agents, improvements in HbA1c was similar between groups. However, in adults taking oral agents plus insulin, the substitution of exenatide for insulin did not improve HbA1c. [full review]

In type 2 diabetes, how does sitagliptin compare to traditional oral agents?

Sitagliptin appears to have a similar effect to glipizide or metformin in improving HbA1c, although more data are needed.

In adults inadequately managed on diet and exercise, the initial therapy of sitagliptin plus metformin (1 or 2 grams/day) improves HbA1c more (1.4 to 1.9%) than either drug alone (sitagliptin, 0.66%; metformin 1 or 2 grams/day, 0.82 to 1.13% respectively).

In adults inadequately managed on metformin, the addition of sitagliptin is as effective as the addition of glipizide or rosiglitazone in lowering HbA1c. Adults receiving glipizide over 52 weeks or rosiglitazone over 18 weeks gained weight whereas those receiving sitagliptin lost weight.

In adults inadequately managed on two oral agents, the addition of sitagliptin improves HbA1c by 0.6% compared with a worsening of HbA1c of 0.3% seen with placebo plus 2 oral agents over 24 weeks. [full review]

In type 2 diabetes, how do the new oral agents compare in adverse effects?

Overall, the new oral agents appear to be well tolerated. The rates of withdrawal from clinical trials due to adverse effects were similar for pramlintide-plus-insulin vs. pramlintide-plus-placebo [full review], sitagliptin vs. placebo [full review], and lower dose exenatide (5 mcg twice daily) vs. placebo. Withdrawal rates were higher with exenatide 10 mcg twice daily than placebo.[full review]

The most commonly reported adverse event was nausea. Nausea occurred more frequently with pramlintide-plus-insulin than with placebo-plus-insulin (nausea declined after the first 4 weeks of therapy). [full review] Nausea was also more frequent in adults being treated with exenatide compared to insulin or placebo (nausea declined after the first 8 weeks). [full review]

Nausea appears to be less of a problem for adults taking sitagliptin - there were no differences between sitagliptin and placebo in the risk of nausea, vomiting, abdominal pain, or diarrhea. [full review]

Reports of hypoglycemia were reported for all three agents. Severe hypoglycemia was more common during the first 4 weeks of treatment with higher dose pramlintide plus insulin than with placebo-plus-insulin, but rates were similar thereafter. [full review]

Hypoglycemia was more common in adults taking exenatide (10 mcg 2 times a day) compared to placebo, but the difference was not significant for the lower dose of 5 mcg twice daily. The rates of hypoglycemia were similar between adults receiving insulin and adults receiving exenatide, and were highest for adults receiving exenatide plus a sulfonylurea compared to a placebo plus sulfonylurea. [full review]

For sitagliptin, the risk of hypoglycemia was greater when it was given with a sulfonylurea such as glipizide. The risk of hypoglycemia from sitagliptin alone is no greater than that of placebo, and when given with metformin, the risk remains less than for glipizide alone. [full review]

Do patient factors influence the effectiveness or safety of the newer oral agents?

In general, age, gender, and BMI do not appear to influence the treatment effects of sitagliptin in patients with type 2 diabetes. A single trial found that Hispanic adults showed slightly larger reductions in HbA1c than other adults did. [full review]

For exenatide, age does not appear to influence the reductions in HbA1c, and hypoglycemia is not common in people aged 65 or older. Data for other patient factors are lacking. [full review]

For pramlintide, evidence is lacking on whether age or gender affects its safety and effectiveness in type 1 and type 2 diabetes. [full review]

One study found that the use of pramlintide in addition to insulin in adults with type 1 diabetes prevented weight gain in adults with a BMI of less than 23, and assisted weight loss in adults with a BMI of greater than 23. In type 2 diabetes, another study found that black adults may have a slightly better response to pramlintide than white and Hispanic adults. [full review]

Drugs included in this review

Further information

Image th-newdiabetes.jpgThis PubMed Clinical Q&A was reviewed by Nancy J. Lee, PharmD, BCPS.

For the full report and evidence tables, please see:
Norris SL, Lee NJ, Severance S, et al. Drug Class Review: Newer Drugs for the Treatment of Diabetes Mellitus: Final Report [Internet]. Portland (OR): Oregon Health & Science University; 2008 Aug. Available at: