Metformin is more effective than thiazolidinediones and second generation sulfonylureas in decreasing weight and body mass index in type 2 diabetes patients, according to an Agency for Healthcare Research & Quality ¹ draft report.

"Metformin, and possibly acarbose [Bayer's Precose ], reduced body weight; in contrast, all other oral diabetes medications increased body weight and BMI," the report concludes based on the review of 212 published articles on oral diabetes drugs.

The draft report, "Comparative Effectiveness and Safety of Oral Diabetes Medications for Adults with Type 2 Diabetes," was prepared by The Johns Hopkins University Evidence-Based Practice Center. Comments are due Aug. 24.

The report, which sought to evaluate short-term and long-term outcomes for different oral diabetes medications, also found that metformin alone decreased BMI/weight compared with its use in combination with second generation sulfonylureas, which either maintained or slightly increased BMI/weight.

Medications in the thiazolidinedione class include GlaxoSmithKline's Avandia (rosiglitazone) and Takeda's Actos (pioglitazone).

GSK additionally markets a combination metformin/ rosiglitazone product, Avandamet , while Takeda will launch a fixed-dose combination of pioglitazone and the sulfonylurea glimepiride as Duetact later this year.

The report adds that thiazolidinediones and second generation sulfonylureas - including Sanofi-Aventis' Amaryl (glimepiride) and generics - "appear to have increased weight gain with higher doses supporting that increased weight gain may also be related to improved glycemic control."

In addition, repaglinide (Novo Nordisk's Prandin ), of the meglitinide class, "may have no consistent effect on body weight" - a result that could reflect use of a sub-maximal dose in the studies reviewed, the report states.

Weight loss is being touted as a benefit of use of a new class of diabetes medications, dipeptidyl peptidase IV inhibitors, in combination with metformin. Merck and Novartis both have DPPIV inhibitors pending at FDA. (² Also see "Novartis Takes Aim: Galvus Promotions To Target Other Classes, Not Januvia" - Pink Sheet, 19 June, 2006., p. 3).

The AHRQ report also compares the effectiveness of oral diabetes medications in reducing glycated hemoglobin (HbA1c). Head to head studies show that "most diabetes medications reduce HbA1c to a similar degree," according to the report.

However, "more direct data is needed on meglitinides and alpha-glucosidase inhibitors compared to other oral diabetes medications to make firm conclusions about these two classes of agents."

The report also states that indirect data show that nateglinide (Novartis' Starlix ), part of the meglitinide class, may not be as effective as other drugs at reducing glycated hemoglobin, and that combination therapies are more effective at reducing HbA1c than monotherapy.

The researchers add three caveats to their conclusions regarding HbA1c: first, certain study results may have been influenced by baseline HbA1c levels; also, "many studies failed to report the significance of between-group differences, thereby hindering efforts to estimate effect size across trials"; and many of the studies were sponsored by industry after FDA approval, "raising the possibility of publication bias."

The draft concludes that all oral diabetes medications have minimal effects on systolic and diastolic blood pressure; however, it also points out that too few studies compare alpha-glucosidase inhibitors, meglitinides and combination therapy with other medications. In addition, thiazolidinediones "may have slightly more beneficial effects" on blood pressure compared with second generation sulfonylureas.

The report points out that "very few published studies compared oral diabetes medications in terms of clinically important distal outcomes, like cardiovascular events and mortality."

In terms of safety, "almost all studies reported rates of hypoglycemia" but most were inconsistent in reporting other adverse events. "There was consistent evidence that second generation sulfonylureas caused hypoglycemia more frequently than most other oral diabetes medications," the report states.

While lactic acidosis is commonly associated with use of metformin, there was no consistent evidence of increased risk of the condition with its use compared to other medications, according to the report. However, "metformin was consistently associated with more gastrointestinal problems than most other oral diabetes medications except acarbose."

Thiazolidinediones were associated with increased episodes of self-reported edema. The class was also associated with anemia, "with an average drop in hematocrit of 1%-3%, which is likely not clinically relevant unless someone already has moderate anemia."

While the label for thiazolidinediones includes congestive heart failure as a potential adverse event, the investigators "found very few randomized controlled clinical trials addressing this issue."

Overall, the report concludes that "oral diabetes medications have similar effects on glycemic control, and slightly different effects on other proximal outcomes." However, the researchers were "unable to draw firm conclusions" regarding their comparative effects on distal outcomes.

In addition, "safety differences do exist among the oral diabetes medications, and deserve further investigation."

Future research should include randomized controlled trials and prospective cohort studies to assess the effects of oral medications on the long-term outcomes of all cause and cardiovascular mortality, cardiovascular disease morbidity, microvascular disease and peripheral vascular disease, according to the report.

"Future prospective cohort studies could improve understanding of the effects of oral diabetes medications on adverse events and distal outcomes if they carefully assess key confounders such as duration of diabetes, adherence to medications, dosing of medications, HbA1c levels, and blood pressure levels," the report adds.

Additionally, studies should "consistently report between group differences with measures of variability, not just changes from baseline" for each drug, the researchers conclude.

Diabetes was one of 10 priority conditions for comparative effectiveness review identified by AHRQ in 2004 (³ Also see "AHRQ Drug Effectiveness Comparisons Will Focus On 10 “Priority Conditions”" - Pink Sheet, 20 December, 2004., p. 11). The agency most recently published a draft report on osteoporosis drugs (⁴ Also see "Osteoporosis Drugs Found To Be Equally Effective – AHRQ Report" - Pink Sheet, 12 June, 2006., p. 25).

Metformin Is Most Effective Diabetes Drug At Reducing BMI – AHRQ Report

More from Archive

More from Pink Sheet