More data needed on drug to delay the onset of diabetes
=======================================================

* Pauline Comeau

A Canadian-led phase 3 trial has shown that taking 8 mg of rosiglitazone (Avandia) daily, coupled with lifestyle counselling, can reduce the risk of developing Type 2 diabetes by 62% in individuals at high risk.

The study results made headlines worldwide, but a drug expert argues that lifestyle changes can do the same thing and that more data are needed before widespread use of the drug.

In the 3-year DREAM (Diabetes REduction Assessment with ramipril [Altace] and rosiglitazone Medication) randomized controlled trial, led by Dr. Hertzel Gerstein of McMaster University in Hamilton, 5269 middle-aged, overweight adults from 21 countries, all with poor glucose regulation but no heart disease, were randomly assigned to receive either rosiglitazone, a drug currently used to treat of Type 2 diabetes, or a placebo.

![Figure1](http://www.cmaj.ca/https://www.cmaj.ca/content/cmaj/175/10/1192/F1.medium.gif)

[Figure1](http://www.cmaj.ca/content/175/10/1192/F1)

Figure. Rosiglitazone has been shown to delay the onset of Type 2 diabetes, but critics say lifestyle changes do the same thing. Photo by: Canapress

Among the 2635 taking rosiglitazone (*Lancet* 2006;368:1096-105), 306 developed diabetes or died, compared to 686 of the 2634 in the placebo group. As well, 51% of those taking the drug had their blood sugar levels return to normal, compared to 30% in the placebo group. Positive liver function results were also noted. The ramipril (15 mg daily) arm of the study (*N Engl J Med* 2006;355:1551-62) showed increased regression to normoglycemia (16%), but no effect on the risk of developing diabetes or dying.

“Five years ago we had no way of preventing diabetes,” Dr. Gerstein told *CMAJ*. “Now we know that we can prevent diabetes through diet and exercise and pharmacotherapy, and this is so far the most powerful pharmacotherapy agent that we have. It has risks associated with it; whether it gets used in an individual patient is part of a doctor–patient discussion and assessment of the risks and benefits for that person.”

But several observers argue that similar results can be attained through diet and exercise, and more research is needed on the drug's side effects. They note that 14 participants on rosiglitazone developed non-fatal heart failure, compared with 2 cases in the placebo group, and the group gained 3% (2.2 kg) more in body weight than the placebo group.

The study authors say heart failure was easily managed with drugs, and conclude the weight gain was “favourable” because it was focused on hips rather than waist (abdominal obesity is a strong diabetes indicator).

But pharmacologist Dr. Jim Wright, managing director of UBC's Therapeutics Initiative, called the results “nonsensical.” Evidence that rosiglitazone lowers blood sugar was known, he says. And while the drug might delay when someone at high risk crosses the “threshold” and is labelled diabetic, the study results do not measure whether diabetes-related problems, such as heart disease, are prevented.

In fact, he says, the study's negative cardiovascular outcomes make him suspicious about the benefits of having people with non-diabetic hypoglycemia take a costly ($2.88/day) drug daily rather than focus on diet and exercise. Wright says hospitalization and mortality rates of the DREAM study participants “must” be followed for the next decade to measure long-term effects from having taken the drug. “Any reasonable person would have major concerns looking at the results,” said Wright.

Gerstein agrees diet and exercise are important, but says pharmacotherapy is the answer now for those unable to comply, or who remain in a pre-diabetes state despite efforts. He says analysis of physical activity data gathered in the study is underway.

Meanwhile, the DREAM study continues. Participants without diabetes at the end of the study have been taken off medication and are being followed for a 2-month “washout” phase to determine if effects are long lasting. Those results will be presented in December. Study participants will also be followed for up to 2 more years to assess the long-term effects of rosiglitazone, and left ventricular function and other indices of cardiac function are being tested on 300 participants.

The trial was funded by the Canadian Institutes of Health Research, Sanofi-Aventis, GlaxoSmithKline PLC (rosiglitazone's manufacturer) and King Pharmaceuticals (ramipril).