December 13, 2006

ADOPT - Avandia "Wins" If You Ignore the Doubled Bone Fractures

Following the publication of the results of the ADOPT study, GlaxoKlineSmith, the makers of Avandia have been touting the finding that Avandia delayed the progression from "Monotherapy" i.e. taking one drug to taking additional drugs better than Metformin.

In the diabetes oral drug sweepstakes, they'd have you believe that Avandia is the "winner." What was missed in this interpretation was that this same study also found that women taking Avandia had twice as many bone fractures as those taking Metformin and three times as many as those taking glyburide.

Here's an excerpt from the story as it ran on several news sites. This one is from Yahoo News.

"After the five-year period of study, commonly reported adverse events across the treatment groups were oedema (rosiglitazone 14.1 percent; glyburide 8.5 percent; metformin 7.2 percent); weight gain (rosiglitazone 6.9 percent; glyburide 3.3 percent; metformin 1.2 percent); gastrointestinal side effects (metformin 38.3 percent; rosiglitazone 23.0 percent; glyburide 21.9 percent); and hypoglycaemia (glyburide 38.7 percent; metformin 11.6 percent; rosiglitazone 9.8 percent).1

[emphasis mine] Recent further analysis showed a lower rate of fractures reported as adverse events in women taking glyburide or metformin versus rosiglitazone (glyburide 3.5 percent; metformin 5.1 percent; rosiglitazone 9.3 percent), most commonly involving fractures of the foot and upper limb bones.1 There was no observed difference among treatment groups in the number of fractures reported in men.1"

This says that women on Avandia (rosiglitazone) had twice as many fractures as those on Metformin and almost 3 times as many as those on Amaryl (glyburide.)

The total number of fractures was not huge, but this finding takes on more importance when linked to a previous study, published in November that also found that Avandia and Actos increased the incidence bone fractures in older women in nursing homes who had been taking it.

Here is a report of that study, as reported in the Diabetes in Control newsletter: TZD’s Can Increase Bone Loss in Type 2 Women

The reason why this might be happening as reported in the nursing home study was that Avandia apparently disrupts the parathyroid hormone that regulates calcium.

This is very disturbing because it may mean that in people who are taking Avandia, it is very slowly and without anyone observing it causing bone to be remodeled in a way that will not become obvious for years to come but will eventually ruin lives.

By the time the bone changes do become obvious, the old ladies who have taken Avandia for a decade will begin to break their hips and die, but it will be too late. Their bones will have already been weakened.

That the drug manufacturer completely fails to mention this new, and dangerous side effects in its hail of advertising to physicians and its press releases to the public is very troubling.

If nothing else, if you have been taking Avandia for any length of time, insist on having your doctor order a bone scan and have it repeated every couple years to be sure that that you aren't developing early signs of osteoporosis that won't otherwise be detectable for many more years. Osteoporosis causes fractures in older people that lead to death in 25% of all cases in people over age 50 and more in the more elderly.


Anonymous said...

just to let you know that a bone scan is not the study indicated to detect osteoporosis. it is the bone densitometry.

thiazolidinediones do not affect parathyroid hormone

Anonymous said...

Just to let you know that the term "side-effect" is only applied to adverse effects that have been shown to be causally associated with the drug. To my knowledge, fracture has not been causally associated with Avandia i.e. no link has been established, which is probably why fracture risk doesn't appear on the label. Your commentary is alarmist. I don't work for GSK, so I have no axe to grind, but I do happen to know a thing or two about pharmacovigilance as well as how to interpret clinical data.

Jenny said...


When more than one study finds a statistically significant rise in some side effect in the group taking a drug, and when there is a mechanism to explain why the side effect occurs, it's wise to treat it as real.

The mechanism appears to be that Avandia and Actos cause the stem cells that should differentiate into bone cells to instead turn into fat cells. Over time this leads to thinning bones as new cells that should strengthen the are diverted elsewhere.

Here is another blog that lists the many research findings.

Note that it reports that FDA's Medwatch has sent out a warning to doctors about this issue. That means it is likely to show up in the PI sometime soon.