Popular Diabetes Drug Actos Worse than Avandia

by Eric Pearson

In 1999, diabetes drug Avandia was introduced. By 2006, it was the most popular diabetes drug sold in America, with annual sales of some $3.4 billion. What happened next has been well documented. In 2007, Dr. Steven Nissen, chief cardiologist of the Cleveland Clinic, conducted an analysis of 42 clinical trials Avandia’s maker, GlaxoSmithKline (“GSK”), had posted on its website. His analysis, published in the New England Journal of Medicine, concluded that the use of Avandia led to a 43% increase in the risk of a heart attack.

As a result of this study, the FDA required GSK to add the strictest possible warning, a “black box” warning, to its packaging. In February 2010, a Senate Finance Committee released findings concluding that GSK knowingly hid the risk of heart attacks associated with Avandia beginning as early as 1999. In June 2010, two prominent medical journals published separate studies confirming the cardiac risks of Avandia and concluding that Avandia should be removed from the market.

Rather than pulling the drug, the FDA instead ordered that its use be heavily restricted. As a result of its well-documented cardiac risks and the restrictions imposed by the FDA, sales of Avandia fell to less than $500 million per year. Numerous lawsuits have been filed against GSK on behalf of patients suffering cardiac problems from taking Avandia, and GSK has been forced to set aside more than $6 billion to address these lawsuits.

Avandia’s bust turned out to be Actos’ boom. Actos, manufactured in Osaka, Japan by Takeda Pharmaceuticals, is a diabetes drug similar to Avandia. Both drugs are members of a class of drugs known as thizolidinediones that decrease the insulin resistance of body tissue and modify the production of cholesterol. As a result of the problems associated with Avandia and its loss of market share, sales of Actos rocketed to $4.8 billion in 2010, accounting for 27% of Takeda’s revenue.

One advantage Actos had over Avandia is that the same studies that had documented the cardiac risks of Avandia showed that Actos carried much less cardiac risk for its users. However, in August 2010, a new study was published that concluded that the cardiac risks of taking Actos were just as severe as those from taking Avandia. The August 2010 study, however, was just the beginning of the ongoing problems for Takeda and its blockbuster drug.

In September 2010, the FDA announced that it was reviewing the interim results of an ongoing 10-year study examining whether the long-term use of Actos could lead to an increased risk of bladder cancer. A review of the five-year interim results of the study, set to conclude at the end of 2012, showed an increased risk of bladder cancer for those patients using Actos the longest and those patients taking the highest doses of the drug.

Specifically, the study showed that patients taking Actos for more than 12 months or at the highest dosage had a 40% higher risk of bladder cancer. The FDA’s press release announcing these findings also noted that studies conducted before the FDA approved Actos showed that male rats receiving a clinical dose of Actos had a higher incidence of bladder tumors and that patients taking Actos had a higher incidence of bladder cancer than patients taking comparable drugs.

In June 2011, the FDA announced that, on the basis of the interim review of the ten-year study begun in 2003, it had concluded that “the use of diabetes medication Actos for more than one year may be associated with an increased risk of bladder cancer.” In the same press release, however, the FDA took pains to point out that the 5 year-review did not show an overall increase in the risk of bladder cancer from taking Actos. According to the FDA, the review showed that “there was no significant increase in the risk for bladder cancer in patients ever exposed to [Actos] compared to patients never exposed to [Actos].”

This assertion, however, was quite misleading. The study results actually showed an overall Hazard Ratio (“HR”) of 1.2, indicating a 20% increase in the risk of bladder cancer from the use of Actos. However, the Confidence Interval (“CI”) for those particular results ranged from .9 to 1.5. What that means is that there was a 95% likelihood that the true HR was somewhere between .9 and 1.5; or between a 10% reduction in risk and a 50% increase in risk. Any time a study has a CI range that includes 1.0 (1.0 means the risk is the same between the two groups being compared), statisticians typically conclude that the results of the study are not “statistically significant.” But a lack of “statistical significance” does not mean there are no adverse effects from taking Actos even short term. By downplaying the results of the five-year review, the FDA may be seriously underestimating the risks of taking Actos and jeopardizing the health of thousands of Americans.

At the same time the FDA was announcing its opinions, health agencies in Europe were doing the same. In June 2011, French authorities announced the results of a three-year study of 1.5 million diabetes patients taking Actos. That study showed results similar to that of the FDA’s five-year review.

Overall, patients taking Actos had a 22% higher risk of bladder cancer (HR of 1.22, with a CI of 1.05 to 1.43). Patients using Actos for more than 12 months had a 34% increase and those using Actos for more than 24 months had a 36% increase. Most significantly, the study showed that patients using the highest doses of Actos had a 75% higher risk of bladder cancer. The French study concluded that there was a “statistically significant association between exposure to [Actos] and incidence of bladder cancer.” Unlike their U.S. counterparts, the French authorities reacted strongly and swiftly to their study, removing Actos from the market completely on June 9, 2011. Germany removed Actos from the market shortly thereafter. The drug continues to be sold in the United States.

So what is the moral of the story? Unfortunately, the history set out above raises more questions than it answers. Among them:

  • Why was Actos approved when clinical trials showed an increased risk of bladder cancer in both humans and mice?
  • What did Takeda know about the risks of bladder cancer from taking Actos?
  • Like GSK did with Avandia, did Takeda hide negative test results from the public?
  • Why hasn’t the FDA removed Actos from the market given both its risks of bladder cancer and its cardiac risks?
  • How many billions of dollars will Takeda have to spend defending itself against the inevitable lawsuits that will be brought by the families of Actos users who died or suffered from bladder cancer?

Only time will tell.