Biotech Values

[DewDiligence]

Non-inferior to what?

[Under existing FDA practice, antibiotics with low efficacy—or no efficacy at all—are relatively to easy to get approved.]

http://online.wsj.com/article/SB115016648349378627.html

>>
Ketek Sparks Calls To Stiffen Antibiotic Trials

By ANNA WILDE MATHEWS
June 13, 2006

Controversy over the Food and Drug Administration's handling of the antibiotic Ketek is focusing new attention on the way the agency evaluates the efficacy of antibiotics.

In many diseases, the FDA insists that drugs be shown to work better than placebo pills in clinical trials. But for some medicines, including antibiotics, the agency allows a different approach: The manufacturer just has to show that its product works about as well as an older drug, not necessarily better.

Such "noninferiority" trials produce less clear-cut results than traditional placebo-controlled tests. The new antibiotic may work as well as an older one, but regulators may not know exactly how much better the older antibiotic works than a placebo. Without knowing that, it's not clear how effective the drugs are compared to a body's own natural defense, at least in milder infections. In short, would patients have gotten better just as well on their own?

"If you don't know drug A works in the first place, and you compare it to drug B, you're not going to learn too much," says Jack Gwaltney, an emeritus University of Virginia professor of medicine who has served on an FDA advisory committee. To resolve lingering questions about the drugs' efficacy in Dr. Gwaltney's specialty, sinusitis, he would like to see an antibiotic tested against placebos with a tap of the sinuses to see if the drug has eliminated the bacteria.

Noninferiority trials are under new scrutiny as part of congressional investigations of Ketek, an antibiotic made by Sanofi-Aventis SA that has been linked to reported liver failures. Senator Charles Grassley of Iowa, who is chairman of the Senate Finance Committee, is examining the issue. [What issues is Grassley *not* examining?] House Democrats Henry Waxman of California and Edward Markey of Massachusetts have questioned the use of such trials to approve Ketek.

The investigations are largely focused on the safety of Ketek and on alleged fraud that arose in one study of the drug. But, the lawmakers say, any worries about a drug's safety are magnified if its benefits weren't fully proven. Ketek was approved for use in sinusitis, bronchitis and pneumonia. Sanofi-Aventis didn't immediately respond yesterday but earlier the company said it "continues to believe that the benefit of Ketek outweighs any potential risks of this drug when used for its FDA approved-indications."

Noninferiority trials have been used in conditions such as AIDS, where doctors thought it was unethical to give patients placebos because they could suffer serious harm. In antibiotics, they became the norm as early as the 1940s, when the drugs were first tested for potentially deadly illnesses such as meningitis. Now noninferiority trials are used even in tests of drugs for less serious ailments, such as ear infections, where most symptoms resolve even without an antibiotic.

The FDA says it plans to broadly re-evaluate its approach to reviewing antibiotics for some uses where noninferiority trials have been questioned, including ear infections and sinusitis. There are "legitimate concerns" about the studies, says Mark Goldberger, director of the agency's office of antimicrobial products. But "it's more complex than it seems at first glance."

Indeed, the drug industry has warned that if the FDA raises the bar too high for antibiotic approvals, demanding challenging new trial designs that would result in larger and more expensive studies, companies won't aggressively develop new treatments. Dr. Goldberger says "that's a factor... that is there in the background" as FDA evaluates its policies.

Some doctors and industry officials say that patients would be reluctant to take part in studies where they risked being given a placebo, even for relatively minor infections. For more serious infections, "it would be immoral to do a placebo-controlled trial where there is an established therapy," says Frank Tally, chief scientific officer at Cubist Pharmaceuticals Inc.

Cubist's antibiotic Cubicin recently won FDA approval for new uses including a form of the serious heart infection called endocarditis with a noninferiority test. Cubicin's results were about equal to other drugs' for that use, but the test involved very few patients with the condition. "In the FDA, there was a lot of debate with the statisticians," Dr. Tally says, but additional evidence factored into the approval.

One problem for regulators is that if they allow too much flexibility -- meaning, the new drug can perform a lot worse than the old one and still be counted as roughly equal -- they risk approving a new treatment that doesn't really work, even if it meets a generous definition of being "not inferior" to an older drug.

Some researchers say drugs may be going on the market without enough proof that they really are effective. "There isn't rigor" in how the FDA approves antibiotics, says Thomas Fleming, a statistician at the University of Washington who has served on agency panels that reviewed the issue. Currently, he says, there is "a model that grants FDA approval for something that is providing safety risks but may not hold tangible benefit to the public."

The FDA's Dr. Goldberger says "we are confident that the antibiotics we approve, based on the whole package, are effective." The agency uses other issues, including test-tube results and findings from trials in serious conditions like pneumonia, to back up the noninferiority studies in areas like bronchitis, he says. The FDA has gotten "pushback" from companies and ethics committees about using placebos in antibiotics trials, he says, and it must also consider other issues, like how to screen the patients allowed into such studies.

FDA officials have conceded that for certain uses, including bronchitis and sinusitis, the extent to which older antibiotics beat placebos isn't clear. An agency statistician told an advisory committee in 2002 that "historic data is often poor" in anti-infective drugs.
<<