Thinking about Life Sciences: FDA Report on Avandia: Implications and a... http://blog.aesisgroup.com//2007/07/26/fda-report-on-avandia-just-issue-...

Thinking about Life Sciences

http://blog.aesisgroup.com

Thursday, July 26, 2007

FDA Report on Avandia: Implications and a "critique" of the critique

According to the New York Times from today (Wednesday 7/25), the FDA issued its formal report on the
Avandia controversy that appears to substantiate the safety concerns raised by Dr. Steve Nissen and others and
apparently also sharply criticizes the manufacturer on a number of points. As I did earlier with Stryker’s hip
resurfacing announcement “The Many Surprising Implications of Hip Resurfacing Implants”, this posting will
do a sentence-by-sentence deconstruction (or “reading between the lines”) of Gardiner Harris’ New York Times
article “F.D.A. Review Criticizes Diabetes Drug and Maker.”
The text in bold-italics below is from the New York Times article; the indented text is the commentary which
references several of the points made previously regarding drug safety and FDA reform. In a sense, the
discussion below provides a contextual backdrop to the broader array of policy issues outlined in these previous
articles. Please note that the discussion below pertains to the New York Times article and not to the FDA report
per se. That being said, some of the implications and takeaways have general value.

“Patients who take Avandia, a popular but controversial diabetes medicine made by GlaxoSmithKline, are far more
likely to suffer and die from heart problems than those who take Actos, a similar pill made by Takeda, according to
federal drug reviewers.”
Kudos to the FDA for intervening which speaks to the core safety mission of the agency.
However, the operative term here is “federal drug reviewers” and one of the major points of
this blog (see “FDA: Tortoise, Hare or Something Else” has been that safety should be almost
a preview (e.g. pre-market) issue rather than a “review” situation.

“Avandia is particularly dangerous to patients who also take insulin. By contrast, Actos can be taken safely with
insulin, according to the review.”
Part of a more robust pre-market safety regime will be to analyze (even if it means more
expensive animal models or computer simulations) drug interactions. Drug interactions such
as this are particularly hard to spot during voluntary post-market surveillance studies. This
has to be done proactively. Of course, it is easy to be an “arm-chair” general and say “I told
you so” but the point is not so much any specific Avandia-insulin interaction but rather to
realize on a conceptual level that post-market surveillance is only a blunt and relatively
unreliable instrument with which to analyze potential drug-interactions.

“The findings likely spell the end of Avandia’s status as one of the nation’s most popular drugs for treating diabetics
who are not dependent on insulin. Last year, more than a million patients in the United States took Avandia, and a
similar number took Actos. ”
When it comes to approving drugs, safety is secondary to efficacy as efficacy is the key
determinant (after routine safety test are done) that creates billions in the stock market.
Investors are becoming increasingly aware that billions can be lost as a result of a safety issue.
The question is then: why isn’t safety given the absolute top priority from the beginning that it
requires?

1 of 3 11/17/2008 3:35 AM
Thinking about Life Sciences: FDA Report on Avandia: Implications and a... http://blog.aesisgroup.com//2007/07/26/fda-report-on-avandia-just-issue-...

“Avandia’s 2006 global sales were nearly $3.4 billion."

Avandia is a blockbuster drug which, by definition, is given to broad populations of patients.
That makes it appealing from a business perspective but also – just on a statistical basis alone
– makes it highly risky from a safety perspective. That realization has not yet percolated into
the types of innovative, proactive safety studies that should be part of pre-market studies.See
“FDA Reform Redux: On Business Models, Regulatory Reform and Safety” for more on the
relationship between the pharmaceutical blockbuster business model and safety risk.

“The report and [sic] charges that GlaxoSmithKline sought to intimidate a doctor who publicly warned about
Avandia’s risks in 1999 could lead to a cascade of lawsuits against GlaxoSmithKline. Indeed, F.D.A. reviewers
were sharply critical of the quality of the studies GlaxoSmithKline has undertaken to test the safety of Avandia,
dismissing the present and future results of an ongoing 4,000-patient trial as unreliable and invalid. ”
As outlined in the blog posting “Avandia, the Drug Safety Debate and How to Get a Nobel
Prize,” scientists risk their careers if they focus their research inquiries on the unglamorous
area of drug safety. They are certainly not considered to be heroes by many, though looking at
the big picture, maybe they should be.

“The report by medical and safety reviewers within the Food and Drug Administration also provides ammunition to
critics on Capitol Hill and elsewhere who claim that top F.D.A. officials have been far too slow to acknowledge
Avandia’s risks. GlaxoSmithKline suggested a year ago that the agency add a note to the drug’s label about
Avandia’s growing heart risks, the report states. ”
There is certainly great furor around potential FDA Reform and it will likely (as indicated by
your own reader poll in “You voted on the FDA. See the results here ...” likely going to be
more along the lines of a paradigm shift at the FDA rather than a tweaking of the system.

“At another point in the report, an F.D.A. safety reviewer, David Graham, concluded that a safety alert released by
top F.D.A. officials on May 21 falsely reassured patients that at least one large Avandia study showed that the drug
was safe. That study, Dr. Graham concluded, provided no such reassurance. ”
As I’ve emphasized before, I don’t believe there is malicious intent, yet, a systematic bias
which does not put safety as absolutely primary (and efficacy as a luxury “item” that’s good to
have but not essential) will result in inappropriate (and potentially false) interpretations of the
data.

“These conclusions come in a 436-page compendium of reviews released in advance of an advisory committee hearing to
be held on Monday to discuss Avandia’s effects on the heart. The F.D.A. intends to ask the committee of independent
experts whether Avandia should continue to be sold. It is far from clear, F.D.A. safety reviewers concluded in the
report, whether taking Avandia, also known as rosiglitazone, is worth the risk. ”
While the risks are “far from clear,” olitics (and life in general) often function on a “backlash”
basis. I would think that the results of Monday’s hearings will be similarly condemnatory.

“A critical question to be resolved in determining appropriate regulatory action is whether the anticipated therapeutic
benefit of rosiglitazone outweighs the demonstrated cardiovascular risk,” one F.D.A. reviewer concluded. ”
Everything in medicine is about risk-benefit ratio. And likewise, the entire point of the FDA
reform proposals outlined in the several blog postings here is that the safety-efficacy ratio also
needs to be aligned more towards the safety side of the equation. That is what our society is
demanding, that is what will need to and what will happen.

“In 2000, the F.D.A. asked Warner-Lambert to remove Rezulin from the market because that drug caused more
liver problems than either Actos or Avandia, both of which were approved for sale in 1999 and provided similar
benefits. ”

2 of 3 11/17/2008 3:35 AM
Thinking about Life Sciences: FDA Report on Avandia: Implications and a... http://blog.aesisgroup.com//2007/07/26/fda-report-on-avandia-just-issue-...

These FDA reform are not about lambasting the FDA or stripping away its powers but rather
shifting its powers away from making efficacy the primary arbiter of drug approval but
towards safety and, in that regard, requiring more safety evaluations and more innovative
approaches to safety studies a part of the pre-market evaluation.

“Avandia’s heart risks are likely to injure far more patients than Rezulin’s rare but serious effects on the liver. Most
diabetics die of heart disease. ”
Again: in my opinion, it would seem that the likelihood of Avandia being taken off the market
is very high - if not explicitly by the FDA then at least over the long-run.
I hope that this “critique of the critique” put some of the ideas about drug safety and FDA reform in a
concrete context. For more on this important topic, see:

You voted on the FDA. See the results here ...

Boston Scientific, Medtronic settle some defibrillator suits; is the device safety storm over?

Avandia, the Drug Safety Debate and How to Get a Nobel Prize

FDA Reform Redux: On Business Models, Regulatory Reform and Safety

FDA: Tortoise, Hare or Something Else

Drug /Device Safety Debate to Yield Big Changes, Grow More Controversial

It’ll be interesting to see how things play out over the coming week …

Ogan Gurel, MD MPhil

gurel@aesisgroup.com
http://blog.aesisgroup.com/

Drug safety FDA reform Avandia Diabetes Risk-Benefit ratio efficacy Aesis Research Group Ogan Gurel MD

3 of 3 11/17/2008 3:35 AM