Sorry!! The article you are trying to read is not available now.
Thank you very much;
you're only a step away from
downloading your reports.

FDA Approval is About Rules


Don't make me come for you!

This was a good week for those willing to be educated in the biotech space. Four diverse oncology drugs met their fate at the hands of the Oncologic Drugs Advisory Committee (ODAC) panel. Three passed - a great batting record compared to recent history.

I want to focus on the one that didn't pass, Abbott's (ABT) Xinlay. This application for approval, which never should have been filed in the first place, was a good example of a drug whose science matches a subset of results from a clinical trial. By the logic most of us operate, matching the science with results from a clinical trial should equal approval. This logic is reinforced by the fact this drug purportedly slows the progression of prostate cancer to bones, preventing prostate cancer patients from experiencing the pain of having cancer splitting apart their bones from the inside. Big need, clear science, demonstration of efficacy.

All that you need for approval, right?

I've written here before that FDA approvals are as much about matching a set of arcane biostatistical rules as they are about a drug's effect on patients. The Xinlay panel was a good example of this.

Clinical trials are run in order to determine, usually with 95% accuracy, that the observed benefit of a drug is not due to random chance. If you are only 90% sure, then you don't get approved.

That's foreign to most people. Raise your hand if you'd pull the trigger on a trade where you were 90% sure of profiting. Yeah, everyone can put their hands down now.

The rules are actually more specific than that. When you start a clinical trial, you have to be very specific about the question you are asking. In Abbott's case, it was whether men with hormone-refractory prostate cancer would benefit from Xinlay by delaying progression or extending survival. As it turns out, the trial told them 'yes', but that they were only about 80% sure the result was not due to chance.

This puzzled the Abbott scientists because Xinlay, for all its other faults, has a pretty clear scientific rationale for why it should work. So Abbott's people went digging through the data collected from this trial and found something they thought was interesting: Men who already had prostate cancer move into their bones benefited from the drug. In fact, they were better than 99% certain it was not due to chance they benefited from Xinlay.

Slam dunk, right? Nope. And understanding why is vital to you being able to handicap FDA decisions in the future.

The original question Abbott asked was whether men with prostate cancer benefited from Xinlay. They did not ask, before the trial started, whether men with prostate cancer in their bones benefited from Xinlay. Different questions. Not much different, but different.

When your trial fails and you dive through the data to find answers as to why, you can always find an answer. When you take that answer and then create a new question, you are engaging in what is known in statistical circles as a "retrospective analysis." It's the statistical equivalent of you making some embarrassing mistake and then saying, "Oh, I meant to do that."

The Xinlay application is a good, though subtle example of that. A less subtle example comes from Cell Therapeutics (CTIC) failed Xyotax trials. Their lung cancer trial for this chemotherapy drug failed horribly to answer their original question (Do patients with lung cancer live longer taking Xyotax). So they went diving back into the data and discovered women were living longer. So, they plan to ask the FDA to approve Xyotax for women based upon this retrospective analysis. (Last I checked, both men and women had lungs and there is no scientific basis for why their drug would act differently in men and women.) They got their answer first (women lived longer than men in the trial), then they created their question (Does Xyotax improve survival in women with lung cancer).

Oops, I meant to do that.

Under a conservative biostatistical viewpoint (which is what the FDA has), retrospective analyses are simply not allowed. While that rule is not absolute, it's pretty close to absolute and anyone who tries to buck the FDA on this does so at his or her peril. Where things get interesting is when the line between retrospective and prospective gets blurred. In those cases, the clinicians on the panel (the panel members who actually treat patients) will agonize over the relationship between the benefit of the drug and its side effects. If they are robust enough, and if the line between prospective and retrospective is fuzzy enough, an applicant can sometimes squeeze by a panel and get their drug approved.

The Xinlay application failed on a number of fronts. First, it was retrospective. Second, the drug had serious side effects. Third, the advantage they showed, even in the retrospective population, was not very large. Fourth, the FDA demands anyone applying for approval based upon one trial have some supporting evidence and the Phase II trial Abbott was trying to use as support was critically flawed. Fifth, Abbott kept changing their story about what was important, giving the FDA reviewers fits and reducing confidence.

Like I said, this application never should have been filed.

It was interesting that many of the panel members indicated they understood the need for a drug like Xinlay. They simply weren't convinced Abbott had proven, by the rules of biostatistics, that Xinlay met the need.

The first question you have to ask yourself before investing in a drug going to the FDA is whether the drug significantly and safely solves some medical need. The second question is whether the drug proved this worth by the rules the FDA set out. If the answer to the second question is "no", then the answer to the first question rarely matters. If the answer to the second question is "maybe" or "mostly", then you must acknowledge the existence of significant risk even if the answer to the first question is. "It's the best drug in the world for patients who really need it."

< Previous
  • 1
Next >
No positions in stocks mentioned.

The information on this website solely reflects the analysis of or opinion about the performance of securities and financial markets by the writers whose articles appear on the site. The views expressed by the writers are not necessarily the views of Minyanville Media, Inc. or members of its management. Nothing contained on the website is intended to constitute a recommendation or advice addressed to an individual investor or category of investors to purchase, sell or hold any security, or to take any action with respect to the prospective movement of the securities markets or to solicit the purchase or sale of any security. Any investment decisions must be made by the reader either individually or in consultation with his or her investment professional. Minyanville writers and staff may trade or hold positions in securities that are discussed in articles appearing on the website. Writers of articles are required to disclose whether they have a position in any stock or fund discussed in an article, but are not permitted to disclose the size or direction of the position. Nothing on this website is intended to solicit business of any kind for a writer's business or fund. Minyanville management and staff as well as contributing writers will not respond to emails or other communications requesting investment advice.

Copyright 2011 Minyanville Media, Inc. All Rights Reserved.

Featured Videos