The Journal Editorial Report | December 10, 2004 | PBS
Visit Your Local PBS Station PBS Home PBS Home Programs A-Z TV Schedules Watch Video Donate Shop PBS Search PBS
The Journal Editorial Report
Features
Front Page
Lead Story
Briefing & Opinion
Tony & Tacky
TV Schedule
For Teachers
About the Series
Archive


Briefing and Opinion
December 10, 2004

Dr. David Graham and Dr. Sandra Kweder
Dr. David Graham, Associate Director of Science for the Office of Drug Safety for the Food and Drug Administration and Dr. Sandra Kweder, Acting Director of the Office of New Drugs, testify before a Senate Finance Committee hearing on "The FDA, Merck and Vioxx: Putting Patient Safety First?" November 18, 2004. (AP Photo/Gerald Herbert)


The decision to pull Vioxx off the market because it increased the risk of heart attacks seems to have had two effects beyond providing lawyers with more business. It's raised new concerns about the way drugs are approved and then monitored for safety, and it's fueled an anti-pharmaceuticals backlash. The question now is whether there will be a return to slower approval of new drugs that might mean relief or even life for people who want them. Dr. Scott Gottlieb, a former senior policy adviser to the Commissioner of the Food and Drug Administration, joins the panel to discuss the issue.



PAUL GIGOT: Scott, we want to get the Vioxx issue later, but you worked at the FDA. You know how that culture operates and how it thinks. What's the reaction of a place like that when you get a big political event like this and criticism from Congress and others about FDA policy?

Scott GottliebSCOTT GOTTLIEB: Drug approvals by necessity are based on the main information. You're testing drugs usually on thousands of patients, sometimes as few as 100 for diseases like cancer and some unmet medical needs. So you can never really understand the full scope of the risks and the benefits of a new drug. The populations in the clinical trials are carefully selected and it doesn't replicate the real world, so you really don't learn about the full scope of the risks and the benefits until the drugs are on the market and given to tens of thousands of patients.

What does that mean if you're a person in the FDA working on drug review, you're a medical reviewer at the FDA? What it means is that you need to extrapolate from this data set and try to anticipate what you think the full scope of the risks and the benefits of the drug is going to be. That's very hard to do and sometimes these are very marginal decisions, difficult decisions, made on limited data where you think there's a benefit, you think there might be certain risks but you feel that the benefits outweigh the risks.

In an environment like this one, where you know there's a lot of scrutiny, you know there's a possibility you might be called before Congress to testify, you don't want to leave any stone uncovered. You want to basically ask for all the information. That also means asking companies to go back and just start collecting more, still more data. It takes additional years and slows down the approval.

FDAGIGOT: So the bulk of the political risk for this person in the FDA who's deciding this is with perhaps too rapid approval. Has there been a Congressional hearing that looked into drugs that weren't approved?

GOTTLIEB: No, that's right. The risk is always that you should be the medical reviewer who approved a drug that went on to have a really severe side effect and you didn't anticipate it from the original data set, or didn't ask the right questions. That's what people are afraid of.

GIGOT: Dan, you've written about this for many years. When you talk about the drug approval process that we now have -- how long does it take, and how much does it cost for a big drug like a Vioxx to get approved?

DANIEL HENNINGER: Well, it typically takes from 10 to 13 years for a drug to get approved. They go through three phases. The first phase establishes the safety of the chemical. It takes several years. The second phase defines the dosage. It takes about two years. The third phase, where you go into these big clinical trials of 1,000 patients or more, can take about three years and that's to decide whether the product works. The filing is usually upwards of 100,000 pages. It's not as though they're dealing with limited data. They'll look at it for months, in the case of a cancer drug perhaps, or several years. At the end of the process, it will cost between 350 million and a half a billion dollars, to get a single drug onto the market.

Rob PollockPOLLOCK: Well I think Vioxx notwithstanding, I don't think there's any question that there are more people dying right now because of good developmental drugs that haven't yet been approved, than they're dying because of drugs that have been approved.

GIGOT: Do you think that's a general rule? What happens? Or is this more like this moment in history, because of new developments in science or technology, or the political backlash from some of these cases?

POLLOCK: No, I think that's generally the rule. As Scott said, there's a general bias among regulators towards caution. And we need to remember that every time the FDA approves a drug and says, "This drug is going to save this many lives next year," that means that many people died last year when the drug wasn't approved.

GIGOT: There have been cases, have there not, though, in the last couple of decades where pressure from people who were sick -- and I'm thinking here of AIDS patients in particular -- pressured, and admirably so, the FDA, to approve drugs faster. Isn't that right?

GOTTLIEB: That's right. The FDA -- well, the real change came back in the late nineties when Congress passed a series of laws to create new mechanisms to speed drugs to the market for these kinds of unmet medical needs -- people who had terminal diagnoses. These go under the names of accelerated approval and fast track. They worked for awhile and the FDA used them aggressively for a period of time to try to get drugs on the market faster. But we've seen them back away from some of this. As recently as last week, we saw the cancer division take a big step back from the accelerated approval of a cancer drug. I think that was a result of the current environment where there's a lot of caution.

Paul Gigot

GIGOT: Let's talk about that, because there really is a lot of excitement -- Rob, you've written a lot about this -- in science right now with developing new therapies for a lot of cancers, particularly for terminally ill patients, that can extend life for some period of time or perhaps, in some cases, cure. What's going on with the cancer division that Scott talks about?

POLLOCK: Well, it looks like they're using some guidelines, which I think Scott can elaborate on in a moment. They're using some guidelines that he was involved in promulgating while in office that were intended to speed up the drug approval process. They're perverting the meaning of those guidelines and using them now to slow it down again.

GIGOT: This is called the accelerated approval process.

GOTTLIEB: That's right. In the laws that created fast track and accelerated approval, they always referred to available therapy. They said, if there's an unmet medical need where there's no "available therapy" or there's an available therapy that is better for one of these terminal conditions, then it could qualify for this accelerated approval. But they never defined what available therapy was. So we put out what we called guidance thoughts about a year ago, a little less, that defined available therapy. It defined it very narrowly. The reason we defined it narrowly was because we wanted to keep these accelerated pathways open as long as possible so as few drugs would be called available therapies, so if a new drug came along it wouldn't have other drugs that it would have to beat. They basically took the definition of available therapy and took some wording that was in this guidance document -- it shows you how careful you have to be when you work in government. We said that available therapy wouldn't be other drugs that are on the market unless there was compelling evidence that that drug worked really well. They said, "Well, since you didn't define compelling evidence, we'll define it the way we want."

GIGOT: Is there a case now where a specific therapy exists for a specific illness that is being affected by this?

Pills

GOTTLIEB: Well, just last week -- I mean this policy came out of an advisory committee meeting, so this is one of the FDA's cancer advisory committees. There is a drug called Marquibo for aggressive non-Hodgkin's lymphoma. There's no approved treatment for this, but there are drugs that are available for other conditions, other cancers, that are being used by doctors "off label" for treating aggressive non-Hodgkin's lymphoma.

What the FDA said, this drug was up for accelerated review, and what the FDA said was, "Okay, the drug has this evidence that it worked in this amount of time, and there's no approved treatment for this condition. But there's all these other treatments that are being used off-label and here's the data on it, and since these treatments in many cases are better than this drug, Marquibo, Marquibo shouldn't qualify for accelerated approval."

GIGOT: Do you think actual cancer patients will be harmed by the denial of this therapy?

GOTTLIEB: You know, it's hard to say. I'm cautious about making a judgment about Marquibo, having not looked at the clinical data myself. But what I will say is that, taking for example the drug was 15 percent effective at shrinking some tumors, even if it wasn't as potent as some of the drugs the doctors are using off-label, it still had a different side effect profile. And cancer patients make decisions based also on the side effect profile. So one patient will say, well, this drug is 30 percent effective but can be toxic to my heart. Maybe I'll choose a drug that's 20 percent effective that will be not toxic to my heart.

GIGOT: Rob, you've worked with the Abigail Alliance, which is a group of cancer patients lobbying for faster approvals. I guess for them, 30 percent, 15 percent, they'll take it.

POLLOCK: Absolutely. As Scott said, the important thing is having alternatives out there. Because we know that some drugs are home runs in certain people and don't work at all for others.

HENNINGER: Shouldn't patients be able to discuss this between them and their clinical oncologist, the expert? Not between them and the FDA?

GIGOT: All right. But let's take this case of Merck and Vioxx, because we're talking about a drug here that was approved, it went through all the hoops that you talked about, but then it developed the side effects with the increasing risk of heart attacks, after it was approved. Isn't this a problem? How do we deal with it to make sure that there aren't people that are harmed?

POLLOCK: Well I think the important thing we need is simply truth in labeling. And personally I think Merck made a huge mistake by withdrawing Vioxx from the market. Vioxx was a good drug that helped a lot of people. It had a certain risk profile that probably meant it shouldn't have been taken as widely as it was, and it should have been relabeled for that reason. But I think by taking it off, they've only given ammunition to the trial lawyers who are now going to destroy them.

GIGOT: The FDA was aware, was it not, of all of the studies that Merck had done showing increased heart attack risk, right?

GOTTLIEB: That's right. It just goes to show you that even if the FDA knows what the risks are and puts it in the label, the company's still subject to all the liability and the law suits.

POLLOCK: FDA approval doesn't protect you.

GIGOT: Well, what, Dan, what's the answer here? I mean there are some proposals in Congress to create a new panel, a new bureaucratic body that would -- in addition and outside the FDA -- that would look at drugs once they're approved and the side effects?

Daniel HenningerHENNINGER: Well, the buzz phrase of the moment is "post-marketing surveillance." That simply means post-approval surveillance of adverse effects over time. We don't do a very good job of that. The one country in the world that does is Holland. The Dutch have set up a software-based system in which general practitioners collect information about their patients, feed it into a central database which is actually controlled by the GP's themselves, and they analyze when the number goes up, and you see a certain drug producing an adverse effect that's significant and ought to be looked at.

Now, in the age we live in, we ought to be able to devise a software system and a technology as we gather more information about patients, that does that rather than telling the FDA to jump on the companies and suppress drug development.

GIGOT: Does that make sense to you, Scott?

GOTTLIEB: That makes a lot of sense. It's something we're working hard to do at FDA, and they're still working hard to do, and they have various pilot programs. But I'm afraid that any of the legislative proposals that are going to grow out of this probably are going to call for something that's more obvious, because that gives more immediate gratification.

GIGOT: A little politically easier.

GOTTLIEB: Yes.