dangerous prescription
homethe fdainterviewsdiscussion
interview: Michael Elashoff, Ph.D.

You worked at the FDA for five years. Can you describe the kind of pressure, day-in and day-out, to recommend approval of new drugs that would come your way for evaluation, versus having an objective opinion?

For the most part, when drug applications came in, there was the presumption that the drug was going to be safe and effective, and was going to be approved. So, when during the course of the review you find something that suggests some problems, typically the other reviewers and the FDA management would not want to really explore those problems, and would kind of view it as a hassle on the way to drug approval. That was a pretty much constant pressure for most reviews.

I think it was pretty well understood that if you were advocating turning a drug down  particularly if it was from a large pharmaceutical company  that that wouldnt be good for your career.

Also, when it would come time to write the official review, there would be times when the management -- it would review my drafts [and] would want to take out various sections where [I] commented on lack of efficacy or safety problems; or [management would] want [me] to tone down the language, so that they could more easily have a written justification for why they're approving the drug.

How general or pervasive was this in review?

It was pretty widespread. You know, many drugs really had pretty good safety and efficacy. So of the remainder, the ones where there are potential problems, I'd say in nearly all the cases I was familiar with -- both that I worked on, and other people in my division worked on -- there would be pressure to approve those drugs, or soften the language in the reviews and on the labels, so that they could have a more easy justification for why they're approving it. So I would say it was very common.

photo of elashoff

A biostatistician, Michael Elashoff was a drug reviewer for the FDA from 1995 to 2000. Elashoff says he found himself marginalized at the FDA after he voiced his concerns about a new flu drug called Relenza. In this interview, Elashoff speaks out about the culture of the FDA's drug approval process and why he felt Relenza should not have been put on the market. This interview was conducted on Feb. 19, 2003.

How subtle or overt was the pressure that was put on you?

I was told very explicitly, "Don't write in your review that you're recommending against approval. Write that the data is unclear, you could go either way." So it was quite explicit. ... It was pretty typical that, at the start of a review, the division director would say, "You know, I don't really see any issues with this drug. Some small things may come up in the course of your reviews. You have to do a careful review, but I don't see any roadblocks on the way to approval." That would set an overall tone for the review, where potential problems were really de-emphasized in the meetings to discuss the data.

There was just an overall expectation from the top that it would be approved. You knew that if you found a problem and wanted to pursue it, it would take a lot of extra effort, a lot of extra meetings, a lot of extra justification for why something wouldn't be approved. Whereas, if you felt the drug was safe and effective, very rarely would someone challenge you on that.

What was the drug company that made Relenza coming to the FDA for approval for? What it was supposed to be?

Relenza was intended to be a treatment for the flu. Glaxo Wellcome did a large number of studies to demonstrate that it would significantly reduce the course of symptoms once you got the flu -- would get your temperature down and all of the other flu symptoms; would get you up and back to work faster. That was the promise of Relenza.

What was your responsibility in reviewing Relenza?

It's actually fairly extensive. My responsibility in the evaluation of Relenza was to look at the clinical trials that Glaxo Wellcome had performed; assess whether they had demonstrated that the drug was safe and effective; try and establish what a reasonable estimate of the treatment effect was, what the risks and benefits were. Basically, [to] review the claims that they're making in their application.

What was your complaint with Relenza, in terms of its efficacy and safety?

Simply stated, Relenza just didn't work in the United States' clinical trials, and really didn't even come close to working. You think the flu can be seven to 10 days, and it maybe knocked half a day or less off the duration. Even that wasn't established statistically. So it was pretty much no different from placebo as far as efficacy, and it had some potential safety concerns, especially in people with underlying lung disease.

Like any new drug, it had potential risks that weren't even characterized. It wasn't studied in very many patients. For a self-limiting disease, where most of the patients that were studied were reasonably healthy, less than one day of benefit for potential risk doesn't really seem like something anyone would want to take.

Glaxo had these expectations for what their drug was going to do in patients who had the flu. How did the drug stack up? In your view, how did it perform in their clinical trials for all of these?

In the largest trial, which was the only one done in the United States, it really had no effect at all on the symptoms of influenza. It had no effect on influenza complications. It had no effect on basically any aspect of the flu, so that you really couldn't distinguish it from placebo in the trial. To me, it was almost a complete wash.

In terms of safety, what were the risks that were seen in clinical trials and that you were concerned about, as far as getting into a large population?

One of the concerns during the review was that Relenza had the potential to cause bronchospasm, or constriction of the airway. It was observed in several patients in the clinical trials. That was a concern, since there weren't really very many patients studied. It was a very controlled population under the strict clinical trial monitoring. So the concern was that, once the drug got under less supervision in the larger market, then incidents of those types of problems would be magnified.

Why is bronchospasm such a concern?

Bronchospasm is potentially fatal, particularly in patients with asthma or other underlying lung condition. So after the drug was released, there were cases of people with asthma or other lung problems who took Relenza, and subsequently died. ...

You weren't too impressed with this drug -- is that fair to say? What was your overall impression?

It was pretty much a placebo that had side effects. I would summarize it that way. ...

What did you present to the advisory committee?

I told the advisory committee that, in my view, the drug was not effective, either for the symptoms of the influenza or for the complications due to influenza, [and] that in the high-risk patient population that was studied, it was actually slightly worse than placebo in terms of efficacy.

The advisory committee had a fair amount of discussion about Relenza, because it had had a lot of advance hype around it. Ultimately, they decided that it was not worth approving, and voted 13-4 to reject the drug.

Were you the only person who spoke at the advisory committee meeting? Were there other people who made presentations for or against?

Well, Glaxo presented, and they were obviously for it. The FDA medical reviewer made a presentation. Her conclusion was it could either be approved or not approved. It was a fairly borderline drug.

Were there other parties that weighed in on this at all?

There were influenza experts on the advisory committee, so there was a lot of internal discussion in the committee meeting about why the drug -- which had looked so promising in earlier studies -- wasn't working in the current study. There was a fair amount of discussion about its safety, its efficacy.

I think there were a lot of people on the committee who had had a lot of experience, both in clinical trials and in influenza specifically. Ultimately, they just concluded that the drug and the clinical trials hadn't established either safety or efficacy.

How much of a deciding factor was your testimony at the advisory committee meeting?

You'd really have to ask the advisory committee members who voted on it. I think my testimony may have been the only one that brought up certain potential problems with the drug. But since they're pretty bright people, I would think that anyone giving [that] presentation -- and I only spoke for about 15, 20 minutes -- would have turned the tide. They had read the results of the clinical trials. So I think even in the absence of any presentations, they would have come to the same conclusion.

You referred to the clinical trials conducted in the United States. Were there more promising or more successful clinical trials that had been conducted elsewhere that had better results?

There were two other studies -- one in Europe, one in Australia -- where Relenza showed some efficacy. One of the issues was, why isn't it showing efficacy in the United States? One of the most promising hypotheses was that people who take other symptomatic relief medication -- Tylenol, aspirin, cough syrup -- in the United States basically masked any effect of Relenza. So Relenza had no impact on symptoms over and above those baseline medications that people take when they have the flu.

In Europe, in particular, there was a very low rate of people taking those other medications. I guess [there's] just a different cultural environment there. So in the context of not being allowed to take anything to relieve your symptoms, you might notice some effect of Relenza. But in the context of a typical flu, where you have to take other things to manage your symptoms, you wouldn't notice any effect of Relenza over and above those other things.

What happened to you after the advisory committee meeting?

The next day after the advisory committee, several people in FDA management told me that they blamed me for the drug getting turned down in the advisory committee; that I wouldn't be allowed to present at the advisory committee meetings in the future for any other drugs.

At the time, I was also scheduled to be the reviewer for another flu medication, Tamaflu. That review was taken away from me. I guess they were worried that I might apply the same level of review to that application as I had for Relenza. So they gave it to someone else.

Where was that message coming from? Who was behind the message to you?

It was mostly coming from the higher-level FDA management, filtered down through my boss, who agreed with my position, but didn't particularly want to fight it out with all of the other higher-level people. ...

When you got that message, what did you think about the place where you were working?

I was really pretty disappointed in the whole place. There had been previous instances where I thought things had got approved where there were really potential problems. But this was, in my mind, pretty much a no-brainer -- where a drug really just had no efficacy at all, had potential safety problems, and where the advisory committee had really voted pretty overwhelmingly against it. You know, I think that was the first time that our own advisory committee had ever voted to reject a drug, but then look like it would be approved.

So it was personally very disappointing, but it was also really hard to see any motive behind it. It was really kind of a mystery why people at the FDA would be pushing so hard for a drug that had so little benefit and risks on top of that, especially when it was being targeted at otherwise healthy people. Even under the kind of most favorable view of efficacy, it would knock about one day off the course of a two-week flu. So, in my presentation, I said I thought the drug had pretty much no effect, and the most favorable view of the data suggested maybe one day of effect.

That's not really too much of a difference. Why would they want to push so strongly to get that drug approved, even when there had been a public advisory committee where experts had really come to the same conclusion that Relenza had no effect? It was pretty puzzling.

Did you have any explanation for why this happened to you? Did you figure out a way to make sense of it?

I guess the fact that Glaxo is a very large and influential drug company, and that they stood to make a lot of money on this drug. If it looked like it was going to be turned down, then Glaxo would go to the highest levels at FDA and really make a big complaint, and start trying to make life pretty miserable for all the reviewers of Relenza. If that's not why, I'd really like to know why.

But I think in the past, when applications from large pharmaceutical companies had seemed problematic, it would be pretty standard for the company to complain to the head of FDA, or the head of the review divisions. Those complaints would make it down to the individual reviewers, basically saying, "Why are you giving this particular drug such a hard time?"

Was it dangerous for reviewers, if they cared about the FDA, to oppose a drug -- especially from a big drug company?

I think it was pretty well understood that if you were advocating turning a drug down -- particularly if it was from a large pharmaceutical company -- that that wouldn't be good for your career, as far as promotions. It wouldn't be good for your career, scientifically, as far as being able to review other drugs in the future that had potential problems.

Did you feel like you were getting punished for your opinions, for expressing what you felt was the truth about a drug?

Yes, I was punished for my views on the drug, which was, in one sense, rather silly, because it was the drug that didn't have the efficacy. It didn't really have anything to do with me pointing it out that it ultimately got rejected in the advisory committee.

But I took it, not so much as punishment, but a strong encouragement to leave the FDA, because in all my performance evaluations and reviews, I still continued to get the highest grades. So it was pretty clear that they weren't going to do or say anything on paper that would indicate they were unhappy with my review. But that they would continue to make my working life more and more difficult -- not doing reviews of controversial drugs, not speaking in front of the advisory committee, having my review subject to extra levels of comment and potentially removing comments from my reviews.

What does what happened to you say about the FDA and the culture there, and the interest in honesty and truthfulness and safety?

What happened to me would be pretty alarming to people who use prescription drugs; that they really shouldn't have a lot of faith that drugs that are approved have really been demonstrated to be safe and effective to the standards that -- at least I thought before I went to FDA -- that they were trying to uphold.

I'm not sure what someone can do about that. I mean, if you're prescribed something, it's difficult to go back and review the underlying clinical trials and come to some other conclusion. So when you're taking medication, you're pretty much taking it on faith that it's been really thoroughly reviewed, and that all of the safety and efficacy concerns have been really explored, have been documented in the labeling, have been communicated to physicians.

In Relenza, when it was ultimately approved, clearly the risks/benefits weren't communicated to people. Relenza was not by any means an isolated incident. I think it was pretty typical for drugs that had problems to get pushed through to approval. As far as the FDA culture, I think people who didn't like that kind of environment ultimately would just leave.

How did your fellow scientists at the FDA react after you made your presentation at the advisory committee meeting?

A lot of the other reviewers at FDA thought that they liked the fact that I made a stand on Relenza, but said they wouldn't have done it themselves, and that it wasn't going to be good for my career at FDA in the future. So in one sense, it was gratifying to have people say that they agreed with me and liked what I did. On the other hand, it was fairly disheartening to know that no one else is really going to step up and do the same thing on drugs that they were reviewing that had similar problems. ...

So you think, to really do your job, you had to stand up? And to not do that would be shirking responsibility?

Officially, that is the responsibility of every reviewer at the FDA. Your responsibility is to review the scientific data, and come to a conclusion about the safety and efficacy of the drug. Just because the unwritten rule is that things should be fast-track to approval doesn't mean -- ultimately, I just felt a responsibility to do a real review. If I felt the drug was going to be potentially unsafe and offer no benefit, then I wanted people to be aware of that.

Was it a moral thing for you? Was it a conscience thing?

I didn't really ever consider not doing it. To me, it's surprising when I heard other reviewers saying they would experience the same types of things and would ultimately back down. To me, I think those are the people who need to explain why they did what they did.

Does the FDA have a problem?

The FDA has a big problem. It's pervasive. It's throughout the entire FDA review culture. I didn't see it getting any better, which is ultimately why I left. ...

Is there room for bad news at the FDA? Or do administrators at the FDA only want to hear good news?

There is no room for bad news, particularly when large pharmaceutical companies are involved. I think with smaller biotech companies, particularly where it's their first drug application, I saw a little more intellectual honesty about the pros and cons of drugs. But for large pharmaceutical companies, it was pretty clear. I mean, it's called the drug approval process. It's not called the drug review process. So that really sets the mindset on what the job is.

Do you have any opinion about why the FDA tends to be that way?

Well, in the past, the drug companies weren't paying a sizeable fraction of FDA's budget. So now when drugs come in for review, there's a large user fee attached to that. I suppose FDA management views themselves as increasingly reliant on those user fees, and if they're too tough on large companies, then maybe a large portion of their budget goes away. When I started there in 1995, I didn't perceive the same level of pressure as I did when I left in 2000. So definitely something changed during the five years I was there, and that also coincided with the User Fee Act.

Can you prove it, or is it just a sense? Because you say this and then the guy we interviewed this morning says, "No, that's not a problem here at all. We don't care where the money comes from. It's all the same to us."

I think most reviewers at FDA you would talk to or most reviewers who have left FDA that you talk to would find it's pretty well understood that the large pharmaceutical companies kind of have a substantial influence on the review process. I don't think anyone is really surprised about that. Someone who would disagree maybe isn't really involved in reviewing drugs, because I think nearly every application I was involved with or I knew about, where it looked like there were concerns about the drug, it was always understood that there would be high-level complaints from the company to FDA management. Those complaints would come down to the reviewers. There would be both subtle and rather overt pressure to change the reviews, change the recommendations, soften the language.

Since it happened numerous times, I think it's pretty clear that they have a large influence. Whether the user fees directly contribute that influence or whether it's indirect through lobbying, I really couldn't say. But it's influence, nonetheless.

We've had this study that suggested that morale was not great among some of the people at the FDA, that there certainly were frustrations. A lot of reviewers had expressed about having their opinions rewritten or being asked to soften language. Yet the person we interviewed this morning said, "No morale problem at all here, everybody is happy, just the weird bird here who is not happy--"

Well, the unhappy ones leave, so it's a self-selection process. The people who stay for the long term are those who aren't unduly upset about the fact that drugs are getting approved that shouldn't be, or that reviews are being influenced either by drug companies or FDA management.

The whole promotion environment is such that people who raise concerns about drugs don't get promoted. So you have a whole set of people at the top who probably didn't have any morale problems, because they didn't see what they were doing as anything different from what they were supposed to be doing.

The ones who had ethical concerns -- there's no reason to stay around in an environment such as that for year after year, when it's really so hard to make a difference. So those people would leave, and the ones who stayed might think this is how the drug approval process is supposed to go. ...

The person we interviewed this morning expressed shock that somebody would be treated like you were, and if it's happening, it shouldn't be happening. You know, "This is just a small problem. We should correct it, but it can't be something that happens a lot here." Is that fair?

That's a joke. I mean, it happens all the time. I guess the only reason it doesn't happen as frequently as it might otherwise is people start censoring themselves. They say, "OK, after the first time I noted a problem and it's been glossed over, well, why bother the second time?" So you have this process where people will stop raising the concerns by themselves, based on what happened to themselves previously or what they perceive happening to other people. It only takes a couple people to be made an example of. Other people see pretty clearly the writing on the wall, and won't let that happen to them.

You may have 20 reviewers in a particular division, most of them potentially concerned about at least one drug they've reviewed in the past. They see, "OK, someone who makes a stand on something is going to really be forced out." So the next drug that comes along and they see problems, I think many people will be less likely to pursue those problems. Then it does become an isolated incident, when it gets as far as actually trying to get a reviewer to leave FDA based on a review they wrote. I don't think the management has to do that a lot of times before everyone gets the message.

Should we, as Americans, be concerned that the FDA is not doing a great job protecting our safety; that safety has been compromised?

Safety has definitely been compromised. Efficacy has been compromised, too. I think the whole drug review process, in a lot of cases, is compromised.

Are Americans being exposed to drugs that are not worth it, that are not safe, that are not effective?

Certainly. ... I think proof that people are being exposed to unsafe or ineffective drugs comes when drugs are pulled off the market for problems, where in most cases, warning signs were seen at the earliest stages of the review. It would just take either more people to die after taking medication, or just such a public recognition of the fact that a particular drug just wasn't effective or had many safety problems. When those drugs are pulled off the market, that's only the tip of the iceberg, as far as what other drugs are causing similar problems that there's just not an awareness of yet. ...

Are you just a complainer? Are you just one of those guys who is never happy with anything? You're always going to try to dot every "i" and cross every "t" and have total perfection and yell and scream when he doesn't get his way? That's kind of how an administrator characterized the people who make a stink.

Well, I reviewed about 20 drugs when I was at FDA and recommended that three of them not be approved. Ultimately one of those was, in fact, not approved. Two of them were approved and are on the market. So in the other 17 cases, I felt that the drug was safe enough, effective enough to be on the market. I think that would argue against someone who says that anyone who recommends turning a drug down is just making too much of small issues.

[According to Steven Galson,] there's not such a crisis in the FDA; they're just little problems. It's like, we could go out and find all the examples of people like you, and it would just not be an accurate picture of what's happening inside the FDA. It would be a distorted picture.

But behind each drug that gets approved when it shouldn't, or each drug that is ultimately taken off the market, you have lots of people who experience serious adverse events or died. So even if it were just a handful of cases, that's something pretty alarming, if those drugs are going through.

It's really not just a handful of cases. It's a significant number of drugs that are approved when they shouldn't be. And for each drug, you have deaths and other problems that could have been prevented. ...

Are people getting hurt as a result of this, as a result of the way the FDA operates?

Certainly. You can point to Relenza, where a year after it was approved, there was new language added to the label, warning of people who had died as a result of bronchospasms after taking Relenza. What do you say to those people when the drug never should have been on the market, and when they weren't actually deriving any benefit from taking it in the first place?

That's just one drug, one example, [with] measurable consequences of people dying as a result. Then you multiply that by the number of drugs, the number of reviewers, the number of companies, and it becomes a pretty large problem. ...

What's sort of the rule of thumb that you advise people on, if any?

I don't know. If I were asked for advice, it's hard to think of an easy answer. But the message is you'd have to view the drugs that are approved and the claims that are made in the label very skeptically. Don't take them at face value, and really do your own investigation into whether the drug is as safe and effective as it's purported to be.

Although as I'm saying that, it's a rather preposterous thing to tell people to do that every time they get a prescription, every time they get a new drug, to go into that level of detail of reviewing the data.

The way the review process is supposed to work is that someone at FDA is supposed to have already done that for people, and already come to a real conclusion about whether the drug is safe to be on the market. So if you can't trust in that and people have to be their own FDA reviewers, I don't know. I don't know what I would tell those people.


home + introduction + hazardous to your health? + defending the fda + politics, profits & pharma + producer's notebook
producer's chat + interviews + faqs & links + a dirty dozen + discussion
tapes & transcripts + press reaction + credits + privacy policy
FRONTLINE + wgbh + pbsi

posted november 13, 2003

background photograph copyright © 2003 getty images - all rights reserved
web site copyright 1995-2014 WGBH educational foundation