BMS Settles On Vanlev

Pharmaceutical firm Bristol-Myers Squibb (www.bms.com) issued a press release last week announcing it had reserved $185 million as payment to settle a class action lawsuit brought on behalf of investors. The settlement, which includes financial terms which have been agreed to in principle and non-financial terms which are being finalized, should resolve the nearly six-year old dispute surrounding Vanlev, a compound BMS was developing for the treatment of high blood pressure. Development and marketing were halted in 2000 when the U.S. Food and Drug Administration informed BMS that it was concerned about side effects of the drug.

According to Labaton Sucharow, the law firm representing the class of investors bringing the suit, the settlement is significant because of BMS' agreement to report publicly on the clinical trials of its products. Not only will information about what clinical trials have been conducted be available, but the results of the research will be presented in on-line databases established specifically to serve as information resources to physicians and patients about BMS drugs.

Is this something new? To a large extent it is. While drug manufacturers have always had to present results of the clinical trials they sponsor to the FDA when they applied for approval of a drug, the availability of that information tended to be somewhat restricted. Research participants and consumers were usually provided with only aggregated data -- that is, the overall results of data after they had been analyzed. And if clinical trials did not demonstrate a drug to be effective, then the results never needed to be made public, or even submitted to the FDA, if the manufacturer decided against marketing the drug.

This settlement is an example of how the game is changing. More and more, results of clinical trials are being made available to physicians, medical researchers, consumers, and, most importantly, research participants. Of course, this also means that competing companies will know more about one another's results. That should be interesting.

In the meantime, I am curious to see how and when BMS will announce the details of the data sharing terms of its settlement. When I checked today, its web site database of clinical trial results did not include information on Vanlev.

-Joe

Related Topics: Sorting Through the Latest Drug Studies, Warning for Blood Drug Hydroxyurea

Technorati Tags: bristolmyerssquib, vanlev

It's Not a Problem

Continuing from last time...

One suggestion was to require clinical trial sponsors to report to the IRB information about prior disapprovals. In this way, it was argued, the IRB considering a protocol would be put on notice that there might be issues requiring special attention. Another suggestion involved requiring sponsors to post information about prior disapprovals on a public web site.

As it turns out, the FDA decided not to regulate IRB shopping in any way - at least for now. Its decision, posted this month (January 17 Federal Register), noted that, according to those who submitted comments on the 2002 FDA notice, IRB shopping did not present "a significant problem." Further, many of those who commented expressed logistical concerns, especially for clinical trials conducted at multiple sites, each with its own IRB. Finally, it was argued that the added burden to IRBs and delays would "not provide any additional human subject protection."

Its conclusion was that "IRB shopping either does not occur or does not present a problem to an extent that would warrant rulemaking at this time."

My opinion is that the administrative and business disincentives for pharmaceutical firms and research institutions were perceived to be much greater than the benefits to the research process or the research participants. At the same time, research participant advocates did not effectively present a case for greater regulation in this area.

So, for now at least, the status quo stands.

-Joe

IRB Shopping

This is the first of a two-part blog inspired by a recent announcement by the Food and Drug Administration (FDA).

Institutional Review Boards (IRBs) are panels of scientists, health care professionals and community representatives charged with overseeing clinical trials. If the research involves a product regulated by the Food and Drug Administration, then the sponsor of the research must submit to an IRB. The original plan (known as the protocol) must be approved, including the consent form to be used when recruiting participants. Periodic reviews are conducted by, and adverse events are reported to, the IRB. If an IRB finds a problem with a clinical trial under its authority, it may suspend or terminate the research in the interest of the safety of the participants. Participants may ask the IRB to intervene in situations of serious disputes with the study team regarding the research. In short, an IRB can make or break a clinical trial.

While IRBs all follow common federal regulations governing research involving human subjects, there is ample opportunity for varying interpretation of the rules. IRBs often develop "personalities" based on the culture of the organization they are in. Some are quite deliberative and conservative in their decisions, while others are relatively quick and efficient. Where the safety and well-being of research participants are concerned, slow and conservative seem appropriate. But, with speed and efficiency being key attributes in the business world, which personality would you prefer if you were a company who wanted to test a product?

These perceptions led the federal government to question a few years ago whether research sponsors, such as pharmaceutical companies, search out or "shop" for the most lenient IRB they can find. Also in question was whether or not research protocols which had been disapproved by one IRB were routinely submitted to subsequent IRBs until one was found which would approve the research.

In 2002 the FDA published a request for comments on the question of IRB shopping - was it a real problem of any significant frequency and how might it be discouraged?

Next time: the FDA's recent decision on this issue...

-Joe
Related Topics: Drug Companies Offer More Details on Clinical Trials, Holes in US Drug Safety Net

Technorati Tags: FDA, IRB, clinical trials

Lessons From the Placebo Effect: A New Design for Clinical Trials?

This is not easy to explain (at least not for me), so please hang in with me...

As I mentioned last time, the standard design for a clinical trial tests a new drug or treatment against a blank or fake treatment - the placebo. But designing a study that way means that at least some of the participants will not receive the treatment. They will only receive the placebo. Because they are usually 'blinded', they know that they may or may not be receiving the actual treatment, but they don't know whether they are or not. Given that now there is a better understanding of what happens when an individual experiences a placebo effect, there is growing support for changing the way placebos are used in clinical trials.

In this new scheme, participants would all be told they were receiving the experimental treatment. Whether they are given the real or placebo treatment, their bodies (and minds) will react to the thought that they are receiving the treatment. In this way there is a better comparison between the two arms of the study. Why? Because both arms will experience whatever placebo effect is associated with the study. Any differences observed between the two groups can be attributed to the drug or treatment itself. The analysis will not be compromised by the placebo effect because everyone will have experienced the placebo effect.

The problem with this approach is this: Suppose most or many clinical trials are designed in this new way. Everyone is told they will be receiving experimental treatment. It's called 'covert' administration, and it would mean that those who actually get placebo are lied to. This would not break the rules governing human subjects research, provided a) it was necessary to the scientific purposes of the study and b) the study was approved with this design by the Institutional Review Board overseeing the research. I suppose it might make some of us feel better if participants were told after their participation was completed that there actually was a placebo arm. I am anticipating a number of annoyed participants, however...

From a strictly scientific point of view, it is probably better science to design trials with covert administration of placebo. We probably should have been designing clinical trials this way all along. If we can get past the perceived ethical hurdle of deceiving participants, it just might become the standard.

I would love to hear from clinical trial coordinators and others on this issue.

-Joe


Related Topics: Are 'Dummy Pills' a Dumb Idea, or Do They Really Help?, Placebo and sham treatment

Technorati Tags: placebo, clinicaltrials, medicalresearch

The Placebo Effect

Many clinical trials are designed to show what effects a drug has. It is important to demonstrate all of the effects of the drug - the good ones as well as the bad ones. In order to do this, participants are often divided into two groups (or "arms"): an experimental group and a control group. Participants in the experimental group receive a dose of the drug, while participants in the control group are given something that looks like the drug, but actually contains no drug at all. This fake or empty dose is necessary because people often react when they think they are being given a drug even though there is really no drug in the dose. The effects they experience may be positive or negative. They may be similar to what the drug is supposed to do or they may take the form of unwanted side effects. These effects, caused by the fake doses, are known as "placebo effects" and the fake doses themselves are called "placebos."

The placebo effect has been known for many years. Occasionally, the placebo effects are every bit as powerful or effective as the drug itself. In fact, before governments began regulating drugs, many medications actually contained no active compounds. But they were cleverly advertised, they were strong smelling or tasting, and they were administered with much care. As a result, patients believed they would in fact be affected by the medications and they showed signs of improvement in whatever the particular condition. Sometimes they showed side effects as well. We didn't know exactly what caused the placebo effect other than to assume it was mental and not physical. But, for the purposes of conducting clinical trials, it really didn't matter. All that mattered was that there were two arms to drug studies so that the effects of placebos were compared with those of the actual test drug in order to show what the real, physical effects of the drug were.

What this means, however, is that any particular participant would be put into either the experimental or the control arm of the study. If they were put in the experimental arm, they received drug; if they were put in the control arm, they received placebo. Of course, they couldn't be told which arm they were being put into because then the true effects of the placebo would not be known. There are ways of designing studies such that participants are put into one arm of the study and then crossed-over onto the other arm, but this is not always possible.

Recent research has shed some light on how the placebo effect is actually caused in the brain. And that information might well change the way clinical trials are designed. More on this next time.

-Joe

Related Topics: Can You Think Pain Away?, Placebo Effect in Multiple Sclerosis

Technorati Tags: placebo,drugtrials,webmd

What To Ask About Preliminary Results

If you're considering participating in a clinical trial or you have already volunteered to participate, you have the right to some very important information. Although the information should be offered to you, if you don't get it - ask!

You are supposed to be informed of all preliminary data that have been generated. If laboratory and/or animal and/or human studies were done previously and those studies showed anything of importance to the clinical trial, you should be informed of it. Examples would include a list of side effects and their frequencies, safety and effectiveness of a drug at the doses tested, and the maximum dose which participants could tolerate, if known. This is a partial, not exhaustive, list.

If the oral or written explanations given to you do not include such information, ask for it. Ask for it directly, using questions such as:

• Has this drug (or treatment or test or device or whatever) been tested in humans or animals before? What did such tests show?
• How were the results of the previous tests used in the design of this study?
• What side effects were seen in previous studies? If so, what were they and at what doses?
• Did anyone die on a previous study? If so, at what dose?

The more you know about what is known, the less chance you'll be disturbed by a surprising result of your own participation. On the other hand, knowing a lot about preliminary results may be discouraging or even frightening. Have you ever actually read all of the possible side effects warnings of even the most common over-the-counter drugs? If you read them closely and then take the medication, you're bound to experience some (or all!) of the side effects.

Finally, if you feel you are not being given all of the available information, don't volunteer for the study. And if you're already in a clinical trial and you believe you're not being told about current findings, then it's probably best to withdraw from the study.

If the article on the results of the Merck-sponsored clinical trial of Vioxx had been more forthcoming with the cardiac problems which were observed, then there might be less finger-pointing (and perhaps less law suits!) directed at Merck.

-Joe

Related Topics: Is a Clinical Trial Right For You?, Clinical Trial Listing

Tags: Vioxx,clinicaltrials,preliminaryresults,webmd

What Has the Vioxx Debacle Taught Us?

Medical journal editors, physicians and medical researchers are all rethinking how results of clinical trials are presented in the scientific literature. Maybe the bar needs to be raised. It's easy to criticize a pharmaceutical company-sponsored clinical trial on the grounds that the desire to present the science in a way that is most favorable to the product being tested outweighs scientific rigor.

That's not really fair, though. True, a company such as Merck is anxious to see favorable data published. But articles in scientific journals are not simply advertisements. They aren't even considered for publication unless they've met several criteria.

First, the article must address a significant question. An example of such a question would be whether or not a new drug is safe and more effective than existing medications for an important condition such as arthritis.

Second, strong methodology must be used in the study. An appropriately large number of participants, good statistics and careful controls all contribute to the legitimacy of a clinical trial's design.

Third, the findings of the study should clearly lead to a conclusion that means something. Less-than-conclusive findings are not very publishable.

Finally, the study design, analysis and presentation must be deemed acceptable by reviewers who examine the article with expert eyes. Only the most significant findings made from well-designed studies are published in top peer-reviewed journals such as the New England Journal of Medicine.

Those articles that are published should be reliable beyond any reasonable doubt. Scientists, physicians and patients should be able to count on the validity of the published results absolutely. And that's exactly the reception the article on Vioxx published in 2000 was given. The fact that the NEJM retracted the article late last year shows that the reliance placed on the integrity of these particular published results was in fact misplaced.

Hopefully, it doesn't happen all that often. But, in this case, it looks like several of the checks and balances which usually prevent less-than-solid findings from being published failed. The authors of the article failed. The reviewers failed. And the NEJM editors only 'succeeded' when they took the unusual step of retracting the article four years after its publication.

We need to do better. Next time - what questions to ask about preliminary results if you're considering participating in a clinical trial.

-Joe

Related Topics: Lawmakers: Doctors Misled on Vixx Safety, Timeline of Pain Reliever Controversy

Tags: Vioxx,clinicaltrial,NEJM,webmd

The Trouble with Vioxx

The New England Journal of Medicine can make or break a new drug. That is, an article published in the Journal which describes favorable results of tests of a drug in humans can dramatically increase sales of the drug. And having the article retracted (withdrawn) by the Journal can cause a company's stock to plummet.

Take Merck's Vioxx, for example. The results of a clinical trial involving approximately 2000 participants were submitted to the NEJM for publication. The article was read and commented upon by reviewers for the Journal -- physicians and scientists -- and, following any necessary revisions, was published in 2000. Because the article not only described how Vioxx was effective in treating arthritis pain but also appeared to cause less stomach problems than existing medications, physicians began prescribing Vioxx for their patients. Stories in the popular medical news cited the article and, of course, people suffering from arthritis asked their doctors about Vioxx. Things looked good. For awhile.

After enough prescriptions had been filled and patients had taken Vioxx for a long enough period of time, it became apparent that one of the side effects of the drug was cardiovascular problems - specifically, heart attacks. As a result, the Food and Drug Administration advised Merck to pull the plug on its very popular drug. Merck did so in September of 2004.

It was bad enough for Merck to lose revenue from its U.S. sales of Vioxx. But then, late last year, the NEJM retracted the original 2000 article. A retraction means that the publisher does not stand by the validity of the results reported. In this particular case, the Journal editors felt the entire article should be withdrawn because there was evidence that the authors of the article failed to accurately report their findings. Specifically, it appears there were at least two problems with the publication.

First, the study which was reported may have been designed so that the negative cardiac effects, which were suspected by Merck, were minimized. Second, the authors failed to mention heart problems which arose in clinical trial participants after the study period was over. It may be argued that those heart problems should not have been counted in the data because they fell outside of the official window of the study, but it has also been suggested that the authors should at least have mentioned them, either in the article or to the editors when the article was submitted for publication.

The scientific record may have been corrected (after 4-5 years) but, in the meantime, people and their doctors made medical decisions based on the results presented in that 2000 article. And they thought the decisions they were making were backed up by sound scientific evidence. Wrong.

Next time: How this experience may change the way in which articles describing the clinical trial findings are presented and interpreted.

-Joe

Related Topics: Ibuprofen, Naproxen: No Heart Risk, Painkiller Update: What You Need To Know

Tags: arthritis,vioxx,nejm,clinicaltrials,webmd

Are Children Taking Untested Drugs?

It's not just drugs. It's also devices such as surgical instruments and artificial organs, as well as new products for diagnosing diseases. Physicians are prescribing, performing surgeries, and diagnosing for their pediatric patients using drugs, devices and test methods that were never tested in children! It's true.

It's true, but there's an explanation... sort of. Most medical advances benefit adults. Because adults comprise the largest segment of the population, it should not be any surprise that most medical research advances the care of adult conditions, diseases, and health. Most clinical trials are designed, therefore, to test new ways of treating adults. They must be designed to adequately prove the safety and effectiveness of the new idea, but are designed as efficiently as is feasible. As a result, the study population is often limited to adults (because that is the population for which the new idea is expected to be used) and children are often specifically excluded from the study. So are pregnant women, but that's another story.

What's the big deal, then? The big deal is that many drugs, for example, are eventually used by physicians on children, often in lower doses. Physicians are not prohibited from prescribing any drug they think will help a patient, including those that have not been tested for a particular disease or condition, unless there is some reason to believe the drug would be harmful. Of course, the physician keeps an especially close eye on the child's reaction to and progress on the drug, but she cannot be sure there will be no unexpected side effects in her pediatric patient because children were never included in the clinical trials of the drug.

If a drug is developed for a childhood disease or condition, then it is always tested in children. The most significant reason for not including children in clinical trials of drugs intended to address predominantly adult medical issues is to avoid subjecting a particularly vulnerable population (children) to what may be uncomfortable or even harmful effects of a new drug during the clinical trial.

It's ironic how our desire to protect children from harm in clinical trials may actually cause more harm down the road when these very same drugs are used on children without the benefit of information learned from clinical research.

-Joe

Related Topics: Clinical Trials: Concerns for Women, Children and Genetic Privacy, Pint-Size Patients Need Pint-Size Prescriptions