Envisioning 2.0

This is the last major article in an ongoing series on this blog titled: “The FDA and the Pharmaceutical Industry.” Posts filed under this category focus on the intersection and relationship between the agency and drug companies.

A few months ago, I conducted an interview with Dr. John Powers, Lead Medical Officer for Antimicrobial Drug Development & Resistance at the FDA. During this wide-ranging discussion, we talked about a number of important topics, including efforts to curb the spread of antimicrobial resistance, the drawbacks of non-inferiority trials and FDA advisory boards. Powers’ extensive commentary appears below.

Interview

Q: Can you briefly tell me why it is so important to preserve the effectiveness of antibiotics?

A: There are two answers to this question:

a. We want to preserve the effectiveness of antibiotics so that when people need these drugs for serious and life-threatening diseases, they will be effective. Using an antibiotic inappropriately can negate its use. If someone receives an antibiotic for a viral upper respiratory tract infection, then when they need that drug if they develop pneumonia, it may no longer be effective. Even more of an issue, if one of their relatives develops pneumonia, the drug may not be effective for them either. It’s hard to think of another class of drugs where the use of a medication affects patient B even though it is being used in patient A. So, the real reason we want to use these drugs carefully is so that they work when we need them.

b. There is another reason to limit the use of antibiotics to situations in which they have proven effectiveness. People widely assume that antibiotics have large treatment effects compared to placebo and relatively few adverse events. However, in self-resolving diseases the benefits of antibiotics may be small, which magnifies any potential issues with toxicities. In addition, the benefits in these self-resolving diseases (getting rid of sniffles or cough) may be of a similar nature to the adverse events caused by the drugs (diarrhea and headache). Many of the serious adverse events associated with antibiotic use occur in people who either do not have a bacterial infection or who have a self-resolving disease where the benefits of antibiotics may be small. This shifts the risk-benefit equation such that serious adverse events become more concerning when the benefits are less. Antibiotics are life saving drugs for serious and life threatening infections, but it is asking a lot of a drug to be “life saving” for a disease that itself is not commonly a cause of mortality or serious morbidity. Antibiotics are prescribed to treat diseases in patients, not merely to treat organisms or isolates, so we need to know how the antibiotics affect the course of the disease in people and that the drug has a benefit for patients. We can preserve the effectiveness of antibiotics by ensuring people are diagnosed appropriately when they have a bacterial infection. There is a desperate need for better diagnosis. We simply don’t have adequate procedures and processes to identify who has an infection, when the infection is caused by a virus, who will benefit from antibiotics even if the infection is bacterial and who will not, and what the resistance pattern of the causative organism, if that even makes a difference in self-resolving diseases. All of this can lead to inappropriate use and unnecessary adverse events.

Q: A recent USA Today article highlighted the dangers of methicillin resistant Staphylococcus aureus (MRSA), yet many people have never heard of it. What is the FDA doing to inform and protect people from/about MRSA and other resistant bacteria?

A: The FDA is doing a number of things. First, FDA is part of the Get Smart campaign, which helps educate the public about the dangers of the inappropriate use of antibiotics so as to preserve the benefits of antibiotics for the diseases in which we really need them. Second, in 2004 FDA put a rule in place mandating that pharmaceutical companies must put a notice in every antibiotic drug label reminding physicians that they should only prescribe if they know the patient has a bacterial infection. FDA also is co-chair of a government interagency task force of 11 different agencies focusing on antibiotic education, research, product development and surveillance. There are a number of drugs under development that appear promising for diseases due to Staphylococcus aureus, and at least in the test tube some of these drugs appear to have activity against methicillin resistant strains of the organism. Of course, we need to know that translates into benefits in terms of treating diseases in people in clinical trials.

Q: Many pharmaceutical companies have gotten out of the business of making new antibiotics. Some have suggested that industry is no longer making antimicrobial medications because the FDA has increased the length, size and cost of drug studies. What is the FDA doing to address these concerns?

A: There are a number of reasons why pharmaceutical companies have exhibited less interest in developing antibiotics. Some of them are economic, in that antibiotics make less money than drugs that are prescribed chronically. On the other hand, antibiotic trials are the shortest and need the smallest numbers of patients of all clinical studies. Antibiotics also have the highest approval rate of any therapeutic area. If you look at overall trends, you will see that drug submission rates in all therapeutic categories have decreased. So, the idea that antibiotics represents a special case is not supported by the available evidence.

Some large pharmaceutical companies have decided to get out of the antibiotic development business all together. That is an economic choice. FDA’s basic requirements to show that a drug is safe and effective have not changed. The methods used to meet that requirement may change as science changes. As science advances, we learn more and that new knowledge should be incorporated into the drug development process. If it costs more, that cost is helping protect the public’s health. Cars didn’t use to have seat belts or air bags, but now they do even though they cost more because the idea if those things prevent harm.

In addition, many companies are less reluctant to do large trials to develop drugs for large markets. Most antibiotic applications to FDA containing several trials have a few thousand people in them – at most. The fewer the patients, the more uncertainty there might be about the drug, both in terms of what we know about effectiveness and certainty in terms of what we know about safety. Changes in FDA regulations can’t turn a $190 million medication into a billion dollar drug.

However, I have to say that it’s certainly a good thing that we don’t have so many infections in terms of public health. The fact that some infections are hard to study means they are not rampant in our country and we have to see that as a good thing. It also means we have to develop more efficient ways to study diseases that are not as common.

Q: What would be the ideal clinical trial of a new antibiotic?

A: According to FDA regulations there are seven criteria that define an adequate and well-controlled clinical trial. Clearly, there are areas where we can improve the methodology used to meet those seven requirements, but a few of these requirements are areas where progress would be really helpful. Specifically:

- We want to be sure that the people we are studying actually have the disease in question. For example in the area of infectious disease, putting people in a trial that don’t have an infection is not very efficient and exposes people to potential harms without the potential benefits. This is why we need better diagnostic tools.

- We must have endpoints that are reliable and well-defined. In serious illnesses, the endpoint in which clinicians and patients are most interested is prevention of death. For less life-threatening conditions, we have to look at other endpoints like symptom relief. For example, in meningitis, we might want to look at clinical endpoints like prevention of death and complications like neurological abnormalities.

- We need to reexamine the questions we ask when we evaluate medications. Is one antibiotic better than another, or is it comparable? If it is comparable, does it do a better job of improving certain areas of the disease than its comparator? Are we making people better in terms of resolution of symptoms or do they suffer from adverse events because the drug did not cure the infection? Or adverse events from the drug itself? In testing medications we have to look at how we ask these questions in clinical trials so that we get answers with more certainty. If trials are designed appropriately and efficiently, we can get more information with the same or less effort.

Clinical trial development and the translation of basic to applied science is an area that the FDA is closely examining. One major project in which FDA is involved is the Critical Path Initiative. Some of the major objectives of this project are to:

- Help develop new ways of predicting whether a novel compound will be successful – i.e., ultimately be used in people

- Develop better ways to assess whether a medication is safe

- Streamline the path from the laboratory to market.

Obviously, a big part of this effort is helping pharmaceutical companies design and field clinical trials that will provide accurate and adequate information. FDA has partnered with a number of companies to move this effort forward. [Editor’s note: To learn more about the Critical Path Initiative, please go to: http://www.fda.gov/oc/initiatives/criticalpath/.]

Q: Over the past few years, the CDC (with the assistance of the FDA) has made educating the public on the dangers of antibiotic resistance a priority. In your mind, how effective have these efforts been?

A: I think that our efforts have been very effective. When you look at the rate of antibiotic prescriptions in areas where people don’t need an antibiotic, like viral infections, you see that they have decreased since we started educating on the appropriate use of antibiotics. We have to continue these educational efforts because if you stop, things revert to where they were before.

Q: In May, the House appropriations committee passed an amendment that would ban any scientist from serving on a FDA advisory panel if they had worked for or consulted with a drug company. Are advisory board conflicts of interest a real problem? If passed, would this new amendment hamper efforts to bring new antibiotics to market?

A: In my mind, one of the most important things we need to do is put people on advisory boards that are clinical trial design experts. We’ve done a good job of this on some advisory committees, less so on others. You want to make sure that you have a range of expertise on advisory board panels, including clinicians who use the drugs and those who have expertise in how to study drugs.

I personally don’t think that the House bill will hamper bringing antibiotics to patients. Advisory boards are only one step – albeit an important one – on the road to drug review and approval. FDA has been dealing with issues relating to conflict of interest for years and will continue to address these issues going forward, to develop appropriate standards. Sometimes it is challenging to find people without conflicts and other times it is easier depending on the topic.

Q: In your recent lecture during the annual CDC Get Smart conference, you talked about the side effects associated with antibiotics. If the FDA were evaluating a new antibiotic, would it tolerate a higher rate of side effects in order to provide medical professionals with a new weapon against bacteria?

A: We don’t develop drugs just to fight bacteria. We develop drugs to help people. Antibiotics usually help people by fighting the bacteria, but they are not always linked. Many drugs that have worked in the test tube end up not working in people, and many of those drugs fail due to safety issues in people. So, the answer to your question depends on what the benefits and risks are of an individual antibiotic developed for use in human beings, and what those uses might be.

I think that holding on to the idea that we are developing drugs only to kill bacteria really narrows our scope. We need a much broader view of antimicrobial development. There may be new drugs or biologics that help people but do so by mechanisms other than inhibiting growth of bacteria.

I’ll answer your question with a question: if you were the patient, would you take a drug with greater adverse effects, even effects that might rarely be fatal, for a self-resolving disease? For example, would you accept a drug whose adverse effect was death to cure a case of the sniffles? Probably not.

On the other hand, would you accept higher adverse effects for a drug that prevents mortality or death? Maybe. It is important for people to understand that we treat humans not bugs. People don’t come in to the doctor’s office and say: “I want my E. coli gone. They want their symptoms treated and to feel better, and they’d like the drug that makes them feel better faster than a different drug or than no drug at all.

Overall, bugs get a bad rap. We cannot live without them. The real issue is that everything is about balance. In drug development, we talk about balancing benefits and side effects. With bacteria, we talk about bugs that are good and those that are bad. If you sterilized your entire body, you wouldn’t live very long, as bacteria in your intestine help make clotting factors for your blood. Exposure to bacteria as a baby help the intestinal lining develop properly. It’s only when bugs get in the wrong place in the wrong numbers that they cause a problem.

Q: Your recent lecture on antibiotic clinical trial development received a number of comments from pharmaceutical company executives. Can you respond?

A: The first comment from industry was that proving whether an antibiotic is superior to another is very difficult. This perception comes from the fact that many people enrolled in clinical trials don’t have resistant bacteria, or they may not have bacterial infections at all. If we had adequate diagnostic tools, we’d do a better job of identifying patients who had bacterial infections including those due to resistant pathogens and selecting people who would be most likely to benefit from a medication. We haven’t done a lot of this – in studies or in clinical practice. With good diagnostic tests, we can eliminate the high numbers of people who are getting drugs they don’t need and suffering from the adverse effects of antibiotic medications without getting any benefit.

The second comment from industry was that the FDA suggests, but does not require changes in clinical trial design. That’s not exactly correct. For example, FDA regulations already state that non-inferiority active controlled trials are not appropriate studies if they don’t ensure that the drug under study is any more effective than placebo. This is the case in some self-resolving diseases like otitis, bronchitis, sinusitis, and uncomplicated skin infections. These regulations have been in place since the 1980’s.

There are also FDA guidelines published by the International Conference on Harmonisation (ICH). ICH-E9 and E-10 have very clear guidance on where non-inferiority trials are appropriate and inappropriate and the kind of information you need to do an appropriate noninferiority trial and these have been in place since 1998 and 2000. Antibiotics are subject to the same requirements of demonstration of safety and effectiveness as other therapeutic areas. The need to demonstrate effectiveness has been part of FDA regulations since 1962.

On September 12 there was an Anti-infective Drugs advisory committee that addressed the issues of noninferiority trials again, as we have for the last several years, and there was a clear message that we need to move on and do superiority trials in circumstances where noninferiority trials are not scientifically justified.

As you can see, the information is out there. Its part of the law, and its part of previous guidance that would supersede any older antimicrobial guidances – it might help if FDA took the old guidances off the website at this point. The FDA has developed general guidance and talked about this issue in relation to antimicrobial development on several occasions. So, this should not be a new issue. Folks should not be surprised at this point – the first meeting we were scheduled to have on the issue of noninferiority trials was 5 years ago. It is clear that noninferiority trials are not an appropriate option when they cannot differentiate a new drug from placebo, even if the effect of the new drug looks similar to an old drug. It is true that FDA is updating it’s indication specific guidances for various infectious disease indications, but in the meantime the more general guidances are available and the principles in those guidances apply to antimicrobials as well as they do to other therapeutic areas. The draft guidances on antimicrobial development for specific indication currently on FDA’s website are almost a decade old, and FDA has been challenged by groups like the Infectious Diseases Society of America to come up with more efficient trial designs. We have been working hard on this.

Sponsors would be well served to discuss their plans with FDA before proceeding with their development programs, and not rely on guidances that they now know are under revision. The question is, what does the appropriate science tell us to do? If the appropriate science tells us that what is in an older guidance is not scientifically appropriate based on what we know now, it does not make sense for public health to keep doing things using older methods while awaiting publication of new guidances.

Now, are we enforcing our regulations? It seems clear we need to do more to get this message across nor does it seem FDA is being clear enough with sponsors if they think that appropriately designed clinical trials are not a “requirement”. Non-inferiority antibiotic clinical trials are appropriate in some situations like meningitis and severe pneumonia, but they are not giving us the information we need to assure drug effectiveness in other diseases like otitis, bronchitis, sinusitis and uncomplicated skin infections.

We need to have clear evidence of an antimicrobial’s effectiveness to balance any potential harms - which include harms to individual patients. Also, without evidence of effectiveness we are contributing to the problem we are trying to solve by causing more resistance and approving medications that we aren’t sure will benefit people. That’s an important public health issue to FDA.

Q: Please provide me with your general comments about the issue of antibiotic resistance and your suggestions for how we can get policy makers and the public to take action on this issue.

A: Overall, I think that people need to be more educated on the issue of antibiotic resistance and that inappropriate use of antibiotics can and does affect all of us. From a clinical trial standpoint, we need to develop ways to better evaluate whether a drug is safe and effective, and how to measure whether potential advantages in a test tube translate into real advantages for people.

This entry was posted on September 19, 2006 at 1:56 pm and is filed under FDA and the Pharmaceutical Industry. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.