TCPR: Dr. Kesselheim, you and your colleagues just published an intriguing paper on generic medications (Kesselheim AS et al., JAMA 2008;300(21):2514-2526). I’d like to start by asking you to explain the generic approval process. What does the FDA do to give its seal of approval to generic medications?
Dr. Kesselheim: A generic drug is a copy of a brand-name drug made by a different manufacturer. It has the same active ingredient, but it may differ in peripheral features such as the pill color or shape, the manufacturing process that it goes through, or any inert binders or fillers. In order to be approved by the FDA as bioequivalent to a brand-name drug, the generic drug has to go through a number of pharmacokinetic and pharmacodynamic tests that establish that the drug achieves an equivalent distribution to the tissues that it is meant to affect. Bioequivalency is usually established on the basis of a serum concentration of the drug, the time until that concentration is reached, and the area under the curve based on serum concentration of the drug as a function of time. Once a generic drug is shown to be bioequivalent to the brand-name drug, it can be substituted freely after the brand-name drug’s market exclusivity period ends.
TCPR: So how long is the market exclusivity life of a branded drug?
Dr. Kesselheim: That varies. Patents last 20 years from the initial application, but this is usually at the very beginning of the discovery process, and drugs need to go through preclinical and clinical testing before getting approved by the FDA. For most drugs, once the drug is actually approved it will typically have about 10-14 years of market exclusivity.
TCPR: Do the generic drug manufacturers ever have to do efficacy testing?
Dr. Kesselheim: They don’t. The FDA established the so-called “bioequivalency pathway” to approval in 1984 with the Waxman-Hatch Act. The Act says that as long as a generic drug-maker can prove bioequivalency to the brand-name drug, then it doesn’t have to repeat the same clinical efficacy trials that the brand-name company went through originally. Such trials are a major expense, and when that sort of testing was required, few generic companies had financial incentives to enter the market. Brand-name companies, even though their patents might have expired, continued to have a monopoly because it was too costly for generic drug makers to introduce their products.
TCPR: How does a generic company actually go about establishing bioequivalency?
Dr. Kesselheim: Usually, bioequivalency studies entail giving the drug to a small series of volunteers whose blood is sampled at various stages after ingesting the pill, like at 2 hours and 4 hours and 6 hours. That is one way a company can determine the serum concentration of the drug and the time until the maximum concentration is reached. Those factors go into the bioequivalency analysis.
TCPR: Is there is a problem with this system? If bioequivalency studies are done in healthy volunteers, can we apply the data to patients who are ill?
Dr. Kesselheim: That is a really good question, because in some cases elderly patients or patients who have certain illnesses may vary in their ability to respond to medications. Actually, it is also true that when a brand-name drug is approved, no one requires the brand-name manufacturer to test its product in the sickest patients, and so most brand-name drugs are approved on the basis of studies in relatively healthy patients who might have, for example, high cholesterol, or major depression, but nothing else. The frail and the elderly are typically excluded from such studies. Physicians do prescribe the drugs for patients who have many comorbidities, but there isn’t a requirement that a brand-name company prove that its drug works equally well in different sub-categories of patients. To answer your question, since the Waxman-Hatch Act, there have been no systemic concerns identified with the generic approval process. Still, one of the reasons we engaged in this JAMA study was to find trials that compared brand-name to generic drugs and looked at real clinical endpoints.
TCPR: So, tell me about your recent paper.
Dr. Kesselheim: We conducted a systematic review of the medical literature from the past 25 years to look at all the studies comparing the clinical outcomes of brand-name and generic drugs used in cardiovascular disease, and then combined them together into a meta-analysis that encompassed over 800 patients. In these studies, patients were randomly assigned to take either a brand-name drug or its generic version, usually in a blinded fashion, and then after a certain period of time, say 4 weeks, the groups would switch over to the other product. So most of the studies would be termed randomized controlled trials with a crossover design.
TCPR: What was the result of your analysis?
Dr. Kesselheim: We looked at 38 randomized controlled trials covering nine subclasses of cardiovascular drugs, most prominently beta-blockers, diuretics, calcium channel blockers, antiplatelet agents, statins, and warfarin. Combining the data, we found that there was no evidence that the brand-name drugs were superior to the generic drugs in the clinical indicators that we were looking at: vital signs (such as blood pressure, heart rate, and urine output), results from laboratory tests such as urine electrolytes or INR (a measure of blood coagulation), and health system usage.
TCPR: What about the drugs that have the narrow therapeutic index? Did you look specifically at those as well?
Dr. Kesselheim: Narrow therapeutic index drugs are drugs whose efficacy range is thought to be very limited. Underdosing or overdosing of the drug, even by a little bit, can cause substantial morbidity. The two narrow therapeutic index drugs that we looked at are the antiarrhythmic drugs and the anticoagulant, warfarin. We found no studies of antiarrhythmics that met our inclusion criteria for the meta-analysis, but we did locate four randomized control trials of warfarin, and found that the results were the same. When patients were switched from brand-name to generic warfarin, the brand-name version was not more effective than the generic drug.
TCPR: What is your conclusion from the research?
Dr. Kesselheim: I would say that our study supports the contention that doctors should consider generic drugs where appropriate for their patients. We know from other studies that they cost less, which promotes adherence, and in our study we found no evidence that brand-name drugs are superior, at least in the cardiovascular realm.
TCPR: Do you think that your findings are likely to apply to other categories such as psychiatric or neurologic drugs as well?
Dr. Kesselheim: Any conclusions about drugs outside the cardiovascular realm would be extending the study beyond what we actually looked at. That said, generic antidepressant and anxiolytic drugs are held to the same bioequivalency standard as cardiovascular drugs, so I would not expect there to be a difference within those drug categories.
TCPR: What about the antiepileptic drugs?
Dr. Kesselheim: The antiepileptic drugs have narrow therapeutic indices, so physicians should be particularly vigilant when switching to a generic. However, we looked at one narrow therapeutic index drug – warfarin – and found that the brand-name version was not superior, so we would expect that to also apply in the class of antiepileptic drugs.
TCPR: There are the data on the one hand, and there are opinions on the other. I know in psychiatry we often hear anecdotal reports from clinicians relating horror stories of patients switching to generics and doing poorly. What is your take on that?
Dr. Kesselheim: As a side study in our JAMA article, we looked at all the commentaries and editorials published in the medical literature about substitution of cardiovascular generic drugs, and found that a lot of these editorials and commentaries espoused a negative view of generic drugs, despite the fact that we found no hard evidence to support this view in the literature. These anecdotes about particular effects of a drug are useful to generate a hypothesis, but do not substitute for randomized controlled trials. There is no way of isolating what caused the bad result simply from these anecdotes. So if there are substantial concerns about a particular psychiatric drug from numerous anecdotes that have built up over the years, then the right thing to do would be to organize a randomized trial and actually test it in a rigorous manner.
TCPR: Overall, do you think that the FDA’s regulations on generics are adequate or do they need to be changed in any way?
Dr. Kesselheim: I think they are adequate. I think that the FDA needs to have more money devoted to oversight of good manufacturing practices. For example, many brand-name and generic drugs are now made in factories overseas and shipped into the United States. So I think to the extent that we need to be concerned about the FDA’s oversight of generic drugs, we need to be concerned about the FDA’s oversight about the integrity of our prescription drugs in general. But I think that the bioequivalency standard has worked well for testing generic drugs for the past 25 years, so I don’t see a need to change that.
TCPR: Has the brand-name pharmaceutical industry criticized your study at all?
Dr. Kesselheim: PhRMA, which is the drug industry’s trade group, did put out a press release response to our paper. They didn’t quibble with our results or criticize the methodology we used in our meta-analysis. They emphasized that there wouldn’t be generic drugs without new brand-name drugs. I think that the reaction has generally been positive and I hope it has encouraged physicians and patients to be comfortable prescribing and taking generic drugs where appropriate. They are safe, effective, and lower in cost, and therefore can help promote adherence to patients’ medical regimens.