FDA Should Add a Comparative Effectiveness Arm to Final Trials

The Food and Drug Administration’s Prescription Drug User Fee Act is up for reauthorization next year, and so is the consumer and drug industry face-off over the contentious issue of comparative effectiveness research (CER). Consumers, patients and some physicians are demanding that CER be required of all new drugs coming to market when there are already FDA-approved therapies for the same condition. They say it will give payers and patients immediate feedback on the relative worth of the latest drugs, which are always more pricey than what preceded them, especially if the older drugs are coming off patent.

Industry opposes including CER arms in final efficacy trials. The companies claim it will place additional costs on the already expensive new drug development process; provide inadequate information for actually divining the relative worth of two competing therapies; and dissuade companies from investing in follow-on drug research, which can turn up drugs that are significantly better than older drugs.

The American Enterprise Institute’s Scott Gottlieb, who served in the FDA during the Bush administration, this week offered a lengthy brief in support of the industry position. Unfortunately, he sets up a straw man in order to knock down what could be a very effective tool for lowering the cost of medicine. It behooves industry leaders to ignore his advice, and to ignore the bleating of their marketing departments’ incessant demand for follow-on drugs.


Requiring comparative arms in Phase III trials in situations where it is appropriate will encourage companies to channel their scarce research-and-development dollars into more medically significant opportunities – therapies for previously untreatable conditions or better drugs where the current options are inadequate. A CER requirement at FDA could go a long way toward turning around the sharp decline in drug industry innovation that has occurred over the past decade.

Phase III trials are also the right time to carry out CER from a societal perspecctive. Physicians and patients are demanding this information, not only because of cost concerns but because they very often they don’t know what works best. The Affordable Care Act created the Patient Centered Outcomes Research Institute and gave it a half billion dollars a year to carry out CER. Given the hundreds of competing therapies that were developed over the last half century when there was no comparative research, PCORI already has a lot on its plate. It makes no sense to add to its burden by not collecting information about the relative medical value of future drugs. The drug development process is when CER is the easiest and cheapest to carry out, since it only requires adding one more arm to an already planned trial — not starting all over from scratch as government-funded CER will have to do.

Gottlieb claims consumers and patient advocates are demanding that “new medicines should demonstrate that they offer clear advantages over older, often cheaper drugs” before they are approved. Some may say that, but it’s a red herring. The FDA standard now for efficacy is that a new drug has to be better than nothing; in other words, it must be tested against placebo (unless it is for a life-threatening condition where there is already an approved therapy on the market; then the new therapy must be tested against that therapy, or added to it, because denying the already-approved therapy in a placebo arm in a trial would be unethical).

I do not support the position of advocates like former New England Journal of Medicine editor Marcia Angell who think new drugs should have to be proven better than what exists before they are approved. If companies want to bring comparable therapies to market, that’s their business. It may even be the case that some me-too drugs work in some sub-populations, but not in others. So if one drug fails to achieve lower cholesterol, or offer arthritis pain relief, for instance, the doctor can switch her patient to the newer drug. But if the new drug is not proven to be better than what exists in a large Phase III trial, then physicians, patients and payers will have the information they need to insist that people start on the cheapest, comparably effective medicine that is available. For most drug classes, that will mean a generic.

These non-inferiority trials do not have to be inordinately more expensive. If companies do not opt to try to prove superiority, it only requires adding a third similarly-sized arm to the final trials. While this could in theory add close to 50 percent to the cost of a final trial (going from two arms to three arms), it won’t require exponentially larger arms since they’re not trying to prove superiority. PCORI could even consider allowing companies to apply for grants to underwrite the comparative arm in a new drug trial. If it is in a therapeutic class or for a condition that the new agency has deemed important, adding the comparative arm or arms to a trial that was largely industry-funded would be substantially cheaper for the government than starting over from scratch.

A word of caution is in order before the FDA requires comparative arms in non-inferiority trials for new drugs. If the condition isn’t life-threatening, it should still require a placebo arm or insist on an adequately sized trial to avoid the “dumbing down” of new drugs. That has been a problem in the antibiotics field, where a succession of non-inferiority trials without placebo arms may have created a situation where new drugs have been approved even though their efficacy is in doubt. This is a fluke of statistics where each succeeding drug is near the bottom of the band that defines statistically significant non-inferiority. Adequately-sized trials with clear endpoints can usually avoid this problem.

Merrill Goozner has been writing about economics and health care for many years. The former chief
economics correspondent for the Chicago Tribune, Merrill has written for a long list of publications including
the New York Times, The American Prospect, The Washington Post and Financial Times. You can read more
pieces by him at GoozNews, where this post first appeared.

Spread the love

33 replies »

  1. Where are you calling from? david grow lexapro label bodies Gordon Jones, the Atlanta History Center’s senior military historian and curator, said it could have been fired by federal soldiers from outside the city in an effort to strike the Confederate’s Railroad Roundhouse, a key military target in the general area where the device was found.

  2. If collecting the data prior to formal FDA approval is too onerous, then why not at least require that post-marketing CER studies be completed and submitted within the first five years after approval? It’s a lot better than what we have to work with in the industry today.

  3. In response to Dr. Greg Maguire, the idea that public opinion and desire is a burden to the FDA goes against the purpose of this very organization. The FDA exists to protect and serve the American public. While its resources may be limited, the requests we place upon the FDA should never be diminished. As with any political situation, the people in charge listen to the constituents with the loudest voices. So if this is something the public and professionals within the public feel is important, I think that this topic should be addressed and decided upon. After all, the whole intent of a renewal of law is to update and improve it.

    Moreover, while I agree that the cost burden already placed on industry development is extensive, as a student pharmacist, I know that many blockbuster drugs on the market are simple twists of a chemical molecule of medication that already exists. These twists sometimes provide little if any additional benefit, and cause patients to pay more for their prescriptions than what may be necessary. Some clinicians create time in their busy schedules to keep up-to-date on new medications and inform patients. However, others may be taxed for time or do not get reimbursed for these cognitive services, which may cause them to make decisions solely based on claims of drug representatives. Educating prescribers on this matter is a daunting task, and if comparative efficacy trials were required in Phase III trials, I believe that information on which drugs work best would lead create more information and lead to less costly choices.

    I already look for these trials. I think Natural Standard, for example, has done a great job with this. On http://www.NaturalStandard.com, a comparative efficacy database exists where disease states are listed. Therapies are graded by their level of scientific evidence. How fantastic would it be if the FDA caught on to what the Natural Standard is doing for integrative medicine and required this for prescription medications? I agree that requiring non-inferiority trials would prevent the industry from making another statin for cholesterol or another ACE inhibitor for blood pressure, because it’s been done. They would not be able to simply repackage their drugs in fancy wrapping paper but instead pump their tight budgets into filling the unmet need in health care. That way, diseases lacking sufficient treatment would have medication improved upon or created. Innovation follows the money; it’s the incentive. That’s why the Orphan Drug Act exists! So why not continue in this direction by adding comparative efficacy data into Phase III trials to create better drugs, reduce health care costs, and help clinicians help their patients?

  4. The average cost to bring a new chemical entity (NCE) to the US market through the FDA approval process is about $1.4 billion. I don’t believe extra trials are warranted before approval, and FDA has already, in the last couple of years, increased reviews of safety and efficacy through an enhanced Phase IV program once the product is in the marketplace. I’ve worked with the FDA in bringing drugs to the market with private corporations, and in my estimation the FDA has many excellent, hard-working, altruistic people working in a realm that is very, very difficult. Let’s not further burden the FDA or the drug companies.

  5. According to Dr. David Graham, Associate Director for Science and Medicine in the Office of Drug Safety at the FDA, the way the FDA is configured, it’s more interested in protecting the interests of industry. It views industry as its client and the client is someone whose interest you represent. Within the Center for Drug Evaluation and Research, about 80 percent of the resources are geared towards the approval of new drugs and 20 percent is for everything else (drug safety is about 5 percent). It is made worse by the Prescription Drug User Fee Act.

    For industry, every day a drug is held up from being marketed, represents a loss of 1 million to 2 million dollars of profit. The incentive is to review and approve the drugs as quickly as possible and not stand in the way of profit-making. The FDA cooperates with that mandate. Add that to the fact that probably two-thirds to three-quarters of the drugs that the FDA reviews are already on the market and are being reviewed for another indication. And the way the FDA’s approval standards are, the drug does not necessarily have to have a very great effect in order to be approved.

    The FDA needs to set up stringent standards of evidence that might lead to the approval of safe drugs that actually have benefits to the population. CER may be able to give them the immediate feedback on the relative worth of these drugs.

  6. The FDA is a Miracules Failure of epic Porportions. The Choir for a Industry who bends to the will of Pharma and gives them extensive abilities to push through acceptance of Farsighted Products that have Nearsighted side affects.
    Often,the FDA act as investors whom work in their own Behalf , being oblivious of needs of citizens. I have noticed the FDA allowing Products that are being protected from being copied. Have found ways to Direct the FDA to further exceed its expirations.
    The FDA is not an agency with any credibility with with US citizenry.
    As they have permitted off label use and prescriptions that have been marketed by Pharma which were inconclusive and deadly.

  7. Increase innovation by increasing the cost and burden of regulation? Quite the opposite is necessary. One reason for the “lost decade” is the increasingly difficult regulatory environment, which reduces investors’ willingness to invest their capital in pharma & bio R&D.

    • I would love to see any actual data to support that claim. I have looked and found none. The best I can find is Mandel’s work looking at numbers of employees. OTOH, there was significant deregulation of the financial sector in the aughts. We are investing more than most (any?) countries in bio/pharma R &D. It is the other areas where we lag.


Leave a Reply

Your email address will not be published.