Revelations that a popular class of painkillers can cause an increase in strokes and heart attacks have raised questions about the nation’s drug-testing and regulatory system, and are another blow to the once-unassailable pharmaceutical industry, according to Wharton faculty.
The latest development in the controversy over the class of drugs known as Cox-2 inhibitors — including Vioxx, sold by Merck, and Celebrex and Bextra, sold by Pfizer — came last week when an advisory panel of the federal Food and Drug Administration ruled that these drugs’ potential benefits outweigh their risks for some patients and should be allowed back on the market, with certain restrictions. Pharmaceutical companies say the drugs, used to treat arthritis, provide pain relief with less risk of gastrointestinal bleeding than aspirin or other medications.
Merck & Co. had withdrawn its blockbuster painkiller Vioxx from the market in September after studies indicated the drug was associated with increased risk of stroke and heart attack. By a close 17-15 vote, the advisory panel decided the drug is still safe for sale. Despite the narrow vote, Merck has indicated it may return Vioxx to the market, based on recent findings that indicate competing drugs, including some over-the-counter medications, pose similar risks.
Pfizer all along has refused to pull Celebrex and Bextra off the market, contending they are safe when prescribed properly. The advisory panel voted 31-1 to keep Celebrex on the market. Bextra won similar approval, but by a tighter margin, 17-13, with two abstentions. The FDA must now decide whether to accept or reject the panel’s report, which also includes a recommendation that the drugs come with strong warning labels — including a ‘black-box’ warning, the FDA’s severest sanction — indicating the risk of cardiovascular side effects. In addition, a panel majority suggested either a ban or some restrictions on direct-to-consumer advertising.
The inherent tradeoff between the health benefits of taking any drug and the risk of side effects is at the root of the Cox-2 inhibitor debate, says Mark Pauly, Wharton professor of health care systems. “The complication here is that for new drugs, even though they are tested as well as can be, there is always a possibility that additional information on an adverse side effect will accumulate over time as the drug is used by millions of people.”
When it comes to blockbuster drugs with sales of more than $1 billion a year, the stakes are high. As more people take the drug, more information about side effects is likely to come out. But as more people take the drug, it grows more important to a firm’s profitability. Vioxx sales hit $2.5 billion last year, before the drug was withdrawn. Meanwhile investment firm William Blair & Co. estimates that sales of Celebrex will drop from $3.3 billion to $1.5 billion a year.
“The best data on the drugs on the market is what the manufacturers get, but they obviously have a conflict of interest,” says Pauly. However, he points out that with almost every new drug there are alarms about potential problems. If all drugs were pulled from the market based on the appearance of side effects, there would be no vaccinations for childhood diseases that once were killers, or for medications to treat HIV/AIDS, he points out. “It’s a judgment call. The best you can hope for as a society is that we have good institutions to help a responsible adult make those decisions.”
As for the panel’s decision on Friday, Pauly says a black-box warning label “will reduce sales and preclude direct-to-consumer advertising since there is no way to put the warning into commercials.” The warning also “seems like the right thing to do, under the rubric that consumers deserve to get true information.” The change in sales as a result of this — as well as changes in death and hospitalization rates due to gastric bleeding, heart attacks and strokes – “have the potential to be very informative about the role of direct-to-consumer advertising,” Pauly adds.
According to Wharton management professor Lawrence Hrebiniak, author of a new book entitled Making Strategy Work: Leading Effective Execution and Change, flaws in the culture and business model of the pharmaceutical industry are illustrated by Vioxx and the other Cox-2 inhibitors. Drug companies, he argues, have become overly reliant on the blockbuster model in which huge investments in research are targeted to a drug that could reach massive sales volume and drive profits for the entire organization.
The model breaks down when a problem arises, he says. If safety concerns crop up, there is little incentive for executives to pursue [these concerns] since the company has invested so much in developing the drug. “The incentive is to ignore them,” says Hrebiniak. “The culture develops around the blockbuster, and if you are the one who says, ‘We spent millions of dollars on research and development and this isn’t working,’ then you are persona non grata.” Drug companies need to develop a culture that embraces error, he says. “If you are going to be an innovator, you are going to make mistakes. If you are not making mistakes, you are not going to innovate.”
Hrebiniak points out that the drug industry has enjoyed a high return on assets for decades, but he predicts tougher times ahead, in addition to significant “change in the industry. Companies have been in bed with the FDA, and the FDA is going to respond by getting a little tougher.”
In response to the uproar over the Cox- 2 inhibitors, the Bush Administration has said it will create a Drug Safety Oversight Board to review safety problems that arise with drugs already on the market. The board would be made up of FDA staff and other government officials. According to Robert Field, a professor of health policy at the University of the Sciences in Philadelphia and a Wharton lecturer in healthcare systems, that is a step in the right direction but will have only limited impact because the board will not be independent of the FDA and will not have the authority to order a drug off the market.
FDA culture and the structure of the regulatory system are bound to create problems with monitoring drug safety after a product goes on the market, Field notes. That is a critical time for drugs because most are only tested on several thousand people before gaining marketing approval — not enough to pick up side effects that would occur in one out of 10,000 cases or even less frequently, he suggests. “There’s an institutional culture within the FDA that presents an inherent conflict of interest,” says Field. “Once the FDA approves a drug, it is an admission of failure to pull the drug. There is pressure within a single agency not to second-guess itself.” In Europe, he says, new drugs are reevaluated after five years and can be pulled off the market if problems develop.
Pauly notes that large health insurance firms could play a role in monitoring drugs after they have received market approval because they have extensive national databases. These insurers might be reluctant, however, to take an active role because of fears they could be held liable if they agreed to pay for a drug that eventually caused harmful side effects.
When it comes to legal liability, Field points out that the Bush Administration is backing new tort reform legislation that would shield drug companies from punitive damages for drugs approved by the FDA. The problems with Vioxx, however, may limit the odds of passage, he says. By the end of last year, 575 lawsuits stemming from the use of Vioxx had been filed against Merck, along with 70 class-action suits. Analysts have estimated that Merck’s liability could reach as high as $20 billion, but those figures may decline following the advisory panel vote.
“If you look at similar product liability cases, such as asbestos and tobacco, what really hurt the defendants were findings that they had concealed information. I think the Merck liability will hinge on whether plaintiff lawyers can show the company actively hid information,” says Field, adding that, ironically, the looming Vioxx liability might be one way Merck can preserve its independence. The company has been considered a take-over target.
Field also suggests that the way in which drugs are regulated, and companies held liable for damages, creates an atmosphere in which a drug is either considered perfectly safe or there is pressure to keep it off the market. As a result, doctors and patients have fewer options. “It seems a shame that our system works in such extremes. There is clearly a role for the Cox-2 inhibitors in the right population with the right warnings.”
In the current system, he argues, “everyone loses: the patients who could be using the drugs safely, the companies, obviously, and if the drugs are not pulled, the people who might experience the side effects. Our system is geared to all-or-nothing rather than a more collegial model in which we can find the best clinical middle ground.”
Quantity vs. Quality of Life
Jack Hershey, professor of operations and information management, says the decision to withdraw Vioxx represents an example of the kind of decision-making that can benefit from reframing the act of commission (taking Vioxx) and the act of omission (not taking Vioxx). “Imagine that the acts of commission and omission were reversed,” Hershey says. “A drug becomes available that causes a significant increase in arthritic pain every day for the next 18 months, but might reduce one’s risk of heart attack or stroke by about 1%. How many of us would take such a drug? Probably very few, particularly if we were told that we would not have the option to quit.”
Hershey argues that removing Vioxx from the market has exactly the same effect as forcing some patients to take this hypothetical drug. “A pill that decreases pain, but may increase the risk of heart attack, is pulled from the market. A pill that increases pain, but may decrease the risk of heart attack, probably doesn’t stand a chance in the same market. Yet, pulling one drug and introducing the other have the same clinical effect.”
Hershey points to research he did with parents asking how much risk of death they would accept to vaccinate their children against a strain of influenza for which the risk of death from the disease was 10 out of 10,000. The hypothetical vaccine would eliminate the chance of getting the flu, but cause side effects, which could be fatal. Parents were asked for the highest death rate from the vaccine they would accept. The majority chose a death rate below 10 in 10,000 — on average 5.4 in 10,000 — indicating they were more willing to accept more deaths from not vaccinating than from side effects of the vaccine.
Some people think that the risk of death should trump any amount of pain, he says. But those living with severe arthritis pain might be more willing to accept the risk because “there is a difference between quantity and quality of life.” Pfizer’s decision to keep its products on the market preserves options, he says. “In that sense I applaud what Pfizer chose to do. And I applaud the recent conclusion of the FDA advisory panel that it is safe to allow Vioxx to return to the market with a black box warning. “
Vioxx, Celebrex and Bextra are not the only drugs caught up in the debate. Aleve, a generic nonsteroidal anti-inflammatory drug sold over-the-counter, has also been linked to adverse side effects. Meanwhile, Novartis may apply for only limited initial FDA approval for one of its painkillers, Prexige, until the drug’s cardiovascular risks are better understood. Prexige has approval to be sold in the U.K., but the company has held off on a launch in the U.S. until the medicine has won wider acceptance across Europe.
Among the FDA advisory panel’s recommendations last week was a unanimous vote to add warning labels to other painkillers — such as ibuprofen — that don’t fall into the Cox-2 class. The FDA is expected to issue its decision on the panel’s report with a few weeks.
Mimicking Real Breakthroughs
David Asch, a physician and professor of health care management and economics at Wharton, says the Cox-2 inhibitor controversy shines a light on the role of marketing in prescription drug sales. “Do Vioxx and Celebrex really offer advantages over the agents that existed prior? I think they were tremendously oversold.”
He agrees with the drug industry’s position that strong profits fuel innovation in a risky and expensive business that is prone to failure. “On the other hand,” he says, “you can question how much innovation we are getting.” He points to the number of Cox-2 inhibitors themselves as an example of so-called “me-too” drugs that mimic true breakthroughs, then are sold largely through intense marketing campaigns often with advertisements aimed directly at consumers. “Tell me exactly how much we got from the introduction of Cox-2 inhibitors,” he says. “For all the great and expensive research that went into this development, we didn’t get much. In fact, we might have hurt ourselves.”
Asch supports the notion that drugs should be tested not just against a placebo, but also against existing therapies. Current U.S. law requires drug manufacturers show only that new medicines are safe and effective in their own right. Pauly, too, says he favors head-to-head trials for new drugs against existing therapies. However, he cautions it would be difficult to structure fair trials because it might not be clear what drug or treatment would be comparable.
Testing against a placebo, Asch argues, is ethically wrong because it requires a patient to take a placebo when there are known cures. Beyond that, testing against a placebo is irrelevant because that’s not the real issue doctors and patients face, he adds. “When I’m facing a patient with arthritis pain, it’s not a choice of nothing or Vioxx,” says Asch. “It’s a choice between Motrin or Vioxx.”