September 26th, 2010
Avandia: Lessons Learned and Unanswered Questions
The following is republished from Pharmalot, a widely read site that provides commentary on the pharmaceutical industry and related litigation.
What should we make of the decisions on Avandia yesterday? There remains some controversy about the risk, but the concerns were sufficient, given the therapeutic alternatives, to bring about action. The Europeans decided to suspend sales of the drug, and the Americans opted to leave it on the market, albeit with tough restrictions. The company has announced it will no longer promote the drug, and the practical result in Europe and the US may be a lot more similar than the decisions at first appear. Usage will stop in Europe, and new use should virtually stop here. But is the story over? There are at least some issues worthy of consideration as we get to the end of the tale.
1. The Delay to Judgment: If the drug has a probable hazard that is sufficient to require severe restrictions in the US and suspension in Europe, then why did it take so long to reach a decision? The basic information about the drug has been around for years. No new trials indicated great risk — all we know could have been known a while ago. Moreover, why did it take more than two months in the US after the FDA advisory committee met (last July) and the vote was clearly signaling a concern to take action? In the interim, sales have been continuing, and patients may have assumed that no news is good news about the drug. We need a more-responsive FDA, one that moves with all deliberate speed — but with speed — particularly when there are life-threatening safety concerns.
And by the way, Avandia has been on the market since 1999 — aside from the recent delays — why did it take so long to assemble all the evidence? And what would have happened if Steve Nissen (who authored the 2007 meta-analysis declaring higher cardiovascular risk with Avandia) had not found the Avandia trial data? And what would have happened if (former New York Attorney General) Elliot Spitzer had not included the sharing of trial data in the settlement with Glaxo over Paxil? Why does it seem fortuitous that we are at this point?
2. The FDA Focus on New Users: Why does the FDA believe that the barriers to prescription in new users should be stronger than those for current users? There are no studies that indicate that the excess risk dissipates over time. Presumably those on the drug, whether new or long-standing, are exposed to the risk. It is not sufficient to determine if people are doing well on the drug since the adverse effect is a silent risk. It seems that those already on the drug should have to go through the same process as new users — and the recommendation should be to change the regimens as soon as possible if the guidance from the FDA is to be sensible. There is no basis for separating the recommendation for new users from those of current users. And if it should be the last option for new users, then it should be the same for current users.
3. The Feasibility of Applying the Restrictions: Can the provisions specified by the FDA be reliably applied? How will doctors ensure that patients are given the information about the risk and that they understand? Will the FDA supply tools? Who will monitor the prescribing? Given the challenges in guiding physician behavior, I was wondering how the program will be implemented. I am supposing that this approach is in effect now — and if I prescribe Avandia to a new user tomorrow, I am wondering what I need to do and what will happen to me if I do not do whatever I am supposed to do. And who is going to tell me what to do? And what about my colleagues who did not notice the announcement? I assume the FDA knows what they are doing, but it is not so simple to leave it on the market, institute restrictions, and expect them to be followed. The FDA has some experience with restrictions — there are roughly 30 drugs now carrying a Risk Evaluation and Mitigation Strategy. But I am not sure they have imposed such a program on a drug as widely prescribed as Avandia.
4. The Failure of the Surrogates: This episode is yet another example of how surrogate outcomes like blood sugar control can be misleading. Knowing how well a drug controls blood sugar, cholesterol, or blood pressure (or any other risk factor) is not enough to know what effect it has on people. Drugs that are better at controlling a risk factor like blood glucose may paradoxically increase risk — we have seen that before. Ask Bristol Myers, the makers of muraglitazar — great glucose control, but it increased cardiovascular risk.
5. Will We Have Another Avandia? Finally, will we learn anything and change anything from watching the misdeeds of the company? There were dirty tricks and lack of transparency, episodes of intimidation of critics, and denial of responsibility. The Senate Finance Committee exposed individuals and the company, and the reports made for interesting reading. But how can we move forward in a way that such shenanigans can be avoided? In the end, the story is as much about Glaxo and the profession of medicine as it is about Avandia. And I fear that the culture has not changed — reputations seem to have survived intact, and the beat goes on. Profits were made: the sales of Avandia were extended by the misdeeds, and billions of dollars in revenues were preserved for a long time, even as many were likely exposed to risks that could have been avoided with a different regimen. Remember that a strong majority of the FDA advisory committee in 2007 agreed that Avandia increased the risk of heart attacks. What truly needs to be remedied is how we all work together to get the best information to the public. Needed studies must be done in a timely way so we can equip clinicians and their patients with the evidence that allows them to make the best choices.