Commentary
Article
Author(s):
In a recent issue of JAMA, a team led by Benjamin N. Rome, MD, MPH, examined how the FDA’s accelerated approval process has moved 5 genetically targeted treatments for Duchenne muscular dystrophy (DMD) through its pipeline despite limited evidence on their efficacy.
In a recent issue of JAMA, a team led by senior author Benjamin N. Rome, MD, MPH, primary care physician and health policy researcher at Brigham and Women's Hospital and Harvard Medical School, and member of the Program on Regulation, Therapeutics, and Law (PORTAL), examined how the FDA’s accelerated approval process has moved 5 genetically targeted treatments for Duchenne muscular dystrophy (DMD) through its pipeline despite limited evidence on their efficacy. The total drug spend for 3 of these treatments alone—eteplirsen (Exondys 51), golodirsen (Vyondys 53), and casimersen (Amondys), all from Sarepta Therapeutics—is an estimated $3 billion.
In this interview with The American Journal of Managed Care® (AJMC®), Rome addresses complexities in the space due to limited preapproval evidence and delayed trials, high costs and the resultant financial burdens, and areas of potential reform.
“If you have drugs on the market that don't work but that patients are taking, that could block people from entering other clinical trials for other diseases because maybe they can't be eligible for the trial because they're already on one drug,” he says. “Ultimately, it has effects on this sort of ecosystem of development for other products in the space. And it has implications for investment in development of the space, too.”
This interview has been edited for clarity and conciseness.
AJMC: Tell us about the research and analysis that led to the publication of your JAMA research letter.
Rome: We were looking specifically at a group of drugs that treats DMD, which is a genetic condition that affects predominantly young boys just because of the way it's genetically carried. Duchenne is the severest form of muscular dystrophy. It is a disease where you have stunted muscle growth and development, and it eventually leads to muscle wasting and death usually in the 20s or 30s—a really severe sort of debilitating, disabling, and deadly and terrible disease.
Starting in 2016, the FDA approved several genetically targeted treatments that aim to increase the level of protein that is not working. The first of these was eteplirsen. But they've been controversial, because the data that supported their FDA approval was quite scarce. They're approved under a pathway at the FDA called the accelerated approval pathway, where they didn't actually show that the children who took the drug actually had better clinical outcomes—meaning they walked better or had better physical functioning—they just showed that when you did muscle biopsies, there was a small increase in the protein level, which may or may not actually correlate with improved clinical outcomes. But in the meantime, while we've been waiting for better data on these drugs, they've been marketed and at high prices, $1 million or more.
We wanted to look at how much we've spent on these drugs since they've been on the market, especially since we've been waiting on clinical trials to confirm the benefits for several years now.
AJMC: What challenges arise from the approval of DMD drugs based on this limited preapproval evidence?
Rome: The 2 biggest challenges are delays in getting the confirmatory evidence, which we've seen with eteplirsen, which, again, was approved in 2016. The confirmatory trial is supposed to be done. The original deadline has passed—it was November 2020—and now the trial won’t be completed until at least next year, if then. So that's a huge problem. If the FDA is going to approve drugs based on preliminary evidence that it might work, if you want to get evidence that it's going to work, you want to make sure that that evidence comes quickly.
The second problem is the cost problem. The FDA doesn't have anything to do with drug pricing. Ultimately, once the FDA has greenlit a drug, the company is allowed to charge whatever price it thinks the market will bear. And that's what's happened here. We often see drugs with very, very limited evidence, but very, very high prices under this program, which is in part why it has been heavily criticized for having drugs that are driving a lot of spending but without evidence that they are effective.
AJMC: What potential implications do these delayed or incomplete confirmatory trials have for DMD, especially because it's considered a rare disease and the price points are astronomical?
Rome: The implications for patients are, if patients are bearing some of this cost or their families are bearing some of this cost, that's obviously a huge, huge burden being placed on them. The real concern is, if you have drugs on the market that don't work but that patients are taking, that could block people from entering other clinical trials for other diseases because maybe they can't be eligible for the trial because they're already on one drug. Ultimately, it has effects on this sort of ecosystem of development for other products in the space. And it has implications for investment in development of the space, too.
Companies argue that they sort of need this accelerated approval in order to stay in business. Sarepta, the company that that created several of these drugs—eteplirsen, golodirsen (Vyondys 53), casimersen (Amondys), and delandistrogene moxeparvovec-rokl (Elevidys)—there's been reports out about the communications between the FDA and Sarepta, where the FDA was actually pretty convinced that Sarepta was going to go out of business if their drugs were not approved. The patient advocacy community was really heavily in favor of getting approval just to keep the pipeline open in case these things panned out.
The downside is, what ended up happening in this story, we partially know the ending at this point. After several iterations of the initial drugs, the gene therapy that came out last year as accelerated approval, the trial for that drug—that is presumably more effective and marketed as potentially curative—was disappointing. We haven't actually seen the trial yet; we've just seen Sarepta’s press release of the trial results. But it did not meet its primary end point. The FDA is currently evaluating whether to fully approve that drug, or presumably, the other option would be to remove it from the market if they don't find the data convincing. We'll find out somewhat soon on that.
It's a really unfortunate situation. It's obviously not the outcome that any of the patients who had been taking these drugs was hoping for and probably not the experience that some patients have who take the drug who think that it works—maybe it does work for individual patients, but just not as a whole. This is a really disappointing, I think, situation. The investment of $3 billion that we as a health care system have now made in this is money that we could have spent elsewhere. You can only imagine if you could just take that money and sort of reinvest it into a broader pipeline of products, I think that would have been a better use of funds. But that's unfortunately not what happened.
AJMC: What recommendations do you have for policy makers about the approval process, pricing, and reimbursement of these genetic treatments for DMD?
Rome: On the approval side, the FDA has already made some changes to the accelerated approval pathway. Congress passed some updates, and the FDA has enacted additional updates through rulemaking. Now, in general, the FDA is requiring that the trial to confirm the drugs’ benefits are underway at the time of approval, and they have more authority to remove drugs from the market if it turns out that they don't work.
In the case of eteplirsen, it's really just a disappointing case of this delay, where the company has been able to argue to the FDA that they've had reasons; for example, it's been difficult to enroll patients [in trials]. You can understand why it's difficult to enroll patients. They can just get the drug. Why would you take a 50/50 chance of getting a placebo, rather than getting an active drug that might actually work? You can understand why the company has struggled. But at the same time, what was the end game of approving the drug? From a policy standpoint, there are clearly reforms that are needed and potentially even additional reforms to make sure that this pathway doesn't allow products to stay on the market for a long time without evidence that they work.
On the pricing side, I think payers, including public payers like Medicare and Medicaid, need better ways to incentivize companies to both complete the data in a timely fashion and to not be on the hook for billions of dollars for drugs that ultimately we find don't work. There have been some suggestions from MedPAC [Medicare Payment Advisory Commission] and MACPAC [Medicaid and CHIP Payment and Access Commission], which are the 2 guidance organizations that advise Congress about policies related to Medicare and Medicaid, that maybe we should pay less for drugs that have accelerated approval. These drugs for DMD would be examples of those where if you've granted that authority, then maybe they would have more ability to either negotiate a price or cap prices up until the company has completed its confirmatory trial. That would provide more leverage to sort of convince the company to go back—and financially they could make more money if they could finish the trials and get full approval—but also prevent the government from paying $1 billion for treatments that don't work. That’s ultimately the main policy lever.
The other one for these gene therapies is to think about outcomes-based contracts with the government, and for other private payers, thinking about contracts that say, for example, “We'll pay now for the drug or we'll pay over time, but if it turns out that you haven't finished the trial in 5 years, we want our money back.” Contracts that build in those clinical outcomes, either for the individual patient or for the trials, into the actual payment mechanism.
There's definitely opportunities for payers, public and private, to take a more aggressive role given that the FDA seems pretty poised to continue using the accelerated approval pathway for cell and gene therapies that are coming down the pike.
Outcomes-based contracts could be on the patient level, where you could say something like, “If this drug doesn't work for this patient, we stop paying.” You have to define what that means. What does work mean? In the case of DMD, what does it mean to say that the drug has worked or hasn't worked? That might be hard to define and measure and you might have disagreement about that. Or you could build contracts around—particularly for Medicare and Medicaid—the actual trial being complete. In those cases, it could just be that we just pay less until the trial’s complete.