Skip to main content

Regulatory Affairs

Sarepta Gets An Approval – Unfortunately

Sarepta got approval yesterday for their Duchenne muscular dystrophy therapy, Exondys (eteplirsen), after many, many twists and turns at the FDA. (When you consider that the advisory committee meeting on this was back in April, you get the idea). I can recommend this summary at Endpoints for what’s gone on with this decision, which has to be regarded as extraordinary.

How so? Well, here’s the 126-page summary review document (and the fact that this can be accurately called a summary tells you something right there). Take a look at the first few pages, which are from FDFA Commissioner Robert Califf. He stipulates that the company had an adequate chance to present all its data, and goes on to say that the underlying science of DMD (and Sarepta’s therapeutic rationale for treating it) are not in dispute. The problem is that the company’s development program was “unusual” in the way that it had a very small number of patients and was “characterized by major flaws in the clinical study design”.

The review goes on to note that everyone agrees that Sarepta increases the amount of dystrophin protein, but after that, the trouble begins. The increase was much smaller than anyone had expected or hoped for, and is well into the range of arguing whether it can possibly be clinically significant. The FDA’s own standards for clinical endpoints call for them to be “reasonably likely” to be meaningful, and this is (necessarily) a judgment call. The FDA’s own review team concluded that eteplirsen was not likely to show any real benefit, while Janet Woodcock, head of the whole Center for Drug Evaluation and Research, disagreed. And in the end, Califf declined to overrule her call, and the drug was approved. Overruling the CDER director, he says, would only be appropriate when the director’s reasoning was clearly flawed, and in this case, both her view and that of the review committee members are tenable (even though they disagree).

I can see why Califf did this, then, but what does that leave us with? Sarepta’s drug does seem rather unlikely to provide much benefit, and is reasonably likely to provide none at all. It is priced in line with other rare-disease drugs, at around $300,000 per year, and that may well be 300 grand worth of placebo. Luciana Borio, acting chief scientist at the agency, was against the idea, saying that approval would open the door to other clinical development programs that were run just as poorly (and she went on to accuse the company of “serious irresponsibility” by selectively publishing data during the whole process). Ellis Unger, director of the Office of Drug Evaluation, was also opposed, saying that the drug was a “scientifically elegant placebo”, and that patients and their families were taking on unknown risks for likely nonexistent benefits. (The small number of patients – twelve – also makes evaluation of side effects difficult).

I find myself agreeing more with Borio and Unger on this one. I think that Sarepta should have run a better trial, and should have gathered more data when the FDA told them that more information was needed. Instead, the company stood pat, called up Duchenne-affected boys and their families to plead with the FDA, and won over Janet Woodcock, and that appears to be enough. Is this going to be the new way to get a drug approved? Run a trial in a dozen people, generate unconvincing data, and then lobby Janet Woodcock? I share the worries that this might open the floodgates, because after all, Sarepta got their drug through.

It’s true that the agency is asking for more data after the drug goes onto the market, and it’s possible that approval could be withdrawn if no benefit becomes apparent (as was the case with Avastin and breast cancer). But that was an awfully difficult process, and any attempt to withdraw Exondys from the market will be met by the same tactics of parading diseased children in front of the cameras. No, in the end, I think that the FDA’s job is to approve drugs that work, and no one knows if this one works or not. The people that want the agency’s approval rules loosened should watch closely, because this is the world that they’ve wanted to live in. Let’s see how it is there.


72 comments on “Sarepta Gets An Approval – Unfortunately”

  1. Running Comment says:

    Fullly agree, Derek; this was bad call from the FDA, and one they will liekly take a lot of deserved flak for. Commissioner Woodcok will have to take the brunt of this criticism, as is her job, but also because of the silly move of going aginst some of her top people. Advocates of eteplirsen have almost invariably used non-scientific arguments to promote its approval, and the fact that FDA decides to cave in to pressure is worrying, to say the least.

    1. Robert Klein says:

      I see a lot of accusations against Eteplirsen supporters of being “non-scientific”, but my reading is the exact opposite. I’ve only seen non-science or at best poor science from the anti crowd. Example: the FDA’s review has a chart comparing the Sarepta kids to placebo controls in the Drisapersen PIII study. The text there states that they all did about the same. Nonsense. The placebo kids only participated for a few months to a year. The placebo effect is greatest at this range, so you can’t compare. But worse, many of the controls started as late ambulatory. You can’t bring a DMD kid off the street at 16 who’s walking a compare him to a Sarepta kid who’s been tracked since he was 13. It’s apples and oranges. The fact is the chart shows most of the walking kids in their upper teens are in fact Sarepta kids. If any conclusion can be drawn (and scientifically, none really can), the trend is that Eteplirsen did in fact make a huge difference.

      There’s a ton more to say, but I assure you if you objectively review the data, you’ll see that the FDA is out of line in its criticisms — something which I’ve come to accept happens far too often. Don’t accept FDA claims at face value. Think for yourself.

  2. Andrew Lundy says:

    Definitely a controversial decision, but then the alternative would have been courageous…!

  3. Erebus says:

    >”The people that want the agency’s approval rules loosened should watch closely, because this is the world that they’ve wanted to live in. Let’s see how it is there.”

    What makes you think this is the case? The Sarepta approval is an unprincipled exception — it does not represent a shift in policy. The approval process for other drugs isn’t going to get any easier, faster, or cheaper; precisely to the contrary, it’s sure to continue to get harder, slower, and more expensive.

    And yet…. Although it was a bad move by the capricious regulatory body, the Sarepta approval is an unmitigated good for the public. Those who want it and can pay for it now have another treatment option at their disposal. Where’s the harm in that? (“Safety concerns” shouldn’t weigh too heavily on people with DMD, which is invariably fatal. And the drug appears to be safe and well-tolerated.)

    1. John Wayne says:

      Your comment reminds me of the system in other countries (Japan?), wherein safety is the primary determining factor towards approval; efficacy takes the back seat. The answer to your question (where is the harm in that?) is one of cost – how much do you think health insurance should cost a society? Do you believe that public funds (including insurance company spends) should be put towards therapies with dubious efficacy? If so, which ones? How much? What percentage of total spend? etc.

      If we go down this path, I’m concerned that health care will become even more of a marketing (vs evidence-based) exercise than it already is. As a scientist working in medical research I appreciate that when looking at data on my projects I am the most easily fooled person in the room – thus, I design (hopefully) careful experiments with lots of controls. If somebody in my family was gravely ill I don’t think my normal instinctive decision-making would be improved. I wouldn’t listen to myself in that situation, so why are we listening to them? A bereaved parent is a tragedy, but they shouldn’t direct medical policy unless our society is willing to bear the costs.

      1. Erebus says:

        >”how much do you think health insurance should cost a society? Do you believe that public funds (including insurance company spends) should be put towards therapies with dubious efficacy?”

        Nobody’s forcing insurance companies to buy Sarepta. Nobody’s forcing governments to buy it, either. Insurance companies and governments can decide for themselves whether or not to cover Sarepta for people with DMD. But now families who are struggling against DMD, who have the money and think that it might help, can, with assistance from their physicians, make the decision for themselves. I don’t think that there’s an ethical argument for prohibiting them from purchasing it.

        I’d add: It costs ~$2B to get a drug approved these days. These costs are always passed on. The reason it costs so much is because efficacy testing requirements — phases II and III — are incredibly demanding. Downright onerous, in fact, particularly now that it has become apparent that lobbying is just as important as data. (See also: flibanserin.) So much for “how much do you think health insurance should cost a society?” The FDA should ask itself that question, and figure out how to streamline regulatory processes.

        1. Hap says:

          But they’ve already had options that are (provably) just as effective, but rubber chickens don’t cost $300K/year.

          If something actually gets to market, then the costs of trials are spread out to the people who use the drugs, but otherwise, the people who invest in the company are the ones who pay. The idea of allowing drugs with insufficient evidence of effectiveness (the reason why those trials are so expensive is to see if the drug actually works). Putting drugs out without such evidence seems like a great way to socialize the costs of drug discovery and privatize the profits. I’m sure that will work out well.

          1. Hap says:

            I can’t edit.

            When evidence of effectiveness is given before trials, then the users of the drugs (and their insurance) pay for the trials, but if they don’t work, then they don’t pay. Under this model, as long as you have enough sick children for PR, users and insurance companies pay for the later trials. This seems like a really good incentive to spend more money on PR and less on finding out if your drug actually works, because if you can get the drug out there, someone else will pay for trials. If it works, you get the profits, but if not, someone else pays for the trials.

            If your disease has no current treatment, then the feeling that you’re doing something is going to be a powerful motivator to spend money, whether or not you (or anyone else) have any idea that it will work. Unfortunately, that means that there will be a strong incentive for companies to throw lots of crap out to see what will stick, since it all will pay, anyway. That seems like a poor way to find treatments that actually do work.

        2. schinderhannes says:

          Yeah sure but the hefty 2bn result from studies wir 40k patients @25 k each for the study centers.
          Going with 12 patients is an insult to science – an saves 1 bn mas or menos….

          1. toxchick says:

            This is a rare disease–not seasonal allergies. I think DMD is one of the less rare diseases, but I’m sure it is still challenging to find enough patients. It can also be challenging & expensive to manufacture large molecule drugs, and there are so few patients so there are not dedicated facilities and economies of scale.
            There is a reason why we have the Orphan Drug Act, before that there were few drugs being developed for rare disease. 12 patients is on the lean side for a rare disease, but most rare diseases I would guess have been approved on the basis of 20-100 patients in clinical trials.

        3. tangent says:

          “The reason it costs so much is because efficacy testing requirements — phases II and III — are incredibly demanding.”

          And the reason is efficacy testing is hard. Whether you do it in FDA approval, by controlled trials run by insurance, by uncontrolled surveys, or by doctors’ empiricism: it’s expensive to do well. It is actually much cheaper to get to any level of evidence by using trials that are designed and controlled for it, versus futzing around with the spurious results you’ll get out of retrospective data-mining.

          And yes, we need to have a good estimate of efficacy, or else you harm patients versus options that are actually more effective. You also waste money, obviously. And consider the economics, the success of an ineffective drug squeezes a potential more-effective drug out of the market, because there’s no visible differentiation. And low market potential starves out the research into that more-effective drug.

          It’s not as if your estimates of efficacy are “good enough” in the absence of trials. Most efficacy trials fail, obviously. If you can predict which will succeed, or which failed spuriously and could be resurrected, I have a job offer for you.

          tl,dr look at what you’re aiming for:
          Remove that efficacy filter, and you will instantly have 5x as many drugs available for any indication. Great. But 4/5 are ineffective, by my estimation, or maybe 3/5 are ineffective by your estimation. They’re all sold by Dr. Oz, and doing science to show effectiveness is not the way to sell more through Dr. Oz. So no way can a company justify doing research to find a more effective drug.

          1. Erebus says:

            I’ll repeat something I wrote here last year. (With some light edits.)

            The FDA’s efficacy requirements:

            -Cannot be standardized — are amorphous and vague at best, when they’re not downright unpredictable — which makes drug discovery more difficult and risky than it needs to be.

            -Have (once again!) buckled under lobbying and political pressure. So what are they worth?

            -Make drug approval a much longer process than it needs to be. It now takes, what, 12 years on average? This hurts patients, payers, and industry professionals alike. It can also, for obvious reasons, stifle innovation in the field.

            -Lead to the prioritization of certain indications over others. This is sometimes warranted (looser efficacy requirements for terminal cancer treatments), is sometimes a happy accident (efficacy trials for acute diseases can on occasion be relatively short and easy), but is sometimes a sort of de facto de-prioritization that can make clinical trials for long-term, chronic, and degenerative diseases excessively onerous.
            It’s well known that Alzheimer’s, for instance, is simply a “tough” target for clinical trials. Anti-senescence drugs for the treatment of aging are, I think, impossible outright — without multiple multi-decade trials, they would make a mockery of “efficacy” trials.

            One could, further, argue that even at the great cost we’re now paying, trials do a poor and uncertain job of actually testing for efficacy. To do a truly good job — especially where chronic and degenerative conditions are concerned — we’d need broader trials, longer trials, and, particularly, trials where the drug candidate is used in combination with a wide number of other pharmaceutical agents. There’s no reasonable way to do this under any existing regulatory framework. The only possible way is to open up access to patients and doctors, and run rigorous post-market analyses.

            As for “there will be too many drugs, and it will be too confusing to sort out the wheat from the chaff!”…

            -Insurance companies and governments will run their own efficacy analyses, based on post-marketing analyses, and will not reimburse for drugs that provide no benefit.

            -Many medical and professional organizations offer treatment guidelines — standard treatment recommendations. For instance, the National Cancer Institute (part of the NIH) offers hundreds of detailed treatment recommendations to physicians, based on the characteristics of the disease they’re trying to treat: Cancer type, stage, patient age, and so forth. These stock recommendations are based on rigorous analyses of patient outcomes, and they are generally followed. I think it’s safe to say that “snake oil” would not warrant a recommendation. I cannot emphasize this strongly enough, as it is a very important point.

            -Physicians are, on the whole, neither stupid nor very easily fooled. Specialist physicians typically understand, at least insofar as is reasonably possible, the molecular and physiological mechanisms behind the diseases they spend their lives facing. They are generally well-equipped to assess treatment options.

            So what’s the harm? That some people may talk their way into prescriptions which do them no good — but, also, which can’t really hurt them, and which they may be forced to pay for out-of-pocket? If, in return, we’re given a more favorable atmosphere for drug development, for innovation in the field and a broader and more competitive marketplace, for the opportunity to more easily attack diseases like aging and Alzheimer’s… I’d take that bargain in a heartbeat.

          2. tangent says:

            @Erebus, your hopes for filtering of effective drugs to be done by insurance companies, treatment guidelines, etc. — those all depend on having enough solid information. Who generated that and how much do you figure it will cost?

            Concrete example: we already have some safe compounds attacking Alzheimer’s by plausible mechanisms. Go back in time and say we put them on the market without running efficacy testing. What *does* the American College of Geriatrics guideline say? Will doctors accurately determine the drugs’ effectiveness by what they see in their own practice? How many units will Dr. Oz move before post-marking surveys might manage to catch that they don’t work? Truly, please paint me your picture of the alternate history of Alzheimer’s drugs (without plunging into generalities), because I’m puzzled where you’re going.

            I guess this was the first thing I said, so my work here might be done…
            “efficacy testing is hard. Whether you do it in FDA approval, by controlled trials run by insurance, by uncontrolled surveys, or by doctors’ empiricism: it’s expensive to do well. It is actually much cheaper to get to any level of evidence by using trials that are designed and controlled for it, versus futzing around with the spurious results you’ll get out of retrospective data-mining.”

            (Or is that you think FDA efficacy testing is too stringent, and maybe we’ve suppressed Alzheimer’s drugs that work?)

          3. Erebus says:

            You’re either ignoring the history of the pharmaceutical industry, and are therefore being disingenuous, or you’re totally ignorant of it.

            As I’ve stated elsewhere, efficacy testing is something of an innovation. It was introduced in 1962 — long after the discovery and commercial introduction of many of the drugs that people rely upon today. How do you suppose they did it in the 1950s, when drug development was roughly an order of magnitude cheaper, faster, and more efficient? Is your position that it’s impossible to repeal bad regulations?

            How much do I figure it’ll cost? Come on, if you think that people — particularly people with degenerative or dangerous diseases — won’t pay to try new drugs out-of-pocket, you don’t understand the first thing about sick people. Don’t forget that without the lengthy, crushingly expensive burden of Phase II and III testing, drugs may become quite a bit cheaper! (Especially if people need to pay for them out of pocket. As things stand now, with regulatory development costs creeping upwards at a rate a hell of a lot faster than the rate of inflation, and with every pharmaceutical company passing the buck onto the US insurance industry, the business of drug development obviously isn’t doing very well.)

            It should be trivially self-evident that postmarketing surveillance can offer a perfectly adequate real-world view of how drugs perform.

            Besides, drug companies — among many others, from medical associations, to insurance companies, to domestic and foreign governments, to those guys at Cochrane, to the NIH itself! — might still be inclined to run efficacy trials. There are lots of ways that these could pan out, and lots of potential funding structures.

            The FDA is simply incompetent at judging efficacy, to such an extent that the decision to approve drugs based on efficacy should not be left in their hands. Their increasingly onerous and capricious regulatory approval process has become a real millstone around the neck of the pharmaceutical industry. It would be preferable to have them stick to their original core mission — to ensure that the drugs on the market are reasonably safe, given the indications they’re designed to treat — and leave them out of determining whether or not those drugs are effective.

      2. Dan Gerszewski says:

        Efficacy can be determined by clinical best practice over time. Safety cannot. By placing both under the same level of scrutiny I think a very basic error is being made.

        If a drug is not being effective worst case scenario is economic loss, and best case is improved clinical outcomes and increased lifespan and quality of life. If a drug is not safe the worst is death and the best is… death still, there is no upside there.

        It’s illustrative to look at the field of toxicology interventions because they have a few unique factors: first, most of the drugs they deal with are generic or predate modern drug testing entirely (sodium bicarbonate, methylene blue, glucose, atropine, insulin, etc), second, they get a lot of clinical experience.

        As a result there are sea changes over time as various traditional antidotes are either bolstered by evidence or supplanted by superior methods. You see new treatments like high-dose insulin for beta blocker overdose come onto the scene and supplant glucagon, or hydroxocoalbumin replace amyl nitrate plus sodium thiosulfate for cyanide poisoning.

        If we adopted the Japanese model of drug approval, we would see the same clinical marketplace of ideas select the most effective treatments over time, as the body of knowledge grows and doctors establish best practices.

        In short, doctors can determine based on their clinical experience over time whether a drug is going to be of benefit to a given patient.

        1. Marie Lutz says:

          I don’t understand the argument that the FDA should only evaluate safety. Efficacy and safety cannot be separated. All drugs have adverse effects. To put it another way, you can’t prove a drug is safe because none of them are. If you don’t know whether a drug works then you don’t know whether the risk of taking the drug is worth it. A drug that can possibly cause death might be worth the risk if it will cure your cancer. But if you don’t know if the drug is efficacious, then maybe you’d prefer trying another type of treatment or do nothing at all. Also, best clinical practice has supported such ineffective treatments as leeches and diet for ulcers.

          1. Erebus says:

            Come on, this is elementary. Drugs are approved for specific diseases and conditions, yeah? So evaluate their safety vs. the standard of care for that indication, or vs. what might be considered “reasonable” in treating that indication. In this way, safety per se can, in a relevant way, be evaluated.

    2. Peter S. Shenkin says:

      As others have pointed out, the issue really is the insurance coverage. But here’s why. Although it is true that nobody is forcing an insurance company to pay for this drug, it’s also the case that nobody is forcing an insurance company to pay for any drug. To a great extent, FDA approval has been regarded as an informed (though never infallible) endorsement of efficacy as well as safety – and endorsement that insurance companies and their customers have relied upon for the provision of medical care. To the extent that the FDA departs from strict standards of efficacy in approving drugs, the insurance companies will feel free to simply deny coverage based on cost. To the extent that the FDA maintains strong efficacy standards, there will be less scope for insurance companies to coverage for efficacious, though expensive, drugs.

      1. Hap says:

        ….which will likely mean that drugs that are actually effective at treating something but which are expensive won’t ever be made because no one will be able to afford them (because insurance companies will balk at the cost and feeling free to balk on that basis, will be able to make the denial stick). In that case, not only will the push for PR-based therapies decrease the pool of available money and time, but it would decrease the likely payout as well, and thus prevent them from ever being made. That would seem to be highly counterproductive for patients – the need for hope would make actual hope a much more expensive commodity, if it existed at all.

    3. T says:

      The harm is that if we approve anything that isn’t too harmful, regardless of whether it works, patients will face a bewildering array of options, most of which don’t help at all, and they (and their insurance companies) won’t be able to tell which ones actually have a good chance of improving/saving their lives. This is the whole point of trials. To find the few options that actually work among the millions that don’t. I have a serious chronic disease and I don’t want the patronising option of “here’s a load of crap you can buy if you want”, I want medication that works. And I want the drug companies to be motivated to find something that works because if it doesn’t work, they can’t sell it.

      1. Erebus says:

        Surely you realize that the “Golden Age” of our industry was from 1940-1960? Over that entire period, the FDA did not test for efficacy. And yet, from chemotherapy to steroids to antibiotics, tremendous strides were made. There was arguably more progress in small-molecule drug development over that 20 year period than ever before… or since!

        Was the market flooded with useless products? Did medication not work? Did anybody ever say “here’s a load of crap you can buy if you want?” No. In fact, the industry was never more meaningfully productive, and never more respected. (For e.g., see George W. Merck on the cover of Time Magazine in 1952. The caption: “Medicine is for people, not profits.”)

        Efficacy testing was a kneejerk response to the Thalidomide debacle. It’s high time it is re-evaluated.

        1. Derek Lowe says:

          I think a better model might be the current “dietary supplement” business. Instead of thinking that the drug industry might start looking like it did sixty years ago, can you think of a reason why it wouldn’t start looking like that one does today?

          1. Erebus says:

            That’s a good question.

            Well… What characterizes the dietary supplement industry? I’d say that the supplement industry is comprised almost solely of:

            1. Products of dubious safety and uncertain efficacy/utility.

            2. Poorly-differentiated commodity products, which compete with each other on basis of price. (Which, of course, subjects all prices to downward pressure, race-to-the-bottom style. This leads to companies cutting corners to gain an edge, which leads to mislabeling, product-spiking, and so forth.)

            3. Products which are marketed in borderline unethical, and sometimes outright illegal, ways.

            With respect to #1, pharmaceutical products would still need to undergo a safety review. So even if the pharmaceutical industry releases drugs of dubious efficacy — which should be the exception rather than the rule — at least they won’t have unacceptable safety profiles.

            #2 would not apply to pharmaceuticals. Whereas supplement products are ostensibly natural or endogenous substances which cannot be strongly patented, strong patents will remain the rule for pharmaceutical products. What’s more, without efficacy testing, a shortened drug development pipeline will actually make those patents much more valuable, in practical terms.
            (And where patents don’t apply — in the case of generics — the benefits should be obvious. We should certainly want lots of poorly-differentiated, low-cost generics.)

            #3 is already the case, wouldn’t you say? Pharmaceutical marketing took a wrong turn about twenty years ago. Efficacy testing has nothing to do with it. Direct-to-consumer advertising has everything to do with it.

  4. steve says:

    There is more than one regulatory hurdle that companies need to go through to get their product on the market. Even though Sarepta got FDA approval they won’t necessarily obtain reimbursement with the data they presented. It will be interesting to see what happens with CMS and other providers.

  5. luysii says:

    It will be like Cognex for Alzheimer’s. After the initial hype about the drug, with clinicians witnessing minimal if any benefit, the rationale became ‘slow the decline’ — something impossible to disprove, and something that made a lot of money for the makers of the drug. Clinically, I never saw any dramatic improvement with the drug. Sarepta is here to stay.

    As the former direct of a Muscular Dystrophy Association clinic, you have no idea how desperate the families of Duchenne and Becker patients are.

    The truly bad thing about Sarepta, is that its usage will block development of truly useful drugs.

    1. CMCguy says:

      luysii I am curious how you think “its usage will block development of truly useful drugs”? Do you feel patients would no longer be willing to participate in clinical trials of other drugs? Based on the questionable results, especially if the cost is high, wouldn’t many continue to seek alternates (since assume they do nor bare the expense as part of a trial)? It may lessen number of naive participants in a study but if desperation is anything like certain cancers patients will often chase any promise out there.

      1. loupgarous says:

        It’ll reduce the chances of similar drugs being developed if, after clinical human efficacy and safety trials in 12 people, eteplirsen kills the next 12 because of an unforeseen sensitivity reaction (or one which could have been anticipated had the study been better designed or run).

        Venture capitalists and corporate decision-makers will see the process leading to the Sarepta approval not as a perfect storm of FDA incompetence at the agency’s highest levels, but as typical pitfalls of funding expensive studies of drugs like eteplirsen with small patient populations.

        They won’t see that it’s up to them not to do what Sarepta did. be unethical, play political games in what ought to be the scientific process protecting patients at FDA, even protecting them from themselves when their only hope has the all the hazards of a pharmacologically active compound whose toxicity can’t have been exhaustively explored and very little efficacy, because patients of diseases like Duchenne’s and their families have little but hope and faith (in this case, misplaced faith and forlorn hope). The decision-makers aren’t being encouraged by FDA to take responsibility for their own actions.

        That Woodcock took the study sponsor’s stock price into account as a decision factor reveals a worse problem, that of a new wave of corporate cronyism at FDA. There’s already a privately-brought Federal racketeering lawsuit alleging undisclosed conflict of interest at the highest level in FDA’s management influenced when and how black box warnings were issued for the fluoroquinolone antibiotics, after two clinical studies showed statistically significant increases in collagen-related adverse events such as tendon rupture, other connective tissue failures, and most ominously, aortic dissections and aneursyms associated with use of the fluoroquinolones.

        The downside of FDA’s decision to approve eteplirsen isn’t going to get much popular press, and neither is the appearance of conflicts of interest in other FDA decision-making. When someone appears before Congress and swears she has no conflicts of interest as FDA administrator, knowing her husband’s hedge firm is massively invested in at least one pharmaceutical firm, and nothing bad happens to her, the press is not just sitting down on the job, they’re complicit in this conflict of interest. For that matter, Congress ought to have investigated more effectively. So ought the administration which promised us “unprecedented transparency and honesty in government.”

  6. tnr says:

    The bad thing about this approval is that it destroys the possibility of completing the ESSENCE placebo controlled trial. Sarepta says they are increasing the duration of that trial now to two years. There is no chance that anyone will ever finish that trial. Frankly, I can’t see how a DMB can even approve a placebo controlled trial now. If they did, the only people who would enter would be subjects whose family’s insurance won’t cover the drug and the family can’t afford it on their own. Once entered, families will find a lab who can assay the subject’s blood levels for the compound – if they are non-existent, then they will withdraw.

    This is truly a terrible non-scientific decision by Dr. Woodcock. I hope the drug does something and maybe in 10 years, someone will be able to show an increase in survival for those on drug versus those not on drug.

  7. David Borhani says:

    Very poorly considered decision, in my view, on the part of the FDA. They have have shirked their legal obligation, dating back to the 1962 Kefauver Harris Amendment to the Federal Food, Drug, and Cosmetics Act to approve on the basis of *efficacy*, not just (perceived) safety. Cf. JohnWayne & his comments.

    Erebus: Why do you think this is an “unprincipled exception?” What does that even mean? “unprincipled” meaning “unethical”, or something else entirely like “lacking organizing principles, and thus sui generis?” Why do you think that the FDA won’t continue to make other exceptions, unprincipled or not, in the future, given sufficient pressure from constituents.

    On the basis of this regrettable FDA decision, I suggest we all invest in Laetrile futures.

    1. loupgarous says:

      It’s about the right time for someone to convince CDER that laetrile just wasn’t studied long enough. Especially a generous political donor.

      “All you need is cash… all you need is cash… all you need is cash, cash… cash is all you need.” – The Ruttles

  8. FrankX says:

    There are a great number of things that went wrong with this development process.
    However, the approval by the FDA was the right decision. Allow me to address a number of misconceptions that have been used as arguments against approval in the comments above.
    If the primary issue with the drug’s approval was efficacy, then demonstrations of efficacy should be seen as supportive of approval. The FDA’s panel and internal decision makers like Drs. Farkas, Borio, and Unger say they did not have proof of the efficacy from WELL CONTROLLED STUDIES. There is no question studies were not designed to the standards of previous drug applications. However, the studies, 201/202 and 301 did show significant improvement over expected, non-treatment outcomes when measured against matched subjects. There is no question that it did produce Dystrophin, which was a primary biomarker. What is also not in question is that the drug produced volumes of evidence of efficacy OUTSIDE the primary endpoints of the study. The battle at the Ad Comm meeting was all about if the approval question could consider any information from outside the sponsor’s studies.

    The FDA’s internal review panel had several noted failures to understand the disease, it’s progression, or mechanisms in their statements and testimonies. These were directly refuted by the doctors who know this disease best. In addition to the review teams deficient understanding, they had two primary concerns. Was the increase in Dystrophin reasonably likely to show clinical benefits and how is the amount of Dystrophin measured? Both are legitimate concerns and open for debate. While less than anticipated, the amount produced seems to have made a difference in the boys in the study, both with the primary endpoints as well as numerous secondary outcomes.

    I would be happy to discuss the ways that I know this drugs works in detail, with examples, and videos. These led one panel member who voted ‘No’ in April to state “It is clear the drug works, but he must vote ‘No’ because of how the question was worded.

    Beyond the science that supports approval, you assert that ‘diseased children were paraded’ in an effort to get approval based on sympathy for desperate parents. This is wrong. Those children and parents were there to provide patient reported outcomes that clearly showed Eteplirsen was making a difference. Beyond that, the leading minds in the field testified that the results of the boys taking the drug were unprecedented in their careers and that Eteplirsen was the reason.

    Lastly, I can assure you that the sacrifices that will be required to use this medication are onerous. Weekly infusions or injections will require significant time or energy that the community would not be willing to waste on something that doesn’t work.

    As Dr. Cardiff wrote there is much that should be learned from this process to improve the approval and development of drugs for rare diseases. However, the approval was based on providing a safe drug, that has shown to be effective when you take the totality of evidence into consideration as required by law. While it is clear that the FDA must adapt its processes to address rare diseases, the learning from Eteplirsen can be used to improve the process. However that cannot be used to justify withholding a safe and effective drug.
    I provided my email and I would be happy to correspond further in a reasonable discussion of the merits of approval and how this might be used to improve the process.
    Thank you.

    1. John Wayne says:

      While I do not agree, you argument is both reasonable and well reasoned. I do have to take exception to the following paragraph:

      “Lastly, I can assure you that the sacrifices that will be required to use this medication are onerous. Weekly infusions or injections will require significant time or energy that the community would not be willing to waste on something that doesn’t work.”

      This is flatly incorrect. Patients, and the parents of patients, are not in a position to objectively judge the effectiveness of a therapy compared with the untreated control unless the efficacy is very large. The arguing about the significance of the clinical data suggests that the magnitude of efficacy across all patients is not that big. Of course, there is the possibility that there could be individual strong responders that show night/day improvements.

      1. FrankX says:

        Thank you John,
        I think we can agree and disagree on your statement:
        Patients, and the parents of patients, are not in a position to objectively judge the effectiveness of a therapy compared with the untreated control unless the efficacy is very large.

        In this case, the effectiveness of the therapy compared to similar boys IS very large and broadly documented with video evidence along side patient reported outcomes that persist beyond any placebo effect. The challenge is that the stark difference were not well captured within the study design by the sponsor. That doesn’t mean the are not there.

        I would like to provide one example. I was having dinner with a boy before he started taking eteplirsen. He was able to move his fingers to text on an Iphone, but not lift it. A year after starting treatment, he was able to lift one arm and apply deodorant with the other. This wasn’t him on a good day or because he was observed. It was a significant change. I personally have dozens of anecdotal stories, which individually can be dismissed, but as a whole, warrant consideration. Beyond that, this community understand the value of documenting benefits and has done so.

        In fair disclosure, I am father to a boy with Duchenne. Although he would not benefit from Eteplirsen, he is amenable to another exon skipping treatment. In that capacity, I can state unequivocally that we would not spend time, which is measured in days, not years, sitting in clinics and doctors offices getting a medicine that we didn’t know worked.

        The potential harm of making the type two error Dr. Woodcock spoke of is that boys would decline and die because we were not sure a treatment worked. You can argue that it is not a bad way to maintain the FDAs purity as the boys were going to die anyway. I find that approach to be wrong for obvious reasons. Approving the drug, gathering more data over larger numbers, with a broader view is necessary and appropriate, and is what is happening. During this time, boys will be on the drug and their disease progression slowed if the drug works or not if it doesn’t. That does not compromise science in general or the FDA specifically. Nor will it deter other drug development. I believe it will have the opposite effect as the path to approval does not require the rigid large placebo controlled studies that are difficult in the rare disease realm.

    2. AC says:

      “There is no question that it did produce Dystrophin, which was a primary biomarker.”

      With such poorly collected data and analysis (and small sample size), I’d question the reliability of the increase in Dystrophin.

      From the Summary Review:
      “”The Western blots submitted by the applicant for Study 201 were oversaturated, unreliable, and uninterpretable.” Because CDER also determined that the conditions under which the original IHC analysis was performed were inadequate, including that the reader was not masked to sequence and time, the Center requested a re-reading of the stored images by three masked pathologists under different conditions. The IHC results from the reread were not nearly as favorable, as compared to the initial IHC results reported by Sarepta.

      The re-read showed a nominally statistically significant increase in dystrophin in response to eteplirsen for the low dose group, but not the high dose group (. . . [T]he type-I error rate was not controlled for multiplicity.) Moreover, for the 4 patients who had received placebo through Week 24 and then switched to eteplirsen, there was no increase in dystrophin at Week 48.

      […] The Western blot analysis resulted in Week 180 dystrophin levels that were small, with a mean increase of only 0.93% of normal dystrophin levels in the muscle fibers. Dr. Unger remarked that the lack of concordance between the IHC and the Western Blot results is “striking” and also noted that FDA did not verify the integrity of the IHC results. As previously noted, each muscle fiber that shows any amount of dystrophin is counted as positive in IHC, regardless of the actual quantity of dystrophin present.”

      1. Frankx says:

        I appreciate the point you make, but there is more to the story. The PRESENCE of dystrophin being produced was not in question nor the assumption that the absence of dystrophin is the cause of the disease. In all the tests, the presence of dystrophin was noted, but the amount of dsytrophin produced was hard to quantify. The review team did not conclude the dystrophin found was the result of the drug whereas other scientists did. This is a viable scientific debate.
        The study had flaws. On of the primary gaps was the measuring of the dystrophin found. The various methods have their benefits and weaknesses, but remember, these are the first drugs that ever produced dystrophin. The measurement techniques at those small levels are still being developed. Further, there is every indication that the production of dystrophin may not be uniform across all muscle tissue. Given that, the amount of dystrophin found in any one sample is subject to luck. The field will need to advance in order to better capture good data. It has and will continue to improve.
        What is clear as day is the observable differences between the boys who get the drug and those who don’t. Ambulation, upper arms utilization, pulmonary function, decrease in falls, increase in stability while walking, ability to regain ambulation after a leg fracture, and others.

        1. AC says:

          “the first drugs that ever produced dystrophin”
          You mention that the drug is responsible for the increase, but acknowledge the methods are unreliable at the level of detection needed to see the effect and that the biopsy sampling may not be representative. This problem is further exacerbated by the use of external baseline samples. The reported increase in dystrophin was also not dose-dependent which conflicts with their primate studies (of course monkeys exaggerate, so that may be expected).

          As for the other outcomes, were the studies sufficiently powered and controlled to support a clinical benefit?
          The FDA review team were not convinced that there were clinical benefits based on the data.

          The use of external controls for the clinical outcomes greatly compromise any effect seen that is without endpoints that are objective (with well characterized baseline and treatment variables that impact those endpoints) or are subject to selection bias.

          From the Summary Review:
          “Even if the 30 mg/kg/week dose were considered to have a meaningful effect on the surrogate endpoint, we already know this dose is sub-therapeutic. We know this because patients who have been receiving this eteplirsen dose for some 3.5 years have been progressing at a rate that is similar to that expected, based on the natural history of the disease”

          “the data from other measures of physical function, i.e., rise time, timed 10-meter run, and North Star Ambulatory Assessment (NSAA), show steady decline in the eteplirsen-treated patients that does not differ substantially from the decline in the external control group.”

          1. FrankX says:

            I worded that poorly.
            The ability to detect dystrophin was solid. The ability to count the dystrophin was not as clear as it should have been.
            The dose response was not found, although it was also not explored sufficiently. That is a significant part of the follow up study requirements.
            The findings that the actual clinical benefit did not vary from expected is an example of the review teams poor understanding of the disease. I wish I could be more polite. They repeatedly were just wrong about what the baseline untreated expectations were. At one point, Dr. Farkas said that 15% of boys with this deletion are still ambulatory at age 18. This was refuted by clinicians and doctors who have treated Duchenne for 25 years as well as the director of the database that the FDA misquoted.

            I appreciate all the attention and very reasonable discussions about the approval. The key here is that we can all see the process was not clean nor sufficiently controlled. There are reasons for this, but those are not relevant. The bottom line is that I, and congress agree with Dr. Woodcock that when considering a treatment for a devastating illness, with no current treatment, evidence from all possible sources should be considered and weighed for its merit.

            Obviously, I have more experience with the boys themselves than science in general. But it doesn’t take a great deal of experience in either to see the effect of the drug. The cab driver that took us to the Ad Comm meeting was able to tell the difference between the treated and untreated boys as they entered the conference center. With all the caveats of ‘crazed and desperate parent’ the amount of evidence visible cannot be dismissed because it didn’t come from a lab.

            The establishment of career FDA reviewers said they were not sure Eteplirsen was the reason for the observable differences between the boys. However, the clinicians who have treated the boys, and were not part of the trial or company predicted that of the twelve that started the trial, 1 or 2 would still be walking at this point, given their starting 6 minute walk results. 10 were still walking with increased stability than would be expected in untreated boys. The FDA cited better Physical Therapy, use of steroids, very motivated mothers, and other reasons why these 10 were walking well beyond expectations.

            The company can be faulted for their study design. At the same time, the FDA review team lead by Dr. Farkas displayed as bad a scientific approach to oppose the approval. The opposition extended to the design of the Ad Com questions and proceedings. I appreciate the nature of the approval is controversial. While many are skeptical of the sponsor, the FDA review team gets a pass for any level of bias, which was clear to many attendees that were not part of the Duchenne community.

            Thank you all and I wish you good evening.

  9. Shanedorf says:

    Thanks to Derek for sharing his views and giving all of us a platform to share ours. And tremendous thanks to FrankX for providing additional facts and insight on the Sarepta approval process from his point of view. These are sterling examples of why this is one of the best blogs on the internet.
    Kudos all around

  10. Trebor says:

    Caveat emptor

  11. anon says:

    People are still debating the approval process, data and science here. The stock has doubled in 3 days. That’s all it matters. Please.

  12. old timer says:

    “Money makes you do things you don’t want to do” — The Godfather

  13. Lindy says:

    But the question remains – will the payers pay for it when it may not be more than a $300K/yr placebo? This approval process shows one of the major flaws in the ‘right to try’ movement. In this case, desperate parents who are watching their children decline are able to pressure FDA to approve a drug off of very low patient numbers and flawed trial designs. This approval is from not much more than anecdotal reports from these desperate families. You would find DMD families ready to testify to the benefits they saw in their children from other drugs which failed in trials with much higher patient numbers and quality trial designs. I have tremendous respect for the research that these families have driven and have worked to help fund, but we can’t let that overrule good science and quality trials.

    1. Thomas says:

      Shouldn’t “right to try” be with a free supply of drug?

      If combined with good record keeping it could even produce meaningful statistics.

      Heck, it can even be seen as a different method of financing trials. The patients finance the ‘standard of care’ for themselves in this model. The drug is free, just as in normal trials. Only problem is: where to get the controls?

    1. loupgarous says:

      At least Forbes is on the case. Time was fast asleep when FDA commissioner Margaret Hamburg resigned in 2015. She got a canonization piece as a going-away present.

  14. Morten G says:

    Ironically, the most affordable Exondys will be from enrolling in other trials where they have to provide the standard of care. You’ll have to give your patients Exondys if you are doing a DMD trial in the US now, won’t you?

    1. CMCguy says:

      Morten I do not think this works that way as even though it is an approved treatment option (yet only one) that does not necessarily make it a standard of care which typically involve much more rigorous and established acceptance in practice.
      Although I am concerned how this played out with a higher up use of power to over rule subordinates it does reflect flaws in the system where an approval largely forced as black & white decisions (which can be later reversed) yet IMO there really should be a Conditional Approval by FDA such as EMEA that may have allowed progression to get the required proof. Scientists always want clear answers from others and can fool ourselves on what the facts are as try to remain objective. Per FrankX I give certain credence to the voice of patients, who can often be ignored due to obviously bias, and although some observations can be born more out of hope and reality, they or their MDs “anecdotal” data should be part of the process because at the end of the day that is why I do want I do with my PhD.

  15. Diver dude says:

    The next signal will be what the directors and officers of the company choose to do with their shares. If they start selling quickly and in large amounts, you can draw your own conclusions.

  16. Eric says:

    I agree with Derek, this strikes me as a mistake. I sincerely do hope that it works, but based upon the presentation at the Ad Comm, I just don’t know if this treatment will do anything to slow the progression of this horrible disease. The increases in dystrophin in the 301 trial (the most reliable data available) were very small. Whether that’s enough to lead to any functional improvement is anybody’s guess at this time. The testimonials and videos are anecdotal observations from parents and children suffering from a horrible disease. They desperately want this to work, which introduces bias. It makes it extremely difficult to separate out potential placebo effects from a true response.

    I’m left thinking that we just don’t know if this treatment works or not. But now that it is approved virtually every parent that can find a way to pay for this treatment will start it immediately. This will certainly slow recruitment for any placebo controlled trials or any trials for other therapies. As a parent I would never consider putting my child in a placebo-controlled study if I could pay for the drug on my own.

    If the treatment does actually work, then it’s major step forward and I’m so happy for children and families that will see some hope, some promise of a treatment. If the treatment eventually turns out to be a failure Sarepta will walk away with billions of dollars in sales, families will have spent their entire savings on a placebo, and DMD research will have been slowed for nearly a decade. It’s a big gamble.

  17. tnr says:

    I hope that the FDA will mandate Sarepta to initiate a patient registry for all patients who will be prescribed the drug. Having that information will allow researchers in the future to examine the data and most importantly, the survival pattern of these boys compared to boys prior to the drug. Its not a randomized trial but at least it will provide some information to see how well the drug works.

  18. DJK says:

    I found the most astonishing part of the Endpoints article to be that Woodcock took the sponsor’s stock price into consideration. Surely the FDA shouldn’t be allowed to factor that kind of consideration into a decision for approval!?!

  19. Anon says:

    Storm in a tea cup. If the drug doesn’t work, then patients won’t use it and payers won’t pay for it. Or they can sue the FDA or the manufacturer for aproving/selling something that should never have been sold/approved in the first place.

    1. Fenichel says:

      Anon would be right, if (a) patients could tell whether drugs work or not (They can’t; that’s why we do clinical trials), (b) it were possible to sue FDA for bad decisions, and (c) anyone had ever succeeded in suing a manufacturer for inefficacy.

      1. Frankx says:

        Thank you, but I call Bullshit.
        You sanctimonious fools are why the FDA is such a mess. To say that Patients can’t tell if a drug work is offensive and ridiculous. You say that is why we need clinical trials to tell us if the boys can still walk, fall less, maintain their pulmonary function, can feed themselves long after their untreated peers lost the ability. That arrogance is why Congress had to step in and will continue to do so if it persists.
        Everyone at the Ad Com meeting could see the trial participants were markedly different from non-treated subjects. It is pure arrogance to say that you (FDAers) are the only capable judge of what does and doesn’t work. With the wider access to the drug, much more will be known. If you think for a minute, the same patient community that you saw in action will sit by placidly if Exondys does nothing, you are as foolish as your comments above. Reasonably likely to produce a benefit is the legal standard, not absolute certainty. Approval to market the drug will show, with abundant clarity if the drug works, measured by actual boys living their lives and not some FDA bureaucrat with a personal agenda like Farkas.
        To the rest of those commenting here, I apologize for leaving the high road, but Fenichel’s thinking has directly harmed too many to count and must be called out.

        1. zero says:

          Not to be obnoxious, but you are aware that the placebo effect is a real thing? Patients receiving sugar pills report a bewildering array of effects that they expected to experience based on the drug they thought they were getting. This by definition is patients not being able to tell if they are being affected by a drug.
          The mechanism of placebo is poorly understood. How is it possible that placebo-controlled patients can show improved survival and disease free progression rates with cancers and other debilitating diseases? Nobody knows for sure but the effect is real. The same applies to these patients; some percentage of them will ‘feel better’ because they are ‘getting treatments’. This may even be backed up with clinical data such as increased protein expression. These anecdotes will sound like the drug is effective, but the reality is these patients could have been injected with saline every week and had the same result.

          Only a sufficiently powerful statistical study can reveal whether there is a true advantage to using the drug. Now that this unproven drug has been approved, such a study will never occur. We won’t know for another decade or so (if ever) whether this company is selling billions of dollars of snake oil.

        2. Eric says:

          Frankx, I politely disagree. If the drug were curative, then yes patients could clearly see if it is working or not. The data presented at the Ad Comm does not suggest eteplirsen is a curative treatment. They presented the 6 minute walk test data for each individual patient and if I recall correctly, every single patient declined over a 4 year time frame. In fact two of the patients on eteplirsen lost the ability to ambulate during the trial. The mean decline in eteplirsen treated patients may be smaller than was previously reported in historical controls, but nonetheless 6 minute walk was getting worse. Furthermore, the variability is huge. In the historical controls some subjects lost the ability to walk at 10 years of age while others were as late as 15 years of age. If the overall affect is slowing the progress of the disease, how can a patient reasonably know that the treatment is working?

          1. FrankX says:

            I appreciate your point. Eteplirsen is not curative. There may be some increase in function in some abilities, but that is not a persistent or prolonged outcome. It does reduce the progression of the declines faced by the boys.
            First, the two that did stop walking did so early in the trial, prior to when the drug is thought is to have sufficient time to be effective to the degree necessary.
            The rest of the boys did have decline over the four years. The key is they decreased significantly less over that time than all the other case history of Duchenne would predict. There is significant variability in what age loss of ambulation occurs. The reasons are not well known. However, and the FDA review team could not grasp this, the boys will follow a relatively consistent decline in 6mwt once they hit the threshold of 300 meters. If may be three or six months when they dip below a certain point, but it is not years.
            The data gathered by the repository and clinics show across large numbers of boys, that only one of the 12 subjects would be expected to still be walking. That is the key. This group of boys had a huge variance from expectations when compared to large numbers of others who were untreated with Eteplirsen.
            The answer is if boys are still walking a year after they hit the 6mwt threshold for predicting loss of ambulation in 3-6 months, it is reasonable to say it works. If the mean age of hitting that threshold increases over time across the larger sample of treated boys, versus untreated or historical data, then the Eteplirsen is mostly likely the cause.
            I absolutely allow for other potential causes that enabled these boys to outperform the expert predictions, such as radioactive spider bites and voodoo magic, but the most reasonable basis to conclude that Eteplirsen may have had something to do with it.

            With regards to the two who did stop walking, both are still functioning at higher than expected pulmonary function, upper limb strength, and range of motion. They are able to perform significant activities of daily life at a higher level than expected when compared to the natural history.

            Listen, I get the value of scientific process and statistical power to move “I think” to “We know.” The path the sponsor took to get to this point was not optimal and should not be repeated. At the same time, the functional differences in the boys who take the drug and those who don’t are absolutely clear and unprecedented in the opinion of not only the patients and parents, but the doctors who have have studied and treated Duchenne across the world, for a quarter century or more. Dosing regime, amount of expression of dystrophin for functional benefit, assay methods, quantification standards are all in need of refinement. Everybody in the community wishes we knew what we know five years ago. Are we certain Eteplirsen will slow the decline? No. Are we reasonable sure that it has positive effects in boys with Duchenne? The people that know Duchenne best are and that is good enough to move forward with wider use that will answer the questions definitively.

        3. Fenichel says:

          Patients knew for many years that gastric freezing cured their ulcers, that ligation of the internal-mammary artery cured their angina, that encainide & flecainide saved their lives after MIs, and that homeopathy was good for everything. It’s a pity that we are wasting time doing science, isn’t it?

          1. NJBiologist says:

            –and in a nice historical irony, it was exercise tolerance that built the case for internal mammary ligation! Right up until Cobb showed that placebo surgery increased it just as much, of course.

  20. Emjeff says:

    It’s clear that this was an approval based on emotion and empathy for these unfortunate boys, and not based on science. As a former FDAer and current industry scientist, I was appalled at the concerns raised by Dr. Woodcock concerning the company’s stock price. Finally, it has reinforced my belief that Janet Woodcock knows almost no science.

    1. Fenichel says:

      I’m a former FDAer too, and this episode leaves my opinion of Janet Woodcock unchanged — it has been stuck at the low end of the scale for many years (see for more on this). The only real surprise here was the decision taken by Rob Califf. When he was an academic trialist, and I was at FDA or (later) an industry consultant, I interacted with him many times, and he was always evidence-based, never afraid to disappoint the sponsors of the trials he ran, or the desperate patients to whom those sponsors had given false hopes. Now this.
      I wish it were not so, but whatever else he does, the eteplirsen decision will probably, in trialist circles, be the decision for which his commissionership will be remembered, like Watergate and the Nixon presidency, or the breast-implant fiasco and David Kessler. It’s a sad business.

  21. Andre Brandli says:

    I do not want to comment on the details of the controversial process that led to the approval of Exondys (eteplirsen) for the treatment of DMD by the FDA. I would rather like to point out the mechanism of action of Exondys ( It is a morpholine antisense oligomer, which triggers excision of exon 51 during the pre-mRNA splicing of the dystrophin RNA transcript. Induction of exon skipping by Exondys restores the reading frame of dystrophin in those DMD patients with this particular class of frame shifting mutations. As a result of Exondys treatment, patients will produce truncated, but functional dystrophin proteins. The truncated dystrophin is similar to the situation in Becker muscular dystrophy, which is a less severe form of muscular dystrophy. This should slow the progression of DMD, which remains to be proven for Exondys as many commentators have pointed out. Now, the important point is that only about 13% of the DMD patients carry mutations that could be treated by Exondys. In other words, 87% or the vast majority of DMD patients will not be able to profit from Exondys. This leaves lots of room for the development of novel DMD treatments. Even if Exondys were to work in patients as advertised, it is not a cure. More research is therefore necessary to develop DMD cures that will be beneficial to those many DMD patients for whom Exondys is not an option.

  22. Sulphonamide says:

    Can anyone comment on how much the drug could actually cost the family with an average sort of health insurance? Putting asides issues of what this precedent does to our ability to work out whether approved drugs actually work or not, for many families $10k / year may seem money very well-spent for a placebo that gives hope; on the other hand $150k / year is effectively blackmailing many families who presumably will do anything they can for their children and will be willing to sacrifice their own financial futures by spending every cent on this drug….and if the drug doesn’t work then the FDA have indeed done something highly damaging with this decision.

  23. a says:

    Any news on BMS reorg? Big announcement tomorrow?

  24. former fda reviewer says:

    I was part of the FDA when Riluzole for ALS was approved long ago. Little efficacy but no other drugs available, so it got a thumbs down at the division level and Woodcock caved into the public hearing overriding the decision.

  25. Emjeff says:

    This may be an unpopular comment, but this reinforces my view that FDA is very bad at evaluating efficacy. It is clear from her review that Janet Woodcock knows almost nothing about either basic or clinical pharmacology. These is no way that increasing anything by 0.3% will have an effect on any biological or disease process. Furthermore,ma good assay will have, at best, a 7-10% CV, so their “positive” increases are well within the margin of error for even a good assay.

    With the way healthcare is structured and paid for these days, an evaluation of safety should be FDA’s primary concern . They simply do not, and will not, have the expertise to evaluate efficacy, particularly for drugs like this. Once a drug is deemed safe, payers , using experts in the disease state under question, can evaluate benefit-risk and decide whether or not to cover the drug.

    1. Fenichel says:

      I don’t know how safety can be evaluated in the absence of efficacy. Suppose that eteplirsen really does slow the downward course in DMD, but only in 1% of the patients treated, perhaps because of some genetic mix that we don’t know how to identify. That is real efficacy, and probably impossible to detect in controlled trials, using the available patient population. But now suppose that eteplirsen also *accelerates* the downward course in 2% of the patients; we couldn’t detect that either.
      In this sort of situation, the FDA standard (NOT what was done in this case) is sensible. Drugs for rare diseases should be made available if there is *good reason* to anticipate efficacy (**not just anecdotes and a 0.3% change in a biomarker**), and as to safety, well, we haven’t seen anything awful yet, and we have no good reason to anticipate anything awful, so OK. Is that just muddling through? Yes. Small numbers do that to us.

  26. MD says:

    FDA agreed dystrophin is a surrogate (no surprise)

    FDA agreed that etep produced dystrophin in a statistically significant manner.

    FDA/Borio/Unger concede that nobody knows what the threshold is for dystrophin correlation to clinical benefit.

    Yet Borio and Unger argue that etep doesn’t produce dystrophin at a clinically meaningful level. How can you argue that point when you concede nobody knows?

    Flawed logic anyone? Think about it.

    Moreover, Dr. Kunkel, one of the leading experts in DMD, stated that etep by WB produced dystrophin at an “appreciable level”, a level he’s never seen in DMD patients.

    A reasonable question to ask, especially in rare and often poorly understood diseases, is why expert input is not more integrated in the process. Yes, there’s potential COI issues but is that any more problematic than having FDA-funded AdCom panelists who clearly don’t understand the intricacies of the disease? And, does anybody else feel uncomfortable with having an ophthamologist (Dr. Farkas) review DMD data and make definitive proclamations that were clearly in scientific error? FDA needs to fix this sooner than later.

    This, in the end, is a dystrophin threshold argument that has to be considered in the context of what FDASIA mandates including patient, expert input (and no, Dr. Unger, that doesn’t mean just hearing patients out as a matter of anecdotal courtesy and then reverting back to business as usual).

    Etep inarguably produces dystrophin and Dr. Woodcock absolutely made the right call in the context of FDASIA, expert and patient input.

    The alternative? DMD children, perhaps the most vulnerable constituents of our society, will face the 100% certainty of rapidly diminished QOL, extreme suffering and early mortality while the scientific i’s are dotted and the t’s are crossed.

    What Dr. Woodcock correctly decided is let’s give these children a chance based on safety and promising data in keeping with what Congress mandated through FDASIA while we figure out the scientific certainties (to a p<=.05 level which by definition has its own set of scientific trade offs).

    In the end, even if the confirmatory data don't pan out, Dr. Woodcock made the right call.


  27. thlizrd says:

    Wow, another educated idiot. The only thing he forgets to mention in his “anti-Sarepta” diatribe is that 10 of the 12 DMD boys on eteplirsen are STILL WALKING after 5 years. Those are the facts. That doesn’t happen with placebo. Is that science enough for you?

    Now, read these comments again while considering this and the fact that Sarepta’s ad board was “rigged” by FDA DNP’s Ron Farkas who was later shown the door (yes, there’s a lot more you don’t know). Eteplirsen is effective and has few, if any, side effects.

    Finally, FDASIA is now the law — it is NOT a request. This law was created to show compassion for those living with a deadly disease who have limited options. I know this isn’t convenient for some of you “numbers only” guys who sit there behind your desks. But, here’s an idea: spend a day with a DMD family before you hit the “send” button,

  28. LegalStockEagle says:

    Having just stumbled across this blog thread, I’d like to thank all who have contributed to one of the most reasonable, thoughtful and balanced investment-related internet discussions I have ever read in my nearly 20 years analyzing public companies across multiple industries.

    Strong on substance, without the non-sensical bullish/bearish rhetoric that usually accompanies these types of discussion is the rarest of exceptions to the rule, but this blog has it in abundance.

    Thanks to all.

  29. Marie says:

    The only thing I disagree with in this article is the statement that no one knows whether this drug works or not. Unless something has been tested there’s no reason to believe it does work. No one knows if vitamin d might work, or snake oil, or chocolate. Given the lack of evidence of efficacy this drug should not ave been approved. Unfortunately, now that the 21st Century Cures Act has passed Congress we’re more likely to see such pharmaceutical driven farces in the future.

Comments are closed.