Skip to Content

Clinical Trials

I’m Surprised That They’re Surprised

This article at the Wall Street Journal should not come as a particular surprise to anyone who followed the Sarepta/eteplirsen story over the last few years. But it’s a good overview of it, and it brings up several questions. (My own take on the drug approval is here – I didn’t think it should have passed).

A brief summary, in case you haven’t followed this one: the company had been developing the exon-skipping therapy for Duchenne muscular dystrophy (DMD) for years, but had very thin clinical data when they went to the FDA. In addition, a broadly similar therapy backed by GSK had failed to show any benefit. But after delays and arguments about accelerated approval (or approval in any form), the drug was approved, to the surprise of many, when Janet Woodcock of the FDA overruled other recommendations.

This is the part that surprises me, I think:

The company, Sarepta Therapeutics Inc., leveraged the emotional appeal of the parents’ stories. Without that, the drug—eteplirsen—likely wouldn’t have been approved, said stock analysts, investors and people close to the matter.

An FDA advisory committee voted 7-6 last year to oppose the drug’s approval, a position later overruled. The Wall Street Journal found that a majority didn’t know how Sarepta’s consultant had guided the patients’ families.

“More might have voted no if they were aware,” said Dr. Bruce Ovbiagele, a committee member, when he learned the role the company consultant played in helping parents prepare for FDA hearings.

I am not surprised at the company using the parents’ stories to make an emotional appeal – this happens all the time in drug approval decisions for rare and serious diseases. I’m not surprised at the company carefully grooming the families to help sway the FDA committee vote, because this happens all the time, too. What I find odd is that the committee members say that they didn’t know that this was happening. Where have you been? What species do you belong to?

Note that in the paragraph above I am not making value judgments about the wisdom or appropriateness of these sorts of maneuvers. That’s another topic. All I’m saying is that patients and drugmakers, since they often find themselves on the same side of a drug approval issue, cooperate with each other to try to achieve their goals. This is exactly how it works with every other key decision taken by a government body, I should also note. Now, I think that this sort of thing should be disclosed, let me make that clear. When some political issue comes up for a vote, and the airwaves are full of ads taken out by the Concerned People Who Want Only Good Stuff For All of Us Committee, I think that the sources of the CPWWOGSFAUC’s funding should be on record for anyone who wants to know where the money is really coming from. And the same goes for lobbying for (or against) drug approvals.

The relationship between companies and patient advocacy groups is complex, and runs along a scale. At one end, you have patient groups that have existed long before any possible therapy has been in the works, and who are likely willing to help out any drug company that might seriously be able to deliver something. At the other end, you have groups that have been created ex nihilo simply to push an agenda. (This same spectrum is found in every other public lobbying effort, and it’s possible to have every shade in between).

These seem, at least to me, to be well known features of public policy, so I’m really baffled by the FDA committee members saying “Had I but known”. The second thing I find odd about that is, what difference should that have made? The decision was supposed to have been based on the data from the clinical trials. I know that the FDA has been getting patients (and patient advocacy groups) more involved in these hearings, but that doesn’t mean that approvals should be tilted to those whose consultants have coached the most wrenching testimony and the most coordinated talking points. We’ve known about these forms of argument for a long, long time now, and the idea is not to elevate pathos at the expense of ethos or (especially) logos. What a mess.

27 comments on “I’m Surprised That They’re Surprised”

  1. Isidore says:

    A number of different issues here. If the FDA is getting patients and patient advocacy groups involved it means that the FDA is suggesting that non-scientific considerations should be included in deciding whether to approve a drug, otherwise why bother. When victims’ relatives give testimony in court at the penalty phase of a trial it is with the expressed purpose of influencing the jury; why would an FDA Advisory Committee meeting be any different? So, if the FDA is giving a formal role in drug approval to patients and their advocacy group, does it need to set guidelines regarding the interaction of such groups with drug companies? This would probably be in violation of the First Amendment if it entailed no interaction, especially with groups that have been around before a drug was in development, so the agency would have to define specifically the types of interactions allowed, and enforcing this would be a nightmare for everyone except the lawyers.

  2. Virginia Jolly says:

    Since personal testimony is given at the penalty phase, it would be appropriate to remind the jurors the nature of the testimony and how it differs from the evidence given by the trial studies and to balance their consideration appropriately. This would be appropriate, and this kind of reminder of the nature of evidence (and indeed a good many things including the consequences of purjery) is done constantly in courts of law. Reminders are in accordance with the First Amendment, even as personal testimony is in any forum.

    That said, I and many other members of the mental health consumer community – particularly if the Bipolar community – were so relieved when the years-long FDA approval for the mood-stabilizing drug Lamictal was finally approved especially with its overwhelming support by psychiatric study. It has allowed many Bipolar sufferers to avoid debilitating manias and allowed treatment of other symptoms to help these people (me included) obtain and retain a life that is productive for them and their community.

  3. biotechtoreador says:

    ACAD’s approval, of a drug that succeeded in 1/3 phase 3 studies, is another example of the FDA being swayed by patient advocacy:

    It’s clear that this is a good, if cynical, strategy to help get questionable drugs approved.

  4. daisyj says:

    He went on to report that members of the committee were shocked–shocked!–to find out that there was gambling going on at Rick’s Café.

    1. Mad Chemist says:

      Your reference game is on point, sick burn. XD

  5. steve says:

    All drug approvals are based on a risk/benefit calculation. It is perfectly appropriate to take into consideration the patients and their families in calculating that risk/benefit. If you had a disabling or lethal disease your perception of risk/benefit might be very different than it would be as someone just reviewing dry data. That said, surely a review panel should be aware of the nature of patient advocacy and exhibit as healthy a skepticism of what is being said on that side as they are about the clinical data under review.

  6. MCS says:

    I fail to see how patient advocacy adds anything except noise to the process. Watching a child die slowly has to be at the top of any parent’s list of nightmares, or any mere person. Of all of the things that they would do to prevent this, throwing great wads of money into a hole, whether their own, some insurance company’s or the government’s, doesn’t even rate.

    The FDA is given the job of trying to insure that there is a rational reason to believe that a drug has value. In all the history of quackery, there has never been a lack of advocacy or people willing to exploit misery. The “ethical” drug industry is supposed to be different. The difference between a plausible conjecture and a marketed drug should be data not feelings or desperation.

    1. steve says:

      The idea that evaluation of clinical data is value-free is nice in theory but doesn’t exist in the real world. The days of doctors in white coats dictating to patients is over. Enlightened physicians – and regulatory authorities – take patients viewpoints into consideration and eschew holier than thou attitudes and the outdated motion that it’s all dry data analysis. If that were true then review boards would never be needed and would never give a split decision.

      1. MCS says:

        An ineffective drug does harm in many ways:

        Most obviously, from side effects and whatever discomfort is associated with administration.

        Especially for rare diseases, charitable money will be diverted from research to paying for the drug.

        The existence of a “treatment”, however ineffective, will be a disincentive to anyone trying to develop something better and an obstruction when it come to testing and approval.

        The bias is always to do something. It seems that this will add a, no doubt, significant cost to treatment without being effective enough to reduce it in other ways. If it had been, it wouldn’t have come down to a coin toss.

        As a country, we have so far avoided having to put a specific value on a added month of life by limiting treatments to those that have objective evidence of effectiveness. This won’t last forever but the reckoning is brought perceptibly closer when the “dry data” is overridden by political pressure or appeals to sympathy.

        Derek’s point about the credulity of the reviewers is spot on. For even a rare disease there is still lots of money at issue, a little coaching and coordination is probably the least sinister manipulation that goes on.

    2. Dan says:

      There are many things about a disease that are not easily boiled down to numerical data. Some things, sure, you can boil down to raw numbers: You can look at diabetes in terms of blood sugar numbers, A1c Hemoglobin count. You can look at side effects in terms of raw measurements like “number of patients that dropped out due to side effects” or percentage of patients that require anti-emetics, and how much. But even in a VERY “number-friendly” drugs like insulin potentiators for Type-2 diabetes you can’t measure everything.

      The theory is that for most drugs the subjective experience of the patients on the drug matters. When you talk about things like quality of life improvements you can try to quantify the numerical reduction in the magnitude of a tremor or increase in grip strength and gait speed (to list a few things I used to measure in a study on health and aging), and that data is important to proving effectiveness, but just as important is the subjective experience of the patient and how the measured results do (or, just as importantly do not) impact their actual daily life.

      It’s NOT looking at these factors, I would strongly argue, that has resulted in vast underestimation of the CNS effects of a lot of common drugs.

  7. It’s appropriate to bring in patients and their stories in order to weigh risks, benefits, and harms properly, for sure.

    However, it is necessary to have a firm grip on these factors beforehand.

    This seems more like trying to get a non zero weight assigned to a benefit that does not exist, than an effort to properly balance actual factors.

  8. johnnyboy says:

    The drug is demonstrably safe. There is no alternative approved treatment. Efficacy data is poor, but the company is required to run another confirmatory clinical trial to maintain approval. Yes the drug is expensive and it’s putting a strain on the insurance companies and therefore all patients, but that’s more a consequence of the shittiness of the american health system than Sarepta’s fault. At some point, if you think you are in the field of treating human beings, you have to get off of your “data is everything” high horse and consider the patient. If Exondys is only a placebo, at least the children might feel a little better before they get to die. That you should care about, more than the sacrosanctedness of the data.

    1. Derek Lowe says:

      If it’s only a placebo, they can feel better for a lot less than they’re being asked to pay Sarepta for Exondys. Right?

      1. Isidore says:

        Doesn’t the fact that the FDA has requested a confirmatory trial imply that the agency has doubts regarding the lack of efficacy shown prior to approval. In other words, does this not suggest that the Agency believes that the Phase III trial was not designed properly and therefore the lack of efficacy shown should not be the last word? Or perhaps it is an attempt by the agency to salvage its reputation by hoping that the confirmatory trial will show some efficacy. Or it may simply by an attempt by those in charge at the FDA to pass the hot potato to their successors who will have to evaluate and act upon the new results.

        1. johnnyboy says:

          I think it was mainly meant as a way to give the patients a chance while more conclusive data was collected. In the absence of any other treatment alternative, for a disease that becomes lethal in a fairly short timeline, asking for a better clinical trial before granting approval meant condemning the current patients to death. If Exondys doesn’t work, they will die anyway, and not faster. If it does work, this will give them a reprieve while better data is collected. To me it’s a no-brainer, in this particular situation. Granting accelerated approval while asking for a confirmatory trial is not an Exondys-only situation, it is also done for other drugs (see Tecentriq in bladder cancer, the confirmatory trial just came up negative, and approval will likely be revoked).

          1. Design Monkey says:

            It is no brainer, as far as parents are fully informed and pay for the drug themselves. Not the government, not the insurance, not charities. With their own money parents fully have rights to waste it on chasing ghosts in whatever way they please. With other peoples money – nope, there they don’t have such rights.

    2. DrOcto says:

      Not to mention that placebos also have a demonstrable theraputic effect, whereas these drugs do not.

      1. Slotermeyer says:

        “…placebos also have a demonstrable theraputic [sic] effect…”

        Do they, now?

  9. MCS says:

    This didn’t start as a discussion about providing convincing placebos and I can’t really see the FDA managing that anyway. It’s about what criteria should be used to determine drugs that are legally saleable to the general public. Efficacy, based on objective evaluation of clinical trials is the accepted criteria. No one believes that this data is perfect.

    The FDA is a government bureaucracy, a very blunt instrument. The best that we can hope for are decisions based on the best available objective and supportable data. It won’t be perfect but all of the alternatives are worse.

    If we turn the process into a pageant of suffering, Fox probably has the equipment to handle the texts for the ultimate reality show.

    Absent an infinite supply of money, every dollar spent is a dollar that won’t be spent on something else. An ineffective drug potentially uses money that would otherwise go to providing genuine good.

    It’s hard to imagine that dis-impacting the American health care system won’t at some point come down to evaluating costs in some objective way. My bet is dollars per month of survival. Every attempt to do this in Great Brittan seems to end as soon as the latest restriction becomes known to the general public. It may come down to a lottery where certain conditions are considered worth treating or not on the basis of politics or which victims are seen as culpable for their condition. Congenital conditions might be big “winners” under that scenario. The rest of us have probably done something to deserve whatever happens to us.

    johnnyboy, I suspect that this hits you close to home. Everyone affected by this has my sympathy (worth very little) as well as my hope that the initial determination was in error and that this or something soon does genuine good. For whatever that’s worth.

    1. steve says:

      It’s very easy to have that attitude until you have to care for someone who is seriously ill. If nothing else has been proven by the recent House health care bill it’s that compassion is sorely lacking in many places. Healthcare is not something that can be done by computer; patients are human beings with a particular perspective based on their experience and deserve more than being termed “a pageant of suffering” to be ignored and denigrated.

      1. MCS says:

        Compassion isn’t the issue, it’s about how scarce resources should be allocated. Transparently, from defensible data or something else.

        Politicians have been compassionate with taxpayers money for a long time. That’s how we got to where we are, the money’s running out and there will never be an end to places that “deserve” support.

        For a look at government compassion, go to a nursing home that depends on Medicaid. I have thought that when the politicians got done fixing health care that we would be lucky if we ended up with something no worse than the V.A. We aren’t that lucky. A visionary politician is one that will admit that water runs down hill, if questioned at gun point. None has ever conceded that could ever reach the bottom. Spending money for compassion just widens and smooths the channel so it will run faster.

    2. johnnyboy says:

      I don’t have personal skin in this game. However my own training leads me to recognize that medical treatment is never an exact science, and in certain situations, you have to show some flexibility. A clinical trial is not a laboratory experiment, it’s subject to a lot of imponderables, a lot of factors that you don’t know you don’t know. Faced with uncertain data, it’s not always the best to hold the trial to the same black and white standards that you’d apply to a lab experiment. Derek and most people in the comments section appear to hold the position that since the clinical trials were small and not very well designed, and thepositivity of the results somewhat equivocal, they should essentially be rejected, or considered as negative. I would agree, in most other disease situations where there are treatment alternatives. One argument against approval of Exondys is that it’s going to open the floodgates and lead to approval of a host of other poorly supported drugs. Time will tell if this is the case, but looking back at history, AIDS activists pushed for approval of drugs before full efficacy was established (eg. Videx), and that didn’t lead to any overall decrease in the quality of later drug approvals. Indeed it may have saved or at least prolonged some lives.

  10. steve says:

    People may want to read what FDA is doing to modernize, including listening to patient’s perspectives on different diseases and treatments (scroll down to section on PFDD).

  11. Erebus says:

    Here’s the thing: The laws and regulations must be applied evenly. The FDA can either approve drugs if, and only if, they are “proven” efficacious — or it can do away with the efficacy requirement entirely. It can’t have its cake and eat it too.

    To selectively disregard the law so that the efficacy requirement applies to certain drugs, for some reason, but not to others… well… those are called “unprincipled exceptions.” And whereas doing away with the efficacy requirement entirely would be a very good thing for the drug industry and for patients, unprincipled exceptions are terrible for the drug industry and they could hardly be worse for patients.

    (Because of the dog-and-pony show some of them go through. The humiliation. The fact that insurance companies are going to think that people who ask for the drug are malingerers and neither need, nor should have, that prove-not-efficacious treatment. Because in a safety-only protocol, efficacy is still an open question — whereas in these cases, drug companies have spent vast sums of development money, which raises drug prices, simply to determine that efficacy doesn’t seem to be there at all. To say nothing of people who have diseases that didn’t qualify for unprincipled exceptions!)

    So we’re either all-in or all-out, and the latter seems preferable. Nobody in his right mind wants anything to do with the FDA’s present review process. Not just Guarente. See, for e.g., Tim Cook’s attitude, which is typical:

    ““Cook hints that Apple may have more plans for the health sphere […], but he doesn’t want the watch itself to become a regulated, government-licensed health product. “We don’t want to put the watch through the Food and Drug Administration (FDA) process. I wouldn’t mind putting something adjacent to the watch through it, but not the watch, because it would hold us back from innovating too much, the cycles are too long.”

    I think it’s time for a broad reconsideration of how things are done.

  12. steve says:

    I’ll just try one more time… Efficacy isn’t usually all or none and the decision is usually not as cut and dry as people are portraying in this thread. Nine times out of ten it’s a judgement call, which is why review panels often are split. There is nothing wrong with taking patient advocacy into account in making the decision on whether to allow a drug that is not either a home run or a complete bust on to the market. Take a look at the history of how the first HIV drugs were approved if you want a good example.

  13. Wallace Grommet says:

    Perhaps the FDA should require drug companies to prove their offerings exhibit no action, either therapeutic or deleterious? Homeopathy, for the win!

  14. Alan Goldhammer says:

    This is a drug that should have been put on ‘Treatment IND’ Let the company recoup manufacturing and distribution costs while the necessary data is collected to support approval but don’t let them make a profit. Approving it with a requirement to collect the necessary confirmatory data is poor policy. What happens if the data shows that it is in fact a placebo? Does any expect that Sarepta will refund the garnished profits back to the American healthcare system?

Leave a Reply

Your email address will not be published. Required fields are marked *

Time limit is exhausted. Please reload CAPTCHA.