Skip to main content

Clinical Trials

Nothing Works (Yet) Against Duchenne Muscular Dystrophy

Duchenne muscular dystrophy (DMD) has always been considered a hard disease to come up with a treatment for. And how we have plenty of evidence to back up that belief: it’s catastrophically hard. Over the last few years, we’ve seen rival therapies go through all kinds of wrenching twists and turns. Prosensa was developing drisapersen, for example, and having a hard time of it, until they were bought by Biomarin for $680 million dollars. That investment seemed rather hopeful (I thought the drug was done for), but as pointed out here, it now looks to be completely wiped out. Biomarin tried to do what Prosensa couldn’t (convince regulatory authorities that the drug actually worked), but this effort has failed, and more than once. Drisapersen missed its primary endpoint in its big clinical trial, and Biomarin’s reanalysis of the data, about which they thought enough of to spend over half a billion dollars, convinced neither the FDA nor the European regulatory authorities. There’s nowhere left to go.

PTC Therapeutics was also working in Duchenne, but they didn’t even get their candidate (atalurenthorough the front door for that indication. Update – it is approved in Europe, but not in the US, or at least not yet. (It’s under development now for cystic fibrosis). That leaves Sarepta and eteplirsen, which has recently been the subject of all kinds of widely covered FDA attention. The agency has delayed its decision, not that I blame them. This is one of those situations where a parade of parents and their critically ill children ask how you could possibly turn down the only drug that could help them – but at the same time, the evidence that the drug actually helps them is so thin as to be unusable.

In that last link, I presented four ways that the agency’s decision could be interpreted, and personally, I fall into the “should be turned down for lack of efficacy” camp. As it stands, we approve drugs in the US after showing safety and efficacy, and we balance those two in making a decision. If we decide to move to a system where we consider safety only, and approve things that have little to no evidence for any actual purpose, then we should be up front about that and change the mandate of the FDA to reflect that shift. Saying that “Oh, we’ll just let this one through” is the worst of both worlds. There are people who are willing to agree with this point but who also feel that eleplirsen has demonstrated enough efficacy to be approved, but I just don’t see how that argument holds up.

What would be so bad about moving to an “efficacy not required” regulatory regime? I think that it’s a flight from scientific evidence, which is the only thing we’ve got. Otherwise, everything starts to look like the “dietary supplement” industry, and what a mess that is. Here, you drop the efficacy requirement and I’ll develop grape juice for Dread Disease X. Not just plain grape juice – grape juice concentrate capsules. Mechanism? It’s the bioflavanoids. Probably. I think that they’re antioxidants, among other things. Lots of things. I can show safety in the clinic, too, so you have to approve my grape juice gel caps: I have a mechanism by which they might work (you can’t prove I’m wrong, can you?), I can show they’re safe, and you’ve eliminated the requirement that I prove that they actually do anything useful. Off to market! The patients who unfortunately suffer from Dread Disease X will, I’m sure, pay a lot for something that might help them. Don’t they have a right to try my antioxidants? There’s nothing else like them on the market, you know.

That’s the world that I worry about living in. And while it may be a reducio ad absurdum, I don’t think we need to reduce things very far to get to the absurdity. Unfortunately.

37 comments on “Nothing Works (Yet) Against Duchenne Muscular Dystrophy”

  1. Hap says:

    Paying for them would also be a problem – if we can’t or aren’t willing to pay for things we say we need that actually work, paying lots of money (and it will be lots of money, because like flies to honey people looking for it will come to sell distilled water at high prices) for things that don’t work is going to create problems. Either the government or insurance companies will have to pay for these things (or people, if insurance companies or the government decide that the FDA imprimatur is insufficient for a “safety only” world), and whoever pays will have a difficult time deciding what to pay for, and the money will have to come from somewhere.

  2. CFaulkner says:

    The grape juice hypothetical is hilarious and spot on.
    You also could have used any of the real-world examples and they would be equally demonstrative of your point. My favourite claim is from the multi-level-marketing brand Isagenix – “Product B: a powerful blend of complex botanicals and vitamins uniquely designed to offer superior telomere support for youthful aging.” (

  3. RandomGuy45 says:

    From what I understand Japan has an “efficacy not required” regulatory regime. Does anyone have any information about what the situation there is like?

    I mostly agree with what Derek is saying, but would it be a bad model if the FDA were to approve treatments based on safety as long as insurance companies will only pay if efficacy is demonstrated? I don’t see it being any different than the current situation with dietary supplements (and homeopathic remedies…ugh). If company X wants to pony up the expense for the clinical trials and reg-CMC work to get a treatment approved by the FDA knowing that insurers will not pay unless efficacy is demonstrated, that is a huge gamble on their part as they would be relying on a small subset of patients that could pay out of pocket to recoup their investment. It would however serve the very specific conditions for unmet medical needs such as what you’ve outlined for DMD patients – if they want and can afford to pay for a potential placebo and know that going in then I can’t really fault them for it, and the companies that have invested the time and money to make it available get to recoup at least a small portion of their investment.

    I’m sure there’s at least one hole in that logic that someone will point out, but it might not be as bad a compromise as it at first seems.

    1. alchemist says:

      Would you really leave to insurance companies the right to decide whether a compound is effective?

      1. Mike says:

        They already do it to a degree. Express Scripts has a long list of products that are excluded from their formulary. They’ve made the call that those products (at their given price) do not offer anything clinically over other products.

        And in addition, they also make clinical calls when it comes to step-edits. You need to take generic Lipitor before taking Crestor because the insurance companies made the call that Crestor offers nothing over Lipitor when judged the by the price difference.

      2. Carl 'SAI' Mitchell says:

        A better choice would be to allow selling compounds once safety is shown, but require marking like dietary supplements (not intended to cure, treat, or prevent any disease, etc) and don’t count them as drugs. Don’t require insurance companies to pay for them as though they are drugs, just as insurance companies don’t pay for dietary supplements (unless they like losing money). Still require prescriptions where necessary. Require informed consent from those taking the compounds, just like a phase 3 trial but without the monitoring/reporting. Prohibit marketing them as cures/treatments. Essentially, create a new class of compounds between supplements and drugs.

        1. Design Monkey says:

          > Prohibit marketing them as cures/treatments.

          Not really possible. Various crap “supplements” already are being subtly and not so subtly advertised as “cures”. And even blatant murder cases like ones done by baking soda charlatan simoncini, dont’t get any significant penalties.

          Also such “not-an-official-drug-because-flunked-efficiacy-test” products anyway will have to be sold at approximately drug prices (otherwise company would not get any profit, especially with fancy, but not working antibodies and other biologicals), and that alone will sort of signal “this shtuff is real drug, which evil FDA is supressing” to a certain category of not so bright population.

    2. Emjeff says:

      Totally agree. The vast majority of FDA medical officers have little clinical experience in the areas where they review, and can have some beliefs that go against established medical beliefs by experts. The payouts want efficacy data and they’ll get it- no company will skimp on that. But, we would be able to design studies with experts without having to compromise based on sometimes un-informed FDA input.

    3. Peter S. Shenkin says:

      I don’t think it’s proper for an insurance company or a national health plan to pay out if there is no convincing demonstration of efficacy. In the realm of private insurance, as in the U.S., you are asking me to pay for someone else’s treatment that has not been shown to be effective. And if a drug company really has paid the big bucks to demonstrate efficacy but has merely been able to demonstrate safety, how can they charge enough to materially affect their bottom line? Insurance pockets are deeper than private patients’ pockets, which is why we have insurance for somewhat rare conditions. And as far as the safety is concerned, you have to regard a verdict of safety as provisional for any newly released drug, particularly for a long-term treatment. It’s not terribly uncommon for safety problems to become visible only in the course of long-term use by a large number of patients, and bearing the responsibility of such a determination would add to the financial burden of the drug company; it does so already, but this would be onerous in view of the limited revenue likely available.

    4. Walther White says:

      In Japan, efficacy is still a requirement. They would even prefer that you prove efficacy within the Japanese population ( there are some drugs that work well for Asians but not Caucasians, and vice-versa, due to small genetic differences). Safety is also a consideration, and new drugs launched in Japan are restricted to 2-week prescriptions (Ryotan restriction) unless there is considerable historical pharmacovigilance data to prove safety.

      Safety-based only approval is completely idiotic and absurd. I can prescribe you a sugar pill for anything then because it’s safe. It doesn’t need to cure you.

  4. bobo says:

    PTCT did get “conditional approval” for ataluren by EMA in 2014. Maybe they will lose their marketing authorization though.

  5. Bernard Munos says:

    In the 1990s, Henry Grabowski and Glenn Lacey at Duke published interesting research showing that the US fostered world-leading innovation among US firms by enforcing demanding standards of efficacy. Drugs that met these standards often became global leaders because they were simply the best. In comparison, countries such as Japan or France, where the regulatory focus is rather on safety, encouraged the production of drugs that could achieve notoriety in their local markets, but often fail to get traction abroad. By relaxing the efficacy requirement, the FDA would seriously harm drug innovation and the competitiveness of the US drug industry.

    The other thing to consider is that safe but useless drugs would not be priced at dietary supplement levels, but would likely fetch tens of thousands of dollars. And society, through our insurance system, would be on the hook. Think what would happen to Alzheimer’s, where plenty of safe drugs have proved to be totally useless. Approving them under the right-to-try argument would create a gold rush for firms that have failed to innovate. Is that what we want to reward? The incentive to innovate would disappear. Why go through the long and expensive slog when one can just push some safe stuff through FDA, and market the heck out of it. If we think Valeant was bad, this would create clones of it by the dozen. The right-to-try would quickly turn into the right-to-fleece.

    1. Design Monkey says:

      The other side of that coin is that there is a significant number of pretty good drugs in Europe, which never will be available through official channels to US patients. Because patents are through or ending, and wasting a zillion for getting FDA approval from scratch just for generic doesn’t pay.

  6. Erebus says:

    Nice strawman argument, Derek.

    The truth is that FDA reform is desperately needed. Consider this article:
    …And the paper it links to by Montazerhodjat and Lo.

    An “efficacy not required” regimen can speed access of drugs to market, and, just as importantly, should dramatically reduce their price-tags. Postmarketing drug surveillance can determine whether or not these drugs are effective, whether or not insurance or governments should pay (and of course insurance companies and government payers should have their own standards for efficacy,) and let’s not forget that professional organizations and physicians can also play a key role in determining which prescriptions ought to be written.

    …Besides, the US Pharmaceutical industry wasn’t that bad before 1962, was it? There was no efficacy requirement then. There was only the requirement that potential drugs be safe.

    1. Snake Oil Crusader says:

      Erebus, I fail to see a strawman argument here. And if you think that the field of medicine (or at least the regulation of drug discovery) was better for society before 1962, you must have a lot of snake oil to sell. There is no perfect strategy to gauge efficacy, but I severely doubt the free markets are capable of fully making those decisions in order to maximize societal benefit. The fact that billions would be spent on drugs that DON’T work doesn’t show up on a balance sheet. The opportunity costs for people buying useless “natural” remedies is already huge.

    2. False Dichotomy says:

      Speaking of strawman arguments, pre-1962 drugs that actually worked (arsphenamine, sulfa drugs, penicillin, chloramphenicol etc) were only found because the first steps in an open field are like shooting fish in a barrel. Once all these low hanging fruits, antibiotics were the phenotypic readout is the patient lives instead of dying (pretty high s/n, if you ask me), were harvested we needed well-controlled trials to even identify whether drugs work at all (the MRC Tuberculosis trial comes to mind as an early example here).

      Pre-Kefauver-Harris history shows that many doctors (both M.D. and Ph.D.) were and continue to be biased idiots when it comes to statistics; if we leave the choice with physicians we will almost certainly see the return of statistically-unpowered “ancedote-based medicine”

    3. CP says:

      Leonard Guarente, is that you?

    4. matt says:

      Among other problems with the paper, it seems to assume the prior probability that a treatment will be effective or ineffective is equal = 0.5 each. This seems incorrect. In fact, we know that most treatments (the vast, vast majority if you go back into animal trials) are ineffective. Maybe you should go back and recalculate with the priors set to 0.9 for the null hypothesis (that the treatment in question is ineffective), and 0.1 for the alternative H1 hypothesis. Or, perhaps you feel that is too conservative, because the FDA has been using these overly strict criteria by your academic consideration: then 0.8 null, 0.2 alternative. It’s extremely unlikely that more than 2x the existing approval rate is desirable. (To their credit, they note this objection, and state they plan to revisit this work with different priors.)

      Furthermore, you have distorted what this paper says to fit your pre-existing stance on the issue. Nowhere do these authors advocate for ignoring efficacy; in fact, the entire thrust of their work is in refining the statistical test for efficacy. And, in fact, they show even with the over-optimistic priors, the FDA might actually want to be MORE stringent on efficacy than it is now on milder severity diseases.

      Yes, if the FDA dropped any efficacy requirement, that would certainly make drug development cost less, but those “postmarketing surveillance” studies are 1. also going to be costly because 2. they are statistically problematic unless you can demonstrate causality, which requires a prospective randomized controlled study, and 3. those costs will be born by the taxpayer and customer rather than by the snake-oil^H^Hahem drug company. So what looked like cheaper prices really just shifted the cost of determining effectiveness, and dumped it on taxpayers and consumers, while freeing drug companies to crank as many possible candidates into the market as possible, further driving up the costs of even problematic surveillance anecdata.

      That takes our current system, where drug companies bear the cost of proving their candidate’s efficacy and receive the benefit of a short term window of high profit if they succeed AND the benefit of the public knowing the FDA has made them jump a pretty high bar, and substitutes instead a system where drug companies profit from shoving as many candidates through a low-cost approval system, and the rest of society struggles to figure out what’s working, and insurance companies (who don’t really care about determining effectiveness but only in maximizing profits and reducing costs) aren’t going to magically pick up the slack. No, thanks.

      In fact, prior experience suggests the overwhelmingly probable outcome is that no adequate efficacy data will ever be generated for most products, and the market will be utterly unable to sort out the competing claims, and legitimate drug companies will be driven out of the market because it is not economic to do the hard work of generating a real drug when the income is the same for a fake but the costs are much less. No, thanks.

      1. Imaging guy says:

        Actually a group of famous clinical trialists have analyzed more than one thousand published and unpublished Phase III randomized clinical trials and found that in about 50% of the time new treatments are better than standard ones. So there is genuine “clinical equipoise” in phase III trials and the prior probability of any drug (being effective or ineffective) before phase III is 0.5. Actually if the prior probability of a drug (being ineffective) before phase III trial were 80%, it would be unethical to conduct that trial.

        ” Medical research: Trial unpredictability yields predictable therapy gains” (Nature. 2013 Aug 22;500(7463):395-6) (PMID: 23969443)

        1. matt says:

          Okay, fair enough, for Phase III trials the 0.5 prior might be reasonable. Or, perhaps you would adjust it according to a running average of effective rates, so oncology according to the more recent post would run a good bit less than 0.5, others might run a bit more.

          I disagree that a prior of less than 0.5 would make the trial unethical, but I suspect we will just have different opinions on this. I think doctors and patients both recognize that even long-ish shots may be reasonable and ethical, given full disclosure and the hope of future medical breakthroughs.

      2. Erebus says:

        -Post-market data analysis can assess benefits and harms across an extremely large population, can be performed relatively cheaply and quickly, and can even uncover relevant effects (e.g. drug interactions) that the FDA might miss. This sort of thing is routinely done right now. It’s not quite that difficult, and it’s really not a bad option.

        -Drug safety testing is amenable to standardization. The agency could use something along the lines of a point system, which would apply to every drug candidate without exception. (Side effects graded by likelihood and severity, and then assigned a certain number of points, with “point allowances” for different indications — for instance, 150 for the treatment of mild anxiety, 2000 for aggressive cancer treatments and “last line of defense” antibiotics.)
        …This sort of thing can be easy, transparent, reproducible, and fair.

        -Testing efficacy is impossible to standardize, it is never straightforward, and the rules vary wildly in each individual case, which makes the FDA look inept and mercurial. Certain indications are de facto prioritized over others — for instance, efficacy trials for acute disease treatments cost vastly less than the long and expensive trials demanded of Alzheimer’s treatments — and nobody has any idea at all how to run meaningful efficacy trials for anti-aging treatments. This deeply perverts incentives.

        -In attempting to rigorously police drug efficacy, the FDA has let the costs related to drug approval spiral out of control. As per Hansen and Chien — writing in Issues in Pharmaceutical Economics, Lexington Books; 1979 — the average cost of releasing a drug between 1963 to 1975 was $119M. (All figures adjusted for inflation.) A different study (DiMasi et al., 1991,) which examined the period from 1970-1982 found that costs nearly doubled over that period to an average of $231M. What are we at now? 2B or thereabouts? Where are we going to be in ten years? 4B?

        -The extremely high costs related to regulatory approval stifle innovation in the industry, as nobody sane should want to pay a fortune to wrestle with FDA bureaucrats for 12 years when there are other options. (See: Guarente.) It also makes it so that only entities like Merck and Pfizer have the resources to carry drugs from pre-clinical to approval. It may be illustrative to mention that the modern strategy among pharmaceutical start-ups is to get bought-out by, or partner with, a larger company after (or even before) Phase I trials.

        -For all that expense, we still don’t have a rational and impartial arbiter of drug efficacy. To the contrary, the FDA is openly corrupt, and transparently buckles under pressure from lobbying groups and the media. See: Flibanserin. Not only does it do a poor (and inordinately expensive!) job, it doesn’t even know what its own standards ought to be.

        The entire pharmaceutical business model is shot all to hell, and patients are suffering for it. It’ll only recover with FDA reform. The question is “how?”, and I think that the easiest way is to remove drug efficacy testing requirements. We’ll get more innovation, a faster path for drugs to market, and a more competitive industry.

        …This is not to say that efficacy testing will never be done, or that the industry will devolve into snake-oil peddling overnight. Let’s not forget that the golden age of drug development — the age when George Merck graced the cover of Time Magazine, and tremendous progress was made in antibiotics, steroids, vaccines, cancer chemotherapy, etc. — was before drug efficacy became part of the FDA’s mission. Yeah, a lot of that was low-hanging fruit. And, yeah, drug companies today aren’t what they used to be. But we can’t ignore the fact that the FDA has done a catastrophically poor job, and is that the best way to revitalize and improve the industry would be to start with meaningful FDA reform.

        1. matt says:

          So you’ve ditched the Trojan-horse paper studying how to adjust the FDA’s statistical bar for efficacy, and just doubling down on your no-efficacy position? I see.

          If post-market data analysis were so cheap and easy and widely done as you suggest, then it should be trivially easy for pharmaceutical companies to assemble a complete efficacy case for off-label indications. Anti-depressants, for example, being prescribed about 50% of the time for off-label indications: we should have rock-solid data on exactly how effective they are for insomnia, rock-solid data for agitation, etc. But instead, pharmaceutical companies do not touch the indications on the label, and all we have are marketing-driven “suggestions of possible use” handed out at dinners paid for by pharma companies by doctors also paid by pharma companies. Where is all that cheap and easy data? It’s opaque, slanted, rife with hidden conflicts of interest, and usually statistically inconclusive.

          –You blame the FDA for the high costs of getting approval, but just automatically approving ineffective drugs as long as they are safe doesn’t magically increase the rate of effective drugs. THAT’s the real problem, and you are shooting the messenger (the FDA) who has to bring the bad news that once again an NDA has proven to be ineffective.

          –That “cheap and easy” post-market data analysis is based on years of people paying full price for ineffective medicines. Now, you will think those medicines will get cheaper now that you’ve eliminated those mean old efficacy considerations in trial design, but there will inevitably be far, far more ineffective medicines out there, because it is infinitely easier to crank out a safe ineffective medicine than to make an effective one. (Again, this is because you haven’t solved the problem of making an effective medicine, you’ve just solved the problem of ineffective medicines being blocked from the market.)

          I’ve got a homeopathic Alzheimer’s cure in mind that will make me billions and would take you many, many years of data analysis before you assemble data to show it ineffective, if I didn’t accidentally lose some patient data before then. Even better, YOU or taxpayers have to pay for the studies to prove me wrong, and if/when you do, I’ll take that treatment off the market, rename that company or start another, pick another homeopathic remedy and start the whole process once again. I can do this forever.

          Lane Simonian will be doing the same thing, only with essential oils, perhaps with a lot more sincerity than I could muster, and a well-honed message about peroxynitrites to sell his remedy. Dr. Oz will be hosting dozens of others doing the same thing: green coffee beans? acai berries? They are probably just the thing to cure Alzheimers now that the evil FDA has been taken off their back.

          The only people trying to make real cures will find it extremely hard to get financing, because frankly it’s a waste of money to pay chemists and biologists when cheaper “cures” are so readily available.

          –How exactly do you or doctors manage the ethics of cost/benefit when you don’t know the benefit yet? Few oncology drugs really qualify “safe” or “non-toxic”–they exist to kill human cells, after all. How do you handle safety?

          –Do you tell patients all new medicines are now drug trials, where they have to pay full price in order for somebody to study whether it helped them, to generate data for a few years so future people will know whether it works better? Sucks if you have cancer, because we may have just blown your treatment opportunity on something ineffective, while your cancer was metastasizing and spreading out.

          –Do you tell insurance companies they have to support paying for more-likely-than-not crap, in the hopes that some data is generated, but in the meantime they may have to pay for 2, 3, 4 different sets of treatment in order to find one that works?

          –I think your charges that the FDA is corrupt are false and defamatory and unsupported. No doubt they do occasionally cave to pressure, but from all I’ve seen, they do a pretty good job of walking the tightwire of too hard and too easy. Good evidence for this is available in the form of harsh criticism from both sides of the spectrum.

          1. Erebus says:

            >”You blame the FDA for the high costs of getting approval, but just automatically approving ineffective drugs as long as they are safe doesn’t magically increase the rate of effective drugs.”

            It would certainly. Reduced development costs would lead to many more firms actively developing drugs. The history of the industry strongly suggests that they won’t all be snake oil salesmen. Sure, we’d get some drugs that don’t work very well, but it’s not like we don’t get those today.

            >”That “cheap and easy” post-market data analysis is based on years of people paying full price for ineffective medicines.”

            Which people will gladly do. See the controversy surrounding “right to try,” see patient statements at the FDA’s Ataluren hearing, see the major lobbying efforts that forced fibanserin through the regulatory process, and see how cancer patients clamor to get into clinical trials. In the latter case, in particular, it is borderline unethical to not allow these people access to medicines which might conceivably help them.

            The continuation of your argument on this point is a restatement of Derek’s initial “reductio ad absurdum“: Everybody but the Top Men at the FDA are idiots, will be hoodwinked into buying essential oils for Alzheimer’s, and things must have been terrible before the FDA started rigorously policing efficacy. (Which sent the price of drugs skyrocketing into the orbit.) It’s just not a very good argument.

            I’m not saying that things will be perfect should the FDA drop that role — but they will be better than they are now. We’ll have:
            (A) Faster access to drugs.
            (B) More competition between firms, more firms actively developing drugs, a healthier pharmaceutical business model. (Aside: This would also solve the patent issue. Right now, with a ridiculous 12-year regulatory process, the 20-year patent lifespan is damned tight.)
            (C) Cheaper drugs.

            Besides, the status quo isn’t doing much for Alzheimer’s right now, is it? As far as I can tell, it’s merely bankrupting the few firms which attempt to tackle the indication.

            >”How exactly do you or doctors manage the ethics of cost/benefit when you don’t know the benefit yet? Few oncology drugs really qualify “safe” or “non-toxic”–they exist to kill human cells, after all. How do you handle safety?”

            How did they do it in the past? I realize that it was sometimes a bumpy road — especially in the early days of chemotherapy — but it’s certainly not intractable. There’s no reason to assume that the medical profession has collectively gotten much stupider since 1962, so we should be fine. Professional organizations and physicians should be able to step up to the plate.

            >”Do you tell patients all new medicines are now drug trials, where they have to pay full price in order for somebody to study whether it helped them, to generate data for a few years so future people will know whether it works better? Sucks if you have cancer, because we may have just blown your treatment opportunity on something ineffective, while your cancer was metastasizing and spreading out.”

            Given how many cancer patients beg to be let into clinical trials — wherein they might receive placebo — this is a ridiculous argument. If you’re facing down a disease that’s likely to be fatal, you’ll grasp at anything which might potentially improve your odds of survival. A faster drug development cycle, and earlier access to drugs, would improve everybody’s odds.

            Besides, in diseases like cancer, professional organizations set prescribing guidelines for physicians. These move rather slowly, and are evidence-based. New drugs of unproven or dubious efficacy are, initially, only likely to be used in cases where the standard treatments, as per the guidelines, are insufficient.

            >”Do you tell insurance companies they have to support paying for more-likely-than-not crap, in the hopes that some data is generated, but in the meantime they may have to pay for 2, 3, 4 different sets of treatment in order to find one that works?”

            This has been covered elsewhere on this comments thread. Suffice it to say that insurance companies don’t “have” to pay for anything. It’ll go on a case-by-case, drug-by-drug basis… as it does now.
            Cheaper development costs should help people who are inclined to pay out of pocket.

            >”I think your charges that the FDA is corrupt are false and defamatory and unsupported. No doubt they do occasionally cave to pressure, but from all I’ve seen, they do a pretty good job of walking the tightwire of too hard and too easy. Good evidence for this is available in the form of harsh criticism from both sides of the spectrum.”

            That’s not good evidence. Good evidence of the contrary can be found in the FDA buckling to pressure from lobbying groups, abandoning the pretense of being an impartial arbiter of efficacy. There’s plenty of that to go around. Flibanserin is the most recent, and most egregious, case. This DMD story might be more of the same.

  7. CMCguy says:

    I do not recall when FDA changes began mandated demonstrations of efficacy rather than just safety (thinking 70s or 80s?) but believe there was a long history of new and useful drugs developed under that very concept. Of course that also was when Pharma was more interested in conducting actual R&D to find and provide drugs to aid patients rather than marketing or profit focused centers therefore IMO inherently was greater awareness that things not only had to be safe but must work to benefit the end-user effectively as a non-reg criteria. There was also more 2nd/3rd generations that built on what was learned to create improved performance that appears less of a mindset today. I do not advocate going back to that one-sided approach however in many cases the ability to reach the currently defined scientific evidence thresholds seems now so high (driving big studies/huge costs to test) I can see a few drugs that might be useful clinically, albeit for limited sub-sets, can not meet today’s rigid statistical standards so the balance between safety and efficacy could be adjusted. This is not black and white where diseases like DMD do warrant more tolerance although knowing what that means at present with answering approvability of drugs in development is tough with limited info on potential benefits that may have been observed during trials.

  8. a. nonymaus says:

    Off-topic, but what happened to comments on the chromium post? Threats from Big Chromium (great robot name, btw), perhaps?

    1. ba-dum-rimshot says:

      What a coincidence! ‘Big Chromium’ is the name of my Metallica tribute band!

  9. Anon says:

    I have no problem with the FDA approving drugs based on safety alone without efficacy – provided that the product claim says something like “safe but totally useless”. It’s about transparency and integrity more than safety and efficacy, then people, payers and providers can choose whether they want to pay in a free market. No ethical issues with that.

  10. Wallace Grommet says:

    Erebus proudly advances the idiot argument while confusing Derek’s sound real world hypothetical scenario with a straight

  11. matt says:

    Returning to a topic that comes up over and over again, the part of the FDA that needs a very judicious and limited tweek is in post-approval marketing studies, and in demonstrating efficacy for widening conditions. In essence, a regulatory way to systematically (and statistically rigorously) condense what is learned by off-label prescriptions into a wider label for a drug. It is occasionally done now, but mostly ignored because it can be and it’s more profitable to do so.

    Once again, I think drug companies would have some incentive to jump through a few hoops for wider marketing approval, if Congress makes it clear those hoops are necessary, and the FDA can appropriately come down extremely hard on off-label marketing.

    The necessary hoops would be prior registry of all drug trials plus mandatory listing of outcomes, and a similar plan of action for approval as an NDA, and appropriate safeguards where the drug companies are not allowed to use or distribute information from trials outside this process. The drug companies would be allowed to sponsor trials or use publicly funded trials, and would be allowed to encourage doctors considering off-label prescriptions to join one of these randomized controlled trials, even allowed to market those trials to doctors or at professional meetings.

    As part of “coming down hard on off-label marketing”, I wonder if Congress could institute penalties for drug companies deriving income in excess of the approved indication market, or perhaps instead of penalties require the drug company sell the drug at nearly zero profit above a certain threshold. (I.e., there are 10,000 patients in your target indication, above about 11,000 patients worth of prescriptions dispensed any additional sales must be at cost, maybe plus 5%.) That would cut down on their eagerness to support off-label prescriptions without widening the indication on the label, and instantly revolutionize their eagerness to run trials for wider indications.

    And yes, that would be an enormous blow to psychiatric (eg depression) drugs…that swamp probably needs to be drained, or hopefully replaced with some decent statistical data.

    Maybe, too, doctors could be required to indicate whether their prescription was on-label or off-label, and if off-label, for what indication. THAT would ruffle some feathers. But the sad truth is, many doctors have abused the leeway they’ve been given to treat their patients, and it is appropriate for the public to demand at least some statistics on the depth of the problem. (I’ve had contact with many doctors who would strongly assert that the free meals and handouts and giveaways and friendly visits by beautiful people haven’t clouded their judgment in the least, while prescribing exactly those medicines marketed by said goodies.)

  12. Pennpenn says:

    Wait, there are places where demonstrating efficacy *isn’t* a pre-requisite for a drug being approved? I mean, I know that kind of crap flies in the supplement industry (actually, it doesn’t just fly, it keeps the whole awful garbage-heap afloat), but in my opinion the only appropriate response to someone suggesting a system of drug approval that doesn’t include at least some efficacy measures is a clinically measured smack to the back of the head.

  13. a. nonymaus says:

    In this specific condition, it seems like prevention is better than treatment, and prevention is a comparatively simple matter of prenatal testing and then keeping abortion safe, legal, and available. We have known this for some time now; why are there fetuses coming to term with a diagnosable and incurable condition in this day and age?

    1. matt says:

      Pre-natal tests are not risk-free. The risk of DMD in the regular population is probably not high enough to outweigh the risk of the test. It would be worthwhile if the mother knew she was a carrier, but most women carrying the recessive gene are unaware. Don’t know if these mothers were tested, or knew they were carriers.

      What I think you are wanting to rail against…is certain mothers would not choose to abort the fetus even if they knew it had Duchennes. And I’ll venture to say, how do you know that’s the wrong decision? I’ll bet none of the mothers at the FDA meeting for eteplirsen regrets the life of her son. They desperately want a cure because they love their sons.

      I know from personal experience that Down’s Syndrome kids offer very powerful and valuable insights on the human condition. The parents, though of course they would deeply wish a full and independent life for their kids, and it is enormously difficult and of course they would “treat” if any were available, have expressed they wouldn’t trade that experience and the uniqueness of their child’s personality for anything. I’m not sure that aborting all the kids with burdensome conditions makes the world a better place.

  14. Paul says:

    “and Biomarin’s reanalysis of the data, about which they thought enough of to spend over half a billion dollars”
    I’ve been in pharma for 15+years, so I know how we spend money like its going out of fashion, but how does the re-analysis of a finished study cost half a billion dollars ? Thats like 5,000 statistician years (I just made that unit up).

    1. george says:

      They bought a company based on the analysis and the company cost them half a billion, not the analysis.

  15. John Thacker says:

    I understand your argument, but life and policy is about tradeoffs. My review of the literature suggests that the FDA shift to a safety and efficacy regime instead of just safety has cost lives on net. While certainly ineffective drugs are a risk and some would get through, the significantly higher regulatory burden does make a lot of even efficacious drugs not worth researching or getting approved.

    It is particularly notable in the case of diseases with a small number of sufferers, or the case of communicable diseases common in other countries, where approval becomes not worth the money for effective drugs mostly used by travelers. The worst off may be Puerto Ricans, who lack access to drugs for diseases endemic to the Caribbean.

    I understand the complaint that the decline in drug discovery may simply be a long term secular decline, but the literature shows a pretty strong step effect with the Kefauver Amendments and others. Your concerns are very understandable but I believe that the science is strongly against you.

  16. DrSAR says:

    I don’t understand how drug approval on safety alone could work. Cisplatin is a bloody toxin and not safe! As are most other useful drugs. A statement about safety is only sensible relative to a benefit. Once you need to prove that you’re back to broad regulatory mandate we have now. Am I missing something about how this safety-based-only approval would work?

Comments are closed.