Skip to main content

Clinical Trials

Gene Therapy for Duchenne

I have said unkind things about Sarepta’s drug for Duchenne muscular dystrophy, Exondys (eteplirsen). That’s because I did not think that there was enough information to approve it at the FDA, and I had trouble believing that its biochemical effects were enough to be meaningful in general. I have no reason to modify those opinions, but I do have reason to modify my opinions about Sarepta itself, because they have just presented startlingly good data for Duchenne MD via a completely different approach.

The key to helping Duchenne patients seems to be production of functional dystrophin protein, which they lack. Exondys was an attempt to do this by changing the way the protein was expressed, via a small molecule antisense oligo, restoring a functional reading frame and producing a less-ruined dystrophin protein. This new method, though, is out-and-out gene therapy, using an AAV vector to insert a new gene. The dystrophin gene itself is far too large a payload, as it turns out, so this uses a “micro-dystrophin”, a truncated form of the protein that should nonetheless be functional (or at least far more functional than the mutated forms that patients themselves produce).

The results in the initial dosing of three children appear to be dramatic. The levels of dystrophin protein increase strongly, and the levels of creatine kinase (an enzyme marker of the disease, involved in destruction of muscle tissue) drop strongly as well. It seems that the patients themselves are experiencing equally strong clinical benefit, if the video (at that first link) of one of them running up a flight of stairs is at all representative. No complete scientific presentation has been made, I need to add, but what we can see is very interesting indeed.

Sarepta is starting a 12-patient placebo-controlled crossover trial, which should help out evaluating this technique. Twelve patients is not a lot, but with an effect size this large, it could be convincing. Meanwhile, Pfizer started dosing their own mini-dystrophin gene therapy in April, and Solid Biosciences, after a clinical hold, is back dosing their version as well. Encouraging word has leaked out concerning a single patient on the latter therapy, for what that’s worth.

In general, though, there has never been a more encouraging time for Duchenne MD patients and their families. Here are three very strong shots at corrective therapy well into the clinic. The thing to hold the breath about, as usual, is safety. AAV is currently of huge interest as a method for gene therapy, and so far seems relatively safe, but the field is young. There’s also a potential problem of antibody formation: everyone’s immune system is different, and some patients may find their own attacking the viral capsid proteins that are delivering the gene. No one’s sure how large a problem this might be yet, or how it might vary across patient populations in clinical practice.

But these questions – efficacy and safety – are finally being sorted out in real patients, and we can hope for the best. Gene therapy has obviously – in theory – been the sovereign remedy for many genetic disorders ever since we understood how these worked in the first place (which would be 1949, and Linus Pauling’s demonstration of the molecular basis of sickle cell anemia). But turning that into reality has been a very long and very hard process – see Jesse Gelsinger for the first example, and not the last. Although Gelsinger’s case is definitely the clearest connection, there have been numerous deaths in subsequent gene therapy trials (although it’s also true that many of these patients have been very sick indeed, trial or not). If we’re closer to getting this to work reliably, the world will be a better place for a lot of patients who have nowhere else to turn (note Bluebird’s recent data on sickle cell and beta-thalassemia). We shall see.

18 comments on “Gene Therapy for Duchenne”

  1. loupgarous says:

    Since one in every 10,000 males gets Duchenne MD (total patient cohort in the US of roughly 11,900), eventually we’ll have studies with some predictive power on the various mini-dystrophin AAV trials. I’ve seen Latin Square studies used in Big Pharma with cohorts that small which were used in NDA submissions and post-marketing.

    Thanks for sharing the good news!

    1. loupgarous says:

      By “that small”, I meant “12-14 patients”.

    2. Josh Argall says:

      Duchenne muscular dystrophy affects approximately 1:5,000 male births, That figure is fairly consistent worldwide.

      1. loupgarous says:

        Thanks for the correction. If your references are more accurate than mine were (they doubtless are) then more statistically powerful clinical studies of treatments for Duchennes are even closer/more likely to happen.

  2. anon says:

    Not a biologist, so asking: For these X or Y linked or other genetic disorders, how long until they become for the large part a thing of the past? I imagine in the coming years it will become standard to have your DNA sequenced (let alone access to family tree data), at which point for Duchenne’s if you are a carrier female you would presumably attempt to do IVF and have embryos screened before implantation. I don’t know how often these types of mutations arise spontaneously, however. Also thinking of other disorders: CF, Huntington’s, etc.

    1. Andre Brandli says:

      Anon, you are right that in vitro fertilization (IVF) and preimplantation diagnostics (PID) can prevent parents who carry each a risk allele from obtaining a sick child carrying the hereditary disease. There are however two problems. First, most parents do not know that they are heterozygous carriers of a disease gene. For instance, I know a couple whose first child has cystic fibrosis (CF). Both parents carry the most prevalent deltaF508 mutation of the CFTR gene. There was no history in either family for CF. They were complete surprised that their first born son developed the disease. Fortunately, they were lucky with the second child, who is healthy. Secondly, many countries such as Germany and Switzerland do not allow IFV with PID to select embryos with a desired genotype. Particularly, the German public is sensitive to this issue given their history. However, things are changing. IFV with PID is in my opinion still the much safer approach than Crispr/Cas to ensure that disease-causing gene variants are not handed down to the next generation.

    2. MrRogers says:

      About 1/3 of DMD mutations are spontaneous, The gene is so large that it takes over a day for a transcript to be generated.

    3. Design Monkey says:

      Besides what Andre noted, main problem would be huge population of religious kooks and just generic stupids, who would resist such tests.

      1. loupgarous says:

        Apart from “religious kooks” and “generic stupids”, there are cultures with very small gene pools.

        I’m from south Louisiana, originally. The place is basically a lot of little towns surrounded by swamps and/or open water. Before roads were built between the towns, someone could live most of their life not travelling more than a dozen miles from home. Income was agrarian, work on plantations, small farms, fishing and hunting. This led to what the rest of the country would call “unwise marriages” with distant cousins marrying much more often than common elsewhere, because everyone was distant cousins. One reason I went to Louisiana State at Baton Rouge right out of high school was to date women I wasn’t even remotely related to.

        My wife went to school in a rural parish (=”county”) with kids from one family, each one of whose four sons had Duchenne’s. That parish has one medium-sized town surrounded by many smaller towns, with very little exogamy, so you see some rare genetic illnesses there. Not because of religious conviction or specific instances of stupidity (though objectively, it’d be wise for anyone from one of those towns to marry away from home).

        There’s another small town in Louisiana called Kaplan. Again, not much exogamy, and a tragically high incidence of what was locally called “lazy baby syndrome”. That turned out to be Tay-Sachs disease, because the acculturated Cajuns in that town had a much higher-than-usual amount of Ashkenazic DNA.

        Unless and until DNA testing becomes cheap, routine and socialized in advance of marriage everywhere, we’ll probably see Duchenne’s and other rare genetic illnesses well into the next century. Cultural patterns throughout the world have resulted in rural populations with small gene pools and not much travel, and not much exogamy, all over the world. You don’t have to be highly religious or notably stupid to run the risk of reinforcing some bad genes in those places.

        1. Design Monkey says:

          Isolated backward incestuous rural populations ain’t exactly known as freethinking centres. Quite typically they tend to be strongly religiously and “traditionally conservatively” brainwashed as well.

          1. Bastiat's Ghost says:

            Plenty of brainwasinging in smug major metropolitan centers and university liberal arts departments too these days, isn’t there Design Monkey? Gosh, I wonder what backwater hell holes have issues with broad based genetic testing? Let’s see, oh yes, Germany, Poland, and Japan…

            Thanks for being so arrogantly superior to the rest of the world, Design Monkey.

          2. loupgarous says:

            My point is that small gene pools exist all over, and it’s not intelligence (or the lack thereof) which causes genetically bad reproduction, but culture and population patterns.

            You can call a culture “dumb” all you like, it’ll persist. There are times when the dumb guys come in handy, because they don’t mind doing the hard and risky jobs.

            My particular group of dumb religious kooks gave the US two Marine Corps Commandants (Camp Lejeune’s named after one of them) and countless ordinary grunts, and supply the deep-sea drilling industry worldwide with technical experts in petroleum engineering and technology. They also produced one hell of a good surgical oncologist who did a tremendous job of unwinding an invasive paraganglioma from around my vagus nerve and stomach wall.

            We’re slow, but we get there.

          3. Hap says:

            Except everyone has genetic and intellectual vulnerabilities – whether they matter depends on circumstance. Diversity works because if you have enough people, someone else’s strengths under a set of circumstances will cancel out your vulnerabilities and vice versa – the system can survive under more kinds of conditions than otherwise.

            Stupidity should hurt, but I wonder that if that were the case how many people would actually be moving and conscious rather than on IV drips of painkillers.

    4. HFM says:

      There’s a push to do basic genetic screening on pretty much everyone, not sequencing necessarily, but a microarray-based test that will find the common diseases cheaply. IMO this will become normal over the next decade or so. The Saudis and the Aussies are already working to test everyone who gets a marriage license.

      Won’t prevent all cases, but could substantially reduce the incidence.

      1. loupgarous says:

        Iceland does that, too – one company has been contracted to do full genotyping on 2,636 Icelanders, and microarray testing of 100,000 others. The kewl part of doing this in Iceland is that almost everyone there can trace their ancestry back to a few common ancestors who settled the place in 9 AD – a very homogeneous population, and a genome with less “noise” so discrete genetic variations can be detected more easily.

        So far, they’ve found a new Alzheimer’s Disease gene, 10,000 discrete genetic sequences among the 100,000 folks who got microarray testing, and 8,000 people who have knockout mutations, which can be studied for their health impact in these folks. This is a case where a small, discrete gene pool is scientifically valuable.

  3. Andre says:

    Derek, thanks for drawing our attention to the new gene therapy approaches to treat DMD. Neutralizing antibodies (NABs) are indeed one of the key problems limiting the use AAV as a vector for transgene delivery in patients as discussed here: “Furthermore, in animals or humans lacking pre-existing NABs, the vector cannot be readministered (unless a different viral capsid is chosen) owing to potent NAB responses that occur within days after the initial exposure to vector, which can persist for months. The inability to readminister the vector because of potent NAB formation upon vector administration is a serious problem in the gene therapy field and has only partially been solved thus far by using alternate capsids, immune suppression and decoy capsid approaches… Typically, 20–40% of patients are excluded from enrollment in liver-directed gene therapy with a specific AAV serotype because of pre-existing NABs6.” These statements are taken from a letter to the editor in the current issue of Nature Medicine ( In other words, after one shot of AAV treatment, patients will develop a massive immune response. This will make subsequent treatments largely ineffective. At least, this is the current thinking in the field. Finally, there is the delivery problem. AAV infection is not uniform and will not reach very cell of a tissue. This is particularly challenging if you are targeting the skeletal musculature (the lung or brain in other applications to treat CF or AD) to gain a therapeutic benefit. Overall, I hope that the ongoing clinical trials will provide more insight and they may prove the theory to be wrong in practice, which would be to the benefit of the patients. However, you should not be surprised if the trials fail.

    1. PV=nRT says:

      There are a number of ways around this, including plasmaphoresis and IVIG from non nAb donors.

  4. Rainer says:

    How does a treatment for Duchenne relate to treatment for FSH muscular dystrophy?

Comments are closed.