Skip to Content

Big Trouble For Zafgen and Beloranib

Zafgen went through a very rough patch back in October when it turned out that a patient in their key Phase III trial in Prader-Willi syndrome died. P-W patients have severe health problems and shortened lifespan, but the question was naturally whether the company’s investigational drug beloranib was a factor. The seemingly slow disclosure of the problem didn’t help much, either.

Well, now they have problems that make those look small: another patient in the trial has died. It’s a bit early to call it, but this could be it for beloranib, and if so, this could be it for Zafgen. We’ll have to wait for more details, but the odds have lengthened, and this is an existential threat to both the drug program and to the company. I’ve followed the progress of this compound from its early days on this blog (go to Google and search “ zafgen” to see the posts), because of its unusual structure. The story is an instructive one, in the “you never know until the end” mode. Nothing is for sure in an investigational drug program, and nothing is in the clear until the last patient has completed the last course of treatment. Sometimes not even then.

17 comments on “Big Trouble For Zafgen and Beloranib”

  1. Anchor says:

    As a medicinal chemist we have been warned to keep away from “epoxides.” as these are only an intermediate. Beloranib, am afraid is bisepoxide and my previous company would have asked me to “sprint” away from the hideous looking molecule. If the site of action is in liver it just begs a question what is going on in there? Rush to clinical trial?

    1. Philip says:

      Isn’t how we got to this deadened innovation in med. chem.? Fearing certain functional groups/molecular weights/numbers of lewis basic sites; just making flat aromatic things with suzuki reactions; listening to what your manager says and not stepping outside the box? Have we forgotten that so much innovation has come from mistakes in the lab? If you want to be an engineer, make compounds that look like everyone else’s. I think we should applaud Zafgen. Regardless, the two causes of death in the study were blood clotting, not liver failure.

      1. Curt F. says:

        If you want to be an engineer

        I don’t know enough about medicinal chemistry to comment on the substance of your post. But this seems like a pretty gratuitous and unjustified swipe and engineers and engineering. Can you actually point to any self-identified engineers who are trying to “make compounds that look like everyone else’s”?

        1. DN says:

          If anything, engineers care for results that are approximately good enough. In fact we seem to delight in putting amazing bodges and shortcuts into mass production.

  2. Anon says:

    Counterpoint: carfilzomib

    1. Andy II says:

      Well, oncology drugs could be some exceptions as carfilzomib is used after two therapies failed in multiple myeloma patients. It is a derivative of a natural product, epoxomicin as you know. In Wikipedia it says: Injected, epoxomicin can induce Parkinson’s-like symptoms in rats. Pladienolide B and D, FD-895 are also natural products with epoxide and are inhibitors of spliceosome.

    2. Design Monkey says:

      In oncology anything goes, that will never be allowed in other therapeutic areas. That’s not too significant.

      A more funny case is natural constituents of OTC valerian extract.

      Valepotriates. Epoxide. Oopsss.

  3. Matthew says:

    Tiotropium is a blockbuster drug – and an epoxide too. The wise medicinal chemist is aware of the flags but preared to break the “rules” intelligently.

    1. Lars says:

      In the case of tiotropium, the reason it clear; it’s based on scopolamine/scopine which has one as well. Wikipedia has a synth scheme for it.

  4. JIA says:

    @Philip: Regardless, the two causes of death in the study were blood clotting, not liver failure.

    Actually several critical coagulation proteins are made in the liver, such as FVIII and FIX. Elevated levels of FIX are a known cause of thrombotic events. (PubMed link in my handle)

    I’m not suggesting a mechanism by which this drug would elevate FIX levels, I’m just making the point that “clotting” as a cause of death doesn’t automatically rule out causation by this drug, or any liver targeted drug.

  5. Morten G says:

    These math problems are getting harder.

    Anyway, 1 in 200 chance of death from gastric bypass operation or complications thereof. So is the drug 1) as effective as or better than bypass? 2) does the drug kill with a greater frequency than the operation? 3) is the drug cheaper than the operation?

  6. SRB says:

    I am not a chemist but a clinical development guy. I appreciate this blog and discussion which I only just found because of revisiting the Zafgen story. Thank goodness for medicinal chemists. I followed Zafgen closely 4-5 years ago during the early clinical development planning. I remember that a top science exec at a major pharma told a recommending team that it looked like rat poison. I thought it was a glib response although at the time the early one month clinical study was too small to say anything about actual weight loss. Somehow i remember this program being worth a shot based on the lack of toxicity so far, based on Tom Hughes thinking about potential mechanism and the key is that they chose a viable development plan focusing on an ultra-orphan indication in Prader-Willi Syndrome. The angst of parents in this ultra-orphan disease means they are organized worldwide to advocate for therapy, any therapy, to help their children with extreme morbid hyperphagia and obesity, which they die from.

    i didn’t realize but it appears that a major cause of death is pulmonary emboli. It looks like the reason even specialty clinical texts don’t highlight this enough is due to the sporadic disjointed reporting from case reports in a rare disease. The recent focus as a high risk obesity model for clinical development and in particular the Zafgen study, may be making clear something that was previously under appreciated. And this possibility is balanced against the sober issue of two deaths. and none can know for sure.
    But a major vulnerability of this kind of development program is that they offer an open-label extension so all participants feel they have the chance to get the experimental drug. Is that entirely ethical? not for me to say but clearly the participants or families are so interested in treatment that they see it as positive.
    So the real questions in my mind now are – is the incidence of pulmonary embolism so high that even 2 deaths in a presumably very sick older prader-willi population will turn out to be actually not surprising. And will the double-blind results actually show an efficacy that will allow FDA to consider the risk-benefit. This is not garden variety obesity at this point so assuming drug prescribing could be controlled (probably mostly by price by default) it is possible that the eventual risk-benefit would allow or even favor approval. On the other hand Zafgen may not make it that far and there may not be a good way to conclude the issue right now. At the very least it is going to depend on the phase 3 results. So its remains a very interesting program indeed, more so than one might expect for a rat poison.

    1. Terry says:

      I appreciate your remarks.
      An additional thought: why did Zafgen choose the P-W population to experiment run the trial with anyway? Profoundly handicapped both physically and in mental development; very high mortality rate – deaths during or soon after the trial were a distinct possibility. And it may not be possible to sort out the causes of death. Or the drug may have caused a chain reaction resulting in death in a way that would simply never happen in a relatively healthy functioning population. Given the success of the drug in previous trials, this all seems more a case of poor design concept

      1. Mel says:

        Terry, just to clear up – PWS has a huge spectrum. Sufferers of it are generally certainly not ‘profoundly handicapped’ – it is true most suffer from some sort of learning disability, but this can range from very mild to more severe. Likewise with the mortality rate – and surely a very high mortality rate is true of us all wouldn’t you say…?!

  7. ROGI says:

    “… At the very least it is going to depend on the phase 3 results. So its remains a very interesting program indeed, more so than one might expect for a rat poison.”

    This is a Phase 3 program.

  8. Srb says:

    Yes this is a small phase 3 based on an orphan indication with the ph 3 topline results due 1q16 but now due to clinical hold the concern is whether the study winds up underpowered. I think everyone can access this link that explains situation from the investor perspective but it seems good
    Best and thanks

Comments are closed.