Skip to Content

Business and Markets

Some Side Effect For An Antibody

Remember back when AstraZeneca was fighting off Pfizer’s ardent, tax-issue-resolving embrace a year ago? One of their weapons was a pitch to their own shareholders about what potential their own pipeline had, and how much of that would presumably go to waste should the deal go through. Even at the time, people thought that their estimates of what was to come might be a bit optimistic. But I can’t really fault them, because if someone were trying to buy me, I’d probably be willing to say all kinds of things to keep it from happening, too.
Well, one of those pipeline assets has just taken a major hit. Brodulamab, targeted against the IL-17 receptor, was part of a 2012 deal between AstraZeneca and Amgen to develop inflammation therapies. Late last November, the companies announced some good clinical results in psoriasis.
But now Amgen has dropped the project, and hard.

The decision was based on events of suicidal ideation and behavior in the brodalumab program, which Amgen believes likely would necessitate restrictive labeling.
“During our preparation process for regulatory submissions, we came to believe that labeling requirements likely would limit the appropriate patient population for brodalumab,” said Sean E. Harper, M.D., executive vice president of Research and Development at Amgen.

That really would be a show-stopper – psoriasis is a cruel disease, but suicide is worse. It’s surprising, though, that an antibody would have this as a side effect (I’ll bet it was surprising to Amgen and AZ, for sure). That’s certainly a real side effect of some drugs (it was one of the big factors that scuppered rimonabant, and its competitor taranabant back when). But those were CNS agents, and that’s the sort of thing you always look out for in a new CNS drug. What’s an antibody to an interleukin receptor doing causing the same problem?
Well, IL-17 certainly has roles in the brain (those recent papers will lead you to others). And given how painfully little we know about what’s going on up there, it’s certainly possible that these pathways could lead to such a side effect – I mean, how do suicidal thoughts form, mechanistically? Right, it’s a black box like all those questions are. But wouldn’t brodulamab have to cross the blood-brain barrier for that to happen?
That’s very unlikely for an antibody, but (as the various efforts targeting beta-amyloid show), not impossible, either. But if that’s what’s going on, what it is is hideous bad luck, because no one is looking for a CNS effect to stop a peripheral antibody target. And if it’s somehow a peripheral mechanism, feeding back to the brain via who-knows-how, that’s hideous bad luck, too. I hope that at some point we find out more about what’s going on here, out of sheer scientific curiosity.

24 comments on “Some Side Effect For An Antibody”

  1. SP says:

    It must be their own fault for not coding 20 hours a day and diagramming all the feedback loops of the circuitry, like a good electrical engineer would do! This is why the software industry is so much more successful than those drug discovery slugs!

  2. SomeGuy says:

    Antibodies can get into certain areas of the brain, something which Pfizer knows well. Check this out, for example:
    http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3659094/

  3. Mark Thorson says:

    Huh! Helper T-cells with IL-17 disrupt the blood-brain barrier.
    http://www.ncbi.nlm.nih.gov/pubmed/19940258
    Throwing a monkey wrench into that mechanism certainly could have unpredictable CNS effects.

  4. NJBiologist says:

    @2 SomeGuy: Don’t assume food intake/weight changes are due to central effects. In fact, weight changes are the most common observation in tox studies of non-CNS compounds (WS Redfern et al 2005 J Pharm Tox Meth (52) 77-82).
    Also, while there have been a number of demonstrations of proteins in the brain following peripheral dosing, there is marked regional specificity: circumventricular organs > hypothalamus > everything else. If I had to pick a target region for suicidal ideation, it would be the cortex.

  5. MLB pitcher and Medicinal Chemist says:

    “I mean, how do suicidal thoughts form, mechanistically? Right, it’s a black box like all those questions are. But wouldn’t brodulamab have to cross the blood-brain barrier for that to happen?” — Who knows. It has something to do with “free will”.
    What’s “free will”? Will one go to hell for committing a “mortal sin” if they take that drug?

  6. Wavefunction says:

    It would be key to find out whether the effect is target specific or drug specific. If it’s target specific then Novartis would need to take another hard look at its own IL-17 antibody. It’s also interesting that Amgen’s antibody targets the receptor while Novartis’s targets the cytokine; I wonder what differences this would result in.
    It’s also worth noting that the psychological effects of acute psoriasis can themselves potentially lead to suicidal thoughts. I really hope the investigators corrected for this confounder in the clinical trials.

  7. alchemist says:

    “It’s also worth noting that the psychological effects of acute psoriasis can themselves potentially lead to suicidal thoughts. I really hope the investigators corrected for this confounder in the clinical trials.”
    Completely agree with Wavefunction here. Was the trial really powered enough to see this effect in a population where notoriously depression can be very intense?

  8. luysii says:

    Time for a bit of neuroanatomy. The brain contains at least two areas lacking a blood brain barrier. The first is the area pastrami ( really postrema –Derek’s automatic spell checker prefers pastrami rather than postrema and so do I) in the medulla oblongata, where stuff you eat that isn’t good immediately makes you nauseated. The second area is the part of the hypothalamus around the third ventricle, which needs to sense things like osmolarity. Among the many things accomplished by the hypothalamus is the control of the pituitary hormone secretion (among them ACTH) which is known to have all sorts of emotional effects.
    So it’s not surprising that an antibody could have psychiatric side effects.

  9. Chrispy says:

    It seems unlikely that a trial designed to look at psoriasis would be sufficiently powered to draw real conclusions about an event as infrequent as suicide. Without seeing the data it is hard to know how strong the connection is (any chance of seeing that, Amgen?). However, a worrisome trend could be enough for a label change, which is apparently the main reason this program was scuttled. @6 Wavefunction — the IL17 axis includes at least 5 different known ligands, so an antibody to the receptor would presumably block all of them while an antibody to IL17A would only block that one. Blocking the whole axis might work better, too, particularly in patients not helped by blocking IL17A alone. It is a pity we’re not likely to ever know now.

  10. Anonymous says:

    Hmmm, the original article doesn’t clarify whether these suicidal thoughts appear in the patients, or the clinicians conducting the study…

  11. acpatel says:

    Brodalumab is an IL-17 receptor A antibody. IL-17RA is used by the receptors for both IL-17 and IL-25. Since anti-IL-17 is safe, this makes me wonder about a role for IL-25 in the observed suicidality. Novartis sees no increased suicide thus far. Placebo group adjusts for the “confounder” / known risk of suicidaility in severe psoriasis — Amgen’s move suggests there was more in the brodalumab group than in the placebo group.

  12. SomeGuy says:

    @4, @8
    luysii’s point is basically what I was getting at. There are certain regions of the brain which antibodies can target and these regions do have connections to higher order brain centers.
    This doesn’t by any means rule out a peripheral effect of the antibody – very well might be the case – but it does leave open the possibility that this could be a direct CNS effect.

  13. steve says:

    An alternative is that when you block IL-17 activity the immune system compensates by up-and/or down-regulating other cytokines. There are a number of cytokines known to affect the CNS – e.g., the first interleukin, IL-1, was discovered because it caused fever by acting on the hypothalamus. Just because an antibody to a receptor caused a certain effect doesn’t mean that the antibody or the receptor are directly the causative agents – the immune system is a complex matrix and tugging on one end could have unforeseen effects on another end.

  14. bhip says:

    CNS toxicity reported with Merck/Schering/Organon anti-PD-1 antibody, Lambrolizumab.
    Behind paywall- http://www.jns-journal.com/article/S0022-510X(14)00393-1/abstract

  15. pipeline says:

    Lycera is doing great work with small molecules which modulate IL-17 activity in the immune system.
    However, I think copy paste work will not shed light inside of black box.
    http://www.lycera.com/press/lycera-to-present-at-immunology-2015-the-annual-meeting-of-the-american-association-of-immunologists-aai-3
    http://www.lycera.com/press/lycera-announces-milestone-in-merck-collaboration-2

  16. Anonymous says:

    Given the pressure they had to push forth crap to fill their pipeline in defending against Pfizer, I would be surprised if AZ gets a single new drug from its pipeline.

  17. pipeline says:

    Taking off the pressure from LYCERA to hit all targets for (LYC-30937) Pfizer and Amgen should consolidate to fill pipeline with new drugs for same targets.

  18. JRosenblum says:

    Not mentioned here, but mentioned elsewhere is the possibility that suicide ideation and completion could be a side effect of loss of efficacy of brodulamab. It a very effective therapy, perhaps more effective than other IL-17 pathway treatments in development. But the benefit is transient. So patients who had their psoriasis “healed” may have become suicide-prone upon remission.

  19. Eric says:

    @7, @9: I think we can answer some of our questions based on whose parsnips get buttered. If the study *weren’t* sufficiently powered to detect suicidal ideation, don’t we think Amgen would have pointed this out? Or if there were any suggestion that the suicidal ideation followed on loss of efficacy, rather than treatment, don’t we think they’d be pointing that out as well? Without seeing the data, these make me think that the study gives evidence that the negative effects are real, without any appreciable loopholes based on lack of power, confounding, or what have you.
    (@6, @7: Also worth noting that suicidal thoughts in psoriasis patients aren’t a confounder — they’re a characteristic of the control group, which I assume is also psoriasis patients. So a proper control group makes that come out in the wash. (It even could be argued, if there is some sort of “saturation point” for suicidal ideation, which some psoriasis patients are pushing up against, that measuring suicidal ideation on a linear scale may mean the study underestimates any effect of the treatment towards increased suicidal ideation.))

  20. Anonymous says:

    Why not ban sad aongs and movies for the same reason?

  21. Anonymous says:

    @11 and @18: just the presence of a plaebo group of psoriasis patients does not ensure that there are enough of them to avoid confounder effects. It depends on which % of each arm had a certain side-effect, and how small or large that percentage is on the total population of the trial.
    @18: also, as an ex-MerckSerono (of cladribine fame) I can assure you that upper management does not necessarily inderstand the concept of powering of clinical trials.
    That said, I don to know the details and the frequency of the suicidal thughts, but the trial was pretty large, so most probably the effect is real (and scary…)

  22. Pete says:

    My wife is an immunologist and when I showed her this article she wasn’t suprised at all.
    Her response follows…please don’t flame me 🙂
    Not really surprising.
    If you consider that IL-17 is the predominant cytokine in T-cell mediated autoimmune responses, and that the 2 symptoms at the top of the list for ALL autoimmune diseases are fatigue and depression, there’s every chance there’s a link.
    It’s becoming increasingly obvious that the CNS and immune systems are intrinsically linked, although noone has quite figured out exactly how and they are finding increasing numbers of hormone receptors expressed on immune cells (i.e. eostrogen receptors on T cells). In fact, the links to hormones and immunity was hypothesised long before the receptors were discovered on the cells due to the strong sex bias for most autoimmune conditions.

  23. pipeline says:

    @18 and @22,
    Agreed, it is not smart to spend when you have already huge deficit. Great spenders are bad lenders. Benjamin Franklin Quotes

  24. matt says:

    Late responding, but it seems to me out of such puzzles Nobel prizes are laboriously unearthed. It just may take a few decades…

Comments are closed.