Skip to main content
Menu

Cancer

An Unexpected Halt in Multiple Myeloma for Venetoclax

 

Venetoclax (ABT-199) is an unusual drug. But now there’s some unusually bad (and unexpected) news about it. That’s the structure at right, and medicinal chemists will understand immediately why it’s a bit of an outlier. With a molecular weight of 868, that structure just keeps on going, with a somefeatures that you don’t usually see, like that dimethylcyclohexene linkage and the ortho-nitro aniline part. No one makes a molecule like that without a reason, and in this case the structure is necessary for it to bind to the Bcl-2 protein, which is involved in apoptosis signaling in lymphoma cells.There’s no traditional small-molecule-ligand site on that one, and protein-protein interactions like that have always been considered a tough place to get anything to work in drug discovery. Bcl-2 keeps apoptosis from proceeding, actually, so blocking that pathway sends the cells to a speedy end – so speedy that earlier in the drug’s development some patients had kidney problems from how fast the lymphoma cells were being cleared.

But AbbVie perservered, even after their original molecule in the field (ABT-737) ran into trouble in the clinic with effects on platelets. That was another whopper compound, but it wasn’t selective enough for Bcl-2 over other related proteins. The more selective ABT-199 did the trick, though, and showed very good efficacy in clinical trials against chronic lymphocytic leukemia (CLL). It’s approved for that indication, and has been under investigation for multiple myeloma as well. In that case, there are particular subtypes of the disease that show high Bcl-2 expression relative to the other Bcl subtypes, and those are just the patients that seemed to be responding to the therapy.

Unfortunately, that joint Roche/Abbvie clinical effort has come to a sudden halt. Just yesterday, it was revealed that the treatment group (venetoclax, bortezomib, and dexamethasone) was showing a substantially higher mortality rate (+20%) than the control standard-of-care group (dexamethasone alone, in refractory patients), and the FDA immediately put a partial clinical hold on the work (no new patient recruitment, for the moment). As well they should. But this bad news was completely unexpected, as far as I can see. Earlier clinical trials in myeloma patients don’t seem to have shown any signs of this problem. This doesn’t affect the drug’s existing approvals for CLL or AML, but it does perhaps show how different multiple myeloma can be from either of those.

There are more trials going on with various venetoclax combinations than I have time to count, and this news has sprayed uncertainty over pretty much all of them. Now to find out what the problem is – if that’s findable – and to see what the clinical path forward might be.

 

ABT-199 Nears Its Finish Line

20 comments on “An Unexpected Halt in Multiple Myeloma for Venetoclax”

  1. Peter Kenny says:

    I’ve always wondered whether the nitroaniline substructure is absolutely essential. I’d expect an intramolecular hydrogen bond and this conformation would look a lot like an isocoumarin.

    1. Hokusai says:

      Hi Derek,
      To answer Peter Kenny’s question, yes the nitroaniline substructure is essential. The substructure contributed to the selectivity of venetoclax for Bcl-2 over Bcl-xl which navitoclax doesn’t have.

      1. Anonymous says:

        Re: the 2-nitro-4-sulphonamide-aniline substructure. It is known that most humans, regardless of where they live or grow up, have antibodies against 2,4-dinitrophenol (DNP). (Theory: DNP is a widespread pollutant found EVERYWHERE and that it is antigenic so almost everyone has circulating DNP Abs.)

        2,4-Dinitroanilines are drug substructures that can also recruit those DNP antibodies. In the cases I am thinking of, the dinitrophenyl is terminal (exposed) and linked via the aniline N to the rest of the drug. Nevertheless, I am wondering to what extent these interior 2-nitro-4-sulphonamide substructures might recruit DNP antibodies and confound the results:
        (a) Ab binding before it reaches the therapeutic target reduces the effective concentration (dose)
        (b) Ab binding after it reaches and binds to the desired therapeutic target can increase potency to kill cancer cells by two mechanisms: small (well, BIG MW 868 small) molecule induced apoptosis PLUS Ab mediated immune response
        (c) Ab binding after it reaches and binds off-target to some site that is selectively resistant to the small molecule mechanism of action but nevertheless is now subject to attack by the native immune system and anti-DNP Abs.

        Can anyone (Wavefunction?) do a quick calculation to compare the properties of RHN-Ph(2-NO2-4-NO2) and RHN-Ph(2-NO2-4-SO2NHR) and make a guess about binding to the same circulating wt DNP Abs?

        Native or induced small molecule Abs need to be considered when going from any model system into humans.

  2. myma says:

    I don’t even know where to start looking at that thing. My eyes start at one end, and think about it. then they wander to the middle bit, and think harder about the middle bit. And then there is even more!
    Does it have a color to it?

    1. TG says:

      Yes, it’s yellow.

    2. CMCguy says:

      When I look at this structure I sense a peptidomimetic type compound thus view from perspective of a chain of connected aminoacid analogs. Not sure is that is the true genesis but would fit with some past Abbott efforts in Renin and HIV inhibition programs

  3. beauty and the beast says:

    Further evidence that what a drug looks like is in the eye of the beholder. Good luck to the ABT-199 team.

  4. Jonathan Stewart says:

    Minor typo nitpick “with a somefeatures”.

  5. Delp says:

    It’s worse than a +20% mortality – 11% of placebo patients died while 21% of treatment patients died.

    1. Carl Bar says:

      @Delp: Wait what they where giving people a placebo for a life threatening condition as part of a trial, isn’t that seriously unethical to effectively leave them untreated?

      1. Eugene says:

        Derek mentions “standard of care” treatment group (dexamethasone). I think Delp misspoke. In situations like this a true placebo would not be allowed.

        1. MrXYZ says:

          I am fairly sure this is the clinical trial:

          A Study Evaluating Venetoclax (ABT-199) in Multiple Myeloma Subjects Who Are Receiving Bortezomib and Dexamethasone as Standard Therapy

          https://clinicaltrials.gov/ct2/show/record/NCT02755597?term=ABT-199&cond=Multiple+Myeloma&phase=2&rank=1

          It’s placebo controlled with respect to Venetoclax: all patients get Bortezomib and Dexamethasone

        2. MTK says:

          Correct.

          The control group receives standard of care. To do otherwise would be unethical and a violation of the Helsinki Accords.

  6. paperclip says:

    Pembrolizumab also ran into trouble in multiple myeloma, while it’s apparently doing fine in the other heme malignancies. Pembrolizumab like venetoclax was also combined with dexamethasone. (Dexamethasone is very common in combinatorial treatments for myeloma.) Makes me wonder whether a drug interaction is to blame.

    Anyway, the news is a pity. (And all that work I did helping a PI draft a clinical trial protocol just went down the drain.)

    1. Klaus Witte says:

      Drug-drug interaction is an interesting thought. Dexamethasone is an induced of CYP3A4, and venetoclax is metabolized by this CYP isoform. Hence, coadministration with dexamethasone could reduce, rather than increase, the bioavailability of venetoclax.
      This may be the reason for the high venetoclax dose, selected for the multiple myeloma trial, 800 mg/d compared to 400 mg/d in CLL. Furthermore, in CLL venetoclax dosing is started with a 5 weeks ramp-up phase until the 400 mg/d dose is reached. The protocol summary of the multiple myeloma trial does not mention any dose ramp-up – looks as if patients received 800 mg/d from day one of the treatment cycle.

      1. cynical1 says:

        Interesting idea but what you are eluding to – lower levels of venetoclax due to cytochrome induction – would explain lack of efficacy not increased mortality in the group receiving venetoclax. It would make more sense if venetoclax was somehow lowering the level of dexamethasone and/or Bortezomib. Then you would see increased mortality versus the arm without venetoclax.

        1. Klaus Witte says:

          Totally agree. That’s why I mentioned the much higher dose selected for the multiple myeloma trial compared to the CLL treatment, and the lack of a ramp-up phase for the dose.
          Hence, I assume that in the MM trial too much of venetoclax was given too fast.

      2. Louis says:

        My thoughts exactly. Additionally I wonder if tumour lysis syndrome was an issue because of this. TLS was an early influence in changing the clinical dosing pattern of Venetoclax in CLL.

  7. Sam Weller says:

    Hi Derek,
    I’m pretty sure ABT-737 didn’t make it to clinical trials and that ABT-263/Navitoclax was the first in this series to make it to the clinic. Navitoclax did cause apoptosis in platelets and this was a rate limiting toxicity that drove the development of Venetoclax
    In answer to myma – well spotted. Both ABT-737 and Venetoclax are a great shade of canary yellow.

    1. Derek Lowe says:

      Yeah, I think you’re right – I’ll make a correction. And the yellow color has got to be due to that nitroaniline!

Leave a Reply to Carl Bar Cancel reply

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

Time limit is exhausted. Please reload CAPTCHA.