Skip to Content

Drug Development

A Long Journey to the Capsid

There’s a fine article at C&E News on Gilead’s capsid inhibitor for HIV. For those not into virology, the capsid is the protein coat that viruses have – it’s their armor, more or less, and disrupting its formation should be a large problem for them. But finding compounds that accomplish that is a large problem for drug discovery, as the article shows. The protein latticework has to fit together beautifully to protect the virus’ cargo, but at the same time it needs to disassemble easily once the infectious attack has begun, so the hope has been that there might be some delicate balancing going on that could be tipped over with the right molecule.

Pfizer published just such molecules several years ago, in a program that had gotten underway from a phenotypic screen. Their molecule, an indole derivative, turned out (no doubt to their surprise) to fit into a gap in the HIV capsid arrangement that had been identified in the 2009 X-ray structure of the capsid proteins. That region is actually where the virus targets a particular human cell protein, CPSF6, that helps it home in on active parts of the genome, increasing its infectious success. The Pfizer molecule didn’t destabilize the capsid armor, as you might imagine, but rather stabilized it and kept it from opening up as readily, which serves to interrupt things just as well. (Pfizer had also discovered some of those capsid-destabilizing molecules as well).

Via C&E News

The Gilead team had been working on capsid-targeting compounds for several years themselves at that point, but without a lot of success. High-throughput screening didn’t give any useful leads, so (as the article shows) the Pfizer publications really gave them something to work with. Shown at right (graphic courtesy of C&E News) is an advanced tool compound that the company has revealed. The Pfizerish core, from what I can see, is at upper left – the indole has turned into that tetrafluoropyrazole derivative, and the acetic acid amide chain coming off it, along with the branch to two aryl groups, is also in Pfizer’s original compound series, albeit in rather different form. But as you can see, the compound has grown off in other directions as well, on its way to picomolar (!) potency and greatly enhanced metabolic stability. (It had better have the latter, since it has approximately as many fluorines on its surface as a nonstick frying pan).

There are a lot of things to raise eyebrows about that compound, and I have to note that it’s not quite the one that’s in the clinic. But you’d have to assume the clinical compound is a similar beast that’s similarly beastly. So we have eight fluorines, <s>two</s> one sulfone (and one sulfonamide), an acetylene, a chiral fused cyclopropane. . .no, this is clearly, as the article says, the product of several thousand compounds worth of analoging. Is it soluble? Well, no, of course it isn’t. And there lies another point of interest.

As Jarvis’ article notes, the HIV treatment field has moved on a great deal over the years, to the great benefit of patients. New mechanism or not, there’s just not as much room as there used to be for an injectable therapy, not when most patients are well-served by an oral pill regime. But it turns out that once you actually administer Gilead’s compound, it remains active for many weeks. That’s led to the idea of some sort of subcutaneous dose that would be once a month, once a quarter – the formulations work is apparently still going on, and those now have an advantage compared to taking X pills per day on a pretty rigorous schedule.

If it hadn’t been for that, you’d have to imagine that this project would have labored long, hard, and with great skill and determination to produce something that was completely outmoded by the time it was delivered. A question that we may not get answered for a while is when the changeover occurred inside Gilead’s team, from “We’re going to target the viral capsid!” to “We’re going to come up with a long-acting HIV therapy!”. Winston Tse, the Gilead chemist quoted, says that they were never thinking about the latter, actually. (And even that field has competition of its own – read the article for more details).

There’s a lot to think about here. First, I like the fact that a molecule this large and hairy (or something similar!) is actually moving into the clinic. Yesterday’s post about all kinds of odd chemical matter could also include these things – classic organic chemistry product, just larger and more complicated (Abbott’s BCL compound Venclexta, venetoclax, is another example). Development of such molecules is definitely more challenging, but for some targets, nothing smaller may ever do the trick. Second, this is really a story of persistence and (I’m sure) of a project surviving a number of attempts to kill it. I doubt if even the Gilead scientists could furnish an accurate count of the number of near-extinction events the project survived over twelve years; you know there had to have been several. (Keep in mind, though, that just keeping a project going for a dozen years is not the recipe for it to succeed – there are many less-celebrated examples out there that can prove that!)

That brings up the third point, the one I mentioned above: this is apparently also the story of a project that might have turned into a vast waste of resources, even had it succeeded. Twelve years is a long time. There’s an old A. E. van Vogt science fiction story (“Far Centaurus“) where a crew of explorers set out on a 500-year trip, made mostly in suspended animation, for Alpha Centaurus. They arrive to find the whole system has been long populated by Earth colonists after a faster-than-light drive was discovered. That’s a real risk, on the time scale of drug development. Your competition, your patients, your pricing structure – all those things might be different by the time you arrive.

(If you look up that story, prepare yourself for a real 1940s SF experience. van Vogt was, objectively, not that great a writer, but in his prime he produced weirdly vivid tales that have an episodic, dreamlike quality to them that seems to have been driven by his rule of throwing some huge plot twist in every few hundred words, no matter what. I first read “Far Centaurus” when I was about eight, in an old paperback Groff Conklin anthology, and I haven’t re-read it in decades – but I can still quote lines from it.)



26 comments on “A Long Journey to the Capsid”

  1. In Vivo Veritas says:

    Derek, you wouldn’t happen to know the formula for adeledicnander force, would you? 😉

    1. Derek Lowe says:

      “He’s trying to tell me that electrons think. . .”

      1. “Omigod. Singing meat. This is altogether too much.”

        1. Thoryke says:

          Thank you, Jesse, for reminding me of how splendidly bizzare those short stories were. I even used them in a science writing syllabus once. “Shut up and press ‘Ann’!”

          1. Eugene says:

            Derek, thanks for the memory jog! Far Centaurus is one of the stories that hooked me into Scifi in the 60’s when I discovered that the local public library had an obscure shelf of Scifi anthologies with short stories by Clark, Asimov, Anderson, Heinlein, Bester and Van Vogt (far right aisle, second to last bottom shelf) and as a bonus several years worth of Analog magazine next shelf over. Spent my summer vacations reading all of it. Later when I came up for air I found the Scientific American magazine collection.

  2. G says:

    Having some trouble with the column width formatting – the word wrap is a little off.

  3. MBB says:

    Any clue how this thing is synthesized?

  4. Nat says:

    Much of van Vogt’s writing (especially most of his full-length novels) was clumsy bordering on incoherent, but I still think “Slan” and “Voyage of the Space Beagle” are some of the best examples of “Golden Age” science fiction. Very influential works, even if some aspects have aged better than others (“nexialism”? please…).

  5. Curious Wavefunction says:

    “Second, this is really a story of persistence and (I’m sure) of a project surviving a number of attempts to kill it.”

    To me that’s one of the biggest problem with these ‘beyond rule of 5’ molecules and novel modalities. Their very nature usually demands an extended period of risky development; not the kind of project timeline that’s supported in the ever shrinking view of investors and CEOs.

  6. Uncle Al says:

    Stipulated: Selective and efficient classes of capsid x-linkers are developed.

    …1) Capsid and sheathed (e.g., tobacco mosaic) viruses as classes are rendered non-reproductive. Do the x-linked protein balls, etc., clog kidneys?
    …2) Gene-gineered nanofabrication of survivable polytopes, including framework-only remnants. Does Big Pharma discover (chiral) catalyst supports, grown photonic crystals, electroless plating toward metamaterials, and syntactic nanofoams?

    A tight pair of shoes is no reason to forget how to dance.

  7. Questions says:

    For new HIV drugs, is an oral medicament preferable and is an oral medicament the most likely to be subscribed? And if viral suppression with an oral treatment is required first, will the end user, long-acting, sub-cut market justify the up-front R&D investment? Is this molecule most likely to be used as a combination therapy? If Gilead doesn’t have a long acting partner they could still be selling a once-a-day treatment. Is patient compliance really an advantage with a long acting HIV drug? A poorly compliant once-a-day may be a poorly compliant once-a-quarter patient. Does the molecule really need to be this big to destabilise the capsid – would a molecule that binds efficiently to the PheAla and indole binding pockets be enough?

    1. AQR says:

      I would think that an injectable given four times a year would be an excellent therapy for treatment in the third world – places where it’s difficult to even find potable water with which to take an oral drug. The local health care workers could administer the drug during their quarterly visit to the villages on their circuit.

  8. HIV Guy says:

    Long-acting HIV drugs (referred to as antiretrovirals) could be useful both for treatment of patients who are infected with HIV and possibly as prevention. Since there are already several very effective oral one-pill-once-a-day formulations on the market, and a few new ones coming, the long-acting antiretrovirals may be best suited for prevention. Why take a pill every day for life to prevent HIV when an injection every few months will work just as well? Approved antiretrovirals for prevention of HIV are currently limited to two drugs (tenofovir and emtricitabine), and there seems to be no need to add a third agent. It will be interesting to see if only one antiretroviral is needed to prevent HIV, or if two are necessary.

    1. Morten G says:


      The main worry with HIV preventive drugs is that those most at risk of contracting HIV have a hard time complying to a once-a-day dosing regimen (whether third-world country inhabitants or drug users).

  9. cynical1 says:

    I think resistance would figure into the long term utility of this MOA delivered either orally or subcutaneously. Either delivery method of an antiretroviral as mono therapy that did not develop resistance, I would guess, could be a game changer for treatment. Any in vitro data out there on this class and resistance?

    1. Dionysius Rex says:

      I am pretty sure the Gilead team will have studied the emergence of mutants.

  10. Mfernflower says:

    It will be a interesting trip indeed! That compound looks like cancer-on-a-stick!

  11. Andy II says:

    Doing frequency (once a day or once a week, for instance) for the same indication would be depend on the pharmacoeconomics issue assuming both are non-inferior in the efficacy/safety per the treatment.

    A long acting drug (looong half-life) for the prevention. Curious to know what the primary/secondary endpoint in the clinical study protocol and what the comparator is used. What the prevention rate is acceptable?

    Everyone is waiting for HIV is another HepC where cure rate is around >98%. Someday…

  12. Barry says:

    Well, only one sulfone (and one sulfonamide). And what the Hell’s a “pyazole”? Did you drop an “r”?

    1. Derek Lowe says:

      I did indeed – going back in now!

  13. Barry says:

    “pyazole” sounds like the misbegotten offspring of “pie-hole” with “asshole”

    1. exAZ says:

      Back in the day, “azole” was code for dysfunctional AZ management…

  14. mfernflower says:

    There is a typo at the start of the post “oleThere’s”

  15. ScientistSailor says:

    Are we sure that structure isn’t a belated April Fool’s day joke?

  16. CPSF6 says:

    The CPSF6 story is interesting for those into virology as well – a capsid protein mediating localization of the viral genome to transcriptionally active regions of the host DNA.
    Who the hell does that capsid protein think it is? Is it lost?
    Viruses take the definition of “multifunctional proteins” to ridiculous extremes.

  17. Chemistofsorts says:

    Sinking significant resources into long acting is betting against cure. When DTG goes off patent in nine years, the expected cost reduction will be difficult to compete with unless you have better efficacy, i.e. Real progress toward cure. Using an alternate delivery method as a winning strategy feels like a loser. Such a flop as PFEs inhaled insulin comes to mind… fundamentally we must be working toward efficacy not convenience. Efficacy sells!

Comments are closed.