Skip to main content


Baricitinib Follow-Up: An AI Prediction for Coronavirus Therapy

I wanted to follow up on something that came up much earlier in the coronavirus pandemic. Back in February, a group at BenevolentAI proposed the kinase inhibitor baricitinib as a possible therapeutic for the coronavirus. They identified this through their company’s machine-learning approach to the medical literature and disease mechanisms, and identified the compound due to its proposed effects on endocytosis. The drug is used in arthritis therapy as a Janus kinase inhibitor, but it’s also known to inhibit adapter-associated kinase 1 (AAK1), which is where the endocytosis comes in (and that’s certainly a candidate for being able to affect viral entry).

Targeting this enzyme and related ones in the pathway the numb-associated kinase (NAK) family, had been suggested several times over the years as a possible antiviral therapy (the earliest paper I know of is from 2007). So it seems that this idea was going to come up one way or another, AI or not. Some of the coverage at the time was a bit breathless, as is often the case with AI stories that hit the popular press. I had some comments on this at the time, mostly to the effect that this was a good example of literature searching and curation of a useful database, and that there’s nothing wrong with that. If software can help us do that, so much the better – the literature is a gigantic shaggy mound, and we need whatever help we can get in extracting actionable things from it.

Since then, baricitinib and other JAK inhibitors have been tried out in the clinic. In August, a paper from Lilly, BenevolentAI, and other collaborators provided more details. Baricitinib did indeed seem to be effective in cellular models, and a case series of patients treated with it showed some promise. The FDA issued an Emergency Use Authorization for the combination of baricitinib and remdesivir, and now we have the data that led to that decision, published in the NEJM.

At the end of this process, what you see is that the patients getting both drugs recovered a median of one day faster than the ones getting remdesivir alone. Differences in mortality between the two groups showed a trend towards improvement in the dual-treatment group. There was evidence that the combination led to less use of oxygen and mechanical ventilation, and all of these differences seemed to be more pronounced in patients who were in more serious condition at the start of the study. The combination actually produced fewer adverse events than remdesivir alone. These numbers sound pretty similar to dexamethasone, but the paper notes that this trial and the RECOVERY one had different designs and can’t be directly compared. You’d have to run a head-to-head between a remdesivir/baricitinib group and a remdesivir/dexamethasone group to sort that one out, and the good news is that that trial is going to happen.

What we don’t know is whether baricitinib works via the mechanism proposed by the original BenevolentAI paper. Remember, it was first identified for the AAK1 activity; this was before we knew so much about damping down cytokine activity as a needed therapy in the later phases of coronavirus therapy in some patients. JAK inhibitors already do that, which is why they’re used in arthritis treatment, and were independently proposed for coronavirus therapy for that reason. But it’s worth noting that another JAK inhibitor (ruxolitinib) just failed to improve the recovery of coronavirus patients in a separate trial. It does seem that of the JAK inhibitors (which all should affect cytokine levels) that baricitinib has the best additional activity on the endocytosis-related kinase targets, so the original proposal is definitely still alive.

We’ll see from the dexamethasone comparison trial, though, how useful it is under real world patient care conditions (especially when compared to such an inexpensive drug as dexamethasone). That’s something you can’t predict with any AI in the world – not yet, anyway, and it’s going to be a long time before that’s feasible. We’ll take what we can get.

28 comments on “Baricitinib Follow-Up: An AI Prediction for Coronavirus Therapy”

  1. David Eugene Young says:

    Another trial that will require 800 to 1400 participants to prove a difference. And that’s fine. What’s not fine is that it will take 3 to 4 months to enroll and complete. With a million new cases a week and some 10,000 or more hospitalized, can’t they find a way to enroll 1,400 participants in just two weeks? It is tremendous work but could they not have 100 hospitals putting on the study each enrolling one patient a day on the average?

    Also, shouldn’t they have a third arm in the study? All three drugs together?

    1. Oudeis says:

      Re: the three-drug cocktail, I was wondering the same thing. It doesn’t look like the mechanisms of action would overlap. (But then, I don’t know anything about chemistry that I didn’t learn on this blog.)

    2. Ken says:

      I was going to make a similar comment, but from a more pessimistic point of view. “The good news is that that trial is going to happen” – but the bad news is that there are still enough cases to allow that. You don’t see very many drug-repurposing trials for the black plague, for example.

  2. JJ Walker says:

    The results of the remdesivir+ IFN-b1a trial should be out within weeks (ACTT-3). Between the EIDD-2801/MK-4882/Molnupiravir trials taking forever and the low availability of monoclonal therapies relative to US current case numbers I am thinking the the ACTT-3 trial is the last trial that could make a dent in the disease before the vaccine is rolled out in large numbers this spring.

    1. Marko says:

      Yep , that’s about the way it looks. Still , look on the bright side. With the next new coronavirus pandemic, we’ll be able to “hit the ground running”.

      Of course , once current Covid-19 cases and deaths start to dwindle, research interest and dollars will dwindle as well. Public health will again become an afterthought. But after another 3, 5, or more new coronavirus outbreaks there will come a time when we’re 100% ready for the next one. You can count on it. I think. Maybe I’m just an optimist.

      1. DrOcto says:

        Predicting 5 or more new corona virus outbreaks and still thinking you’re an optimist….

        1. Oudeis says:

          I think that was the joke. 馃檪

    2. Rob says:

      What’s happened with the Merck drug? For the formerly top pharma in the world, their performance has been disappointing throughout the pandemic. Some glitch? A lack of urgency or leadership? Nothing contributed to date and the window is closing.

  3. anonnymous says:

    With Covid19 vaccine distribution and delivered to patients and people alike, the discovery of Baricitinib is now, a damp squib.

    1. Pedwards says:

      Seeing as we still easily have ~5-6 months before vaccination rates will be anywhere close to widespread, anything that can potentially lessen the severity of the disease and push down the burden on ICUs (and morgues, for that matter) is still very worthwhile. Not to mention increasing our understanding of these things for when the next COVID makes the jump to humans.

      1. rondi says:

        There are very successful therapies available with common drugs. Nobody cares about them tho. Hopefully after the recent Senate hearing–something can get done. In the meantime—

  4. Jason P says:

    So, Baricitinib makes how many ‘repurposed’ drugs that have “some” efficacy with COVID-19 and its results/effects?

  5. Anonymmous says:

    An article just published in Nature points to a causal link between high expression of the JAK kinase TYK2 and severe COVID disease. A selective TYK2 inhibitor, deucravacitinib, is in late stage clinical studies for autoimmune diseases. Presumably the sponsor company, BMS, could initiate a trial. Does the emergence of the vaccines make the prospect of new clinical trials for COVID-19 moot point?

    1. rtah100 says:

      I’d like to see the results of the studies of fostamatinib, which was identified by researchers in the Netherlands as a potential repurposing candidate, by its action on the SYK pathways.

      I’m not familiar with either JAK or SYK in any detail – there appears to be some cross-talk between them but I would be interested in an expert view on which is the better target (or whether they are complementary).

  6. James N says:

    Good Afternoon Dr Lowe,

    I may have missed it, but has anyone commented on the effectiveness in the vaccine of preventing spread? It’s been pushed on us all through this situation that being asymptomatic isn’t enough to keep from spreading the virus. A vaccine hopes to prevents you from getting hospital sick may not be enough to slow the spread of the virus itself. I know where I live the only reason people are willing to take the vaccine is because they want life to return to normal, and they’re thinking once they have the vaccine, they won’t need to wear masks and social distance anymore.

    Thank you for keeping up with your blog over the years, It’s nice for an IT guy to have a place to look for relevant and knowledgeable information on the pharmaceutical and chemistry fields without having to have an associates degree just to read the titles.

    1. MrRogers says:

      At the VRBPAC meeting, Pfizer indicated that data on whether the vaccine reduces the likelihood of infection (vs. “just” eliminating disease) should be ready to present “early next year”. Whatever the outcome of those studies, it seems to me that if infection doesn’t lead to disease, then life can go back to normal once everyone who wants it can get the vaccine.

    2. Michael says:

      I’d bet my life savings on Pfizer/Moderna vaccines significantly reducing transmission.

      To name just three reasons:

      1) Asymptomatic secondary attack rate is much lower than symptomatic/pre-symptomatic:

      2) It is very likely that the vaccines will reduce the likelihood of infection;

      3) It is very likely that the vaccines will reduce the volume and duration of shedding of asymptomatic infections, probably even beyond unvaccinated asymptomatic infections.

      If the data shows a certain number of asymptomatic infections, the media will likely refer to it as being proof of transmission or spread, but evidence of vaccinated asymptomatic infections alone won’t clearly prove that.

    3. sort_of_knowledgeable says:

      Asymptomatic is often confused with presymptomatic where the individual wasn’t exhibiting symptoms at the time but later developed symptoms. Presymptomatic spread will obvious be reduced since vaccination drastically reduced the percentage of people developed symptoms. Assuming behavior doesn’t change then asymptomatic transmission is reduced compared to symptomatic transmission.

      1. Charles H. says:

        “Assuming behavior doesn’t change”, however, is clearly a false assumption except in cases where they didn’t change their behavior in the presence of a pandemic.

    4. Adrian says:

      Derek discussed this recently here in the blog:
      “What, for example, is its effect on transmission of the disease? We should be seeing more on that early next year, but based on these figures, it seems very likely that this vaccination would cut down the transmission rate sharply. But that has to be proven; there are plenty of things that have seemed very likely over the years in drug development that haven鈥檛 worked out.”

  7. H贸lmsteinn J贸nasson says:

    cd147 and sars coV 2

  8. theasdgamer says:

    “CDC recommends early treatment with covid antivirals for high-risk patients”

    Oh, wait, I got the headline wrong:

    “CDC recommends early treatment with flu antivirals for high-risk patients”

    I’m sure this has an easy explanation. Flu and covid antivirals work differently. Covid antivirals must be given late, amirite?


  9. cancer_man says:

    ChromaDex’s 100 person trial of NR (Nicotidamide Riboside) with Covid-19 patients with moderate symptoms went well as recovery time decreased 30%. The results of a similar 300 person trial should be known within a few weeks.

    1. John Liu says:

      Nicotinamide riboside also corrects social deficits in mouse model of autism, reduces inflammatory cytokines in stage D heart failure, prevents light-induced retinal damage in mice, increases milk production in preterm mothers, and that’s just the last 6 months of press releases. The most impressive nutritional supplement ever.

      1. Mister B. says:

        Does it also prevent too much mother-in-law presence in men鈥檚 life ?

        There is a tremendous demand for such drug.

      2. theasdgamer says:

        Unfortunately, one of my female relatives passed away recently after testing positive for covid and failing to take any antivirals early.

        You all with your fake science really ARE killing people. Just stop.

  10. H贸lmsteinn J贸nasson says:

    CD147-spike protein is a novel route for SARS-CoV-2 infection to host cells

Leave a Reply

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

Time limit is exhausted. Please reload CAPTCHA.

This site uses Akismet to reduce spam. Learn how your comment data is processed.