Skip to main content

Clinical Trials

Lilly’s Monoclonal EUA

I’ve been meaning to write about what has to be called a regulatory mystery. Eli Lilly obtained an Emergency Use Authorization for its monoclonal antibody (bamlanivimab) against SARS-Cov-2, but (as detailed in this post) the dosage that they applied for was 700mg. Which is one-quarter of the dose that showed any efficacy against the primary endpoint in the actual clinical trial. The approved 700mg dose did not reach statistical significance (nor, it should be said, did the even-higher 8g dose) in lowering viral loads at day 11 vs. placebo.

Here’s BioCentury on this, which has to be the only EUA ever granted for a dose that failed in the clinic. To be sure, the company has said that there was no difference between the three doses on various secondary endpoints, and in fact says that they’re not sure that they’ve found the lowest effective dose yet. The FDA’s Janet Woodcock has said the same thing. That’s definitely looking on the bright side, but it suggests that the original endpoint in the trial was either too optimistic or aimed in the wrong direction, doesn’t it?

The best spin I can put on this is that perhaps the viral loads (the primary endpoint that failed) are not well correlated with later hospitalizations (where there was a significant difference). In that case, it may be that those hospitalizations have more to do with the immune over-response to the virus than they do with the virus itself, although that’s sheer speculation on my part. But it seems clear that the mAb treatment needs to be aimed at people who are at high risk, particularly those who don’t seem to be mounting a good antibody response on their own. In that population, I think Lilly’s therapy (and Regeneron’s two-mAb cocktail) could do some real good, but that’s definitely a smaller subset of the total coronavirus population,

Not that we have enough of the mAbs to treat the total coronavirus population, anyway! The government has contracted to buy 300,000 doses of bamlanivimab this year, and Lilly is supposed to be shipping 80,000 doses this week. We just had 163,000 cases reported yesterday, to put that in perspective. Now, it’s for sure that not all of these people would be getting a monoclonal antibody, even if we had barrels of it sitting around, but there’s clearly a need for targeting the supplies we do have. I’m not sure how that’s to be done, though: you want to catch people early in the infection who are having trouble mounting a vigorous antibody response, but how many people are being tested in a manner (that early on) to pick that up? I really don’t know, and so far I haven’t seen any information addressing this issue. But it’s surely something that’s being worked on now that the mAb is starting to appear – I’ll update as the situation becomes clearer.

47 comments on “Lilly’s Monoclonal EUA”

  1. Aaron says:

    Did Lilly analyze if seronegative patients responded to their monoclonal antibodies?
    That was Regeneron’s main finding.

  2. Tom says:

    Is this also a case where ‘it can’t hurt to use it’, where it may help some but not all? I’m not aware of the safety profile of the Lilly mAb though…?

  3. Matthew says:

    Anti-viral therapies generally only work very early after initial infection.
    Everyone should know that by now.

  4. Giannis says:

    IgG antibodies work better for stopping systemic viral spread including lungs. In contrast IgG don’t work very well for stopping viral replication in the upper respiratory tract, and that’s where “viral loads” are measured.

    1. eub says:

      If this stuff does work, that seems plausible as a reason why. But then what bozo let nasopharyngeal load be the primary endpoint?

    2. Barry says:

      IgG antibodies work better than what?

  5. SteveM says:

    “The best spin I can put on this is that perhaps the viral loads”

    Lilly’s marketing strategy is no doubt based on the Remdesivir model. I.e., deliver a very high cost treatment with marginal therapeutic utility and they’ll clean up anyway by getting Dr. Fauci to front for it.

    So why not?

    P.S. Lilly also has experience with this strategy as a corporate “Best Practice”.

    1. MagickChicken says:

      I’m not sure Fauci saying that it reduces hospitalizations by 4 days on average and has no statistical impact on mortality is “fronting.”

  6. confused says:

    From the data in the last mAb post, it sounded like the Regeneron one did better than the Lilly one… does anyone know why the latter got the EUA first?

    (Maybe something to do with avoiding the appearance of political influence, since the Regeneron one was used on the President & then he advocated it???)

    Is an EUA still expected for the Regeneron one?

  7. navarro says:

    has the standard for emergency use authorizations always been “well, at least it does no harm even if it does no good” or is this a recent development? have they specified that criterion anywhere? does anyone involved in decision making on this own very much lilly stock?

    asking for a friend . . .

    1. former Wyeth says:

      no one at the FDA is allowed to own stock in any company regulated by the FDA.

      1. navarro says:

        that is actually a relief to hear.

        this decision is still very odd.

        1. SteveM says:

          But FDA employees are free to leave and join the Pharma companies regulated by the FDA. When they are at the agency the crony skids are greased more opaquely for a downstream play later.

          That of course is the way revolving door regulatory “sausage” is made across all agencies. (And especially the hyper-bloated and corrupt MIC).

      2. Irene says:

        Wouldn’t that mean they could hardly own any mutual funds? or does the ban on stock only refer to specific individual purchases of a particular named stock?

        1. The ban is interpreted aggressively, or at least it was when I worked for FDA (1988-2000). For example, I couldn’t own shares in CVS, because a non-zero portion of CVS’ income came from a mix of products regulated by FDA. I think broad index funds were OK, but anything like a sector fund was excluded. There is a list of forbidden investments (see, but I remember being told that I had to sell something because the gnomes making those decisions thought it was inappropriate to hold it, even though it wasn’t on the list.

  8. Marko says:

    With the limit in supply , I would target everyone over the age of , say , 70 or 75 , who was detected positive early enough in the course of disease where the MAb could do some good. Treat absolutely everyone in nursing / long-term care homes regardless of age , if caught early. Do a rapid antibody test and don’t waste treatment on those who’ve already mounted an Ab response.

    The lower dosage might actually make sense , given the limited supply. Also , as another commenter pointed out here , the lower dose would avoid the potential problem of loss of avidity due to swamping of binding sites with excess antibody. I’m optimistic that if used early on the right risk-stratified population , the number-needed-to-treat to save a life might actually be fairly low , and that this treatment might end up having a measurable effect on fatality rates, though not likely a major effect.

    1. John Wayne says:

      I hope you are right; and I agree that the risk/reward equation looks reasonable. Is there a test that doctors can use to select for patients more likely to respond based on how their immune system is doing?

      1. Marko says:

        Nothing quick and easy , as far as I know. Hopefully , most of the elderly who are severely immunocompromised would be aware of it from previous medical evaluations. The main thing to check for would be antibodies to COV2 , which can be done with a point-of-care test. If someone has already seroconverted , it’s probably a waste to treat them with MAbs given the current supply limitations. You want to save the MAbs for those who are pre-seroconversion in the disease course.

    2. confused says:

      Yeah, maybe so. People over 75 are not that large a fraction of the population (but represent over 50% of all COVID deaths).

  9. metaphysician says:

    So, silly question time. What evidence is there for the claim “antivirals work, but only early in the infection”? I keep seeing this presented as if its self evident truth. Is it actually verified fact, or is it just the same special pleading that we constantly see everything an Alzheimers treatment fails?

    1. cynical1 says:

      I think this is more for the viruses that are cleared by the host in a relatively short period of time. There is evidence that SARS-CoV-2 is cleared in the majority of people in a relatively short period of time, usually within 10 days of onset of symptoms. Some more severe cases take 10-15 days to clear viral replication but that’s more common in immunosuppressed patients. Average time to death for COVID-19 is 18 days. So the virus is gone at that point but you’re still dead. If you wait until patients start heading south, it’s too late. Hence the logic behind early treatment. That’s also what has been observed with neuraminidase inhibitors with influenza.

    2. anon says:

      All of the people still alive after receiving anti-retrovirals long into the course of their HIV infections give the lie to this. As do the people being cured of HepC after decades of infection. It may be true to say of the full course of a viral infection is only a few weeks it’s important to intervene early.

      1. Dionysius Rex says:

        Well, the argument you make here compares acute infections with chronic, which doesn‘t make much sense.

  10. hoosierphd says:

    Using day 11 viral load as an endpoint is ridiculous. The Ct values will be extremely low for the placebo control, in addition to the various treatment groups. I think the argument can be made that the 700mg dose works well at reducing viral loads on day 3-4. Either way, the combos will be a much better option, albeit more expensive to make and less available.

    1. hoosierphd says:

      Should clarify that I meant the Ct values will be very high, and viral load very low.

    2. eub says:

      From the article: “Additional data reported by Lilly on Oct. 7 showed viral loads were not significantly lower at days three or seven in prespecified analyses of pooled data from the three bamlanivimab groups.”

  11. PDINV says:

    Not surprising. Remdeaivir has not demontrated reduction of viral load in humans with any virus, ever. It got full approval, EUA is for amateurs.

    Luckily, we have our Lord Savior Albert Bourla, Hero of Humanity, Father of Vaccines, the Unburned. The science was so strong that the unthinkable happened. Vaccine efficacy has been demonstrated after less than 2 motnhs (median) following last dose. Of course we have to wait until next year to find out what this efficacy refers to exactly when they said they will publish, but rejoice! Approval is next week. I tried to find a vaccine that produced negative data after less than 2 months following last dose, but I could not. It might be me. Or it might be that only powerful science can get such approval.

    I suggest we name year 2020 as “The Year of Science and Magic”. The year we realised we can cure everything, if we can just produce enough press releases that say so.

    1. MrRogers says:

      A quick Google search will provide you with the Pfizer analysis plan which details exactly what the trial endpoints are and how they are analyzed.

      1. PDINV says:

        Looooool. Are you saying they are going to seek approval after 2 years? It’s next week. My problem isn’t with the plan, my problem is the plan doesn’t matter. All praise A Bourla. Savior of humanity.

  12. Barry says:

    Sounds like Lilly’s regulatory department pressed forward for and approval/revenue w/o regard to the science. That belongs to a different department.

    1. PDINV says:

      On the contrary. The more diluted, the better, a combination of science and homeopathy. Synergy guaranteed. Powerful stuff. However, as out host here has mentioned before, as a preventative it could be useful, but the haven’t done the trial for that.

  13. steve says:

    Lilly’s (and Regeneron’s) antibodies are neutralizing antibodies. That means they block the binding of the virus to the receptor, not necessarily clear viral particles from the system. You simply cannot rely on historical data on antivirals like Tamiflu as some are doing to this virus. The pathogenesis of SARS-CoV-2 is dramatically different from any respiratory virus in human history. For one thing, it causes auto-antibodies to interferon to be formed. It attacks the vasculature throughout the body. It may well be that blocking viral binding to ACE2 has effects very different from the simplistic models used in this thread. The bottom line will be if it reduces hospitalizations, not whether it affects surrogate markers like viral load.

    1. Doppelkupplungsgetriebe says:

      As a layman, when I read statements like that I tend to think:
      “Nature is a greater engineer than any lab, yet that still sounds like an interesting experiment. Perhaps the interesting experiment leaked out with lab animals that were sold off rather than cremated.”

  14. Sure, there may be 163,000 cases in a day, but there’s only ~1,000 of those that are actually fatal. It doesn’t really matter if you can’t treat the people who are asymptomatic or just get a mild cough. If this reduces the death rate, 80,000 doses per week should be enough to treat those where it matters, right?

    1. Marko says:

      Right. If it works at all , the trick that remains is to distribute the limited supply efficiently to the highest-risk patients at the optimum time after infection to achieve max benefit.

      It’s too bad we don’t know if it has a prophylactic benefit yet. If it does , you could save a lot of lives over the next 3 months or so just dosing every nursing home resident once or twice to get them thru the winter pandemic surge. I think AstraZeneca has a prophylactic trial pending for their MAb , so maybe we’ll get some info on that before too long.

    2. confused says:

      Not at all an expert, but I think the issue is that if it has to be given before hospitalization, you have to *guess* which patients are likely to be life-threatening cases vs. mild ones.

      Of course, there are strong trends (eg. the oldest people / people in long-term care facilities are at vastly greater risk than the general population)

  15. Lane Simonian says:

    A seeming conundrum is how you can have too little immune response early on and too much immune response later on. Part of the answer may lie in T cell dysfunction due to nitration, especially in those with pre-existing conditions (such as in neurodegenerative diseases, cancer, and in diabetes).

    The flip side to this conudrum is how can certain plant compounds (in panax ginseng, for example) both “boost” your immune system (by partially reversing T cell dysfunction) and protect against cytokine storms (by inhibiting oxidation). Identifying the most effective of these compounds could potentially lead to a treatment for both the early and late stages of a coronavirus infection.

    1. Lane Simonian says:

      Wrong subject, sorry.

  16. Scientist somewhere says:

    Probably better than nothing. I would take it. Sounds a bit like Lilly’s baricitinib story though

  17. Tony M says:

    I think it worth repeating the statistically significance of the results wrt ER visits and hospitalisations which are quoted in the Biocentury link provided in the article above:

    “The earlier results did suggest, however, that the mAb decreased ER visits and hospitalization rates, producing a statistically significant effect on the combined metric in the pooled treatment arms vs. placebo (1.6% vs. 5.8%, p=0.02). In addition, the low, medium and high doses of bamlanivimab significantly lowered symptom scores on a 24-point scale over 11 days, relative to placebo (-7.9, p=0.009; -6.35, p=0.038; and -7.86, p=0.011, respectively).”

  18. DrOcto says:

    Remember back in April when doctors were so desperate to find a treatment, they were using any old thing they could find off the shelf? Well now they are still desperate, but have nothing worthwhile left to test.
    Having a few new ‘maybe something’ drugs from the pharma industry at least lets our overworked hospital staff feel like they have some hope to contribute to the people dying in front of them.

  19. Urban Achiever says:

    Oh, Black Betty
    Whoa, Black Betty

    1. Michael says:

      That joke easily warrants an EUA, and is well on track to full approval.

    2. Derek Lowe says:

      I hope Leadbelly’s ghost haunts you for that one.

  20. J N says:

    If I had COVID I’d be worrying more about the severity of my symptoms or the possibility of hospitalization than viral load.

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.