Skip to main content
Menu

Analytical Chemistry

Antibody Tests for the Coronavirus

Let’s talk antibodies. Mounting an antibody response is crucial for anyone to overcome a challenge from an infectious pathogen, and the immunity that can result is crucial for entire communities and populations. Determining who has such immunity is furthermore going to be crucial for us as we come out from under the current pandemic. If someone is truly immune to re-infection (and if the infectious agent isn’t mutating out of range) then they should be able to freely leave their homes, go to work, encounter groups of people and life that normal life that is such a desirable state these days.

Here’s some immunology background for those who could use a refresher. As always, you can assume that for every detail mentioned here that there are further details and complications behind them – that’s always true of human biology, but it is never more true than when you’re talking about the immune system. First off, there’s a big division into the innate immune system and the adaptive immune system. The first one is evolutionarily older, and is the general-purpose always-on component. I’m going to give it very little detail here in the interest of space, and because it’s adaptive immunity that is the key issue in the current outbreak, but it involves the familiar and extremely important processes of inflammation.

The innate immune system relies on general “that ain’t right” signals, such as the ability of toll-like receptors (TLRs) to recognize unusual double-stranded RNA floating around as a sign of viral infection. The RIG-I receptor family also recognizes several viral RNA motifs – viral replication is necessarily going to involve lots of production of such species, so a constant . Naturally, over the eons viruses have also evolved methods of their own to disrupt these recognition signals and their downstream events (many mediated by type I interferons). Another sign of viral infection is cells that have deficiencies in their major histocompatibility complex (MHC) surface proteins. The MHC is famously the body’s “identification-friend-or-foe” (IFF) system for recognizing “self” versus “nonself”, and is the basis for things like transplant rejection. NK (natural killer) cells are constantly patrolling for cells that are displaying “insufficient self” and destroying them, and that can mean cells whose functions have been disrupted by viral replication going on inside them.

These are all valuable functions, but there are many situations where they won’t be enough. For one thing, as mentioned, the innate immune pathways have been around a lot longer, and pathogens have had more time to undergo selection for strains that have stumbled into ways to evade or interfere with these mechanisms. The adaptive immune system (AIS) is a more recent development, if by “recent” you can accept 500 million years or so, about the time that fishes with jaws started showing up in the fossil record. There have been some refinements since then, but it’s basically sharks on up. The AIS builds on the innate immune system and adds some extraordinary power and specificity.

The big puzzles in immunity were figuring out how the body could recognize such a wide variety of pathogens, how it was able to “scale up” a response (which can take a few days), and how the memory of such infections is maintained afterwards. The key was realizing that all of us are carrying, at all times, a gigantic combinatorial library of specialized Y-shaped glycoproteins (estimates vary, but probably around ten billion different ones out of a possible suite of a trillion or so) whose function is to hang around until something shows up that one or more of them can bind to. These are the antibodies. They’re carried around on the surface of B cells, and every human being has a somewhat different collection of them. The mechanisms by which our rather limited genome is able to produce such a huge variety of proteins are beyond the scope of a single blog post, but they’re pretty damned impressive, and medicinal chemists who have done combichem and/or DNA-encoded libraries will find a shock of recognition awaiting them when they read about what goes on.

A substance (generally a protein or something bound to a protein) that binds to one of those antibodies is called an antigen, and it’s estimated that only a few B cells respond at first to a new one. But that sets off another really impressive part of the process, clonal selection. When an antigen binds, that surface antibody/antigen complex gets taken back into the B cell, and the antigen itself is chopped up and sent back up to the surface of the B cell as part of its MHC presentation. Then when a helper T-cell binds to that primed B cell, the B cell gets stimulated to divide rapidly. Some of these become plasma cells, which produce large numbers of copies of the antibody that set off the whole process (and release these into the bloodstream and the lymphatic system), and some of which hang around as memory B cells.

And there you have some answers to the earlier immune puzzles: the body is able to recognize so many antigens because we constantly carry around a ridiculously huge variety of antibodies on our B-cells, far more than will ever be activated in any individual’s lifetime. And we “ramp up” the response to such antigens by selecting out the ones that hit and having their carrier cells multiply and produce huge numbers of those particular antibodies. And finally, some of those cells are specifically designated to stay behind, surviving for decades, as a repository of Stuff That Worked That One Time, just in case that particular pathogen should show up again.

I’m leaving out (for now anyway) ungodly amounts of immunology, such as the various functions those antibodies have and the many cell types that carry out their duties in response to them. That’s because I want to go to the things that are making headlines now, the antibody-based blood tests that many companies are working on and presenting to the FDA for approval. The quick point-of-care ones are generally lateral flow assays, whose appearance will be familiar to anyone who’s taken or seen a pregnancy test. You can set these up several different ways, but they all depend on antibody recognition and some sort of colorful indicator.

For determining whether or not a person has developed antibodies against the new coronavirus, a typical test would work like this: drops of blood are absorbed onto a “sample pad” at one end of the test device. That soaks up red blood cells and the like and lets the plasma soak along a laminate of what’s essentially paper. The first thing it encounters is a zone that has known coronavirus antigens (such as pieces of the spike proteins, etc.), which pieces are also linked, in the most common form, to tiny particles of colloidal gold metal. If the plasma has antibodies to the coronavirus proteins, those will bind to the test antigens and carry them (and their colloidal gold particles) along up the strip. Then it runs into three zones on the paper, narrow strips that are impregnated with “antibodies to antibodies” (yep, that’s a real thing).

I didn’t go into the various subclasses of antibodies in my quick explanation above (and yes, by immunology standards that was about as short as it gets!) But the tests are looking for two antibody subclasses, IgG and IgM. The IgM ones are the first that get produced in an immune response, mostly coming from the spleen, but they’re also relatively short-lived, with a half-life of five or six days. So detection of IgM against coronavirus antigens indicates a recent (or still active) infection. The IgG antibodies are more numerous in the end, though, and for many infections (measles, chickenpox, mumps, hepatitis B and more) they indicate that a person is now immune to re-infection.

So on that paper strip, the plasma will hit a band of anti-IgM antibodies, bound to the paper, and then a band of anti-IgG antibodies, and finally a band of control antibodies that react with human antibodies in general. Remember, the plasma is carrying the test patient’s antibodies that are holding onto antigens with colloidal gold particles tied to them. When these hit one of those antibody-to-antibodies zones, they’ll come to a halt there, and the colloidal gold particles will pile up enough in that zone to show you a red-pink color. So the test strip can show red lines for either IgG or IgM, both, or neither, but if there’s no red line in the control strip then something has gone wrong and the test needs to be discarded and run again with a fresh kit.

You can realize, then, that if a person shows positive for IgM only then they may well be actively infected. And if they show only IgG, they may well have gone through an infection and could be immune (more about that in a minute). Showing both, well, you’re probably on the back end of an infection? And showing neither (but with a valid control line) could mean that you haven’t been exposed to the virus at all. But wait! There are complications, because there are always complications with the immune response. It should be mentioned that if a person was infected with SARS a few years ago that they would also probably show positive in this test; I don’t think they are specific enough to distinguish although I’d be interested to hear more details. On the other side, a negative result really doesn’t mean much, because there’s always the chance that a person generated antibodies that don’t recognize the antigens that the test kit has built into it for detection. You can’t rule it out. It is also quite possible that a person has been infected but hasn’t had time to generate enough antibodies for the test to detect yet. All such kits will include a warning that negative result can’t be used to say that a person isn’t/hasn’t been infected. And they’ll also include a warning that such a kit can’t be used as the last word even if they come out positive, although to be sure it is a pretty strong indicator. And obviously, you’re not getting any information about the actual levels of antibodies (past “enough to show a red line” anyway) or how those levels might be changing.

A big question is whether this coronavirus infection will provide lasting immunity: is a person who has “seroconverted” and shows IgG against coronavirus antigens safe to go out without fear of re-infection? And that we don’t quite know yet. The record with past coronavirus pathogens is mixed. We’re going to know eventually, and it could be a key to get past this whole epidemic, but we need more data to be sure. We also don’t know how long such immunity will last, obviously. Months? Years? How many? There’s no way to speed that data collection up; we’ll find out as time goes on. An example is that many vaccinated people my age are still immune to rubella but not to measles. I had myself checked last year because of the increase in cases in the US and got re-vaccinated because I found out that my measles immunity had vanished. (Mumps I became immune to the hard way in about 1967!)

So that’s the antibody test situation. There are several companies that have developed point-of-care tests as described above and have FDA allowance (not full approval yet) for use subject to those qualifications mentioned above. Those are definitely useful here in the short term, and as we learn more about the serology and the long-term immune response could be very useful indeed. We need time, and data. As always.

98 comments on “Antibody Tests for the Coronavirus”

  1. Barry says:

    We’re going to need widespread testing w/ BOTH the current PCR-based kits (to detect active infections) and with the Ab-based kits (to evaluate potential vaccines, and to sort the naive from the immune)
    Just as childhood vaccines are currently required to get a child (or teacher) into school, evidence of immunity may be required for workers in some fields (healthcare?) in the foreseeable future.
    Wow. Good job on explaining immunology while standing on one foot.

  2. JasonP says:

    THANK YOU Derek! EXCELLENT!

    Just the primer needed to bring this topic to light!

    I’ve always heard that those who truly “know” something can explain it so anyone can understand it and this is perfect proof!

    Kudos!

  3. luysii says:

    So sad that this was obvious 3 weeks ago and we’re just getting to it now. https://luysii.wordpress.com/2020/03/12/some-sanity-and-hope-about-the-wuhan-flu/

    Just imagine how much worse things would be given our current and past states of ignorance, had not flights from China (15,000 passengers/day) been stopped 31 January. This was decried at the time as racist and not effective.

    1. Derek Lowe says:

      Oh, it’s been obvious for much longer than that, sadly. People were talking the same way about SARS, about any other new viral infection.

      1. selene says:

        hi Derek i recently had one of these tests at home to help researches, a faint igm line showed up after the 20 minutes time frame, and progressively became darker. could that be interpreted as a positive for igm or would that be a normal reaction of the kit itself?
        thank you
        selene

        1. Vivek says:

          Thank you so much for your simplified explanation of the immune system. It was like I was watching a movie. This was so simple that non science background person will also be understand.
          I have a question how can we reuse this if result shows negative? As S protein is unused in negative result, it might work if we are able to recover it.

    2. Aleksei Besogonov says:

      The decries actually were right. The travel ban delayed the spread by at most 5 days: https://science.sciencemag.org/content/early/2020/03/05/science.aba9757

      For travel bans to work they have to be complete, total and enacted very early.

        1. Aleksei Besogonov says:

          China indeed does total ban. EVERY visitor gets escorted into their home where they are isolated for 14 days, with people from “neighborhood watch” making sure you don’t go out (they glue a paper on your doorframe, if it’s torn then you’re in trouble).

          These kinds of bans work. Not the ones enacted by the US or Europe.

          1. Nick K says:

            Astonishingly, the British Government, despite decreeing a severe domestic lockdown, has not bothered to ban incoming flights from China or Northern Italy, or even check the temperatures of incoming passengers.

          2. MagickChicken says:

            I see this being called a “ban,” but I think “quarantine” is a much more apt description.

            And THAT’S why it works.

    3. aim says:

      How many US citizens and others not covered by the travel ban continued to enter the country carrying the virus? There was a long list of people excluded from the ban. The idea that it was only Chinese citizens carrying the virus is ludicrous.

      1. Barry says:

        The only people allowed in to the Country were US Citizens. How do you stop people who live here and are Citizens, like you, from being able to come into their own country?

    4. kismet says:

      Not that they can’t be effective but travel bans are probably among the least effective and most expensive measures — unless they are extreme and early. The world was successfully seeded with cases long before people considered a ban. Look at Korea, there was no magic travel ban against Chinese citizens. Most successful countries only required quarantine or banned high risk arrival (Wuhan/Hubei). Most successful countries in Asia managed chains of endemic transmission without travel bans just by testing-tracing and mild social distancing. Of course if you get too many arrivals from a high risk country this may overwhelm test and trace.

  4. charlie says:

    The reason you want to do antibody tests it to see how many people have this. Actually gives you numbers to plug into various models.

    https://www.kunc.org/post/amid-national-testing-delays-one-colorado-county-offers-covid-19-tests-everyone

    (Yes, it is a PR stunt. But this is what we need to be doing).

  5. Me says:

    Word is that IgG is detectable during active infection with SARS-CoV-2 around the same time as IgM 🙁

    Sensitivity of Ab tests is also typically lower than molecular tests.

    We’ll definitely need additional Ab tests over period of days or a confirmatory PCR before a person can be declared immune and uninfected.

    1. Alan Goldhammer says:

      @Me – there is a pre-print from China showing their serology test was as good as the RT-PCR test. “The IgM-IgG antibodies-based test exhibited a comparative superiority to the NAT for COVID-19 diagnosis, which provides an effective complement to the false negative results from NAT for SARS-CoV-2 infection diagnosis.” They looked at 130 virus positive patient as judged by clinical symptoms.

    2. x says:

      “Word is that IgG is detectable during active infection with SARS-CoV-2 around the same time as IgM”

      Nothing problematic about that. We know what it means.

  6. Alan Goldhammer says:

    According to the FDA website on EUAs, there is only one serological test listed; the rest are RT-PCR tests. https://www.fda.gov/medical-devices/emergency-situations-medical-devices/emergency-use-authorizations#covid19ivd To me it’s a crime that we have not figured out how to do massive testing here. To get people back to work reasonably safety several million tests need to be run DAILY, not just in a week. Nobel Laureate Paul Romer (economics) has spoken to this. “If” the background level of viral infection is x times (pick a number between 5-15) it means that there are a lot of people out there with antibodies who have recovered from SARS-CoV-2 who can get back to doing stuff. That number will grow every week and those people don’t need to be retested. Yes, this is expensive but it is far cheaper than keeping things shut down.

    there is a huge failure of imagination here.

    1. charlie says:

      and the Cellex test was just given an EUA yesterday (april 1).

      No CDC test.

      According the Brix not a priority of the CDC.

    2. Tony says:

      Except we don’t know how long that immunity could last or how bad reinfection will be (will it be less? Will it be worse? ) or how fast the virus could mutate beyond the immunity meaning anything. I don’t disagree that we need more testing, but we also just plain need more data.

      1. Charles H. says:

        It’s even worse than that. There are reports from China of people who have recovered from COVID-19, and then become infectious again without becoming sick.
        I don’t know *how* infectious, the article just said that after 49 days this one person (one of those “recovered”?) was still shedding live virus particles, until he was given serum.

        Now of course this may be rare, but as of now we would have a very hard time detecting it, so I think one needs to assume it isn’t *that* rare. As a result my projection (FWIW) is that eventually everyone is going to be exposed to active virus. So while testing is very important, we need to really focus on treatments and, if possible, a vaccine that’s at least moderately durable.

    3. MrRogers says:

      Yesterday, Deborah Birx challenged academic labs to get on it and start testing their clinical colleagues immediately. Apparently the appropriate antigens are now commercially available. I’ve seen multiple ads to that effect in my inbox in the last few days anyway.

  7. Mark says:

    Thanks for being a calm voice of reason.

  8. Amita says:

    Thanks for this great analysis, Derek! As of this morning (4/2), 54 serological antibody test companies have “FDA allowance” to distribute their test within the U.S. These tests have early validation in some form (although FDA does not claim to have reviewed it) and a good number were used in China or Europe. FDA requires these (non-home) tests are NOT used for sole basis of diagnosis along with many other disclaimers.
    (Source (FAQ #6): https://www.fda.gov/medical-devices/emergency-situations-medical-devices/faqs-diagnostic-testing-sars-cov-2#5e848af0ddac4

    Currently, Only ONE serological test –an IgG/IgM test made by CellexInc, a N.C. biotech– has FDA Emergency Use Authorization (EUA) which means FDA has reviewed their data and found it agreeable. This test is listed on the FDA EUA website alongside the 23 other tests with FDA EUA which until now, were all nucleic-acid based (ie the CDC test, the 5 min Abbott test, Roche, Qiagen etc).
    https://www.fda.gov/medical-devices/emergency-situations-medical-devices/emergency-use-authorizations#covid19ivd

    Based on Derek’s post and caveats, it’s pretty fun to actually read CellexInc test’s instructions with full IgG/IgM sensitivity results. Their 15 min test requires 10 uL plasma, serum or whole blood and correctly ID’s (using either IgG or IgM as positive) 120/128 (94%) of positive samples and 240/250 (96%) of negative samples, compared to a qPCR method.

    https://www.fda.gov/media/136625/download

    So far, it is the antibody test that has FDA EUA but others should also soon receive FDA EUA (although it is mostly their manufacturing/distribution capacity that will impact their adoption, not EUA).Among others, the Biomedomics test is promising for wider adoption given their partnership with both BD and Henry Schwein–a major healthcare distributor. Other tests receiving a lot of early attention are Healgen (U.S. branch of the Chinese OrientGene) and Diazyme. It appears from online that Diazyme is selling kits of 400 test for $400 ($10/test).

    For pics of CellexInc data analysis: https://twitter.com/amitaarizona/status/1245751795042758657

  9. LKN says:

    yay for tests!

  10. Zach says:

    Are you at all concerned by the fact that all but one of the 50ish antibody tests offered in the US today (listed under Q6 here: https://www.fda.gov/medical-devices/emergency-situations-medical-devices/faqs-diagnostic-testing-sars-cov-2) haven’t had their validation data reviewed by the FDA?

  11. Tuck says:

    “A big question is whether this coronavirus infection will provide lasting immunity: is a person who has “seroconverted” and shows IgG against coronavirus antigens safe to go out without fear of re-infection? And that we don’t quite know yet.”

    Indeed, there’s an added wrinkle to the whole antibody story, that may affect the usefulness of these tests:

    “Antibody-dependent enhancement (ADE) is a mechanism through which dengue viruses, feline coronaviruses, and HIV viruses take advantage of anti-viral humoral immune responses to infect host target cells. Here we describe our observations of SARS-CoV using ADE to enhance the infectivity of a HL-CZ human promonocyte cell line.”

    “Antibody-dependent SARS coronavirus infection is mediated by antibodies against spike proteins”

    https://www.sciencedirect.com/science/article/pii/S0006291X14013321

    The speculation is that this may explain why people don’t develop permanent immunity to coronaviruses which don’t mutate like influenza. The virus uses low levels of antibodies to infect cells.

    1. Toni says:

      if this were true, would it possibly mean that a 2nd infection might be more severe with another corona suptype?

      1. toni says:

        Is COVID-19 receiving ADE from other coronaviruses? :
        https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7102551/

        Abstract
        “One of the most perplexing questions regarding the current COVID-19 coronavirus epidemic is the discrepancy between the severity of cases observed in the Hubei province of China and those occurring elsewhere in the world. One possible answer is antibody dependent enhancement (ADE) of SARS-CoV-2 due to prior exposure to other coronaviruses. ADE modulates the immune response and can elicit sustained inflammation, lymphopenia, and/or cytokine storm, one or all of which have been documented in severe cases and deaths. ADE also requires prior exposure to similar antigenic epitopes, presumably circulating in local viruses, making it a possible explanation for the observed geographic limitation of severe cases and deaths.”

        For the development of a vaccine this does not mean good either.

        1. Toni says:

          Medical Countermeasures Analysis of 2019-nCoV and Vaccine Risks for Antibody-Dependent Enhancement (ADE): https://jvi.asm.org/content/94/5/e02015-19

          Findings: (…) Prior studies involving vaccine candidates for FCoV5,6 SARS-CoV-17-10 and Middle East Respiratory Syndrome coronavirus (MERS-CoV) 11 demonstrate vaccination-induced antibody-dependent enhancement of disease (ADE), including infection of phagocytic antigen presenting cells (APC).

        2. kismet says:

          The premise is wrong isn’t it? Hubei cases were not particularly severe as far as I know. Look at Italy, same thing, mortality even higher than Hubei. High mortality is probably suggestive of lack of testing rather than anything else, plus overwhelmed hospitals as a secondary factor. We need to test more to get at the true fatality rate, as it so happens testing and contact tracing will also reduce or eliminate the spread.

          1. Toni says:

            you’re probably right.

    2. Scott Stewart says:

      This is definitely a worry. Years ago there was a vaccine available for the Feline CV that under certain circumstances would be fatal to cats. A few years after the vaccine was available, the data came out that in clinical settings, the cats actually had a higher infection and death rate than unvaccinated cats.

      1. johnnyboy says:

        FIP (feline infectious peritonitis) is a deadly disease of cats due to an abnormal immune reaction to feline coronavirus. While feline coronavirus infections are generally asymptomatic or subclinical, a minority of cats develop a raging, chronic inflammatory reaction that is refractory to any anti-inflammatory or immuno-suppressive treatment, and invariably ends up killing them. I haven’t kept up with the (admittedly limited) science around mechanisms for this to happen to certain cats and not others, and it’s not a feature of other animal coronavirus infections, but it would be something to watch for as a possible sequelae of COVID-19, or as a reaction to vaccination.

  12. Daniel Barkalow says:

    Given that we’ve had plenty of Ab tests for other antigens for ages, I assume that development and validation of new tests consists of determining that the test maker got antigens that match SARS-CoV-2 and don’t match the minor coronavirus strain we’ve probably all had, and that their processing of antigens in general doesn’t somehow mess up these new antigens. On that basis, it seems like it would be quick to verify that someone who has some existing approved Ab tests (for other things) has made a valid SARS-CoV-2 Ab test without requiring the full approval process you’d want for a new manufacturer. Is that right? What actually goes into figuring out what to use as the antigen in the test and producing it reliably at scale?

  13. Zee Bendelstein says:

    Another thing: we are in the dark as to the estimated prevalence of “immunity” as defined by having the requisite antibody set. But unless our test has extraordinarily high specificity and/or the prevalence of immunity is extraordinarily high, the proportion of positive test results that actually correspond to true immunity is going to be astonishingly small (only 50% if you assume 95% sensitivity, specificity, and 5% prevalence of immunity)….That’s my math on PPV (positive predictive value). Am I missing something?

    1. Zee Bendelstein says:

      So that’s 50% of people you’re telling are safe to go out and mingle yet are not safe to go out and mingle. Of course contingent on my assumptions being correct which they may well not be. If we have 99%+ specificity and the true immunity rates are far higher this point is more or less moot.

      1. Helen Colhoun says:

        I agree as long as the true herd immunity is low then the false positive rate of current ab tests is too high for using random testing to generate estimates of the true infection fatality rate. What are your ideas for how false positives occur and how this might be improved

    2. johnnyboy says:

      In its initial application, the antibody test would most likely be dedicated to populations with a much higher incidence of the disease (hospital/care staff, people in contact with known PCR+ patients, etc…). So the prevalence assumption of 5% would in fact be much higher.
      As infection spreads, so will the prevalence of antibodies through the general population, so in time this will also get higher.
      Generally I think you’d want higher specificity and sensitivity than 95% for a proper diagnostic test, and that’s fairly easy to get with a proper lab-based ELISA or such; though for a quick-result, mass-produced test one like Derek describes, lower specs might be the necessary compromise.

      1. Zee Bendelstein says:

        Makes sense, thanks.
        I guess repeated testing might nip some of these issues in the bud too, though would of course be onerous & costly.
        And assumes some degree of independence between false results in successive tests which might negate its benefit?

  14. luysii says:

    The following article appeared in the Wall Street Journal on Thursday, 19 March 20. I thought it was important enough that I spent a few hours typing out the article verbatim to send it to family and friends (with no omissions) because it’s behind a paywall. The responsibility for posting it here is mine, and Derek is certainly free to take it down.

    It does explain why we’re in the mess we’re in. Read it and weep.

    Authors: Christopher Weaver, Betsy McKay, Brianna Abbott

    Title: America Needed Tests. The Government Failed.

    A series of blunders blinded the U. S. to the outbreak’s scale

    When cases of the new coronavirus began emerging several weeks ago in California, Washington state and other pockets of the country, U.S. public-health officials worried that this might be The Big One, emails and interviews show.

    The testing program they rolled out to combat it, though, was a small one.

    Limited testing has blinded Americans to the scale of the outbreak so far, impeding the nation’s ability to fight the virus through isolating the sick and their contacts, public health officials say. As of Wednesday evening (18 March), about 7,800 people in the U. S. had tested positive, data compiled by Johns Hopkins University show; but the Centers for Disease Control had reported only about 32,000 tests conducted at its facilities and other public-health labs. The CDC last updated its data on Tuesday (17 March), its website shows, leaving out an expected uptick in testing in recent days.

    Limited testing is also keeping patients like Justin LaBor in the dark, despite recent improvements., Mr. LaBor, 36, said he went to the emergency room at AtlantiCare Regional Medical Center in Pomona, N.J., Monday with a fever and dry cough, symptoms typical in a coronavirus infection. Doctors admitted him, but he hasn’t been tested for coronavirus, he said on Tuesday, gasping for breath over the phone.

    “Everyone just told me there were no tests and I didn’t check all the boxes for the state,” said Mr. LaBor, a social-media marketer from Elwood, N.J. referring to the criteria that state labs require patients to meet before running tests. A New Jersey health-deparment spokeswoman said the state “has sufficient lab capacity to test those who meet the testing criteria.” An AtlantiCare spokesowoman said the hospital system “is not testing patients in a widespread manner for coronavirus.”

    While the virus was quietly spreading within the U. S., the CDC had told state and local officials its testing capacity is more than adequate to meet current testing demands”, according to a Feb. 26 agency email viewed by The Wall Street Journal, part of a cache of agency communications reviewed by the Journal that sheds light on the early response. The agency’s data show that it tested fewer than 100 patients that day.

    When the CDC first dispersed test kits in early February, it shipped them to a network of state and local government labs and restricted to people with virus symptoms who had recently traveled to China, where the virus first emerged or had been exposed to a known case. Federal officials hoped the virus could be contained—even as they disputed alarms from those on the front lines that the CDC’s guidelines weren’t keeping up with the outbreak’s spread, emails between the U. S. Agency and local officials show. The Government left other laboratories on the sidelines for crucial weeks.

    The narrow effort is “a failing”, said Anthony Fauci, a government doctor who has become the de factor face of the administration’s coronavirus response, in Congressional testimony last week that for many in Washington was a wake-up call.

    Problems still persist, but more labs are beginning to do tests and manufacturers are ramping up production. “We can expect to see a marked acceleration of the availability and implementation of testing.” Dr. Fauci said in a Friday interview.

    Botched Test Kit

    CDC officials botched an initial test kit developed in an agency lab, retracting many tests. They resisted calls from state officials and medical providers to broaden testing, and health officials failed to coordinate with outside companies to ensure needed test-kit supplies, such as nasal swabs and chemical reagents, would be available, according to suppliers and heatlh officials.

    When the U.S. Food and Drug Administration, also involved in the response, finally opened testing to more outside labs, a run on limited stocks of some supplies needed for the CDC-developed test quickly depleted stores, lab operators and suppliers said. Hospital and commercial lab operators said the Government didn’t reach out to enlist their help until it was too late.

    “This was kind of a perfect storm of three separate failures,” said Tom Frieden, who directed the CDC from 2009 to 2017, citing the botched test, overstrict FDA reules and sidelined private labs. He cautioned that he didn’t have direct knowledge of details.

    Now the U. S. is testing far fewer patients than public-health and infectious-disease experts say is necessary and just a fraction as many as other countries that rolled out wide-reaching diagnostic programs. South Korea as of Tuesday (17 March) was testing up to 20,000 patients a day, more than half the total of U.S. patients who have been tested since the outbreak began.

    The test shortage hurt U.S. efforts to contain the virus, said Neil Fishman, chief medical officer at the Hospital of the University of Pennsylvania and an infectious-disease specialist.

    “If we would have had a true understanding of the extent of the disease several weeks ago, implementation of social-distancing measures could have prevented the escalation of the disease”, Dr. FIshman said, and demand for the test is now huge.

    Health-care officials say the current state of testing reflects both technical and planning failures, as well as a broader failure of imagination. Leaders including President Trump and Health and Human Services Secretary Alex Azar to envision a crisis of the scale that has now emerged, and no one stepped up to effectively coordinate among federal agencies or the private-sector labs, medical providers and manufacturers needed for a large-scale testing push, they say.

    An HHS spokeswoman said Brett Giroir, a deputy Mr. Azar put in charge of testing last week, would assume that role. She said the FDA began working with private test developers in January by sharing information about the process for approving tests. The CDC said in an email to the Journal on Monday that its work with public-health labs is meant to fill “the short-term gap until experienced commercial diagnostic manufacturers come to market.”

    ‘It Will Disappear’

    Mr. Trump repeatedly dismissed the threat of a broad U.S. outbreak, saying in late February, “One day it’s like a miracle, it will disappear.” . The next day, the first reported American death tied to Covid-19, the disease caused by the coronavirus, occurred in Washington state.

    Some White House aides learned of complaints about the availability of testing from the media, not the public-health officials in their own government, an Administration official familiar with the matter said. Only in the first week of March did discussions in a White House coronavirus task force about the testing shortfall take on a sense of urgency, the person said.

    Even the, Mr. Azar defended the testing program in television interviews including twice on ABC News that week, citing the low number of confirmed cases—at a time when almost no tests were available to detect them.

    The White House didn’t respond to requests for comment. The HHS spokeswoman said health officials are focused on efforts to increase test accessibility and point to a series of recent statement on the topic.

    At the CDC, the tone was more dire. “While leaning forward aggressively with the hope that we will be able to prevent community spread, we also are preparing for the worst,” the agency told state public health officials in a Feb. 20 email.

    The Government Accounting Office had waned federal officials in early January that its readiness for something like a pandemic fell short, a GAO official said. GAO invetigators found crisis plans that didn’t fully account for the huge role the private sector would have to play, documents show.

    The HHS spokeswoman said the agency’s coronavirus response was guided by other “well-practiced operational plans, and not the strategic one the GAO reviewed. An HHS official told the GAO in a letter dated Jan 31, that the agency had addressed their concerns and put in place policies that would “prevent early implementation challenges from becoming institutionalized.”

    In the weeks ahead, however, those very challenges did become institutionalized. The FDA first announced labs seeking to perform testing would have to submit a special application to get permission to start on Feb 4. That initially deterred some hospital and other lab operators— which normally aren’t required to submit any application—from developing tests, experts say.

    “We had considered developing a test but had been in communication with the CDC and FDA and had been told that the federal and state authorities would be able to handle everything,” Alan Wells, the medical direct for the University of Pittsburgh Medical Center’s clinical laboratories, told reporters over the weekend. He said in an interview on Monday that it later became clar that the CDC and state were overwhelmed.

    Once the CDC launched its initial test in the first week of February, the response was quickly stymied by setbacks, including flaws that forced the CDC to claw back many of the kits it had already sent out to state public-health laboratories, according to the agency and public health officials.

    An email to state public-health-lab officials later in February from the CDC said some labs had encountered “sporadic reactivity in the negative control of one of the three assay comonents.” That means the test in some cases wrongly indicated that it had detected coronavirus in samples of laboratory grade water.

    The CDC on Monday said it “has not yet determined if the problem involves the assay design or contamination. It could have been either.

    Despite news and official reports heralding the rapid spread of the virus in Japan and Hong Kong, on Feb. 22 CDC officials told state officals to refer for testing only patients showing symptoms who had travel histories in mainland China.

    When a top Minnesota epidemiologist pointed out in an email to CDC officials that agency director Robert Redfield had recently tweeted that doctors treating patients who had visited Hong Kong and Japan should consider #COVID19, the CDC’s deputy incident commander responded: “This tweet is being taken down,” And, in fact, it was. The CDC hasn’t responded to requests for comment on why the tweet was deleted.

    Hawaii’s top epidemiologist, Sara Park, chimed in on the thread, saying the bar for testing should be lowered. Dr. Park pointed out that the CDC’s own travel website said travelers from Japan with symptoms should be considered for coronavirus even though the guide lines precluded testing.

    Rather than expand testing the CDC replied that it was considering changes to its travel notices. Dr. Park didn’t respond to request for comment.

    Outside Help

    As the CDC sought to get the network of state labs up and running, it finally turned to an outside manufacturer, Integrated DNA Technologies Inc., to order a run of custom reagents —substances used in a chemical reaction — needed to identify the genetic imprint of coronavirus in late February according to the company and a timeline provided by the CDC. IDT said in a statement that it shipped the CDC’s first order on Feb. 26.

    The CDC said it signed a contract with IDT to supply reagents on Feb. 20. Contracting records show one IDT order from that day, for only about $90,000 ot testing supplies. The CDC said that contract was for coronavirus-test material. IDT denied that contract was related to the coronavirus tests.

    With IDT creating special coronavirus test kits, the federal government abruptly began to make move that would open the door to more and broader testing, including expanding its criteria for whom to test. On Feb. 26 Nancy Messonnier, a top CDC official, promised in a call with reporters that commercial labs would “be coming online soon, and a couple of days later, the FDA allowed some labs seeking to use the CDC’s testing method or developing their own to jump through fewer hoops.

    IDT would produce millions of tests worth of its coronavirus-detecting reagent over the next couple of weeks, according to the company’s statement.

    Even still, the wave of private labs joining the fight against the virus didn’t arrive on schedule. One reason was that many of the off-the-shelf supplies used in the CDC’s testing method weren’t readily available on the scale needed, the Journal found. That included both the simple products like synthetic swabs—cotton interferes with readings—used to collect mucus samples, and complex ones. Because labs copying the CDC’s test method have to use its exact chemical recipe, there has been a run on manufacturer Quiagen NV’s reagent for separating viral RNA from human mucus, one of the products used by the agency.

    The spokeswoman said the agency was providing labs with “information on alternative sources of reagents, extraction kits, swabs and more,” she said.

    Through mid-March, Qiagen spokesman Thomas Theuringer said the Dutch biotech has shipped more than twice as many units of the product in question as in the whole of 2019 to U. S. clients.

    “It’s like queuing up in line to buy toilet paper at the grocery store,” said Richard Scanlan, the medical director of the Oregon Health and Sciences University Hospital laboratory, comparing the lab’s predicament to the empty shelves consumers are facing around the country.

    Qiagen said it was rationing test kits to its “most critical customers” and acknowledged the company was struggling to meet demand. Dr. Theuringer said the factories producing the kits had ramped upto “three shifts working seven days a week.”

    Virus researcher Scott Weaver, at the University of Texas Medical Branch at Galveston, said he sent out coronavirus RNA samples needed to do the validation studies to around 50 labs. Nonetheless, he said, some of those labs were delayed as they waited for Qiagen to fill orders in order to begin testing.

    In recent days, the FDA has tried to confront the testing shortage by approving new test designs by manufacturers such as Themo Fisher Scientific Inc. and testing firms like Laborary Corporation of America Holdings Inc., and relaxing requirements for labs to prove their tests actually work and stick strictly to the CDC recipe. Thermo said Tuesday it has 1.5 million tests ready to ship.

    FDA Commissioner Stephen Hahn maintained the agency’s insistence on test accuracy has been vital. “If you don’t have that check on the test findings,” he said, “you run the risk of inaccurate test results which means you aren’t truly assessing the full scope of the outbreak.”

    Now, in a concession to demand, the agency is letting labs run tests first and prove they are accurate later—within about two weeks. On Wednesday, Mr. Trump said in a briefing that, “in case we need it,” he would invoke a Korean-War-era law called the Defense Production Act that allows the federal government to force U.S. companies to produce needed supplies.

  15. Edgar Cayce says:

    Derek you have such a way of talking about these complicated topics that makes it informative as well as entertaining to the layman. Thank you so much

    1. philip alabi says:

      True. I just sit here as I read your blog post every day, I wonder how you know so much. I want to be “like you” when I grow up (lol). I guess it is a mix of experience and just reading so much. Happy to hear any advice you’d have for a grad student.

    2. Derek Lowe says:

      Thanks! That’s just what I’m aiming for when I do a post like that.

  16. luysii says:

    The following article appeared in the Wall Street Journal 20 March ’20. It was behind a paywall, so I typed it out verbatim to send to family and friends. It explains why we’re in the mess we’re currently in. The responsibility for posting it here is mine, and Derek is certainly free to take it down.

    In the meantime read it and weep

    The following article appeared in the Wall Street Journal a week ago today on Thursday, 19 March 20. I think it’s important enough that I spent a few hours typing out the article verbatim (with no omissions) because it’s behind a paywall.

    Authors: Christopher Weaver, Betsy McKay, Brianna Abbott

    Title: Amterica Needed Tests. The Government Failed.

    A series of blunders blinded the U. S. to the outbreak’s scale

    When cases of the new coronavirus began emerging several weeks ago in California, Washington state and other pockets of the country, U.S. public-health officials worried that this might be The Big One, emails and interviews show.

    The testing program they rolled out to combat it, though, was a small one.

    Limited testing has blinded Americans to the scale of the outbreak so far, impeding the nation’s ability to fight the virus through isolating the sick and their contacts, public health officials say. As of Wednesday evening (18 March), about 7,800 people in the U. S. had tested positive, data compiled by Johns Hopkins University show; but the Centers for Disease Control had reported only about 32,000 tests conducted at its facilities and other public-health labs. The CDC last updated its data on Tuesday (17 March), its website shows, leaving out an expected uptick in testing in recent days.

    Limited testing is also keeping patients like Justin LaBor in the dark, despite recent improvements., Mr. LaBor, 36, said he went to the emergency room at AtlantiCare Regional Medical Center in Pomona, N.J., Monday with a fever and dry cough, symptoms typical in a coronavirus infection. Doctors admitted him, but he hasn’t been tested for coronavirus, he said on Tuesday, gasping for breath over the phone.

    “Everyone just told me there were no tests and I didn’t check all the boxes for the state,” said Mr. LaBor, a social-media marketer from Elwood, N.J. referring to the criteria that state labs require patients to meet before running tests. A New Jersey health-deparment spokeswoman said the state “has sufficient lab capacity to test those who meet the testing criteria.” An AtlantiCare spokesowoman said the hospital system “is not testing patients in a widespread manner for coronavirus.”

    While the virus was quietly spreading within the U. S., the CDC had told state and local officials its testing capacity is more than adequate to meet current testing demands”, according to a Feb. 26 agency email viewed by The Wall Street Journal, part of a cache of agency communications reviewed by the Journal that sheds light on the early response. The agency’s data show that it tested fewer than 100 patients that day.

    When the CDC first dispersed test kits in early February, it shipped them to a network of state and local government labs and restricted to people with virus symptoms who had recently traveled to China, where the virus first emerged or had been exposed to a known case. Federal officials hoped the virus could be contained—even as they disputed alarms from those on the front lines that the CDC’s guidelines weren’t keeping up with the outbreak’s spread, emails between the U. S. Agency and local officials show. The Government left other laboratories on the sidelines for crucial weeks.

    The narrow effort is “a failing”, said Anthony Fauci, a government doctor who has become the de factor face of the administration’s coronavirus response, in Congressional testimony last week that for many in Washington was a wake-up call.

    Problems still persist, but more labs are beginning to do tests and manufacturers are ramping up production. “We can expect to see a marked acceleration of the availability and implementation of testing.” Dr. Fauci said in a Friday interview.

    Botched Test Kit

    CDC officials botched an initial test kit developed in an agency lab, retracting many tests. They resisted calls from state officials and medical providers to broaden testing, and health officials failed to coordinate with outside companies to ensure needed test-kit supplies, such as nasal swabs and chemical reagents, would be available, according to suppliers and heatlh officials.

    When the U.S. Food and Drug Administration, also involved in the response, finally opened testing to more outside labs, a run on limited stocks of some supplies needed for the CDC-developed test quickly depleted stores, lab operators and suppliers said. Hospital and commercial lab operators said the Government didn’t reach out to enlist their help until it was too late.

    “This was kind of a perfect storm of three separate failures,” said Tom Frieden, who directed the CDC from 2009 to 2017, citing the botched test, overstrict FDA reules and sidelined private labs. He cautioned that he didn’t have direct knowledge of details.

    Now the U. S. is testing far fewer patients than public-health and infectious-disease experts say is necessary and just a fraction as many as other countries that rolled out wide-reaching diagnostic programs. South Korea as of Tuesday (17 March) was testing up to 20,000 patients a day, more than half the total of U.S. patients who have been tested since the outbreak began.

    The test shortage hurt U.S. efforts to contain the virus, said Neil Fishman, chief medical officer at the Hospital of the University of Pennsylvania and an infectious-disease specialist.

    “If we would have had a true understanding of the extent of the disease several weeks ago, implementation of social-distancing measures could have prevented the escalation of the disease”, Dr. FIshman said, and demand for the test is now huge.

    Health-care officials say the current state of testing reflects both technical and planning failures, as well as a broader failure of imagination. Leaders including President Trump and Health and Human Services Seretary Alex Azar to envision a crisis of the scale that has now emerged, and no one stepped up to effectively coordinate among federal agencies or the private-sector labs, medical providers and manufacturers needed for a large-scale testing push, they say. (Well I did as of 27 Jan as the scale was quite evident to the Chinese — see https://luysii.wordpress.com/2020/01/27/what-to-do-about-the-wuhan-flu/ —I am not sure that Trump should have seen the information, but Azar certainly should have.)

    An HHS spokeswoman said Brett Giroir, a deputy Mr. Azar put in charge of testing last week, would assume that role. She said the FDA began working with private test developers in January by sharing information about the process for approving tests. The CDC said in an email to the Journal on Monday that its work with public-health labs is meant to fill “the short-term gap until experienced commercial diagnostic manufacturers come to market.”

    ‘It Will Disappear’

    Mr. Trump repeatedly dismissed the threat of a broad U.S. outbreak, saying in late February, “One day it’s like a miracle, it will disappear.” (I’m not reading about anyone who told Trump how serious it would be, and amazingly, there have been no leaks from people who say they did, considering the leakathons of the past 4 years). The next day, the first reported American death tied to Covid-19, the disease caused by the coronavirus, occurred in Washington state.

    Some White House aides learned of complaints about the availability of testing from the media, not the public-health officials in their own government, an Administration official familiar with the matter said (oy vey, an anonymous source ). Only in the first week of March did discussions in a White House coronavirus task force about the testing shortfall take on a sense of urgency, the person said.

    Even the, Mr. Azar defended the testing program in television interviews including twice on ABC News that week, citing the low number of confirmed cases—at a time when almost no tests were available to detect them. (What staggering incompetence!)

    The White House didn’t respond to requests for comment. The HHS spokeswoman said health officials are focused on efforts to increase test accessibility and point to a series of recent statement on the topic.

    At the CDC, the tone was more dire. “While leaning forward aggressively with the hope that we will be able to prevent community spread, we also are preparing for the worst,” the agency told state public health officials in a Feb. 20 email.

    The Government Accounting Office had waned federal officials in early January that its readiness for something like a pandemic fell short, a GAO official said. GAO invetigators found crisis plans that didn’t fully account for the huge role the private sector would have to play, documents show.

    The HHS spokeswoman said the agency’s coronavirus response was guided by other “well-practiced operational plans, and not the strategic one the GAO reviewed. An HHS official told the GAO in a letter dated Jan 31, that the agency had addressed their concerns and put in place policies that would “prevent early implementation challenges from becoming institutionalized.”

    In the weeks ahead, however, those very challenges did become institutionalized. The FDA first announced labs seeking to perform testing would have to submit a special application to get permission to start on Feb 4. That initially deterred some hospital and other lab operators— which normally aren’t required to submit any application—from developing tests, experts say.

    “We had considered developing a test but had been in communication with the CDC and FDA and had been told that the federal and state authorities would be able to handle everything,” Alan Wells, the medical direct for the University of Pittsburgh Medical Center’s clinical laboratories, told reporters over the weekend. He said in an interview on Monday that it later became clar that the CDC and state were overwhelmed.

    Once the CDC launched its initial test in the first week of February, the response was quickly stymied by setbacks, including flaws that forced the CDC to claw back many of the kits it had already sent out to state public-health laboratories, according to the agency and public health officials.

    An email to state public-health-lab officials later in February from the CDC said some labs had encountered “sporadic reactivity in the negative control of one of the three assay comonents.” That means the test in some cases wrongly indicated that it had detected coronavirus in samples of laboratory grade water.

    The CDC on Monday said it “has not yet determined if the problem involves the assay design or contamination. It could have been either.

    Despite news and official reports heralding the rapid spread of the virus in Japan and Hong Kong, on Feb. 22 CDC officials told state officals to refer for testing only patients showing symptoms who had travel histories in mainland China.

    When a top Minnesota epidemiologist pointed out in an email to CDC officials that agency director Robert Redfield had recently tweeted that doctors treating patients who had visited Hong Kong and Japan should consider #COVID19, the CDC’s deputy incident commander responded: “This tweet is being taken down,” And, in fact, it was. The CDC hasn’t responded to requests for comment on why the tweet was deleted.

    Hawaii’s top epidemiologist, Sara Park, chimed in on the thread, saying the bar for testing should be lowered. Dr. Park pointed out that the CDC’s own travel website said travelers from Japan with symptoms should be considered for coronavirus even though the guide lines precluded testing.

    Rather than expand testing the CDC replied that it was considering changes to its travel notices. Dr. Park didn’t respond to request for comment.

    Outside Help

    As the CDC sought to get the network of state labs up and running, it finally turned to an outside manufacturer, Integrated DNA Technologies Inc., to order a run of custom reagents —substances used in a chemical reaction — needed to identify the genetic imprint of coronavirus in late February according to the company and a timeline provided by the CDC. IDT said in a statement that it shipped the CDC’s first order on Feb. 26.

    The CDC said it signed a contract with IDT to supply reagents on Feb. 20. Contracting records show one IDT order from that day, for only about $90,000 ot testing supplies. The CDC said that contract was for coronavirus-test material. IDT denied that contract was related to the coronavirus tests.

    With IDT creating special coronavirus test kits, the federal government abruptly began to make move that would open the door to more and broader testing, including expanding its criteria for whom to test. On Feb. 26 Nancy Messonnier, a top CDC official, promised in a call with reporters that commercial labs would “be coming online soon, and a couple of days later, the FDA allowed some labs seeking to use the CDC’s testing method or developing their own to jump through fewer hoops.

    IDT would produce millions of tests worth of its coronavirus-detecting reagent over the next couple of weeks, according to the company’s statement.

    Even still, the wave of private labs joining the fight against the virus didn’t arrive on schedule. One reason was that many of the off-the-shelf supplies used in the CDC’s testing method weren’t readily available on the scale needed, the Journal found. That included both the simple products like synthetic swabs—cotton interferes with readings—used to collect mucus samples, and complex ones. Because labs copying the CDC’s test method have to use its exact chemical recipe, there has been a run on manufacturer Quiagen NV’s reagent for separating viral RNA from human mucus, one of the products used by the agency.

    The spokeswoman said the agency was providing labs with “information on alternative sources of reagents, extraction kits, swabs and more,” she said.

    Through mid-March, Qiagen spokesman Thomas Theuringer said the Dutch biotech has shipped more than twice as many units of the product in question as in the whole of 2019 to U. S. clients.

    “It’s like queuing up in line to buy toilet paper at the grocery store,” said Richard Scanlan, the medical director of the Oregon Health and Sciences University Hospital laboratory, comparing the lab’s predicament to the empty shelves consumers are facing around the country.

    Qiagen said it was rationing test kits to its “most critical customers” and acknowledged the company was struggling to meet demand. Dr. Theuringer said the factories producing the kits had ramped upto “three shifts working seven days a week.”

    Virus researcher Scott Weaver, at the University of Texas Medical Branch at Galveston, said he sent out coronavirus RNA samples needed to do the validation studies to around 50 labs. Nonetheless, he said, some of those labs were delayed as they waited for Qiagen to fill orders in order to begin testing.

    In recent days, the FDA has tried to confront the testing shortage by approving new test designs by manufacturers such as Themo Fisher Scientific Inc. and testing firms like Laborary Corporation of America Holdings Inc., and relaxing requirements for labs to prove their tests actually work and stick strictly to the CDC recipe. Thermo said Tuesday (17 March) it has 1.5 million tests ready to ship.

    FDA Commissioner Stephen Hahn maintained the agency’s insistence on test accuracy has been vital. “If you don’t have that check on the test findings,” he said, “you run the risk of inaccurate test results which means you aren’t truly assessing the full scope of the outbreak.”

    Now, in a concession to demand, the agency is letting labs run tests first and prove they are accurate later—within about two weeks. On Wednesday, Mr. Trump said in a briefing that, “in case we need it,” he would invoke a Korean-War-era law called the Defense Production Act that allows the federal government to force U.S. companies to produce needed supplies.

  17. steve says:

    A couple of points from an immunologist.
    1) Testing for antibodies against SARS-CoV-2 is not enough (BTW, the virus is SARS-Cov-2; the disease is COVID-19). You want to know that the antibodies are neutralizing.
    2) When the SARS epidemic hit the first vaccines that were made caused more severe disease through a mechanism called immune enhancement. This is similar to what happens in Dengue; the first time you get infected it’s not so bad but the second time your immune system goes haywire and you get this horrible hemorrhagic fever. We need to know that immune enhancement doesn’t occur in people with antibodies to SARS-CoV-2.

    1. The control line is anti-viral spike antibody: uncaptured viral protein coated colloidal cold (or coloured latex) wicks past the antibody-class specific test lines and pitches up on the control line.

      1. Heh..meant to post this to the question below….

    2. Brian R says:

      thank you Steve,
      This point number 2 really needs to be pronounced from the rooftops, and/or whispered into DT, MP, JK’s ear (from a very USA point of view, but replace with the initial of your local leader of interest), which maybe Fauci is doing. It’s not just about whether the vaccine is safe, it’s about whether it is effective. I mean, this method that Moderna is using, with an RNA based vaccine, is, in my mind, a bit of an unproven lark. I mean, I know that’s not the only option, and enough smart immunologists are thinking about this I hope one of the approaches works. Badly. I’m holed up in the epicenter of NYC, healthy but anxious. And the SARS and MERS vaccines never made it far enough along. Except to have scepticism.

      But I mean, obviously some people are able to make neutralizing antibodies, if you believe the convalescent plasma data which I am prone to, but if we don’t have a properly designed vaccine, that can elicit neutralizing antibodies, we’d be better off injecting nothing, since the NON-neutralizing antibodies to a pathological insult are considerably worse than NO antibodies… Anyway, I know you know this, but i just think MORE people need to know this!

  18. antibody guy says:

    “The key was realizing that all of us are carrying, at all times, a gigantic combinatorial library of specialized Y-shaped glycoproteins (estimates vary, but probably around ten billion different ones out of a possible suite of a trillion or so)…”

    The theoretical diversity of the human antibody repertoire is actually 10^28. Trillions of possibilities exist via V-D-J recombination with the remainder possible through somatic hypermutation.

    1. NJBiologist says:

      The same thing jumped out at me. The deeper you dig, the *more* interesting affinity maturation gets.

  19. just curious says:

    Thank you for another very informative post.

    I have one question: can anybody explain if the virus spike coated gold particle can only be picked up by IgG and IgM, how does the control band with general human antibody display the red/pink color?

    1. The control line is anti-viral spike antibody: uncaptured viral protein coated colloidal cold (or coloured latex) wicks past the antibody-class specific test lines and pitches up on the control line.

      1. emba says:

        Something doesn’t make sense, Derek has said here that control are to ensure that the kit can detect the presence of *any* antibody, if that is negative something has gone wrong.

        If the “control line is anti-viral spike antibody”…that doesn’t make sense to me, what would that control for? It would be positive or negative depending on the presence of the spike and wouldn’t tell you if the panel is defective or has been damaged.

        1. The “control line” is to indicate that the test has functioned properly, specifically that the viral protein coated colloidal gold or coloured latex has wicked along the complete length of the nitrocellulose or paper substrate and over the class-specific anti-human immunoglobulin lines.

          In a pregnancy test, HCG in urine is first bound by mouse monoclonal anti-HCG bound to latex or gold particles before wicking along the substrate to meet a test line of anti-HCG and a control line of anti-mouse immunoglobulin. Antigen versus antibody detection, but same form of control for a correctly performed test.

          1. emba says:

            So then when Derek says:

            “…finally a band of control antibodies that react with human antibodies in general…if there’s no red line in the control strip then something has gone wrong and the test needs to be discarded and run again with a fresh kit.”

            must be incorrect. You are saying (do I understand correctly?) that it is capturing uncaptured viral protein:

            “The control line is anti-viral spike antibody: uncaptured viral protein coated colloidal cold (or coloured latex) wicks past the antibody-class specific test lines and pitches up on the control”

            That doesn’t make sense to me since you might or might not have uncaptured viral protein around depending on the sample. The documentation for a kit itself says that some NPs are coated in rabbit IgG and the control line is anti-rabit IgG. That makes sense to me:
            https://www.biomedomics.com/?fldl=3050

          2. Richa Sharma says:

            so this colored latex is plain latex or carboxylated latex that is used and what is the particle size
            As chemistry has been my subject, I am interested to know this

      2. Richa Sharma says:

        so this colored latex is plain latex or carboxylated latex that is used and what is the particle size
        As chemistry has been my subject, I am interested to know this

  20. abelpharmboy says:

    Derek, another question is whether the antibodies detected in a positive test for IgM or IgG are actually viral-inactivating antibodies. I suspect that the test antigen used in these tests is most likely some portion of the Spike (S) protein that’s required for cellular entry and the assumption is made that a positive antibody test means those selected antibodies inactivate the virus, but that’s not always the case, correct?

    1. Derek Lowe says:

      Yep, that’s my impression, too. Neutralizing antibodies are what we care about in the long term (or not so long term), and we’re not there yet characterizing those.

  21. abelpharmboy says:

    Oh, and similarly, the presence of a positive test in these capillary paper methods is qualitative and not quantitative. In looking at the package inserts for some of these tests, it seems there’s no guarantee that a positive test equates to the subject having an antibody titer that would be sufficient to ward off an infection. I recall my research institute requiring not only that I be vaccinated for hepatitis B when beginning work with human tissues, but that I also then had to be tested for abundance of hep B antibodies weeks later to assure that I met the standard for a protective antibody response. Do we run the risk of inferring that healthcare workers with a qualitatively-positive anti-SARS-CoV-2 IgG test are immune to infection before mobilizing them for the more risky procedures in healthcare?

    1. NJBiologist says:

      Qualitative tests also make assumptions about affinity/avidity of binding that may or may not be valid.

  22. Correct on both counts. Screening assays are a useful but blunt tool, detecting the entirety of the antibody response specific for the selected viral antigen. Neutralising regions tend to be conformationally dependent and may not be recognised by neutralising antibodies when the protein is presented electrostatically/hydrophobically bound to a plastic substrate.

    Antibody titre is not a reliable indicator of protection, which is dependent on the nature of the antibodies (neutralising, and of a subclass to interact with immune effector cells), and on the strength of binding – “functional avidity”. As far I’m aware, we are still a long way from identifying what constitutes a protective response with respect to antibody characteristics. Here’s hoping that analysis of the convalescent immune response will point the way forward.

    1. Derek Lowe says:

      Good point about the neutralizing antibodies’ detection by laminar flow assays. Same problem that you can have in chemical biology when you bind a protein to a surface for an SPR assay or to beads – you have to be careful that you haven’t buried or altered the binding sites and surfaces you’re interested in. . .

  23. matt says:

    Wandering off the testing path, I wonder about

    1) using perfluorinated compounds like the departed Fluosol or Oxygent) to alleviate ARDS. Here’s a paper that talks about babies in respiratory distress being treated with PFCs, on the way to trying intravenous infusions:

    https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3905038/

    I’m not thinking intravenous, I was thinking either aerosolized particles or straight liquid by invasive ventilation. I don’t have any expertise on the subject, and I wonder how much of this if any has ever been tried on humans. Also, if you had virus-binding, neutralizing antibodies or IL-6 inhibitors or other “chicken soup for the lungs” ideas haha, they might be tried later as additives to the PFC emulsion or mixture.

    I can imagine there’s not much opportunity normally to test an intervention like this, and there isn’t much time for patients/families to confront likely death and choose what may be a really risky opportunity to learn and develop a treatment…but maybe now is the time.

    2) The second question is closer to subject material of this blog…what about inhaled interventions generally? The inhaled insulin was absorbed higher up in the nasal passages, is that right? I realize normally no one really wants to mess around with the lungs (see vitamin E oil vaping). In this case, though. if you are delivering some medicine intended to help prevent or ease lung distress, doesn’t delivery by inhalation perfectly target your medicine?
    Is it possible to bind the viral spike protein tighter than it binds ACE2? Is it possible to deliver angiotensin1,7 directly to the lungs in quantities comparable to the loss from the virus occupying all the ACE2 sites? Or does the virus not inhibit formation of angiotensin1,7.

  24. JP Leonard says:

    Great overview of the subject, Derek!
    Re: “if a person was infected with SARS a few years ago that they would also probably show positive in this test; I don’t think they are specific enough to distinguish.”
    Question. They are saying that the spike protein on Covid19 is unusually well formed to match up with human cell receptors. Could that feature be exploited to select unique proteins for the antibody test?
    I have seen one test (Diazyme) with disclaimers about accuracy and another company (HenrySchein) says its test is rapid and accurate, without disclaimers AFAICS.

  25. Omar from the wire says:

    Yeah boi, big pharma be pushin thees sheet bruh, muthafackas be drankin that koool aid

  26. Juliet says:

    My PCR exam was positive, then i stayed at home and did another PCR exam 21 days after at the same time as IGG and IGM exam. The PCR is still positive, but IGM was negative and IGG positive. Am I transmissitting it ?

  27. Kate says:

    What are everyone’s thoughts re the hypothesis that the covid virus might be binding to hemoglobin, leading to ARDS and oxygen deprivation? http://www.smalldeadanimals.com/index.php/2020/04/05/wuhan-flu-21/
    Researcher in Italy also noticed patients had extremely low levels of Vitamin D3/25OHD.

    1. loupgarous says:

      Diverging slightly from your question, a clinical trial of Farxiga (dapaglifozin) to help protect the heart and kidneys in Covid-19 is underway.

      Since dapaglifozin has been shown to protect the heart and kidneys from damage in diabetes, the study is looking at how useful it would be to protect them from such damage in Covid-19. Quoting from clinicaltrials.gov:

      “Recent information on patients at risk for developing serious complications, including death, in the setting of COVID-19, indicate that those with cardiometabolic disease (hypertension, type 2 diabetes, atherosclerotic cardiovascular disease, heart failure, and/or kidney disease at baseline) are at much greater risk (Arentz et al 2020, Grasselli et al 2020). Moreover, a large proportion of these patients develop cardiovascular complications; many have markedly elevated NT-proBNP levels, and a substantial proportion have evidence of acute myocardial injury and/or acute kidney injury. SGLT2i have previously been demonstrated to have potent heart and kidney-protective effects in patients with type 2 diabetes, heart failure and/or chronic kidney disease, and may afford protection of these vital organ systems in the setting of COVID-19.”

      You got me thinking of that by asking about levels of vitamin D3, which I’ve had an endocrinologist prescribe for my own NIDDM.

      We might want to think about aggressively treating “metabolic syndrome” and the damaging processes surrounding poorly-controlled NIDDM in Covid-19 patients who have them. We already know that they are high risk factors for Covid-19, and perhaps removing their contribution to the disease process would help as much as immune modulation.

    2. loupgarous says:

      Let’s carry your thought a little farther. HbA1C might be interesting to look at in Covid-19 patients (since it measures glycated hemoglobin). Poorly-controlled blood glucose is itself a sign of trouble to come in Covid-19 patients. Does the virus have an affinity for glycated hemoglobin?

  28. Thoryke says:

    It looks as if antibodies to SARS don’t hang around even 5 years; if SARS2 evokes a similar immune response, we may be looking forward to a lifetime of booster shots if/when a vaccine becomes available. The article which makes me think this is ‘Response of Memory CD8+ T Cells to Severe Acute Respiratory Syndrome (SARS) Coronavirus in Recovered SARS Patients and Healthy Individuals’, DOI: 10.4049/jimmunol.175.1.591

    URL is in the name link above

    1. Shelagh Wilson says:

      I dont suppose there are any data from survivors of the 2002 SARS outbreak and whether they have shown any immunity to infection with Covid-19??

    2. Shelagh Wilson says:

      I dont suppose there are any data from survivors of the 2002 SARS outbreak and whether that has conferred any immunity to infection with Covid-19??

  29. luysii says:

    Testing those on the front lines (nurses, docs, EMTs, orderlies, housekeeping, cops etc.) for antibodies to the 4 or so coronaviruses which cause 15 – 30% of all common colds might lead to a way of stopping the pandemic. If these antibodies were in any protective, we’d know in a few weeks or sooner (because in this high risk group some are almost certain to get infected), then just give everyone colds and slow or stop transmission. Shades of Edward Jenner

    For more please see — https://luysii.wordpress.com/2020/04/05/a-way-to-end-the-pandemic/

  30. Mayank says:

    I am working on a COVID research, and had related dataset as well. But the dataset lacks information about igg, igm. They are important for the research perspective. Do anyone have such dataset, which have data related to igg, igm related info for covid(corona). If yes, please contact me at mamomakxnx@gmail.com. The information may turn very valuable for a noble cause. Thanks in advance.

  31. Rezia says:

    “The RIG-I receptor family also recognizes several viral RNA motifs – viral replication is necessarily going to involve lots of production of such species, so a constant .” ? Did this sentence get cut off?

  32. Yansen says:

    Hi Derek,
    I am currently developing an immuno method for detection of sera COVID19 IgG and IgM, the preliminary results are excellent compared to RT-qPCR on swabs. Both IgG and IgM are detection strongly on patients of newly diagnosed and at convalescent stage. But I have just got a question from a friend far away from me saying his neighbor was negative on 03/30/2020 for both IgG and IgM, but then on 04/10 IgG became positive and this is repeated on 04/11. The IgM however, never showed up. PCR swab test on 04/03 was negative. Does this make sense?

    1. Mark F. says:

      Yansen,

      I was wondering about this as well. How is this possible?

  33. MELISSA ARVIN says:

    How will these tests work on people that have CVID (common variable immune deficiency )? I am deficient in Igg proteins which I take monthly infusions for . However would the assumption be a negative result because I don’t produce enough of this protein naturally?

    1. Lisa Carlson says:

      I was wondering the exact same thing….or will this test assume I have had it when I may or may not have?

  34. Vayare says:

    That’s really cool content

  35. Sam says:

    Over here in the UK they’ve made a antibody test that has an “EC-Declaration of Conformity”, which I think is roughly the equivalent of having FDA approval. The company has donated the first batch if them to NHS workers – https://www.alphabiolabs.co.uk/2020/03/20/statement-from-alphabiolabs/ They say the tests are between 91-99% accurate

  36. Kevin Curry says:

    Doesn’t most people already carry antibodies to the corona virus already ? Not speaking directly about covid19 but the corona virus has been around for some time previous to the new mutation. I was wondering how would you differentiate the new from the old. If the body already carries antibodies to the corona virus how would you know that this is the new mutation of virus?

    1. Kevin Curry says:

      Could someone answer the previous question posted by myself about antibodies already existing. Curryfamily4@att.net.

    2. Derek Lowe says:

      The neutralizing antibodies bind so tightly to their targets because they recognize very specific parts of them (“epitopes”, in the lingo). And that means that they are usually extremely selective. The current coronavirus is one that has never infected humans, from everything we’ve seen. The closest relative that has done so is the SARS virus from 2002-2003 – there are very few people in the US who were exposed to that one, and it looks like people who were only had immunity for a few years, so we don’t know if that even would give anyone protection from the 2019 virus.

      There are coronaviruses that are known to infect humans, including some that cause some of the “common colds” every year. But those are not that closely related to the COVID-19 virus; antibodies that anyone might have developed to those are of no use in recognizing the current pandemic.

      So when you say “the corona virus has been around for some time”, the problem with that is that there is not just one corona virus. It’s a large family of viruses, and they infect all sorts of animals, including humans. Once in a while one of them that’s been infecting some particular species manages to jump to another, and that’s what’s happened to us this time. No one had antibodies to this one, for that reason.

      1. Layperson says:

        Apologies for a stupid question: since this coronavirus already infected bats (and has infected cats / dogs), is it possible to isolate the antibodies from that species and inject them into humans? Or do antibodies not jump species?

        1. sgcox says:

          Funny thing, we still did not isolate the animal host of this particular coronavirus, AFAIK
          The closest in the bats is 96% identity but it is still quite far from the actual transmission candidate.
          Previous cases, SARS and MERS, involved the intermediates but what specious were involved now is still a mystery.

          1. loupgarous says:

            What about civet cats? They’ve been mentioned as possible animal hosts, too?

            There’s also the issue that this coronavirus (like its relatives) poorly conserves its genome. If we had patient zero’s blood samples to work with (the human in which the original jump from a zoonosis occurred), the genetic sequence might be even closer than we’re seeing now. Selection pressure in humans probably works differently than it does in bats.

        2. Derek Lowe says:

          You run a much greater risk of setting off antibodies to the antibody when you do that, unfortunately. The first monoclonals used in human therapy had a lot of mouse background, and immune response was definitely an issue. There was a lot of effort to make fully human ones (and even those can generate an immune response in some patients). Isolating them from other species can give you some really good tools for in vitro assays, but they have big headwinds toward becoming drugs.

          One weird exception are the antibodies from camels, llamas, alpacas, etc. They have their own “chopped-off” antibodies that are missing some of the parts of the rest of mammalian ones, the so-called “nanobodies”. There’s been one approved for human use so far, and others are in the works. My impression is that overall they’re less immunogenic. They also have their own behavior in pharmacokinetics (blood levels, half-life in plasma, and so on).

  37. Miles says:

    Civit cats? You had better stay clear of that strange coffee

    1. sgcox says:

      Civet cats are for SARS. SARS-CoV-19 is more genetically diversed from SARS than from many other bat coronavius strains.
      Current virus clearly took other root – which one we still not know.

  38. Yasmeen says:

    What a refresher to immunology!! The most comprehensible science article I have read till date. Thank you so much….

  39. Lala Liya says:

    That was really good content

  40. Dai software says:

    Thanks for sharing such an amazing article, really informative.
    home services app development

  41. Vivek says:

    Thank you so much for your simplified explanation of the immune system. It was like I was watching a movie. This was so simple that non science background person will also be understand.
    I have a question how can we reuse this if result shows negative? As S protein is unused in negative result, it might work if we are able to recover it.

  42. Gdax says:

    Really thanks for sharing this great article and I love this post.

Leave a Reply to JP Leonard Cancel 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.