How to use, and not use, the new blood antibody tests
Thanks to everyone who made it through my initial long essay. I promise this post will be shorter.
As I write, we await the new “home” blood antibody tests for SARS-CoV-2. They should start to showing up this month.
As I argued in my long essay, these tests, which are designed to be self-administered (but are not approved for home use in the U.S. yet), are our best hope for breaking the testing logjam and, ultimately, the pandemic itself.
Will two tests be better than one? Yes. But….
The interplay between them is going to be complicated.
To the anxious public, this may not matter much. They want a binary Yes or No from a doctor or other authoritative health care professional. This is understandable; people are annoyed when told the weather forecast says 65% chance of rain, when all they want to know is if to carry an umbrella.
The blood antibody tests are not going to replace the current PCR test for clinical diagnosis, except as a stopgap.
In the long-term, however (three to six months?) the blood antibody tests will be invaluable.
But seeing that requires knowing exactly what the two tests are testing.
The PCR test, fuzzified
Let’s start with what goes on it the current PCR test, using a whiteboard-style simplification.
The PCR test takes a specimen from the nose or throat with the swab; washes it; puts it through a sort of DNA/RNA blender; after which the polymerase chain reaction amplifies millions of copies of little snippets that have the tell-tale signature of the virus RNA. At the end of the process, a laser counts the number of snippets.
The count of particles per milliliter is used to estimate the patient’s viral load, usually expressed in a more usable number, cT. “The viral load is a measure of how bright the fire is burning in an individual,” Edward Parker of the London School of Hygiene and Tropical Medicine, told the New Scientist.
So here’s a rough curve for viral load from time of infection to four weeks out:
Looking at this, we can break down a victim’s illness into stages:
- Incubation phase. This is the immediately post-infection. The virus is just starting to replicate and build up steam. Its RNA is not yet detectable and the victim is not classically infectious.
- Asymptomatic but detectable. The virus is replicating and, at some point, its RNA will be detectable by the PCR test. But, it’s entirely possible to get a false negative by testing too soon. This explains the policy of forcing people to wait until they have symptoms before getting a test. With no retro-viral drug available, that policy makes sense in terms of clinical medicine, and conserving scarce tests. But, as I argue in my essay, we need to test a large random sample that includes the asymptomatic — or simply everyone in a limited area, such as Vo, Italy or San Miguel Country Colorado. Otherwise, in terms of the ultimate scale of pandemic, we’re still flying blind, or rather on a wing and a prayer.
- Asymptomatic infectious phase. At this point, the virus has replicated to where it is beginning to “shed,” meaning it escapes from the victim’s lungs, nose and throat in sneezes, coughs, and (more controversially) talk and breath.The victim is unaware of any symptoms, or dismisses the few he or she has. Peak infectiousness is just before the onset of noticeable symptoms.
- Symptomatic infectious phase. As far as symptoms go, this is it. The PCR test returns positive. Counter-intuitively, the viral load starts to decrease after this point.
- Late-onset Danger phase. I’m inventing this phase myself, because of the growing number of anecdotes about catastrophic late-onset failures, especially in underlying-condition patients. This suggests to me some sort of immune-system storm that appears like a wildcard in the disease progression. We don’t know; obviously, research should continue.
- Recovered but still infectious phase. After a few weeks or so, the patient starts to feel better. However, until a PCR test proves otherwise by returning negative, the assumption has to be that they are still infectious.
- Recovered and clear. The patient is much better. The PCR test detects no virus RNA in nose or throat samples. Two negative tests, spaced more than 24 hours apart, are considered proof the patient has cleared the virus. A Louisiana man, Mark Frilot, 45, whose case the New York Times reported, was pronounced clear and sent home 25 days after symptoms and 40 days after he thinks what was his exposure in a Mardi Gras crowd.
Infectiousness closely tracks viral load. But here are some additional data points on infectiousness. It happens early.
- The median duration of viral shedding was 20 days in survivors.
- The longest observed duration of viral shedding in survivors was 37 days. [Both from The Lancet Infectious Diseases].
- Infectiousness started from 2.5 days before symptom onset and reached its peak at 0.6 days before symptom onset.
- The proportion of transmission before symptom onset was 44%.
- Infectiousness was estimated to decline relatively quickly within 7 days of illness onset. [PDF of article].
Enter the blood antibody test.
That’s already a lot to take in.
And now let’s complicate it with this:
If the tests get approved for home use, here’s how you use it:
What are IgM and IgG? Obviously you didn’t read my essay. But for those of in a hurry:
- IgM (“Ig” is short for “immunoglobulin”) is a general antibody whose presence in your blood means you have some kind of infection. Big but coming up: IgM is not specific to SARS-CoV-2. Meaning it could be something else, like an ordinary cold or the flu. Or not.
- IgG is specific for SARS-CoV-2. That line on the test means you have the antibody for the virus. And that it all it means. It does not imply definitively that you are about to get sick. Worry about that if you also test positive on the PCR test.
Here’s the grid of 4 possibilities, assuming the test is valid (“C,” for control):
- IgM negative, IgG negative: Not exposed, or too early.
- IgM positive, IgG negative: not good; a recent infection, but may not be by SARS-CoV-2.
- IgM positive, IgG positive: the bad news; recent infection SARS-CoV-2.
- IgM negative, IgG positive: Previous, distant infection.
The last one, in keeping with my immunity theme, is a really interesting one. That results means you’ve already had it, and are most likely immune.
Yes, I know. Debate on this to follow.
If widespread blood antibody testing turns up a lot of “previous, distant infections,” (“distant” doesn’t have to be that distant), it’s going to rewarp our conception of Covid-19 — how it got established, and how best to fight it. Just as we fear silent carriers walking among us, there may be immunes among us, too.
So let’s mess up the whiteboard:
I’ve taken the liberty of drawing the light purple arrow straight out into the future. We don’t know for sure how long the IgM antibody persists in the body at levels that confer immunity and thus prevents reinfection. So far, it seems to be at least a month or two, and counting.
A note on the Wuhan re-infection anecdotes
The fuzziness of the tests, and what the tests are testing, exactly, need to be kept in mind in considering a report on NPR on March 27, 2020, that “some Wuhan residents who had tested positive earlier and then recovered from the disease are testing positive for the virus a second time.”
Elsewhere the NPR reporters made clear that the Wuhan residents got three PCR tests: positive-negative-positive.
From which it was irresistible to speculate that they caught the disease for a second time, all over again. Which would be very depressing.
But it’s also possible something went wrong with the testing. A first, most obvious, possibility that the Wuhan residents were positive the entire time, and the second test was a false negative.
But a third possibility is, in my mind, more likely. Recall that the PCR test finds fragments of viral nucleic acid.
If it finds RNA fragments, it doesn’t follow that the virus is “live.” (Viruses aren’t alive, so “live” means intact enough to be able to replicate.)
The polymerase chain reaction is very powerful, and amplifies everything, including debris and antibody-neutralised virus.
So “post-infection false positives” on the PCR test are going to increase as more people go into recovery. An IgM negative on the blood antibody test could dispel their fears.
The classic method of determining if a virus is “live” is to grow it in tissue culture, in the modern lab equivalent of the old petri dish. This has actually been done for SARS-CoV-2. The Wuhan scientists grew the virus in samples of lung tissue taken from non-infected patients, and were able to observe, under a microscope, cytopathic effects on the lung cells.
The report is disturbing, be sure, but not enough to go on.
I’ll leave it there for now. Soon, an update on the actual status of the blood antibody kits.