For the first time since I started writing this newsletter, I’m writing with positive developments that give me some optimism. The past few days have given us two significant pieces of good news - approved serologic tests and encouraging data on remdesivir.
Let’s talk about the serologic test today and we’ll get to remdesivir later this week.
What is a serologic test?
A serologic test is an antibody test. You may also see the term serology used. It means the same thing.
An antibody test means that your body produces antibodies to SARS-CoV-2, the virus that causes COVID-19.
There are different types of antibodies, IgM and IgG being the ones most likely to be tested for in serologic tests.
To [over]simplify what these mean, producing IgM means you’re in the acute phase (read: infectious) phase of disease. Producing IgG means that you’ve had past disease.
So why didn’t you just call it an antibody test?
I don’t know. I probably should have. But you’ll see both terms, so I figured I’d explain.
I thought we already had a bunch of antibody tests
The antibody tests that we’ve had available so far have been a mixture of inaccurate and unreliable tests that didn’t undergo proper FDA vetting.
It’s worth taking a step back to think about the way that we evaluate the accuracy of testing. There are two major concepts here to think about: sensitivity and specificity.
Sensitivity is a measure of the likelihood of diagnosing a disease that’s present. Low sensitivity means a lot of false negatives. Example: a sensitivity of 80% means 1 out of 5 negatives is a false negative.
Specificity is the extent to which a diagnostic test is specific to the disease in question. Low specificity means a lot of false positives. Example: a specificity of 90% means 1 out of 10 positives is a false positive.
The hodgepodge of antibody tests that we’ve been dealing with so far have sensitivity and specificity numbers around 80-90%.
Simply put, that degree of accuracy isn’t good enough for us to use in our decision making, either on a societal or an individual level.
The development and approval of a serologic test from Roche means that we can finally determine whether someone has been infected with that SARS-CoV-2 virus. The new test is reported to have a 99.8% specificity and a 100% sensitivity.
These are the kinds of numbers we can make decisions based on.
Abbott also has an antibody test with reported sensitivity of 100% and specificity of 99.5%. But their test isn’t FDA approved, and so until it is, I’m going to keep a healthy skepticism.
No word yet on whether Elizabeth Holmes has an antibody test in the pipeline.
Does a positive antibody test mean that I’m immune to COVID?
This is really the important question. I wish I had a simple answer.
The short answer is: it is very likely that a positive antibody test means some degree of immunity.
Unfortunately, we don’t really know what that degree of immunity will be. There are a number of different possibilities.
Best case scenario: sterilizing immunity
Sterilizing immunity means that your body is able to completely prevent a reinfection with the virus. With sterilizing immunity, no amount of exposure to COVID secretions can reinfect you. Sterilizing immunity also means that you can’t spread the disease to others.
Sterilizing immunity is the goal of vaccine development.
More limited immunity possibilities
There are other types of more limited immunity that can either protect individuals who are reinfected, public health, or both.
We can have limited immunity that leads to less viral replication once we are reinfected. This probably means both fewer symptoms and decreased spread because we produce fewer viral particles.
Limited immunity can also mean fewer symptoms but no decrease in viral replication. This would mean fewer symptoms for reinfected individuals - which is a big deal! - but not much impact on disease spread.
It’s also possible to have immunity that leads to less viral replication but no decrease in severity of disease. This would mean each person who is infected can still have an incredibly severe - and possibly life threatening - infection, but would spread the virus with less efficiency.
We’re not there yet with the science to know much about what spectrum of immunity we’ll see on a societal level.
To reiterate: it is very likely that those who develop a positive antibody test will have some degree of immunity. It’s just unclear what the nature of immunity will be, and it’s likely there will be some individual variation.
It is possible that a quantitative assessment of antibody levels (antibody titers) will predict degree of immunity, but we just don’t know yet.
How long does immunity last?
I’m sick of saying that we don’t know. But we just don’t.
Hopefully, the antibody response to COVID-19 infection provides a similar immunity duration as that for the SARS virus, which lasts for about 2-3 years.
A much less optimistic scenario would be that SARS-CoV-2 behaves more similarly to the usual coronaviruses that cause a seasonal cold, and last for less than a year, although encouragingly, this type of immunity appears to come with less severe symptoms and less shedding of the virus.
It’s likely that achieving sterilizing immunity requires activation of more parts of the immune system than just the part that makes antibodies.
I’m going to conclude this newsletter installment with optimism
We now have reliable antibody testing, which I didn’t feel comfortable saying a few weeks ago.