Thread by @andreaprinzi

Thoughts on upcoming #COVID19 #serology tests:
This is actually quite a challenge! (#Diagnostics often involve a lot of complexities). There is a lot of pressure to roll these tests out, but they need to perform well, or we do more harm than good. #MicroRounds (A thread)

WHY we need these soon:
1. Contact tracing.
2. Can be used to test if a vaccine is working during a clinical trial (70 of them ongoing right now, I believe)
3. Inform public policy makers about rate of asymptomatic cases + previous infections/exposures =informed decision making

How is developing a #PCR different than developing a #serology test?
1. #PCR tests detect viral RNA/DNA (in this case RNA) and can be pretty straight-forward in terms of development
2. #Serology relies on knowing about the #SARSCoV2 structure and how the human body responds.

When someone becomes infected with #SARSCoV2 , their immune system will respond to proteins (antigens) on the viral coat, and subsequently form antibodies against these (in theory). In the case of #SARSCoV2, that pretty little spike protein is the most obvious choice...

But as described in this Lancet article (thelancet.com/journals/lance…), it's not that straight forward. There are several parts of the spike protein that could be used to develop a #diagnostic #test. Which one is best? The answer is likely whichever part of the virus is most UNIQUE:

It is important to remember that while this virus is "closest" to SARS, there are 4 seasonal #coronaviruses that circulate and give us colds every year (not 19, despite what the white house says). Having cross-reactivity with any of these would be a bad thing...

Once the viral protein that will be used for a test is decided on, it has to be produced in the lab. That, in itself, is not particularly easy either.
So, for all those out there screaming, "I don't care, I just want to be tested!!", what are the consequences of a bad test?

1. Having #SARS_CoV2 antibodies doesn't necessarily mean you're #immune. How long do the antibodies stay? Do they protect from #reinfection?
2. False positives: people are wrongly re-assured. They return to work, school, life...potentially spreading disease.

Importantly, the specificity of these tests matters and it matters WHICH populations you test. Using these tests in populations where the #prevalence of #COVID19 is low can be problematic. Check out the calculator on this site to play with the numbers: npr.org/sections/healt…

Once these tests are available, their suggested use needs to be explained well. This is where #diagnostic #stewardship comes into play. These tests will likely not be helpful for acute infection, and the timing of testing will be important to detect IgG...

For example, this study (wwwnc.cdc.gov/eid/article/26…)
demonstrates that reliable IgG seroconversion didn't happen until 2 weeks after disease onset (although this seroconversion has remained difficult to determine).

There is a lot more to unpack here, but it's important to remember that with #diagnostics, simply having the test is not enough. We have to shift our mentality away from, "there's a drug, let's use it!" and "There's a test, let's use it!" Doing this often leads to additional harm

It is the perfect time to lean into your infectious disease, microbiology and diagnostics experts at this time.
#IDtwitter #MedTwitter #MedEd @ASCLS @RodneyRohde @odie0222 @richdavisphd @ASMicrobiology @JClinMicro @GermHunterMD @SJLovesMicro #Laboratory @jesscataldi