Charles Corfield’s COVID 19 Weekly Assessment: COVID-19 Antibody tests

Today, nVoq is sharing the second installment of Charles Corfield’s weekly update to the pandemic crisis. Charles covers testing this week and what it means for positive and negative results. Also, Charles looks at where we should be headed for clinical trials. Please check back on Monday’s for additional updates from Charles.

The topic of testing seems to be on everyone’s mind. There are two types of test kits available, one for whether you have an active infection, and the other for antibodies to see if you have had an infection in the past. Note, there is a time delay between the start of an infection and the appearance of a measurable number of antibodies (a reasonable guess is three weeks). Last Friday, a team of researchers published preliminary evaluations of 14 antibody test kits, which illustrates the problems which lie ahead. No test is perfect, and you must stop and ask yourself the question “What are the consequences if the result is mistaken?”. If you get tested by one of the antibody test kits and it says that you have antibodies, when in fact you do not, then your risk is that you stop social isolation, pick up the virus and spread it to vulnerable community members, without even realizing that you were infectious (most cases are asymptomatic).

This type of false positive is obviously problematic. If the test says you don’t have antibodies, when in fact you do, you will (likely) continue social distancing and avoid contact with vulnerable folks. There’s not much downside in this case, since you are no more likely to get it and pass it along than you were before the test. Let’s take a real-world example: suppose you live in Santa Clara County (estimated prevalence: 3-5%) and you get tested with a kit that has a 10% false positive rate, and it says you are positive. That is a pretty useless result, since 10% of the population will be rated as having had the virus, when we know that only 3-5% have it. The best you might conclude is that there are, perhaps, three test kits worth considering (i.e., < ¼ of the number on offer), or you might equally conclude that none of them are ready for prime time and we will have to wait.

I took a look at the list of registered clinical trials. What strikes me is the number of trials which should be there but aren’t. It did not take long for researchers to scour drug databases and run computer programs to identify existing compounds (i.e., ones we know how to manufacture, and we also know their side effects) which disable critical components that the virus needs to enter host cells and replicate. Why haven’t a good number of these candidates been registered for clinical trials? This suggests that we are lacking a central clearing house with the authority to parcel out promising compounds to groups of researchers to run preliminary “sniff” tests on small patient cohorts to sort out which ones should be dropped, and which should be pursued at larger scale. We need to get a crisp process in place for rapidly triaging such candidate compounds ahead of the next pandemic (e.g., suppose an as yet unknown virus learns similar tricks to smallpox?) That would be orders of magnitude worse than SARS-COV-19, whose IFR (Infection Fatality Rate) is < 1%, and time would be very much of the essence).