Widespread testing for COVID-19 antibodies, which our immune systems make to fight off infection, can help scientists track the spread of the disease and may eventually determine who has protective immunity and is safe to go back to work. To that end, dozens of antibody tests from different manufacturers have become available.
But preliminary results from a new study at UCSF and UC Berkeley show that when it comes to accuracy, not all tests are created equal.
The study preprint, which still needs to go through peer review, shows variability in the number of false positives and negatives between the 14 different antibody tests the researchers compared. The performance evaluation used blood from known COVID-19 patients as well as samples from people who did not have the virus.
Four of the tests performed "quite well," said Dr. Caryn Bern, a professor of epidemiology and biostatistics at UCSF. “Then we had a few that had real specificity problems. So they were having 10 or 15 percent false positive results in our pre-COVID specimens. And to me, that would be a huge problem if you're doing a population survey.”
The researchers caution that more research needs to be done to determine if the presence of COVID-19 antibodies will give patients lasting immunity from the coronavirus.
— Peter Arcuni (@peterarcuni)