As I have mentioned, two large US commercial labs have been offering antibody tests to the public. In total they have performed at least 4 million of these tests, the results of which are being provided to the CDC. The first study based on these tests has now been released. (US AB Study) The tests were conducted in 6 locations from March 23 to May 3rd, but with a varying time period by site, on over 11,900 persons. Note that this is not random sampling but is based on people who sought out an antibody test, but the test results were standardized to the age/sex mix of the areas from which the samples came. The sites were Puget Sound, NYC, South Florida, and all of Missouri, Utah and Connecticut. Based on the testing, 1.13% were positive in the Puget Sound area, 1.85% in South Florida, 2.18% in Utah, 2.65% in Missouri, 4.94% in Connecticut and 6.93% in NYC. The locations with later testing dates had higher prevalences. The authors note that in most places there likely have been ten times more infections than there are reported cases. The under-reporting of cases ranged from 6 times in Connecticut to 24 in Missouri. It is particularly interesting to note the possibility that early in the epidemic in NYC there may have been a 1000 times more cases than were reported. I would assume that an updated report with more recent tests and based on a larger population would reveal even higher prevalence. And once again, if you aren’t testing for T-cells, you are likely missing a large number of infected persons and people who had pre-existing immunity.
Here is another study of antibody prevalence, this one from Spain, looking at health care workers. (Spain AB Study) Over 7500 health care workers were included in the testing and over 10% were positive for antibodies, which was far more than shown by infection testing alone.
And another good study on the important role T-cells may play both in clearing a coronavirus infection and in building lasting immunity against re-infection. (T-Cell Study) The researchers, based in Germany, examined 180 recovered infected individuals and 185 non-infected controls. The researchers were focussed on development of T-cell responses because prior research on the original SARS and MERS had shown their importance in controlling re-infection. They first looked for protein fragments that might be recognized by T-cells. The people who had been infected all had strong T-cell reactions both to protein fragments more unique to the current strain and to fragments similar to those from the common seasonal coronaviruses. An amazing 81% of the uninfected donors, however, also had T-cells that recognized the similar fragments between seasonal and COVID-19 virus, although at a lower intensity than for the infected group. Also of interest is that some of the infected people did not show measurable antibody response, but did have the T-cell response. Antibody response was correlated with disease severity, but T-cell response was not.
This study is extremely important. It demonstrates a strong adaptive immune response to infection from T-cells and it demonstrates that a large percent of the population has T-cells from past coronavirus infections that are capable of aiding in the fight against the current strain. This strongly suggests that not all the population is susceptible to infection, indeed only a minority may be, and it is the best explanation for so many asymptomatic and mild infections.