In the last week we have seen an antibody test for Covid-19 finally being approved by public health authorities and another going on sale through a high street pharmacy, yet the president of the Rockefeller Foundation, warns against using antibody tests to alter behaviour:
“If we start using the results in a way that’s not backed up by science, people could end up doing some risky things. It is not a licence to do anything that you’re not already allowed to do.” (https://on.ft.com/2ZnNmB)
So where do we stand with antibody testing? Is it of any use or is big and little pharma companies milking the crisis?
The test produced by Roche that has been recognised by the UK government has quoted sensitivity and specificity close to 100% for detection of anti-COVID-19 IgG. However, existing antibody tests have an overall 98% sensitivity and specificity and are widely used elsewhere including, amongst others, the Dutch military and the Vatican so what is the problem with using these?
The main problem and also the main limitation in antibody testing is the low level of positive tests found in the general population whichlimits the accuracy of any test with below 100% specificity. But by concentrating on the performance of a test at low detection levels we are in danger of not seeing the “wood for the trees” – if only 2% of the population are showing positive antibodies it is impractical to base a return to work or travel policy on positive antibodies as most of us would not have them!
It is estimated that to be a meaningful tool in allowing return to work in excess of 2/3 of the population should have immunity and we are a long way from this – even in New York state the current epicentre of the pandemic only 1 in 8 people are estimated to have antibodies.
Irrespective of the limitations of using positive antibodies to make public health decisions, a fundamental flaw in their use for individual clinical decision making is the uncertainty as to what level of immunity having antibodies affords. For an individual buying a test from Super Drug what does it mean for their risk of getting the disease now or in the future?
Until last month we had all thought that that antibodies equated with immunity but the detection of second infections has cast doubt upon this. Another problem is that many people with known infections do not develop antibodies. The WHO has very clearly stated that they cannot recommend a COVID-19 passport based upon antibodies.
Does this mean that testing for antibodies is not worthwhile?
Although there does not appear to be a large role for antibody testing in terms of clinical decision making at the present moment things are constantly evolving – compare this to government advice 6 weeks ago. The presence of antibodies does allow us to track likely contacts and may in-turn inform any fact-based decision of returning to a semblance of normality. Antibody testing allows a better interpretation of PCR data looking for the virus. The PCR test will tell us at any one point in time if the virus is present whereas the value of antibody tests are that they give us an idea of the timeframe for that infection. Using the two together provides more information for the individual and also for future planning and contact tracing. For these purposes a test with 98% sensitivity and specificity is pretty good and all the discussion in government circles around a couple of percentage points of accuracy is background noise and deflects from what should be the main point of discussion ie the role of antibody testing in general. Similarly the value of a one off antibody test, purchased from a pharmacist, is unlikely to be very meaningful unless interpreted together within the wider sphere of things.Leave a reply