Serological tests to detect antibodies against novel coronavirus (SARS-CoV-2) could improve diagnosis of COVID-19 and be useful tools for epidemiological surveillance. These have been seen as a tool to issue immunity passports or certificates so that already-infected people can move around freely. There has been increasing number of serological tests, and many are being marketed for point-of-care use.
A systematic review and meta-analysis of 40 studies of antibody testing for novel coronavirus (SARS-CoV-2) has found “major weaknesses” in the evidence base for serological tests. The “evidence does not support the continued use of existing point-of-care serological tests for COVID-19”, says a study published in The British Medical Journal.
Risk of bias
The study found that the available evidence on the accuracy of serological tests is characterised by risks of bias, and estimates of sensitivity and specificity are unreliable and have limited generalisability. The evidence is “particularly weak” for point-of-care serological tests.
The team led by Dr F. Ahmad Khan from the Respiratory Epidemiology and Clinical Research Unit, Research Institute of the McGill University Health Centre, Montreal, Canada has said that caution is warranted while relying on serological tests for clinical decision making or epidemiological surveillance. And they say, “current evidence does not support the continued use of existing point-of-care tests”.
The study warns: “Our findings should also give pause to governments that are contemplating the use of serological tests — in particular, point-of-care tests — to issue immunity certificates or passports.”
A linked editorial also echoes the authors’ views on using serological tests for decision-making. The editorial says: “The key message of the review aligns with the conclusion of another systematic review published last week: serologic assays for SARS-CoV-2 antibodies, especially point-of-care tests, are not ready for widespread use by clinicians, the general public, or policy makers.”
The primary outcome of the analysis was to evaluate the overall sensitivity and specificity based on the method of serological testing — ELISA, lateral flow immunoassays (LFIAs), or chemiluminescent immunoassays (CLIAs), and immunoglobulin class (IgG, IgM or both).
The secondary outcomes of the analysis were to evaluate the sensitivity and specificity of the tests within subgroups defined by study or participant characteristics, including time since symptom onset.
The study found high risk of patient selection bias in 98% (48/49 studies) of assessments, and high or unclear risk of bias from performance or interpretation of the serological test in 73% (36/49) of studies. Only as little as 10% (4/40) of studies included outpatients.
However, the pooled sensitivity and specificity were estimated for each serological tests. The pooled sensitivity of ELISA measuring IgG or IgM was just 84.3%. But for all methods of serological testing, the sensitivity increased — 69.9% to 98.9% — when the testing was carried out at least three weeks after symptom onset compared with within the first week (from 13.4% to 50.3%).
But they warn that even when the sensitivity estimates were higher at later time points — third week or later — important false negative rates were found. “In people with COVID-19 who are tested three weeks after symptom onset, ELISA IgG will misclassify 18% as not having been infected and LFIA IgG will misclassify 30%,” they write.
“For each test method, the type of immunoglobulin being measured — IgM, IgG, or both — was not associated with diagnostic accuracy. Pooled sensitivities were lower with commercial kits and in the first and second week after symptom onset compared with the third week or later. Pooled specificities of each test method were high,” they write.
Word of caution
But the editorial cautions that pooling sensitivities makes it “difficult to determine how well tests perform at detecting antibody” in the course of illness whether undertaken early or late in the illness. The ability to identify individual tests that might perform well in testing algorithms is also hindered.
The study found many shortcomings in the case of LFIA serological tests, leading to say that LFIA should not be used beyond research and evaluation purposes.