But the results are yet to be peer reviewed. The study was posted on the MedRxiv (pronounced med-archive) preprint server, founded by Cold Spring Harbor, Yale University and the British Medical Journal.
It has also been submitted to the Journal of Infectitious Diseases for publication.
If confirmed and published, this test could help to estimate how many people have had COVID-19. This would provide a better grasp of spread and the effectiveness of community testing. It could also help to determine evidence of herd immunity.
One of the study authors is Dr Nick Coatsworth, a deputy Chief Medical Officer of Australia, who enabled access to 3000 patients who tested negative before elective surgery and who had no known contacts with the disease.
The study screened their 3,000 blood samples and after accounting for false positives, found eight had antibodies indicating they had been infected.
Co-led by Associate Professor Ian Cockburn and Professor Elizabeth Gardiner of the John Curtin School of Medical Research, the testing was done between eary June and mid-July, ahead of Melbourne’s outbreak and before testing increased in response to the second wave.
ANU’s high-throughput robotic capacity enabled the rapid evaluation of thousands of samples.
“Our best estimate is that around 0.28 per cent of Australians – one in 350 – had been infected with SARS-CoV-2 by that time, ” says Professor Cockburn.
“But that was after the first wave, when everything was still ramping up and people were still being turned away from testing. Many of the misses were likely asymptomatic.”
He says the blood test is more accurate because it analyses responses for two proteins on the surface of the virus where others only go for one. As a result its specificity and sensitivity are comparable, to or better than the best commercially avaibale tests.
Professor Gardiner says overall, to the end of June, even with the undetected cases, there was low community transmission in Australia.
“The data suggest low case numbers in Australia reflected low community transmission not inadequate testing.
“This is supported by the subsequent outbreak in Melbourne in July/August which emerged from breaches of hotel quarantine of overseas travellers rather than undetected community transmission.”
“Once vaccines come online, this test will enable us quickly to monitor responses in individuals. But then, not all vaccines will raise an antibody response.”