They say this could help speed up new vaccine development, and lead to quicker approval.
University of Oxford researchers have identified the levels of antibody protection required to prevent symptomatic coronavirus.
They've released their findings about the so-called correlates of protection – the level of antibodies needed to prevent infection.
Researchers say the correlates of protection provide enough information for regulators to assess the potency of any new vaccines that are developed, without the need for a phase three clinical trial.
Professor Sir Andrew Pollard, director of the Oxford Vaccine Group, and chief investigator on the Oxford vaccine trial, said: “There is an urgent need to increase supply of vaccines for the world, but development and approval of new vaccines takes many months.
“We hope that the use of correlates by developers and regulators could speed up the process.”
Dr Merryn Voysey, lead statistician in the Oxford Vaccine Group at the University of Oxford, said: “These results are important as they allow estimation of vaccine efficacy using blood samples from much smaller clinical trials than have been needed previously.”
Using an analysis based on Covid-19 cases detected in the UK, and immune system data from the blood samples of volunteers who took part in the AstraZeneca trials, they compared antibody levels in vaccine recipients 28 days after their second dose, and Covid-19 cases that occurred more than seven days after the blood sample was taken.
The pre-print, published om MedRxiv, suggests that higher levels of certain antibodies were identified in the blood samples.
This allowed the researchers to gauge the level of antibodies linked to different levels of protection against the virus.
It also provided estimates for a range of vaccine efficacies from 50% to 90%, using three different tests.
The study also confirms previous indications that there is no single level on any of the binding or neutralising antibody assays used, that provides full protection against Covid-19.
While the results link immune responses to expected population protection after two doses, they cannot be used to check protection of an individual who has been vaccinated or protection conferred by a single jab.