By Alexander Winning
JOHANNESBURG (Reuters) – Efficacy results may not be ready for months from a trial of AstraZeneca and Oxford University’s coronavirus vaccine in South Africa, the principal investigator of the South African study told Reuters.
AstraZeneca published findings this week from trials in Brazil and Britain showing its vaccine was on average about 70% effective in preventing coronavirus infection. It is seeking regulatory approval in several countries for its shot.
However, the company has faced calls to provide more data, after it found that the efficacy was higher in a small sub-group of participants given a smaller initial dose.
A larger U.S. trial has been delayed, potentially slowing down approval there of what has long been seen as one of the leading vaccine candidates. The published findings also did not include efficacy data from South Africa, where around 2,000 people are participating in a trial.
Shabir Madhi, professor of vaccinology at the University of the Witwatersrand, said too few participants in South Africa had contracted the coronavirus at the right time — at least two weeks after receiving two doses of either the vaccine or a placebo control — to analyse the data.
In order to demonstrate 60% efficacy, about 46 trial participants needed to test positive, he said. As of early last month fewer than five had tested positive more than two weeks after the second dose, known as an “endpoint case”.
“Right now we are still somewhat thin in terms of the number of endpoint cases to do an analysis for an efficacy readout. In all likelihood we would require another resurgence, which is currently in the making in South Africa, before we would accrue an adequate number of endpoint cases,” Madhi said.
Coronavirus cases had been surging in South Africa rapidly when the study was launched in June, and a number of participants ended up testing positive too early to say whether the vaccine might have protected them. Cases then dropped off at the time when they would have produced helpful data.
Moderna and Pfizer were able to analyse their rival vaccine candidates so quickly because most of the vaccination took place before infections surged, so when the surge happened endpoint cases accrued swiftly, Madhi added.
“The time to be enrolling to these studies is not when the peak is upon you, but a good few months before that,” he said.
AstraZeneca-Oxford’s interim analysis of its British and Brazilian trial data was based on a total of 131 COVID-19 cases. Pfizer and Moderna analysed their data after just over 90 infections.
Madhi estimated his study could reach the required number of endpoint cases around February or March next year. In July he had expressed hope that researchers would be able to analyse efficacy by the end of November.
(Reporting by Alexander Winning; Additional reporting by Kate Kelland; Editing by Josephine Mason and Peter Graff)
