The data, disclosed at a news conference Sunday, showed that once the variant became dominant in the country in November, the vaccine provided no significant protection against illness — although all the cases of disease were mild or moderate. There were 19 cases of covid-19 caused by the variant among people who received the vaccine and 20 cases among people who got a placebo. That suggests the vaccine was 10 percent effective, but the difference could have been due to chance.
"That is largely disappointing news," said Shabir Madhi, a vaccine expert at the University of the Witwatersrand in Johannesburg, who presented the data.
Because many vaccines, including other coronavirus vaccines, are more effective against severe disease, there's still hope that it could protect against the worst outcomes. But because the participants in the trial were young and healthy — their median age was 31 — researchers couldn't test whether the vaccine still works against severe disease.
“The vaccine clearly does not work against this variant for mild and moderate disease,” said Larry Corey, a virologist at the Fred Hutchinson Cancer Research Center. “Does it work against severe disease? The answer is: We don’t know.”
The preliminary data is a potential setback and bolsters concern that some of the emerging variants of the virus might be able to elude at least some of the new vaccines.
Vaccine developers say they are creating “libraries” of tweaked vaccines that they could quickly test against emerging viral variants. They say that new and improved versions of their vaccines could be tested and released within the year, if necessary.
The variant first identified in South Africa, known as B.1.351, appears to be more transmissible. It has subtle but important changes to its telltale spike protein, which the virus uses to attach to and then enter human host cells. The variant has been detected in three U.S. states, including, on Friday, Virginia, and more than 30 countries.
Thousands of new variants are circling the planet, but only a few have risen to the level of “variants of concern,” because they are more transmissible, more lethal or are suspected of being able to dodge the antibodies produced by vaccination.
The data released Sunday was limited. Scientists eagerly await a detailed preprint or publication in a scientific journal.
Researchers from the universities of Witwatersrand and Oxford examined adults between the ages of 18 and 64 across seven sites in South Africa last summer and fall. Half the participants received at least one dose of the Oxford vaccine; half received at least one dose of a placebo.
Given the median age of the participants, researchers could not produce statistically robust conclusions about whether the vaccine protected against severe covid-19 cases, hospitalizations and deaths. Few young adults get severe illness, and the number of participants in the study was small.
The Oxford vaccine looked very promising until November, with about 75 percent effectiveness — in line with its performance in other trials. But once the mutated B.1.351 variant became dominant, its protective abilities were severely eroded, at least against mild and moderate disease.
In contrast, vaccine trials from Novavax and Johnson & Johnson have shown that vaccines don’t work as well against B.1.351, but they do still work, with 50 to 60 percent efficacy.
Andrew Pollard, chief investigator on the Oxford vaccine trial, said the study “confirms that the pandemic coronavirus will find ways to continue to spread in vaccinated populations, as expected.”
Oxford vaccine developer Sarah Gilbert told the BBC that the vaccines now in use worldwide “have a reduction in efficacy against some of the variant viruses.”
“What that is looking like is that we may not be reducing the total number of cases, but there’s still protection in that case against deaths, hospitalizations and severe disease,” she told the “Andrew Marr Show.”
Gilbert said her team is developing a vaccine to protect against the variant identified in South Africa.
“It’s not quite ready to vaccinate people yet,” she said. “It’s easy to adapt the technology, develop a new vaccine, which will have to go through a small amount of clinical testing, not nearly the same amount as we had to go through last year.”
Public health officials have been relieved in recent days by preliminary reports that the Novavax and Johnson & Johnson vaccines were up to 60 percent effective against the variant identified in South Africa. That efficacy is good but substantially lower than against the original virus.
That variant contains a worrisome mutation, at a site on the virus RNA called E484K, which has drawn close scrutiny from infectious-disease experts, who have nicknamed it “Eeek.”
The “Eeek” mutation has been seen in variants in Brazil and Britain. It has also been identified in recent days in a handful of cases in the United States.
On Friday, the Oxford-AstraZeneca team reported that its vaccine might help keep people from spreading the virus, offering a hopeful but uncertain answer to one of the great remaining questions of the pandemic.
In a preprint of an article under review at the Lancet medical journal, the Oxford vaccine developers reported that based on follow-up studies of their clinical trials in Britain, which found the vaccine safe and effective, there is also “the potential for the vaccine to reduce transmission of the virus.” That data is preliminary; independent researchers await more information.