Science’s COVID-19 reporting is supported by the Pulitzer Center and the Heising-Simons Foundation.
In the third in a series of conflicting announcements about a Chinese-made COVID-19 vaccine candidate, Brazilian researchers reported today that the results of their efficacy trial were less impressive than they claimed last week. When analyzed by stricter criteria than used earlier, the vaccine’s efficacy against all forms of COVID-19, including mild cases, dropped from about 78% to 50%.
The vaccine still appeared to give nearly 100% protection against disease severe enough to require hospitalization, although the trial amassed too few of those cases for that result to reach statistical significance. Commenting at the press conference today, microbiologist Natália Pasternak Taschner, president of the Question of Science Institute, a Brazilian nonprofit that aims to support public policy based on scientific evidence, said it was a “clear and clean study” and stressed that she wants the vaccine for herself and her parents. “We do not need to say this is the best vaccine in the world,” said Pasternak, who was not involved with the trial. “We have to say that this is our vaccine and it is a good vaccine to start the process of pandemic control.”
The vaccine, made by China’s Sinovac Biotech, contains a chemically inactivated version of SARS-CoV-2, the virus that causes COVID-19. At last week’s press conference that announced the 78% efficacy, Dimas Tadeu Covas, head of the Butantan Institute—a state-owned vaccinemaker in São Paulo that is conducting the study with Sinovac—offered few details. (Information was even sketchier in a press conference last month, when the team claimed only that the efficacy was greater than 50%.) But when pressed by journalists, he responded that this number came from about 220 cases of symptomatic disease. Today, researchers revealed that the 78% figure actually derived from analyzing COVID-19 cases in which people sought assistance for symptoms and tested positive for the virus—31 such cases among placebo recipients in the trial and seven in the vaccinated group.
A more detailed analysis that included trial participants who tested positive but had symptoms so mild they did not seek help found 167 cases of COVID-19 in the placebo group and 85 in people who received the vaccine, for an estimated efficacy of 50.34%. Even allowing for statistical uncertainties, that is considerably below the greater than 90% efficacies of three other already authorized COVID-19 vaccines.
But Esper Kallas, a study investigator based at the University of São Paulo’s main campus who spoke at the press conference, notes that because definitions of a positive case have varied among trials, “There may not be an accurate comparability between the studies.” In the Brazil trial, the liberal definition of a case “put things to the most difficult test,” said Ricardo Palácios of Butantan, who served as the clinical director of the study. “It is like comparing someone who runs a kilometer on paved ground versus someone running 1 kilometer with obstacles.”
World Health Organization guidelines and those of several countries say a COVID-19 vaccine that has 50% efficacy with a lower bound of 30% meets the standards for initial authorization. “It has all the requirements to justify its emergency use,” Covas said. Brazil has one of the most devastating COVID-19 epidemics in the world, and Covas and several other speakers at the press conference said they hoped the Sinovac vaccine—which the Butantan is now producing—would promptly go into widespread use. No vaccine has yet received authorization in Brazil, but regulators are also reviewing efficacy data in one made by AstraZeneca and the University of Oxford.
Covas said the reporting of the new data on the Chinese vaccine shows “our complete commitment and transparency in this process.” Earlier, Sinovac had asked Butantan not to release details of the study so the company could more closely analyze the data, leading to assertions that the Brazilian researchers had been muzzled by the company. Sinovac has not responded to repeated requests for comment from Science.
Kallas notes the current case analysis only goes through 28 November 2020 and “much more” data are now available. Efficacy trials of the Sinovac COVID-19 vaccine are also underway in Turkey and Indonesia and have presented data that are even more preliminary. Researchers in Turkey reported on 24 December that they found 91.5% efficacy with its first 29 cases, and Indonesia reported 65% efficacy with 25 cases. Indonesia yesterday gave the vaccine emergency use authorization.
Epidemiologist Ethel Maciel at the Federal University of Espírito Santo said the choice to release the vaccine data in different press conferences with vague numbers was harmful. “People started to be suspicious and were left with a lot of doubts that encouraged public mistrust on the vaccines,” says Maciel, who works on the federal vaccination plan.
“We need more than ever to reduce moderate and severe cases, and for this purpose, the vaccine is very good,” she says. “But we will need a lot of people vaccinated to reduce transmission.”