As coronavirus vaccines hurtle through development, scientists are getting their first look at data that hint at how well different vaccines are likely to work . The picture, so far, is murky.

Yesterday, US biotech firm Moderna revealed the first data from a human trial: its COVID-19 vaccine triggered an immune response in people, and protected mice from lung infections of the SARS-CoV-2 coronavirus. The results — announced in a press release by the Cambridge, Massachusetts-based firm — were widely interpreted as positive and sent stock markets surging. But some scientists say that because the data aren’t published, they lack the details needed to properly evaluate those claims.

Tests of other fast-tracked vaccines in monkeys have prevented lung infections in animals exposed to SARS-CoV-2 — but not in some other parts of the body. One — a vaccine being developed at the University of Oxford, UK, that is also in human trials — protected six monkeys from pneumonia, but their noses harboured as much virus as unvaccinated animals, researchers reported ¹ last week in a bioRxiv preprint . A Chinese group reported similar caveats of its vaccine’s early animal tests this month ³ .

Despite such uncertainties, all three teams are pressing ahead with clinical trials. These early studies are meant mainly to test safety, but larger clinical trials designed to determine whether the vaccines can actually protect humans from COVID-19 could report as early as this summer.

Still, the early data offer clues as to how coronavirus vaccines might generate a strong immune response. Scientists say that animal data will be crucial for understanding how coronavirus vaccines work, so that the most promising candidates can be quickly identified and later improved. “We might have vaccines in the clinic that are useful in people within 12 or 18 months,” says Dave O’Conner, a virologist at the University of Wisconsin-Madison. “But we’re going to need to improve on them to develop second- and third-generation vaccines.”

Immune response

Moderna’s vaccine, which is being co-developed with US National Institute of Allergy and Infectious Disease (NIAID) in Bethesda, Maryland, began safety testing in humans in March. The vaccine consists of mRNA instructions for the coronavirus’s spike protein; human cells churn out the foreign protein, alerting the immune system. While easy and safe to develop, such RNA-based vaccines haven’t yet been licensed anywhere in the world.

In its press release, the company reported that 45 study participants who received two doses of the vaccine developed a strong immune response to the virus: they measured virus-recognizing antibodies in 25 participants. These antibody levels were similar to or higher than those found in the blood of people who have recovered from COVID-19.

Tal Zaks, Moderna’s chief medical officer, said in presentation to investors that these antibody levels bode well for the vaccine preventing infection. “If you get to the level of people who had disease, that should be enough,” Zaks said.

But it’s not at all clear whether the responses are adequate to protect people from infection, because the data aren’t published, says Peter Hotez, a vaccine scientist at Baylor College of Medicine in Dallas. “I’m not convinced that this is really a positive result,” Hotez says. He points to a 15 May bioRxiv preprint ² that found that most people who have recovered from COVID-19 without hospitalization did not produce high levels of ‘neutralizing antibodies’, which block the virus from infecting cells. Moderna measured these potent antibodies in eight participants and found their levels to be similar to those of recovered patients.

Hotez also has doubts about the Oxford team’s first results, which found that monkeys produced modest levels of neutralizing antibodies after receiving one vaccine dose (the same regimen that is being tested in human trials). “It looks like those numbers need to be considerably higher to afford protection,” says Hotez. The vaccine is a made of a chimpanzee virus that has been genetically altered to produce a coronavirus protein.

He says that the vaccine being developed by Sinovac Biotech in Beijing seems to have elicited a more promising antibody response in macaques that received three doses, in data presented ³ in a 5 May Science paper . That vaccine is comprised of chemically inactivated SARS-CoV-2 particles.

No one yet knows the precise nature of the immune response that protects people from COVID-19, and the levels of neutralizing antibodies made by the monkeys may be good enough to protect people from infection, says Michael Diamond, a viral immunologist at Washington University in St. Louis who is a member of Moderna’s scientific advisory board. If not, a second shot would likely boost them to much higher levels. “What we don’t know is how long they’ll last,” he adds.

Animal studies

Still more questions hover over experiments showing that vaccines can protect animals from infection. Moderna said its vaccine, which was also tested in mice, stopped viral replication in the animals’ lungs. The rodents were infected with a version of the virus that was genetically modified to enable infection of mouse cells, which are not ordinarily susceptible to SARS-CoV-2, according to Zaks’ presentation. But the viruses are mutated in the protein that most vaccines, including Moderna’s, use to stimulate the immune system, and this could affect the animals’ response to infection.

The Oxford monkeys didn’t develop any signs of pneumonia after infection. But their noses had just as much SARS-CoV-2 genetic material as unvaccinated animals.

This could be because of the extremely high dose of virus the monkeys received after being given the vaccine, says Sarah Gilbert, an Oxford vaccinologist who co-led the study with Vincent Munster, a virologist at NIAID’s laboratories in Hamilton, Montana. Such doses ensure that the animals are infected with the virus, and might not replicate natural infections. The study did not measure whether the virus was still infectious, Diamond says, and it could represent virus particles inactivated by the monkeys’ immune response or the viruses the researchers administered and not an ongoing infection.

Still, the result is “a concern” that raises the possibility that vaccinated people could still spread the virus, says Douglas Reed, an aerobiologist at the University of Pittsburgh in Pennsylvania. “Ideally, you want a vaccine that would protect against disease and against transmission, so that we can kind of break the chain,” he says.

One way to find out whether vaccines can prevent transmission would be to study them in animals that are naturally susceptible to the virus and seem capable of spreading it, such as ferrets and hamsters, says Reed. He and other researchers also wonder whether vaccines should also be trialled in animals that develop a more severe disease than macaques, whose symptoms are mild.

Safety signs

While assessing vaccines’ potential to work is difficult, the latest data are less equivocal about safety, say researchers. The Moderna vaccine caused few severe and no lasting health problems in volunteers. The Oxford and Sinovac monkeys did not develop more severe infection after being injected — a key fear because an inactivated vaccine for SARS (severe acute respiratory syndrome) showed signs of this in the animals.

Stanley Perlman, a coronavirologist at the University of Iowa in Iowa City, says that the animal studies conducted so far can tell vaccine developers only so much. “People are doing as best they can,” he says. None of the data that he’s seen should dissuade developers from pressing on with human trials in humans to determine whether they work.

Moderna will soon begin a phase II trial involving 600 participants and hopes to begin a phase III efficacy trial in July to test whether the vaccine can prevent disease in high-risk groups, such as health-care workers and people with underlying medical problems. Zaks said further animal studies, including in monkeys, were under way and that it wasn’t yet clear what animal would best predict whether and how their vaccine works.

The Oxford team has already enrolled more than 1,000 people in its UK trial. This has been giving some volunteers a placebo, which could allow researchers to determine whether the vaccine works in humans in the coming months. The team’s monkey study was reassuring because it did not find any safety problems, Gilbert says.

“We don’t really need any more data from animal trials to continue,” she says. “If we get human efficacy, we’ve got human efficacy, and that’s what matters.”