In early April, as covid-19 cases and deaths in New York City were rising to horrifying numbers, Tal Zaks, the chief medical officer of Moderna, a Cambridge-based biotech company, was concerned about time. In just three months, his company had created an experimental vaccine to inoculate against covid-19, and begun to inject the vaccine into humans, under the guidance of the National Institute of Allergy and Infectious Diseases, in a Phase I clinical trial involving forty-five healthy men and women. This kind of speed in vaccine development was unprecedented, and largely derived from the revolutionary—and yet, on a large scale, untested—biomedical technology behind the Moderna vaccine. Still, if there was any chance of getting the vaccine federally licensed, and then manufactured into hundreds of millions of doses, in twelve to eighteen months—as Anthony Fauci, the director of the N.I.A.I.D., has said is the fastest possible timeline—Zaks knew that the company would have to be able to prove the vaccine’s potential, or an “expectation of benefit,” as he put it, by this summer.

To accomplish this, Moderna would have to demonstrate three things: first, that the vaccine causes no significant adverse side effects in the healthy people dosed; second, that the vaccine can prevent disease in other mammals, such as mice and monkeys; and, third, that the vaccine induces neutralizing antibodies in trial participants’ blood, which is tested by adding inoculated blood to a petri dish and seeing if the virus is prevented from infecting and killing cells in a tissue culture. “In virology,” Zaks told me, “generating neutralizing antibodies is a pretty good surrogate of your ability to eventually protect the human being from becoming sick.” Once you have these Phase I results, “it becomes a judgment call,” he said. “When does that expectation of benefit actually become strong enough to warrant exposing more and more people to an unknown risk?”

On May 15th, Zaks received his answer in his in-box—a hundred-and-forty-page report analyzing the initial results. As he read, his smile got bigger. “It was really, really reassuring,” he said. After two doses—a first shot and a booster shot—eight of the trial participants produced neutralizing antibodies at or exceeding the levels seen in the blood of people who have recovered from covid-19. (Moderna is still awaiting data on the other thirty-seven participants from N.I.A.I.D. and its academic partners, which are carrying out the dangerous experiments in Biosafety Level 3 labs.) Across the entire group of forty-five trial participants, all of whom were between the ages of eighteen and fifty-five, the vaccine was safe and tolerated, with almost no adverse side effects. Among the two lower-dosage groups, one person had some redness around the injection site, and, in the high-dosage group, three people experienced short-lived, mild flu-like symptoms. In mice, the vaccine provided full protection against viral replication in the lungs, which researchers tested by giving mice the vaccine and then deliberately infecting them with the virus. (Scientists place a small drop of virus-laced fluid over the mouse’s nostrils. Similar trials on rhesus macaques are about to start.)

With these results, Moderna and the N.I.A.I.D. could continue with their wildly ambitious plan to start Phase II trials almost immediately, in six hundred participants, with no limit on maximum age, and then, by July, to start Phase III efficacy trials, which will likely involve upward of ten thousand people. After finishing his review of the report, Zaks jumped on a call with other Moderna executives and scientists, and also investigators at N.I.A.I.D. “The sense of excitement was palpable,” he told me in a video call on Monday, still smiling, wearing a blood-red tie. “There was clapping.” But he emphasized that there was still a long way to go. Mice are not humans. “I have to say that the degree of excitement is the same as the degree, or sense, of responsibility,” Zaks said. “Now we’ve got to get this next bit right. It’s on us.”

Stéphane Bancel, the C.E.O. of Moderna, said last week that, pending the results of the Phase III efficacy trial this summer, the vaccine could be ready for approval and licensing as soon as the fall. In the Phase III trial, one group will get a placebo and another will get the vaccine. Everyone is given the same precautions to avoid infection. Then the investigators have to wait for people to stumble into the virus or get infected; once there are significantly more cases among the placebo group, to prove that the vaccine is working, an independent review board will analyze the data and decide whether the trial is done. “The biggest risk is how quickly we get enough people infected to have significant data,” Zaks said. “Designing the trial and picking the right place to run it is going to be key now.”

New York City, where there likely still will be some community circulation of the virus this summer, is one of the locations under consideration. Zaks could not confirm yet how many cases of infection would be needed to prove that the vaccine is effective. Often, in Phase III case-driven trials of this kind, investigators and regulators want to have between sixty and a hundred cases of infection with clinical symptoms (the kind that make a person go to the doctor) in the placebo group. Ideally, there would be zero cases of infection among the vaccinated group. “Fifty-to-zero would be great. Thirty-to-zero would be good,” Zaks said. “The higher the ratio, the sooner we’re likely to be able to show that, indeed, it is a statistically significant difference, and therefore we got it right.” But most vaccines are not a hundred-per-cent effective. “Even fifty-per-cent or seventy-per-cent efficacy would be helpful here,” Barney Graham, the deputy director of the Vaccine Research Center at the N.I.A.I.D., who has been deeply involved with the Moderna vaccine’s development and clinical trials, said.

Usually, vaccine development takes a decade or longer, and the statistical chance of failure, historically, is more than ninety per cent. But Moderna, which was founded a decade ago, has invested billions to create vaccines (and also therapeutic drugs) using messenger RNA, which is, essentially, the code—like the zeroes and ones that drive computing—that tells cells how to build new proteins (i.e., how to grow life). To make these vaccines, all that Moderna’s researchers need to know is the atomic-level genetic sequence—the messenger RNA—of the protein they want our bodies to create. Everything else about the vaccine—the upstream supply chain (the enzymes and nucleic acids, the lipids, the plastics) and the downstream processes (purification, tests for stability and integrity of the vaccine)—is the same as it would be for any of their other vaccines. That’s why they can move so quickly. With this coronavirus, the vaccine’s mRNA instructs our cells to make the now famous spike protein—the part of the virus that is particularly adept at binding to our cells—prompting our immune system to create antibodies that can disarm it. The company has some of its own manufacturing capacity, which allowed it to make the early doses. In April, the U.S. Biomedical Advanced Research and Development Authority (barda) awarded the company nearly five hundred million dollars to accelerate development, and, on May 1st, Moderna announced that it would be collaborating with Lonza, a Swiss multinational chemical and biotech company, to manufacture up to a billion doses of the coronavirus vaccine in 2021.

Still, this vaccine will be Moderna’s first to go into Phase III trials, a historic first for mRNA vaccines and therapeutics. Fauci, during Senate testimony on May 12th, said, “There is no guarantee the vaccine is going to be effective.” He also pointed out that there have been at least two vaccines in the past that have had adverse effects; the vaccines induced the wrong sort of antibodies, and people exposed to the virus got even sicker. That is unlikely in this case, Zaks said, thanks to years of research by Barney Graham and others at the Vaccine Research Center. But it is still a theoretical possibility. At this stage, almost anything could happen. For Moderna, there will be pressure from all sides—the crisis of the pandemic itself, the Trump Administration, a depressed economy—to quickly show that the vaccine works. Despite the first round of promising evidence, Zaks knows that the company must wait for proof. “There’s going to be a tension between our ability to generate data for safety and effectiveness and our wish to use the vaccine as we’re ramping up manufacturing,” he said. “We’re going to have a pile of vaccines ahead of having actual data that proves its safety profile and benefit.” For a desperate world, the wait will be painful.