A clinical lab technician at the UW Medicine Virology laboratory in Seattle processes patient samples for COVID-19 on March 11. Tribune News Service

Tom Avril, Tribune News Service

Researchers around the world started work on a coronavirus vaccine on Jan. 11, the moment their Chinese counterparts published the microbe’s genetic code.

At Inovio Pharmaceuticals in Plymouth Meeting, Pennsylvania, scientists made a preliminary vaccine in just a few hours — a speed that was unheard of a decade ago — and they plan to start human testing in April. And a vaccine developed by Cambridge-based Moderna Inc. proceeded to human tests even faster: four volunteers received injections last week.

But don’t expect to roll up your sleeves any time soon.

As with any new drug, a manufacturer must demonstrate both that a vaccine is safe and that it works. Add in the time needed to produce a vaccine on a mass scale, and government scientists have been predicting the whole process could take 18 months.

That would still be a remarkably short time, representing the fastest development of a vaccine yet. And despite public pressure to speed up the timetable even more, infectious disease specialists say it would be a grave mistake to cut corners. Here’s what happens during each crucial step.

Children’s Hospital of Philadelphia physician Paul A. Offit helped develop a vaccine for a microbe called rotavirus, which causes diarrhoea and vomiting. In countries with limited access to clean water, the virus used to kill hundreds of thousands of children a year as a result of dehydration.

By the time the vaccine won Food and Drug Administration approval in 2006, the entire process of research and development had lasted 26 years. Asked about the projected schedule for a coronavirus vaccine, he chuckled ruefully.

“I’m a little jealous of the turnaround time,” he said.

Those working on the coronavirus vaccines have a big advantage over Offit and his colleagues, whose vaccine was derived from a strain of rotavirus found in calves. The rotavirus team spent years working with stool samples from sick animals, purifying virus strains that could be combined with human genes to yield a safe, effective product.

On the other hand, no animals were needed to make the coronavirus vaccine candidate at Inovio. Unlike most vaccines on the market, it is not a killed or weakened form of a virus, but simply a package of genetic instructions that teach human cells how to make a fragment of the virus. This chemical package was “printed” — assembled by machine — in just a few hours.

Still, the end goal is the same: to give the human immune system a taste of an invader so it learns to fight back when faced with a real infection. And the steps to test it for safety remain largely unchanged.

Researchers start by studying a vaccine in a lab, both in individual cells in a dish and also in animals, checking to see if it provokes the proper level of immune-system activity.

Then they inject a small number of healthy human volunteers with a small dose of the vaccine, and study them for two kinds of responses. Scientists want to see that the volunteers’ immune systems will produce protective antibodies, and they want to make sure there are no harmful side effects. That could include some unforeseen overreaction by the immune system, potentially leading to a dangerous cascade of inflammation.

If all seems OK after 42 days — the amount of time mandated by regulators — researchers try again in more people, gradually increasing the dose of the vaccine. At least four or five rounds with progressively larger doses are typically needed, each time lasting another 42 days, said Offit, now director of the Vaccine Education Center at CHOP. Additional time is needed to recruit volunteers and analyze results.

“You never, never know whether something is going to work, or whether there’s a problem, until you put it in people,” he said. “You have to be humble when you go into this.”

That is especially true for vaccines, unlike drugs aimed at sick people. The eventual goal is to inject large numbers of people who are not sick and who would not become very sick even if they were to become infected. A preventive measure is no use if it turns out to be worse than the disease itself.

If early evidence suggests a vaccine is safe, researchers test it in even larger groups for what is called efficacy — essentially determining whether people who get the vaccine are less likely to become infected than those who do not.

But who gets it? For some vaccines, researchers perform a “challenge” trial: inoculating a select group of healthy volunteers, then deliberately exposing them — challenging them — with the actual virus.

For ethical reasons, that is unlikely to happen with coronavirus vaccines, as the infection can have severe consequences in a minority of patients, Offit said. Instead, a drug company would administer its vaccine to a carefully selected sample of people in an area where the virus was circulating. That typically takes longer, because the virus takes time to spread, and because researchers must follow a larger number of people. One approach is to vaccinate healthy people in a ring surrounding a geographic area with a concentration of sick people.

And as they study volunteers for signs that the vaccine is working, researchers continue to monitor them for any adverse events. Safety remains an essential part of the process throughout.

Setting up a vaccine manufacturing facility from scratch would take years. But the various companies working on coronavirus vaccines likely will get around that hurdle in several ways, and they are starting to plan for mass production well before the drugs are shown to be effective.

Inovio, the Plymouth Meeting drug company, plans to make its coronavirus vaccine using a “platform” that it has used successfully before: a package of genetic instructions called a plasmid, said chief executive officer J. Joseph Kim.

In South Korea, for example, researchers are testing one of Inovio’s plasmid-based vaccines against MERS, a different kind of coronavirus.

In addition to using proven technology, another way to speed up the process is to repurpose equipment that is idle or being used to make other drugs, said Christa Myers, a process engineer at CRB, a Kansas City, Mo.-based architecture, engineering, and construction firm that has designed or built more than 75 vaccine facilities.

Drug companies also may work with contract manufacturers for some portion of the process.

Using its network of manufacturers, Inovio could deliver up to one million doses of its coronavirus vaccine by the end of the year, said Kim. Using a two-dose regimen, that would be enough to test 500,000 people.

Among those who have forecast that the entire process could take as little as a year to 18 months is Anthony Fauci, a prominent infectious disease specialist at the National Institutes of Health.

Asked if that timetable were realistic, Kim deferred to Fauci on making a firm prediction. “Traditionally, new vaccines can take more than 10 to 15 years to get an approval,” he said. “But this is not a traditional time.”