NEW YORK (NYTIMES) - The United States Food and Drug Administration's (FDA) upcoming review of the Pfizer-BioNTech coronavirus vaccine for young children is without precedent in recent history.

Next week, scientific advisers to the agency will decide whether to endorse two doses of the vaccine for children six months to four years old, before clinical trials have shown the full course - three doses - to be effective.

Such an authorisation would be a first for the agency, many experts say.

In fact, interim results from the trial suggested that two doses of the vaccine did not produce a strong immune response in children aged two to four years. Results from trials of the third dose are expected in a few weeks.

The companies applied for authorisation at the urging of the FDA, also highly unusual. The fast-moving Covid-19 pandemic has forced federal health officials to make important decisions with limited data before, and they argue that it is important to begin vaccinating young children now, before a new, potentially more dangerous variant arrives.

But the agency's review of incomplete data as a basis for authorisation has alarmed some experts.

"We've never done that before; that's what gives me some pause," said Dr Gregory Poland, founder and director of the Mayo Vaccine Research Group in Minnesota and editor-in-chief of the journal Vaccine. "I don't like that there isn't more data."

The third dose is likely to build up immunity in young children, Dr Poland and several other experts said, but it is not guaranteed to do so. With the ebbing of the Omicron surge, many scientists feel that the agency could afford to wait for results on the third shots, which are expected in just a few weeks.

Authorising a vaccine before that may undermine the public's trust in the regulatory process and deter parents who are already anxious about immunising their children, they warn.

What if the third dose simply doesn't work and millions of parents have already given their children the first two doses?

Although children generally do not become seriously ill when infected with the coronavirus, more of them have been hospitalised during the Omicron surge than at any other point in the pandemic.

But multiple studies have shown that children who are hospitalised with Covid-19 tend to have conditions that predispose them to severe illness, including diabetes, chronic lung disease or heart problems.

Instead of authorising the vaccine for all 18 million children who are six months to 4 years old, the agency might consider recommending it only for children at high risk until more evidence becomes available, some experts said.

Still, the spate of infections during the Omicron surge has left some parents eager for a vaccine.

"On one side, parents are desperate to get their kids protected. On another side, there is extreme distrust," said Dr Natalie Dean, a biostatistician at Emory University in Atlanta. "The whole process will need to be approached with care and a lot of transparency."

Scientific advisers to the FDA will meet on Feb 15 to weigh the current data, which will be released on Feb 18. The Centres for Disease Control and Prevention (CDC) could recommend the two-dose regimen for the youngest children shortly after that.

The Joe Biden administration has promised to respect the recommendations of the advisers.

"Please know that the FDA will not cut any corners in their review process," Dr Vivek Murthy, the surgeon-general, told reporters last week. "They know that they are the gold standard that all of us rely on."

As with the booster recommendations for all adults, the push to immunise children is part of the administration's plan for the future, according to two federal officials familiar with the discussions: Omicron may be on its way out, but children should be protected before the next variant arrives.

"We are also concerned by the notable increase in reports of children experiencing Covid-19 long-haul symptoms, including in some cases children developing autoimmune diseases and Type 1 diabetes after having had Covid-19," said FDA spokesman Stephanie Caccomo.

Even if vaccination of young children begins in April, it will be summer before they have had three doses, noted Dr Diego Hijano, a paediatric infectious disease specialist at St Jude Children's Research Hospital and an investigator for the Pfizer-BioNTech trial. "For sure, by summer we may have a variant of concern that's spreading around."

But other researchers said preparing for the future was not a compelling enough reason to get ahead of the third-dose clinical trial. The risk-benefit calculus for young children now is very different from that of adults at the start of the pandemic, Dr Poland said.

"When we're making these considerations for kids, we're not making it in the smoke and fog and chaos of war," he said.

"I would, as a vaccinologist, just have to sit and think about it a little bit," Dr Poland added of the FDA's decision. "I can just guess that puzzling is going to take a lot longer for the majority of America's parents."

Vaccine trials

Evaluation of the Pfizer-BioNTech vaccine has proceeded in stages. First came the large trials of adults that delivered an efficacy of 95 per cent, laying the groundwork for the vaccine's swift authorisation for Americans aged 16 years and older.

The companies then tested the vaccine in adolescents aged 12 to 15 but opted for a 10-microgram dose, one-third of the dose for adults.

In the youngest children, the companies tested three doses - 3, 10 and 30 micrograms - and chose the lowest dose because it seemed to be safe and yet strong enough to fend off the virus.

Unlike the adult trials, the paediatric groups were too small to gauge efficacy by comparing the rate of infections in those who got the vaccine versus just saline water. The FDA instead set antibody levels in people aged 16 to 25 as the benchmark the vaccine must meet in these children.

This method, called immunobridging, is commonly used; it was the basis for the vaccine's authorisation in adolescents.

As of Jan 20, the trial had enrolled 1,570 children aged six months to two years, and 2,328 children aged two to four years, according to Pfizer. Roughly twice as many children in each group got the vaccine as received the placebo.

In results that Pfizer-BioNTech announced in December, children aged two to four years did not produce as many antibodies as adolescents and young adults - meaning that the trial did not meet the bar that the FDA had set. The investigators decided to test a third dose in all the children.

But as the trial continued through December, some children became infected with the Omicron variant, giving the companies insight into the vaccine's performance in the real world, according to a federal official familiar with the data who spoke on the condition of anonymity.

Those numbers suggest an efficacy of about 57 per cent in children aged two to four years, the official said. But there were only about 50 infections in the vaccinated and unvaccinated groups combined, so the margins of error are most likely wide. The FDA set 30 per cent as the lower bound for efficacy in its criteria.

"If the numbers are small, the lower bound may not be above the 30 per cent threshold that was used for the adult trials," Dr Dean said.

More data in support of a third dose may soon be available. "Barring any unusual situations, I think we'll be finished with the three-dose study by March," said Dr Yvonne Maldonado, an infectious diseases physician who oversees the trial at Stanford University.

She and other scientists believe that the third dose will work - that it will augment the immune response in these children.

"But what would happen in the unlikely scenario that it doesn't is unclear," said Dr Florian Krammer, an immunologist at the Icahn School of Medicine at Mount Sinai in New York. "I think the approach is messy."

If the third dose fails, the companies may need to add a fourth or even fifth dose to the regimen - as is the case for the five-dose diphtheria, tetanus and pertussis vaccine - to create a strong enough response.

The vaccine-makers could have tested two doses of 10 micrograms, the amount given to older children. But in the safety testing, that dose produced fevers in about one in five children aged two to four years, according to a federal official who is familiar with the data.

Although that rate is comparable to the rate observed in adolescents and young adults after the second dose, fevers in children younger than five years are much more concerning.

Young children with high fevers often end up in emergency rooms, and the visits may entail antibiotics, invasive tests and hospitalisations, Dr Maldonado said. Conversely, parents may mistakenly attribute a fever to the vaccine when it may be caused by something else that warrants medical attention.

"You don't want it for any age group, but especially not the little ones," Dr Maldonado said of a vaccine that produces high fevers. "If you extrapolate that to every under-five-year-old, that's a lot of fevers."

The vaccine has been safe in other age groups, and even the rare risk of heart problems in adolescents is unlikely in prepubescent children. Still, vaccine hesitancy is running strong among parents of younger children.

The FDA authorised the Pfizer-BioNTech vaccine for children aged five to 11 years in October. But only about one in five of the 28 million children in that age group in the US has received two shots of the vaccine, according to CDC data.

A vaccine for the youngest children is likely to be even more fraught, and any regulatory missteps could be used to discredit the vaccines, said Dr Angela Rasmussen, a virus expert at the Vaccine and Infectious Disease Organisation at the University of Saskatchewan in Canada.

Evaluating the vaccine now advances the authorisation only by a few weeks, she noted.

"Isn't it worth taking the time to make sure regulatory decisions are strongly supported with data, given the potential long-term stakes for vaccine confidence and uptake?" she said.