The leader of the U.S. government’s coronavirus vaccine program said AstraZeneca Plc found that 16 participants who received a placebo in its clinical trial contracted severe COVID-19, a sign that the shot could block the worst cases of disease.

The British drugmaker and its partner, the University of Oxford, said earlier Monday that none of the trial participants who received the vaccine had become severely ill, and that none of the patients in that group were hospitalized.

“That’s very important,” said Moncef Slaoui, chief scientific adviser to Operation Warp Speed, in an interview with Bloomberg News. “It’s exciting.”

Slaoui revealed the placebo-arm data after Astra and Oxford said their vaccine prevented 70 per cent of people from getting COVID-19 in large trials in the U.K and Brazil. However, many observers were confused by findings showing the shot was 62 per cent effective when given in two full doses, but 90 per cent effective when patients were given half a dose, followed later by a full dose.

“We need to understand exactly why there are those differences and hopefully identify means by which we could make all the vaccine perform at 90 per cent efficiency,” said Slaoui, an expert in immunology and former head of GlaxoSmithKline Plc’s vaccines division. While the differences could be related to a chance event, he said, “the big message is overall reasonable efficacy.”

The findings are “no disappointment,” but he added that more analysis must be done to see how the vaccine stacks up against candidates from Pfizer Inc. and Moderna Inc. Shots from those companies were shown to be roughly 95 per cent effective in preventing symptomatic disease.

AstraZeneca spokeswoman Michele Meixell said clinical-trial data will be published in a peer-reviewed medical journal in due course.

Slaoui said in the next two to three weeks the U.S. would learn more about the immune response generated by the Astra-Oxford vaccine, and would better understand why the lower dose resulted in a more significant response, which he described as counterintuitive. Typically, the higher the dose, the greater the effect on the immune system, Slaoui said.

The Astra-Oxford vaccine candidate is also being tested in the U.S., where a larger late-stage trial has seen 10,500 participants injected with both doses of the vaccine. Results from that separate study could be key for an emergency-use authorization from U.S. regulators.

The Food and Drug Administration is likely to evaluate data from both the U.S. trials and those abroad when it makes its decision, according to Slaoui, who said he spoke with Peter Marks, the director of the Center for Biologics Evaluation and Research at the agency, about the Astra-Oxford results and the path to an emergency-use authorization.

“It’s very likely that, like us at Operation Warp Speed, the FDA will want to understand why there is a difference” in efficacy levels, Slaoui said. “Clearly we’re going to need to understand that before such decision is made.”

The Astra-Oxford vaccine uses a harmless virus to carry some of the pathogen’s genetic material into cells to generate an immune response. It’s made from a weakened version of a common cold virus that’s genetically changed to make it unable to grow in humans.

Johnson & Johnson is also making what’s called an adenovirus-based vaccine. But Slaoui said the two programs are not to be confused, as the underlying science between the Astra-Oxford and J&J shots is different, as is the way in which they are being manufactured.

J&J’s experimental shot, which is also in late-stage trials in the U.S., looks better in terms of generating a higher immune response in humans, Slaoui said. “But today’s data are good in that they show a live-vector vaccine can work, and can work to a very high level,” said the Moroccan-born Belgian-American scientist.