In the Trump administration's efforts to secure coronavirus vaccines, two key bets turned out to be mistakes

• The Trump administration purchased 300 million vaccine doses from AstraZeneca in May, but declined to purchase additional doses on top of the 100 million it had arranged with Pfizer, the New York Times reported.
• The Pfizer contract was contingent on the shot being authorized by the FDA, so it was relatively low-risk. But the shot is more expensive.
• Now that Pfizer's vaccine is authorized, the US government seems to have made the wrong bet in hindsight, since other countries have bought up Pfizer's early doses.

Operation Warp Speed was always predicated on risky bets.

The US initiative to deliver coronavirus shots bankrolled the production of six promising vaccine candidates in large quantities before clinical trials were finished. That way, the US would be prepared to distribute doses right away once any candidate - which federal officials referred to as "horses" - crossed the finish line.

But the Trump administration didn't bet evenly on all six vaccines.

It bought 300 million doses from AstraZeneca in May, but initially ordered 100 million doses each of the other five. Then in the summer, the government declined to purchase more doses of Pfizer's vaccine, even when the company's trial started pulling ahead, the New York Times reported. By the fall, Pfizer's candidate was the clear frontrunner - but the government again didn't buy more.

In hindsight, those bets might have been mistakes. The Food and Drug Administration authorized Pfizer's vaccine for emergency use on Friday. AstraZeneca's US trial, meanwhile, hasn't finished; it was delayed more than a month this fall due to a serious neurological illness found in a UK participant.

Fewer Pfizer doses means a slower process of vaccinating Americans. Federal officials expect it will take through January 2021 to vaccinate America's 21 million healthcare workers and 3 million residents of long-term care facilities. That's far short of Operation Warp Speed's goal to produce and deliver 300 million vaccine doses by the start of January.

The Trump administration is now negotiating with Pfizer to order more doses - but even if that deal works out, there's a chance the company won't be able to deliver more in the US before June 2021, the New York Times reported.

"This has been a challenging process because there have been multiple conversations happening as recently as November," former FDA Commissioner Scott Gottlieb, who is now on the board of Pfizer, told CNBC. "Now they're coming back and wanting to restart those conversations when other commitments have been made in the interim."

Pfizer gave the US a low-risk, high-reward opportunity

Operation Warp Speed's contract with Pfizer was relatively low-risk compared to the rest. The government ordered 100 million doses for $1.95 billion in July, but didn't pay up front (unlike other Warp Speed contracts). Instead, the agreement was contingent on the shot being authorized by the FDA.

"It was hard to know up front which vaccine would eventually be successful," Amesh Adalja, a senior scholar at the Johns Hopkins University Center for Health Security, told Business Insider. However, he added, "it would have been a low-to-zero-risk proposition for the government to secure more doses, as payment would only be on delivery of a successful vaccine."

Pfizer also didn't accept government money for research or development - only for manufacturing and distribution. Even then, it opted to use its own delivery system, centered in Michigan and Wisconsin.

By contrast, the government gave Moderna $483 million to develop its vaccine, $472 million to advance its clinical trial, plus $1.5 billion to buy 100 million doses. With AstraZeneca, $1.2 billion in US funding went toward clinical studies, manufacturing, and the purchase of 300 million doses.

Members of the Trump administration have said they chose not to purchase additional doses of Pfizer's vaccine in July because they weren't convinced the company would meet the government's speedy distribution goals. Alex Azar, Secretary of Health and Human Services, told PBS NewsHour that Pfizer "refused to commit" to additional production or delivery timelines.

"I'm certainly not going to sign a deal with Pfizer, giving them $10 billion to buy vaccine that they could deliver to us five, 10 years hence," Azar said. "That doesn't make any sense."

The administration spoke again with Pfizer about purchasing additional doses in October, Azar told PBS NewsHour, but he said "they still resisted giving us any date by which they would do it."

Plus, Pfizer didn't reveal its late-stage trial data by the end of October, despite saying it would. Overall, the company's vaccine also seemed like a riskier bet than many others, since no country had ever authorized an mRNA vaccine before.

But in the end, Pfizer's trial results turned out to be better than experts anticipated. The two-dose regimen was found to 95% effective in preventing COVID-19 among more than 43,000 volunteers. The side effects also appeared to be temporary and mild.

So now, the US only has 100 million doses (enough for 50 million people) of its one authorized vaccine, but an agreement to buy three times that amount of a vaccine that has hit a string of setbacks.

Betting on a cheaper, easier-to-store vaccine

The US had valid reasons to purchase more of AstraZeneca's vaccine than of Pfizer's. For one, the AstraZeneca shot was cheaper: around $3 to $4 per dose. Pfizer's vaccine, meanwhile, cost $19.50 per dose.

AstraZeneca's vaccine should also be easier to store. The shot must be kept at 36 degrees to 46 degrees Fahrenheit, whereas Pfizer's must be kept at minus-94 degrees Fahrenheit. That requires vaccination sites to have deep freezers or dry ice.

In the spring, AstraZeneca also seemed to be a frontrunner in the vaccine race. The shot's creator, University of Oxford scientist Sarah Gilbert, is a veteran vaccine researcher who'd been studying the MERS coronavirus. Gilbert's MERS vaccine advanced to a phase-one clinical trial in December 2019, just as the novel coronavirus was starting to spread in China.

By April, AstraZeneca's candidate became the first to enter a human trial in Europe. Around that time, Gilbert said she was 80% confident her team could produce a successful vaccine by September. By the time the Trump administration purchased the doses in May, Warp Speed officials estimated that the first doses could be delivered as early as October.

But then in September, AstraZeneca halted its trials due to an illness in a participant.

The FDA was blindsided by the announcement, the New York Times reported. Just hours before the news became public, FDA officials had met with AstraZeneca representatives, who didn't breathe a word about the trial delay.

Then came another problem: In November, AstraZeneca's Brazil and UK trials yielded puzzling results. When the vaccine was given as intended -two doses one month apart - it was found to be 62% effective. But by mistake, 2,700 volunteers were given a half-dose, followed by a full dose. For them, the effectiveness was 90%.

AstraZeneca's US trial, which was previously just testing two full doses, will now test the half-dose regimen as well. Still, the potential for FDA authorization is murky.

"Frankly, unless there is a very clear explanation based on facts and data for what's behind those two numbers, it's very likely that that package would not be sufficient for an approval," Warp Speed's chief adviser, Moncef Slaoui, said earlier this month.

Learning from past mistakes

The Trump administration still has the option to purchase 500 million additional doses of Pfizer's vaccine, but other countries have already nabbed their own doses. The European Union finalized a deal for 200 million doses on November 11, two days after Pfizer's late-stage trial results came out.

"Contracts are forward-looking, that means we could have (and did) sign contracts with other manufacturers that reserve future capacity when it became available," Rena Conti, a health economist at Boston University, told NPR. "We should have [been] including language in every contract reserving the rights to more quantity in advance at a given price."

Moderna's mRNA vaccine is likely to get authorized later this month, which would give the US enough doses to vaccinate 20 million people before the end of the year, Slaoui estimated.

That vaccine is by far the most expensive: Moderna said in August that it would charge "some customers" $32 to $37 per dose during the pandemic. Like Pfizer's vaccine, Moderna's shot also has to be stored at subzero temperatures, though not quite as low.

An FDA advisory committee will review Moderna's trial data on Thursday. Those numbers show that among more than 30,000 volunteers, the vaccine had a 94.5% efficacy rate. Last week, the government purchased another 100 million doses of Moderna's shot - its first vaccine order since August.