How a secretive Pentagon agency seeded the ground for a rapid coronavirus cure
By PAUL SONNE | The Washington Post | Published: July 30, 2020
Stars and Stripes is making stories on the coronavirus pandemic available free of charge. See other free reports here. Sign up for our daily coronavirus newsletter here. Please support our journalism with a subscription.
The scientists were working through the night over a weekend in February in their Vancouver offices, running a blood sample from an early American COVID-19 survivor through a credit card-sized device made up of 200,000 tiny chambers, hoping to help save the world.
Their mission was part of a program under the Pentagon's secretive technology research agency. The goal: to find a way to produce antibodies for any virus in the world within 60 days of collecting a blood sample from a survivor.
Established years before the current pandemic, the program was halfway done when the first case of the novel coronavirus arrived in the United States early this year. But everyone involved in the effort by the Defense Advanced Research Projects Agency (DARPA) knew their time had come ahead of schedule.
The four teams participating in the program abandoned their plans and began sprinting, separately, toward the development of an antibody for COVID-19, the disease caused by the coronavirus.
"We have been thinking about and preparing for this for a long time, and it's almost a bit surreal," said Amy Jenkins, manager of DARPA's antibody program, which is known as the Pandemic Prevention Platform, or P3. "We are very hopeful that we will at least be able to have an impact on this outbreak. We want to make a difference."
In that program and others, DARPA has quietly been seeding the ground for the United States to produce a rapid cure for a pathogen like COVID-19 for years.
The U.S. government's response to the pandemic has been impugned as slow and haphazard, with flawed test kits, limited contact tracing, insufficient protective gear, late encouragement of masks and at times baffling messages from President Donald Trump.
But DARPA's story is a counterexample of U.S. government foresight, one that began more than a decade ago with the aim of finding super-fast ways to protect American troops if they were to confront a deadly new virus in the field.
If it weren't for DARPA's investments over the past decade and earlier, largely outside the glare of Washington's partisan politics, the American race toward a vaccine and antibody therapy to stop the coronavirus most likely wouldn't be moving as quickly as it is today.
"Being at DARPA at this time ... is exciting in some ways because we get to see the research work that was funded that was done ten to fifteen years ago now really starting to pay off," acting director Peter Highnam said in a discussion with reporters on Thursday.
The first company in the United States to enter clinical trials with a vaccine for the virus was funded by DARPA. So was the second company. And the P3 program has already led to the world's first study in humans of a potential COVID-19 antibody treatment. If successful, antibody treatments would offer up to three months of immunity against COVID-19. Unlike vaccines, they could also help heal people already infected with the virus.
Some of the vaccines and antibodies linked to DARPA could be ready later this year, which would mark one of the speediest responses to a global pandemic in the history of medicine. Experts say it normally takes four to 10 years to devise, test and produce a vaccine for a new pathogen. Antibodies, which the body creates to combat a virus, have taken years to discover, let alone produce.
DARPA is far from exclusively responsible for the fast pace. Other countries, including China, where the virus originated, are also moving quickly toward a cure. So are firms unaffiliated with DARPA.
Still, the Pentagon agency played a significant role in advancing the science that is making the quick pace possible and setting a North Star for researchers.
"I think their role is very important," said James Crowe Jr., director of the vaccine center at Vanderbilt University, one of the four participants in the P3 program. "The reason is they have catalyzed more rapid progression than otherwise would have happened. And the reason they accomplished it was they were willing to state ... grand challenges."
Crowe said he and his colleagues laughed when DARPA put out its request in 2017 for a system that could produce a human-ready antibody from a convalescent blood sample in 60 days.
"Somehow, in setting that aspirational goal, after the first brainstem reaction of, 'That's ridiculous,' the next step is, 'Well, how close could we get?' " Crowe said. "Then, you start buying in to the belief that it could be possible."
Sense of urgency
Established in 1958 in response to the Soviet Union's launch of Sputnik, DARPA was created by President Dwight Eisenhower out of a sense of urgency.
Washington could have sent the first satellite to space, but Moscow got there first — and it wasn't because the United States lacked the science. The American government simply didn't move fast enough.
DARPA was the answer to that problem.
The nimble military science research agency wouldn't invent things itself. Rather, its officials would look across the American scientific landscape — to universities, military labs and defense contractors — and channel emerging technologies into risky mega-endeavors to prevent another Sputnik. The agency's pie-in-the-sky projects would have a high risk of failure, but if successful, would transform the U.S. military and possibly society, too.
Over the years, DARPA-funded projects have created the building blocks of GPS, the first computer mouse and the protocols that underpin the modern internet. The agency pioneered stealth technology that made American fighter jets all but invisible to enemy radar. And it advanced a bevy of new weaponry, including drones.
In the years after the attacks of Sept. 11, 2001, a series of anthrax attacks, combined with overseas intelligence about potential biological threats, heightened fears of bioterrorism and drove DARPA to invest in faster ways to respond, including technology to accelerate vaccine development, spot emerging viruses and speed up pharmaceutical manufacturing.
A decade ago, a brainy Air Force doctor named Dan Wattendorf helped push rapid pandemic response further to the top of DARPA's priority list.
Regularly citing the 1918 flu pandemic, the DARPA program manager saw how a novel pathogen, whether from another species or an enemy's lab, could cripple the American military in the field.
"If we need to deploy someone in harm's way and it's a new virus, you don't have time to wait for a new vaccine," Wattendorf said. "That could be a decade."
Wattendorf had ideas for a solution. In 2010, he took to a conference room at DARPA headquarters in northern Virginia with notes scribbled on his hand to make a pitch.
At the time, the Obama administration was emphasizing the need to step up pandemic response capabilities in the wake of the H1N1 outbreak, and DARPA was increasingly focusing on biology — an emphasis that would lead to the agency's first biotechnology office in 2014.
In the conference room, Wattendorf outlined his ideas to agency higher-ups. Regina Dugan, the DARPA director at the time, ribbed him for the writing on his hand before greenlighting his proposal.
The result was a program called ADEPT, which invested $291 million from 2011 to 2019 in an array of technologies — including a credit card-sized device for rapid antibody discovery developed by the Vancouver-based firm AbCellera — that, taken together, could significantly reduce the timelines for vaccines and antibodies.
"It may turn out to be the most important program from my time at the agency," said Dugan, who ran DARPA from 2009 to 2012.
Chief among Wattendorf's targets for the program: delivering vaccines and antibodies by implanting their genetic code.
Traditional vaccines inject what's known as an antigen — usually a piece of live or deactivated virus that is sufficient to provoke the immune system into a protective response. Antigens are typically manufactured in a long process that involves growing live virus in chicken eggs in bioreactors.
Wattendorf hoped to short-circuit it. He wanted to inject genetic code that would prompt the human body to create the antigen in its own cells, cutting out the manufacturing process. The immune system would recognize the antigen in the cells and launch a protective response.
By 2010, scientists had tested the idea using DNA with mixed results. Wattendorf wanted to try its single-stranded sibling RNA.
If successful, RNA could be used to develop both vaccines and antibodies, shortening development timelines from years to days before clinical trials, he thought. It also offered a one-size-fits-all approach; in the future, scientists would need only the genetic code of a virus to create a vaccine.
At the time, many considered it a fool's errand. Being so ephemeral, RNA is unstable in the environment and highly susceptible to degradation. It was unclear how to get it into a human cell. Over at the National Institutes of Health, where Wattendorf previously worked, research into DNA vaccines was presenting enough hurdles. Few wanted to take the risk of trying RNA, too.
"Skeptics cited the lack of evidence that it would work, and Dan cited the lack of evidence that it wouldn't," Dugan recalled. "That's very typical of a DARPA program."
By 2019, a project DARPA funded at the Massachusetts-based company Moderna demonstrated in a Phase 1 clinical trial that RNA could indeed deliver an antibody to humans and provide protection against the mosquito-borne virus chikungunya. It was an affirmation of Wattendorf's bet that came after years of DARPA funding the effort.
Today, RNA vaccines, although still experimental, are among the fastest-moving candidates in the race to stop COVID-19. In March, Moderna was the first company in the United States to enter Phase 1 trials with a COVID-1 vaccine using RNA. The company injected its first test into a human 66 days after receiving the virus's genetic code. Phase 2 trials began in May, and Phase 3 began on July 27, making it possible that the vaccine could be available by the end of the year.
In addition to Moderna, two other pharmaceutical companies — Pfizer and CureVac — are pursuing RNA vaccines, as is a small laboratory at Imperial College in London and the People's Liberation Army Academy of Military Sciences in China. CureVac was also funded by DARPA.
"Is an RNA vaccine going to potentially be made available at scale?" Wattendorf said. "We are seeing very well that could happen. It demonstrates the role DARPA can play in creating these capabilities."
Wattendorf also continued DARPA investments in vaccines delivered using DNA.
Inovio Pharmaceuticals, funded by DARPA, entered Phase 1 trials for its DNA-delivered COVID-19 vaccine in April, making it the second company to enter trials in the United States. The Pennsylvania-based firm, which began trials 80 days after receiving the virus's genetic code, is looking to begin its Phase 2 and 3 trials this summer, pending regulatory approval.
DARPA also funded other technologies for rapid vaccine development, including companies that manufacture vaccines by growing proteins in tobacco-like plants, as well as a "self-assembling vaccine" platform at Massachusetts General Hospital. Mass General's vaccine center used the platform to develop a more traditional COVID-19 vaccine that entered animal testing in early July.
"DARPA comes off as a visionary organization that was ringing people's doorbells and saying you have to prepare for this," said Mark Poznansky, director of the Mass General Vaccine and Immunotherapy Center. "Some people would say we have enough to worry about without that."
Threat of infectious diseases
From the start, Wattendorf and the DARPA team knew that fast vaccines on their own wouldn't solve the threat that infectious diseases posed to American troops.
It can take weeks for a vaccine to give a person protection, and even then, follow-up booster shots are sometimes needed.
Instead of forcing the body to produce antibodies using a vaccine, why not just inject the best antibody directly? The DARPA team began to pursue that aim in parallel. Wattendorf called the rapid delivery of an antibody using RNA "the more aspirational dream."
The idea was to take the blood of a virus survivor and quickly identify the best antibody out of thousands in the bloodstream. Then, the genetic code of that antibody could be injected into troops to give them temporary protection against the virus immediately. Protection could range from a few weeks to a few months — enough time for a deployment.
In a pandemic, DARPA envisioned using such antibodies as a "firebreak" — an obstacle that slows the rapid spread of a conflagration.
For example, if one person in a nursing home tests positive, the antibody could be given to all the other residents to prevent the spread of the illness.
Critically, unlike vaccines, antibodies can also treat those who have already fallen ill.
DARPA had funded the development of rapid antibody technologies for years. Then, around 2016, DARPA Director Arati Prabhakar wanted to weave them together into a production line and test it.
"Very few interesting problems just have a miracle, single-threaded solution," Prabhakar said.
The result was the Pandemic Prevention Platform, which Prabhakar signed off on before leaving DARPA in January 2017. The goal of the four-year $96 million program was to develop an antibody for any virus within 60 days of receiving the blood sample of a survivor.
When COVID-19 arrived in the United States, the program's participants — AbCellera, Vanderbilt University, Duke University and AstraZeneca — had already done test runs with various viruses to see where they could cut time on their quest to hit the 60-day goal.
As they pivoted to tackle COVID-19, Jenkins, the DARPA program manager, knew the participants wouldn't meet the 60-day timeline but thought some could come close to 90 days, and potentially help end the global pandemic.
Some of the participants obtained a blood sample in February from one of the first American COVID-19 patients to return from China. But the sample wasn't great; the patient had recovered fairly recently and therefore didn't have a sufficiently mature immune response from which to draw good antibodies.
At AbCellera, chief executive Carl Hansen forged ahead with the sample anyway.
On Feb. 28, AbCellera's employees began working round-the-clock over a weekend in their Vancouver offices, ultimately finding 550 unique antibodies using their tiny device.
Hansen contacted the pharmaceutical company Eli Lilly and came to an agreement, announced March 13, whereby Lilly would manufacture the best antibody and take it into clinical trials.
But first they had to decide which antibody was the winner.
Daniel Skovronsky, Lilly's chief scientific officer, said the pharmaceutical company threw the normal years-long process out the window and immediately began scaling up to make the top 100 antibodies to save time, even though only one would proceed.
The company worked with AbCellera, the National Institute of Allergy and Infectious Diseases, and academic researchers to conduct experiments on the antibody candidates. By late April, they had to pick which, if any, should proceed to the last and most expensive stage of scale-up.
The best candidate, they determined, was antibody No. 555.
"This was a tough decision," Skovronsky said. "There were mixed views."
The first patient was dosed on May 29, 91 days after AbCellera received the blood sample. It became the world's first study of a potential COVID-19 antibody treatment in humans, according to Lilly. The national registry of clinical trials indicates that Phase 2 is expected to be completed in August.
At Vanderbilt, Crowe wanted a better sample than the initial one obtained in February. By mid-March, his team found two people in the United States who had been infected 50 days earlier in China.
After screening the samples, his team narrowed their list to the 30 best antibodies and then interacted with companies interested in producing them.
IDBiologics Inc., a Nashville, Tenn.-based biotechnology startup that Crowe co-founded, will begin human trials with one of the antibodies in August, he said, with possible availability in the United States under emergency use early next year if all goes well.
After licensing six antibodies from Vanderbilt and screening their own in house, AstraZeneca picked two antibodies to take into clinical trials this summer as a pair, said Mene Pangalos, executive vice president for biopharmaceutical research and development.
The global pharmaceutical industry can produce billions of vaccine doses, but it lacks the capacity to manufacture antibodies at such a large scale. At least initially, the antibodies won't be delivered using RNA, although Duke University plans to manufacture an RNA version of its antibody, meeting the original DARPA vision for the program.
Wattendorf, who has left DARPA and now works at the Bill and Melinda Gates Foundation, said that looking back 10 years, the agency was trying to solve the problem of speed to protect American forces.
"These things were funded to be fast," Wattendorf said. "They were not actually funded to be global scale."
Other DARPA efforts did take aim at the scale question. The agency, for example, funded technology to produce vaccines using plants instead of chicken eggs — an approach that has the benefit of easy mass scale-up. One of the firms DARPA funded, Quebec City-based Medicago, began Phase 1 clinical trials in mid July with a vaccine produced in a tobacco-like plant and plans to enter Phase 2 and 3 trials in October.
Mass production and its costs are now a problem that governments the world over are looking to solve, all while watching the results of the fastest-moving clinical trials.
"While we're all hoping our therapies work, at the end of the day, we all hope somebody's therapy works," Pangalos said. "Because we all want to get back to some semblance of reality."