He built a MERS treatment in 6 months and then the best Ebola drug. Now Christos Kyratsous turns his sights on Covid-19
Tuesday, February 18, 2020
Source: Endpoints News
In 2015, as the Ebola epidemic raged through swaths of West Africa, Kristen Pascal’s roommates sat her down on their couch and staged an intervention.
“Are you sure this is what you want to be doing with your life?” she recalls them asking her.
Pascal, a research associate for Regeneron, had been coming home at 2 am and leaving at 6 am. At one point, she didn’t see her roommate for a week. For months, that was life in Christos Kyratsous’ lab as the pair led a company-wide race to develop the first drug that could effectively treat Ebola before the outbreak ended. For Pascal, that was worth it.
“I’m ok, I don’t have Ebola,” Pascal told them. “I see that death toll rising and I can’t not do something about it.”
Last August, Regeneron learned they had succeeded: In a large trial across West Africa, their drug, REGN-EB3, was vastly more effective than the standard treatments. It was surprise news for the company, coming just 10 months into a trial they thought would take several years and a major victory in the global fight against a deadly virus that killed over 2,000 in 2019 and can carry a mortality rate of up to 90%.
For Kyratsous and Pascal, though, it brought only fleeting reprieve. Just four months after the NIH informed them REGN-EB3 worked, Kyratsous was back in his office reading the New York Times for updates on a new outbreak on another continent, and wondering alongside Pascal and senior management whether it was time to pull the trigger again.
In late January, as the death toll swelled and the first confirmed cases outside China broke double digits, they made a decision. Soon they were back on the phone with the multiple government agencies and their coronavirus partners at the University of Maryland’s Level 3 bio lab. The question was simple: Can Kyratsous and his team use a process honed over two previous outbreaks, and create a treatment before the newest epidemic ends? Or worse, if, as world health experts fear, it doesn’t vanish but becomes a recurrent virus like the flu?
“Christos likes things immediately,” Matt Frieman, Regeneron’s coronavirus collaborator at the University of Maryland, told Endpoints. “That’s what makes us good collaborators: We push each other to develop things faster and faster.”
Kristen Pascal (Regeneron)
The first time Regeneron tried to respond to a global outbreak, it was something of a systems test, Kyratsous explains from his office at Regeneron’s Tarrytown headquarters. Kyratsous, newly promoted, has crammed it with photos of his family, sketches of viral vectors and a shark he drew for his 3-year-old son. He speaks rapidly – an idiosyncrasy his press person says has only been aggravated this afternoon by the contents of his “Regeneron Infectious Diseases”-minted espresso glass – and he gesticulates with similar fluidity, tumbling through antibodies, MERS, the novel coronavirus, Ebola-infected monkeys.
When Kyratsous joined Regeneron in 2011 as a staff scientist, after a postdoc stint at NYU, the company had recently completed their “VelociSuite” platform: A series of technologies designed to speed up the notoriously laborious pace of drug development. About a year and a half later, the MERS virus broke out in Saudia Arabia, and Kyratsous had an idea.
“We thought it would be a good challenge to see if we can put all these technologies side by side to accelerate the time and respond to emerging infectious disease,” Kyratsous told Endpoints. “It would be a very nice proof of concept.”
Kyratsous is a thin man in a large office, dressed in one variation of the same knit-sweater-and-button-down he wears in every work photo. He’s a modest man, too, or at least one who is corporately inclined. Without fail, he speaks in “we,” crediting a motivated lab team and a senior management willing to listen to any idea. Pascal, who joined Regeneron from Roche two months after MERS began, is more forthcoming.
“It was very hectic, it was really just me and Christos,” Pascal said, noting there was also support from around the company.
VelociSuite includes several technologies designed to ease the transition between mouse models and humans. Regeneron genetically engineers the mice, swapping out the genes that code for antibodies against a given pathogen for their human equivalent. When these modified mice are infected with a virus – or with the safe, inert “pseudo-viruses” Regeneron uses – they will generate human antibodies. Those antibodies can then be collected from the mice, screened to select the best, and put into bioreactors that produce enough for clinical trials.
This tech, which helped lead to blockbusters like Dupixent, was particularly useful for MERs, a virus that doesn’t infect normal mice and thus proved difficult for most developers to study. After about 6 months, they had an antibody that could be put in clinical trials, although one was never intended. For Kyratsous, who as a kid toyed with becoming an engineer and got into virology because it showed what the host can do, it offered the perfect systems test.
“You know where you can take a calculated risk,” he said. “The processes you can run in parallel to save time.”
He wouldn’t have to wait long before the system was put in action. In the spring of 2014, the Ebola virus migrated from a Guinea village across that country and into Sierra Leone and Liberia. Kyratsous and other top scientists and executives kept watch as it spread, periodically meeting in the boardroom to debate if they should launch a drug discovery effort. Unlike MERS, Ebola wasn’t a new virus. It had broken out before and petered out quickly every time.
But then, the virus reached Conakry, Monrovia and other major cities – a dire sign for a deadly virus that, like most, spreads fastest in close proximity.
“That’s when things get out of control,” Kyratsous said. “Our CEO [Leonard Schleifer] sent an email around saying, ‘it’s time, let’s do this.'”
The Ebola project followed roughly the same trajectory as the MERS program, but in key ways it set the stage for their newest coronavirus effort. Most notably, Regeneron showed their MERS data to the US Government’s Biomedical Advanced Research and Development Authority, who funded and collaborated on the Ebola project. The government had been particularly concerned about the Ebola virus because they viewed it as a bioterror threat.
Kyratsous, Pascal and the rest of their team were able to get drugs ready for non-human primates in the same period as they had with MERS, but Ebola involved bigger risks. After creating their engineered mice, infecting them, collecting their antibodies and then screening those antibodies for the strongest, they were left with 20. In most scenarios, Kyratsous said, they would test each of those thoroughly. But with the virus still raging, they did limited testing, selected 3 to turn into a drug cocktail (cocktails are preferred for antivirals to prevent the pathogen from easily mutating resistance) and walked into the boardroom to present the data to Schleifer and CSO George Yancopoulos.
“I also recall how stressed I was,” Kyratsous said. “It was risky, we had a discussion with senior management and we were like, ‘These are the 3, let’s see what happens.'”
They sent the cocktail to a bio level 4 lab at Emory University that administered them to Ebola-infected monkeys, and waited. In a presage of things to come, they didn’t have to wait long. The investigators called Kyratsous before the trial was unblinded, and told him “It’s very, very obvious.”
“They’d never seen monkeys recover from Ebola virus infections so fast before,” Kyratsous said.
Developing drugs for emergent infections is a dissonant exercise; you put all your energies into building something you hope will never be used. And by the time the antibody was ready for a human trial — about 12 months after Kyratsous began — the epidemic was over. The drug, REGN-EB3, went on the shelf for three years.
When Ebola returned in the fall of 2018 in the DRC, world health authorities wasted little time. They sent the antibody to Congo, along with three other treatments, including two drugs developed in part by the US government and Gilead’s remdesivir, which would later become one of the first drugs tested against the novel coronavirus outbreak. They launched a large clinical trial testing three new treatments against the standard-of-care drug, an antibody called ZMapp other. Kyratsous and the rest of Regeneron thought they wouldn’t hear results for years as each arm of the study needed hundreds of patients to be statistically significant.
So on one August Saturday 9 months later, most of Regeneron’s leadership was away from the headquarters and Kyratsous was at home with his family preparing to go on his own vacation Monday, when they got an email: The drug had worked. The outbreak had continued long enough to enroll patients and those patients performed so much better on Regeneron’s drug than on the others that they had to halt the trial. Patients given Regeneron’s drug survived 71% of the time, compared with 47% on remdesivir. (mAb114, developed by the NIH, had a similar 66% survival rate.)
The team canceled or delayed the vacation plans and returned to the office Monday.
“It took us a few days to believe exactly what was going on,” Kyratsous said.
The novel coronavirus Covid-19 spread through February 16 (Dr Edward Parker, The Vaccine Centre, London School of Hygiene & Tropical Medicine )
By the time of the Covid-19 outbreak, Regeneron hadn’t launched an infectious disease response program in 4 years. But the previous two programs and the recent success of the Ebola program had left its imprint far beyond Kyratsous’ team. A project like Ebola subsumed the entire organization, becoming the the top priority not only on Pascal and Kyratsous’ desks, but also on those who support them: the sequencing, immunization, and protein expression teams. And as news of the new virus spread, employees across the organization began asking the question: Are we going to respond?
Kyratsous was asking the same question. He first read of the outbreak in the New York Times in December, but he couldn’t just launch the entire organization into an emergency response based on that.
“There are all sorts of novel viruses,” Kyratsous said. “You can’t start working on everything the moment you hear it in the media, otherwise you’re going to go crazy.”
He proceeded slowly as the news unfolded. After the virus was sequenced on January 11, he decided to synthesize the spike “crown” proteins that give coronaviruses their names and that they use to infect cells. It could be used to generate the proteins in mice.
“It was something that was easy, doesn’t take a lot of time,” Kyratsous said. “And then everything exploded.”
With patient numbers rising rapidly, Kyratsous, Pascal and Regeneron’s leadership decided to launch the program. In the lab last week, Pascal was incubating a pseudovirus and testing if it was expressing the right proteins that they will eventually test the antibodies on. These proteins generally produce light under a microscope, and they’ll know the antibodies are working when they go dark.
They haven’t yet reached the 6 am to 2 am stage – Pascal says she hasn’t been in the lab past midnight – as they wait for the immunization team to make sure the mice will express the right human antibodies, a process that can take several weeks. And they say they’ve learned key steps to fast-forward the process: Looking at the viral sequence on a computer before they acquired it, ordering the virus at first in pieces and then piecing it together themselves.
The work may be just as crucial as the Ebola program. Infections have slowed in recent days and the virus could dissipate by summer, but experts are concerned it could continue to expand or never go away entirely, becoming a recurrent virus like the flu.
Regeneron is cautious on giving a timeline for development, saying only that the Ebola program offers a good framework: 6 months to get drugs in non-human primates and a trial in 12 months. But Kyratsous smiles easily amid the emergency and has a quiet confidence to him. He notes they can build off their work on MERS, also a coronavirus, and with their BARDA and University of Maryland connections, have formed a “coronavirus network.”
His office is full of trinkets – a race track from Monaco, small colorful cubes – but displayed most prominently is a bright red Staples “easy” button. It was a gift from his team after the Ebola program, a joke on his favorite word to describe his and their work.
“Every virus is different. I don’t think there’s any guarantee of success but we’ve done it now several times,” Kyratsous told Endpoints. “That gives you the confidence to think: Maybe I can do this again.”