Xin Yin was excited to be visiting his parents in the Shandong Province of China last December. Yin, a researcher in the Immunity and Pathogenesis Program at the Sanford Burnham Prebys Medical Discovery Institute in California, hadn’t been home in five years. So much had changed since then — a new high-speed rail station now connected his hometown with Beijing, for example. On Jan. 1, Yin was at the Beijing station, headed back to Shandong after seeing friends, when someone texted him about an unusual cluster of pneumonia cases in Wuhan. Pneumonia, an infection of the lungs, is typically caused by the flu or another virus. Yin, whose Ph.D. is in molecular virology, was intrigued. He texted a friend who lives in Wuhan, as well as several others who work in hospitals, to see if they had heard anything about the cluster. It wasn’t the flu, they told him.
On Jan. 7, he got a message saying that the virus appeared to be related to the severe acute respiratory syndrome coronavirus, a highly contagious pathogen first identified in 2003 and responsible for the SARS outbreak that sickened some 8,000 people worldwide and killed nearly 800 of them. “At that moment, I am shocked,” Yin says. He emailed his boss, Sumit Chanda, the director of the program, and told him he thought the situation in Wuhan might be graver than was currently being reported.
Chanda’s lab studies respiratory and other viruses, including Influenza A, in order to design drugs that combat them. This process can take a decade or more to succeed (or fail definitively) and can cost hundreds of millions of dollars. But his lab also tests already existing drugs to see if they can be repurposed against pathogens — the dengue virus, say — for which there are no effective vaccines or treatments. A drug that has already been tested for safety can proceed to human clinical trials for new uses in weeks or months, rather than years. Alarmed by the potential of the novel coronavirus to set off a pandemic, and knowing that a vaccine could take years to develop, Chanda decided “to go all in,” as he puts it, and use his lab to look for a drug that could be deployed against it quickly.
Finding a way to prevent or treat Covid-19 has since become the overwhelming focus in medical-research facilities around the world. Keeping track of the number of projects and their progress is all but impossible: Reportedly, some 90 projects had pursued vaccines, and a handful are proceeding to clinical trials. At the same time, scientists are searching for intermediate treatment options that could reduce the numbers of people becoming critically ill and dying. This effort includes repurposing existing drugs. By mid-April, the Food and Drug Administration website had listed 72 active Covid drug trials and 211 programs in planning stages; 950 drug-development proposals or inquiries had been submitted to the agency. So far, only one drug, remdesivir, has been shown in a major, randomized control trial to lessen the recovery time of Covid patients, according to preliminary results.
There is no standard way of arriving at drugs to test. The choice of remdesivir was based on established understandings of how both it and coronaviruses are known to work. Some groups are using computer software to model how the virus interacts with human cells and to search for compounds that might inhibit that action. But because there is still much that researchers don’t know about how the virus behaves, actually testing thousands of drugs simultaneously can point to potential therapies that logic couldn’t predict. ‘‘It’s a real nonbiased experiment,’’ says Ajay Nirula, vice president for immunology at Lilly Research Laboratories, a drug-development arm of the pharmaceutical company Eli Lilly. “You get unexpected insights here, and it’s probably going to teach us something about the viral biology and generate some hypotheses.” Indeed, because researchers have studied the mechanisms by which known drugs interact with cells, those that fail in screenings can potentially reveal just as much about how the virus behaves as those that succeed, says Matthew Hall, a scientist at the National Center for Advancing Translational Sciences, which is also screening drugs against the coronavirus.
On April 17, Chanda, Yin and 29 co-authors from institutions around the world published a paper detailing their findings on the preprint server bioRxiv.org. (The paper is currently being peer-reviewed for publication in a journal.) They tested 12,000 drugs and identified 30 that appear to stop the virus from destroying human cells. Of those, four have already been clinically tested for other uses, including rheumatoid arthritis, Crohn’s disease, osteoporosis and cancer, and they seem to neutralize the virus in doses that people are likely to tolerate.
What makes the work done by Chanda and his colleagues noteworthy is the number of drugs they assessed and the speed with which that happened. It was, Nirula says, “a technological tour de force.” In the U.S., Hall estimates, only a few dozen labs have the capability of testing so many compounds at once, a process called high-throughput screening. Yet how Chanda’s lab went from a text message on a train platform to a handful of strong drug candidates in four months required as much cooperation as it did machinery. The collaborations they formed, spanning multiple unrelated institutions, are unusual in the competitive world of drug development. “These networks we’ve set up are all ad hoc, calling up a friend of a friend and putting it together,” Chanda says. “My hope is that we take the infrastructure we’ve built in an ad hoc way and reinforce it so that it will be there for the next time. There will be a next time.”
In January, after hearing from Yin in China, Chanda urged him to return as soon as possible. Yin arrived back in California on Jan. 21. Two days later, Wuhan went into lockdown. By then, the Centers for Disease Control and Prevention had obtained a sample of the virus from what was then the first known case in the U.S., outside Seattle, but it was unclear when researchers would be able to get samples. It would be quicker, Chanda thought, to partner with a Chinese lab that already had the pathogen but was unable to perform high-throughput screening; such a lab wouldn’t be able to send him a viral sample, but maybe a scientist from his lab could go there. He soon got word that a team at the University of Hong Kong led by Yuen Kwok-yung, the microbiologist whose work was instrumental in identifying the SARS coronavirus in 2003, had cultured the novel coronavirus and was willing to collaborate.
On Jan. 24, Chanda reached out to Arnab Chatterjee, vice president for medicinal chemistry at Calibr at Scripps Research. The two had met nearly 18 years earlier, when both were scientists at the Genomics Institute of the Novartis Research Foundation. The institute was established in 1999 and soon began to study, among other things, the interaction of drug molecules with the tens of thousands of genes in the human genome, whose sequencing was close to completion. Until then, scientists had mostly looked at one gene at a time; the institute’s founding director, Peter Schultz, a chemist and entrepreneur, was one of the first to utilize the sort of robotics used in automotive-assembly lines to look at thousands of genes at once. (He got the idea from a former college roommate who was an engineer at the car company Saturn.) Eventually, Schultz left Novartis, and in 2012 he founded Calibr, where Chatterjee joined him. They planned to partner with commercial enterprises to find existing drugs that could treat rare or neglected diseases for which pharmaceutical companies have little financial incentive to develop therapies.
The concept showed promise early: A leprosy drug, clofazimine, appears to be effective against a parasite that is a leading cause of infant death from diarrhea worldwide; an arthritis drug, auranofin, looks as though it might be a possible tuberculosis treatment. But only about half of the molecules that have safety records for use in people were commercially available to screen. In 2015, Schultz pitched the idea of compiling a complete collection to the Gates Foundation, which gave Calibr $20 million. By 2018, it had bought 7,000 compounds and hired 500 chemists, who spent 18 months synthesizing 5,000 more compounds. The ReFrame library, as it is called, stores 20 milligrams of each compound, in powder form, in vials stacked in filing cabinets and freezers. It is the largest drug-repurposing library in the world and is available free to all academic and nonprofit labs, like Sanford Burnham Prebys, so long as they make any resulting data publicly available.
Chanda told Chatterjee he wanted to send all 12,000 compounds to Hong Kong, along with high-throughput screening equipment. A lab member named Laura Riva had planned to accompany the shipment and run the screening. ReFrame keeps 1 milligram of each of its compounds dissolved in a solvent on a “mother plate,” a piece of plastic not much larger than an eight-track cassette, with 384 wells sunk into it. When someone wants the library, an acoustic dispenser, roughly the size of a coffee maker, uses sound waves to transfer nanoliters of each compound to a “daughter plate.” The entire library, about 40 plates’ worth, can fit inside a briefcase. While the plates were prepared, Riva, who is from Bergamo, Italy, began applying for a visa. On Jan. 31, however, President Trump announced a ban on most non-U.S. citizens arriving from China, meaning Riva almost certainly wouldn’t be able to return. The only solution, she and Chanda decided, was to ship the equipment and the library to Hong Kong and have Riva talk the scientists there through the setup over her iPhone.
In preparation, Riva set up a dispenser at Sanford Burnham Prebys, documenting each step of the process with photos and video. Paul De Jesus, a senior lab manager, bubble-wrapped another, along with extra plates, adapters and power converters, and shipped it to Hong Kong. When it was accidentally sent to mainland China, initiating a customs delay, De Jesus realized that the same thing could happen to the library, which had to stay frozen. So he found a company called World Courier that promised to refill the package with dry ice every few days.
For the next month, Riva watched over WeChat as Shuofeng Yuan, a scientist at the Hong Kong lab, and his colleagues unpacked the boxes and set up the dispenser. Because they could not take phones into the biosafety chamber with the virus, they practiced each step before entering so they could troubleshoot with Riva. Yuan had a 3-month-old daughter at home whom he was barely seeing. Riva was worried about her parents in Bergamo, which was becoming the center of the crisis in Italy. There was no time to share personal stories, but Riva felt a kinship with Yuan. “They were as restless as our side,” she says. “OK, let’s keep moving. We can do it.”
After several days of practice, Yuan and colleagues set up the dispenser in the isolation chamber and began the screening. They combined monkey cells with the virus and each of the library’s 12,000 drugs (along with drugs from another smaller library); the dispenser apportioned the mixes to the drug wells in a matter of minutes. After three days, they added another substance, called a reagent, that would glow when it detected ATP, an organic compound that provides energy to living cells. If the virus had killed the monkey cells, no ATP would be present. If it hadn’t, “there will be a luminance,” Yuan says. “A signal to tell us whether the cell is happy or not.” The screening, run twice, identified about 300 compounds that seemed to have kept the cells alive.
By mid-March, the C.D.C. was sending coronavirus samples to U.S. labs. On March 9, around 10 a.m., Laura Martin-Sancho, a scientist in Chanda’s lab, got a call from the receiving department: It had a hazardous package for them. Martin-Sancho, who has worked in biosafety labs with influenza viruses, would be the one to open the box and replicate the virus so they would have enough of it to test. A safety team deposited the box in the biosafety lab anteroom. Once Martin-Sancho was fully dressed in protective gear, she took the box inside the biosafety chamber. Inside was a second box, which she placed in a sterile cabinet. Inside that box was a metallic tube. Inside the tube was another tube — inside of which was a vial that contained, as she puts it, “a few drops of the virus that everyone is fearing.” Martin-Sancho adds, “I remember Laura Riva looking at me like, If you cannot do it, nobody can.”
Using very small amounts from the vial, Martin-Sancho tested 18 ways of cultivating the virus, staying in the chamber as long as she could manage. On the afternoon of March 13, she sent an email to Chanda, Riva and Yin: “We do have cytopathic effect!!!” The virus was multiplying, in other words. Soon Martin-Sancho was testing the 300 compounds identified by the Hong Kong group on infected human cells, while Riva tried to replicate the group’s results in monkey cells. As a check, they used different reagents to detect the virus, which had been sent by Adolfo Garcia-Sastre, director of the Global Health and Emerging Pathogens Institute at the Icahn School of Medicine at Mount Sinai in New York. Garcia-Sastre and colleagues had developed the reagents for the original SARS virus. When that outbreak ended, instead of throwing them away, he stored them in a freezer; now, nearly two decades later, they proved effective at revealing the presence of the novel coronavirus.
On March 19, California’s governor, Gavin Newsom, ordered residents to stay at home. Riva, Martin-Sancho, Chanda and Yin were deemed essential workers and continued to report to the lab. From home, Paul De Jesus worked with vendors to keep Riva and Martin-Sancho supplied with the protective gear that was running short nationwide. In Bergamo, Riva’s father fell ill with Covid-19; hospitals there were overflowing, and doctors were being forced to decide whom to turn away. Martin-Sancho tried to take a 15-minute walk every day to call her mother, who lives in Valladolid, Spain, and was under strict lockdown; she and the rest of her family wrote to her 99-year-old grandmother, who is in a nursing home there and is hard of hearing, explaining why they couldn’t visit her. Martin-Sancho’s niece sent her pictures of the virus, which she drew as a Pac-Man ghost with eyelashes. Yin, too, called his parents daily. “Just stay at home,” he told them. “Clean your hands. Clean everything.” The three have logged long hours trying to determine not only which compounds work against the virus but also at what doses and in which combinations they are most effective. “We support each other,” Yin told me in early April. He added, “We are so tired.”
Their preprint, the largest screening of drugs against the virus to have publicly produced results to date, has sparked collaborations with large pharmaceutical companies like Eli Lilly, as well as academic and government labs. “They’ve gotten really promising hits,” says Micholas Dean Smith, a biophysicist at the University of Tennessee, Knoxville, who has used computer modeling to search for potential Covid drugs. When Eytan Ruppin, chief of the Cancer Data Science Laboratory at the National Institutes of Health’s Center for Cancer Research, saw the preprint, he turned the computer network his lab uses toward analyzing some of the drugs that did best in the screening to predict which are likely to succeed in actual patients and in what combinations. At Kansas State University, Juergen Richt, a professor of veterinary science, has begun testing several top targets on animals. The maker of the drug that performed best in the screening, AI Therapeutics, was already investigating it for use against the virus. The compound, apilimod, was initially being developed as an anti-inflammatory. By inhibiting a particular enzyme, it may disable basic cellular processes that the virus uses to access the cell and replicate. The company thinks it may work prophylactically to prevent infection.
It hopes to begin F.D.A.-approved clinical trials in the next few months, says Murat Gunel, a scientific adviser at the company and chair of the neurosurgery department at the Yale School of Medicine; he believes results from Chanda’s study and others could strengthen its application. AI will soon have about 200,000 pills available and is processing the materials needed to make five million more. “If it’s effective, we won’t have to wait any time, because the pills are ready,” Gunel says. “If the clinical trials fail, we will have a lot of pills.”
Indeed, there’s no guarantee that any of those drugs will succeed in patients. “A lot of these screens will throw up a lot of false positives, but they’re likely to be completely ineffective,” says Munir Pirmohamed, the director of the MRC Center for Drug Safety Science at the University of Liverpool. And because viruses mutate, shape-shifting to evade a particular drug, it might ultimately be a cocktail of several drugs, as is used to treat H.I.V., that emerges as the eventual treatment.
Calibr says that some 20 other groups are now screening the ReFrame library. Like skydivers leaping from a plane, Chanda says, the slightest difference in how the screenings are set up will mean landing in different places. “So it’s really important that more people do this.” If his lab’s initial screening yields one drug ready for patients by October — when he and others anticipate that there may be a second wave of infections — “that would be a resounding success.”
It’s still unclear how the pandemic will affect other scientific pursuits. Many labs have been forced to shut down or redirect their focus, away from their usual subjects and toward Covid-related research. On the other hand, that “team science” approach could also, Pirmohamed thinks, enable researchers to tackle longstanding global health issues like cancer and heart disease more efficiently. “When we come out the other side,” he says, “I hope that collaboration continues.”
Kim Tingley is a contributing writer for the magazine.
Sign up for our newsletter to get the best of The New York Times Magazine delivered to your inbox every week. A version of this article appears in print on May 17, 2020, Page 18 of the Sunday Magazine with the headline: An international team quickly screened thousands of drugs in search of Covid-19 treatments — an intriguing new model for collaboration in science. Order Reprints | Today’s Paper | Subscribe