Development of effective treatments for covid-19, the disease the virus causes, would be one of the most significant milestones in returning the United States to normalcy. But the massive effort is disorganized and scattershot, harming its prospects for success, according to multiple researchers and health experts. Researchers working around-the-clock describe a lack of a centralized national strategy, overlapping efforts, an array of small-scale trials that will not lead to definitive answers and no standards for how to prioritize efforts, what data to collect or how to share it to get to answers faster.
“It’s a cacophony — it’s not an orchestra. There’s no conductor,” said Derek Angus, chair of the department of critical care medicine at University of Pittsburgh School of Medicine, who is leading a covid-19 trial that will test multiple therapies. “My heart aches over the complete chaos in the response.”
The global biomedical research establishment could be one of most powerful assets in the campaign against the new virus, with experts all over the world — and especially the scientific and medical powerhouse of the United States — in rare alignment in their focus on a single enemy. Some large trials designed to be definitive have launched. But with more than 500 human clinical trials worldwide, the lack of coordination puts the world at risk of ending up with a raft of inconclusive and conflicting studies and little idea of what interventions work for the next wave of illness.
Francis Collins, director of the National Institutes of Health, the nation’s largest biomedical research agency, acknowledged researchers’ frustrations but said in an interview Wednesday he has been working behind the scenes to launch an unprecedented public-private partnership to address the problems. He said the framework involves top pharmaceutical companies such as Pfizer and Johnson & Johnson, domestic and international government agencies including the European Medicines Agency, and academic research centers.
Collins said the month-long discussions have been kept under wraps to ensure buy-in for an approach likely to require sacrifices of personal recognition, scientific credit and profit — a centralized decision, for example, not to proceed with tests of one company’s drug in order to move faster on a competitor’s.
“I think we have the necessary clout to steer this whole complicated ecosystem,” he said. “When you look at some of the things that are happening sporadically, we may be unlikely to learn what we need to from such disconnected, small trials. The whole point is to replace that with a coherent, evidence-based approach. … I want to know what works, and I want to have it answered by June or July.”
Agency officials said further details would be released in coming days.
Anthony S. Fauci, director of the National Institute of Allergy and Infectious Diseases (NIAID), said in an email Tuesday that the partnership led by Collins is the “functional equivalent of a National Strategy.”
While Collins was working on developing that strategy, hospitals, drug companies, government labs and individual doctors were flooding the system with proposals for drugs and other interventions to test against the virus — an outpouring that reveals how siloed and fragmented the research enterprise remains. For example, there are more than two dozen separate U.S. trials listed for the anti-malarial drug hydroxychloroquine, all with different designs. Some use the drug as a preventive, others as a treatment. Some use it alone, some with other drugs or vitamins, and some have no comparison group to tell if the drug was responsible for the outcome. That will make it more difficult to conclude whether, or in what circumstances, the drug may work.
Collins said a working group is addressing this problem by sifting through about 100 possible covid-19 treatments to decide which are the six to eight most promising drugs to move forward in large-scale trials. Those will be deployed in large clinical trial networks.
The new federal effort is motivated in part by what happened in China. Clifford Lane, deputy director for clinical research and special projects at NIAID, traveled to the origin of the outbreak in February as part of an international delegation to help the world learn from the Chinese experience. He was troubled by the lack of a strategic plan to prioritize and fast-track the most promising treatments, leading to a mosaic of inconclusive findings.
“We do have to have a bigger strategy than every university, every institute and — to be blunt — every country” working on their own research efforts, Lane said in an interview.
At the heart of the problem is the basic question of whether a drug really works. Typically, drugs and medical interventions are first tested in small clinical trials that establish safety before the most promising ones are funneled into bigger trials, in an iterative and years-long process. These trials, which typically randomly assign patients to receive either a drug or a placebo, prove that medicines, vaccines and medical procedures are effective and safe. But with the urgency of the coronavirus threat, timelines have been squeezed, doctors are doing uncontrolled experiments as they administer regular care, and the typical model for research is too slow.
David Boulware, an infectious-disease physician and scientist at the University of Minnesota, has gotten at least 50 emails from companies and researchers with treatments they want to test. The urgency to find something — anything — for patients who have nothing other than supportive care has led researchers to pull everything off the shelf: a mix of existing drugs that show promise, stem cell treatments and brand new compounds designed specifically against covid-19.
The energy is remarkable, but it needs to be channeled. Clinical trials, whether for an HIV drug or a brand new medicine, compete for many of the same patients. If there are too many trials at a hospital, none of them may enroll enough patients to get clear results. If there are too many similar small trials running in parallel, their results individually may be inconclusive and the data could have so many differences they may not be able to be pooled.
“There’s all sorts of people wanting to try anything, because people are desperate,” Boulware said.
He and others who were unaware of Collins’s plans argue that national leadership — whether guidance on how to prioritize trials, a central body coordinating efforts, or a mechanism to play matchmaker among institutions working on similar ideas — could help channel the ubiquitous scientific desire to make a bigger, faster impact.
Some large-scale efforts are already underway: The World Health Organization has organized a massive trial in 90 countries of four promising therapies. The National Institutes of Health is conducting a test of the antiviral medication remdesivir at more than 50 institutions and last week launched a large trial for an anti-malarial medication. A $50 million effort at the Duke Clinical Research Institute will test hydroxychloroquine in 15,000 health-care workers and create a registry that can be tapped to speed up future trials, such as for a vaccine.
“People should not be fatalistic that we’re going to have a paucity of evidence for things that provide benefit,” Collins said. “I hope we’ll have three to four of those [treatments] by the summer.”
But as the federal effort has proceeded largely in secret, individual institutions have scrambled to set up committees of experts who evaluate which trials make sense to move forward. At the University of Pennsylvania, a weeks-old committee gives priority scores to trials based on criteria such as whether they will compete with existing efforts and how they are to enroll all the patients needed to get a result. A task force at Duke University does a similar review.
“There’s a lot of stuff bubbling up. It would seem like a sensible thing to do would be to align everyone around the same trials, not one trial for each context and not have each institution do its own thing and at the end of the day everyone has done a small trial … and we don’t know what to make of it,” said Steven Joffe, a bioethicist at the University of Pennsylvania. “Let’s get to an answer.”
Many proposed trials overlap — using the same drug, such as the anti-malaria treatment hydroxychloroquine that has been touted by President Trump, his personal lawyer Rudolph W. Giuliani and conservative talk show host Laura Ingraham.
Boulware is halfway through one such hydroxychloroquine trial, which examines whether the drug is effective at preventing the disease or in treating people with mild cases. People who participate will receive either the drug or a vitamin in the mail.
He said he is motivated by his experience working on Ebola, when by the time a well-designed trial was up and running, the outbreak was dying down. He plowed ahead with his hydroxychloroquine trial weeks before he heard — last Friday — the National Institutes of Health had declined to fund it. He found international collaborators through chance and social media when some Canadian researchers emailed to ask if he would share his trial design with them. He ended up connecting the Canadians with each other and is now working on overcoming the complex legal requirements to share data.
But now, his trial is potentially competing with a bunch of others that also test hydroxychloroquine all across the country — and enrollment has slowed in recent days.
Risks of inconclusive evidence in the social media age
Small trials and even anecdotal reports of treatments that appear to have worked on small groups of patients already are being shared, sowing both hope and confusion about the evidence.
A study published Friday in the New England Journal of Medicine reported on 53 people who took remdesivir, a failed Ebola drug Trump has praised and many families have tried to access outside of trials. The results were impossible to interpret, though, since some of those patients might have gotten better on their own and there was no comparison group of patients who did not receive the drug. Hydroxychloroquine, the cheap and readily available anti-malarial drug, also already has been widely used in patients despite only suggestive evidence it might work.
As more small-scale studies are designed, the risks of inconclusive but suggestive results multiply — and paradoxically, they could make it harder to conduct well-designed clinical trials that get to the bottom of whether a treatment works. Well-designed clinical trials require patients to be willing to be randomly assigned to receive the treatment — or a placebo.
Emma Meagher, chief clinical research officer at the University of Pennsylvania’s Perelman School of Medicine, said her institution’s study of the malaria drug in severely ill patients does not have a comparison group that receives a placebo because the media around the drug has made it the standard of care despite the lack of evidence. Every meeting, she said, begins with a discussion about how to prevent the next experimental therapy from becoming like hydroxychloroquine.
In some ways, designing research studies when clinicians have an imperative to give their patients the best care possible is an inherently tricky situation. Outside of top-tier research hospitals, patients may not have access to trials, so clinicians may have little option but to give them drugs in what amounts to an uncontrolled experiment. The Infectious Diseases Society of America released guidelines last week recommending that clinicians should give experimental drugs only in trials, but safety-net and rural hospitals are less likely to have access to those.
“Do we really want to have [some] people trying new different things and the rest of you sit and wait?” asked Benjamin Linas, an infectious-disease physician working on oversight of clinical trial protocols at Boston Medical Center.
Many researchers have said they are hopeful a national strategy will help unify and speed up the search but lament the time already lost. The United States did not have a pandemic clinical trials network ready to be activated, but existing clinical trials networks, such as those used to test HIV treatments, are now being repurposed. Collins said he had never seen research move faster, but in a pandemic that can still feel slow.
“We have imperfect networks. I mean, there are some there, but they’ve never been tested in this way,” said Adrian Hernandez, vice dean for clinical research at Duke University School of Medicine. “Having a common infrastructure that can do rapid-cycle trials — that would be beneficial.”
Collins said he began partnering with private companies, research institutions and other agencies in mid-March. Had he started sooner, he said, the urgency of the situation might not have been clear to those groups he had to convince to work together. He disagrees significant time has been lost and noted the efforts were able to piggyback on an existing framework for pursuing drugs with industry collaboration called the Accelerating Medicines Partnership.
Partnerships that previously took a year and a half to build were put together in a week, he said.
“I think it is a world record for anything of this sort. It might have been difficult to get full unanimous agreement to what is clearly unprecedented — for a willingness to give up control,” Collins said.
As with other areas of pandemic planning, many hope the United States will learn from this lesson, that it needs a preparedness plan not only for allocating essential supplies and scaling up testing but also in plotting research.
“The problem is we need to remember to invest in preparedness, at times when we’re not affected with a pandemic or epidemic,” said Barbara Bierer, director of the regulatory foundations, ethics and the law program of the Harvard Clinical and Translational Science Center. “And it’s hard to repurpose or commit resources to something that doesn’t appear immediate on the horizon.”