The FDA’s decision this week to allow convalescent plasma as a treatment for COVID-19 will make it harder to conduct rigorous studies of its effectiveness, because patients will demand the therapy rather than agreeing to studies in which they might get a placebo, experts said Monday.
The “emergency use authorization” issued by the Food and Drug Administration on Sunday allows doctors to infuse COVID-19 patients with plasma from people who have recovered, even if the treatment is not part of a study. The theory is that the recovered person’s antibodies can help a sick person fight off the illness.
The decision came at a politically fraught moment, the day before the Republican National Convention began, amid tweets from President Trump accusing the FDA of deliberately slowing approvals for vaccines and therapeutics until after the election. The New York Times reported that National Institutes of Health officials implored the FDA not to approve the therapy because the evidence supporting its use was too weak.
“The announcement shakes the confidence of the medical community in the rigor of the FDA decision-making process,” said Dr. Howard K. Koh, a Harvard professor who has served as US assistant secretary of health and Massachusetts public health commissioner.
Until now, plasma treatment was allowed only for patients in studies. And none of the completed studies has met the gold standard of medical research — randomized clinical trials in which half the patients receive a placebo.
“This announcement will hamper the ability to do clinical trials that are necessary,” Koh said. “Every patient with COVID will now be requesting this treatment and will not be willing to be randomized to any placebo arm [of a clinical trial].”
But at least one Massachusetts researcher is fine with that, saying enough is already known about the benefits of plasma.
Dr. Jonathan M. Gerber, chief of the Division of Hematology-Oncology at UMass Memorial Medical Center, is leading a study in which all participants receive plasma. He has been impressed with the results.
UMass has been treating critically ill COVID-19 inpatients with convalescent plasma since April 18. So far, 146 people age 23 to 95 have received the plasma, and 47 have died. Given the severity of their illness, Gerber said he would have expected at least half to die.
Some made “very substantial, sometimes very dramatic improvements soon after receiving plasma,” said Gerber, an associate professor at the University of Massachusetts Medical School. In as little as 24 hours after getting a dose, some stopped needing extra oxygen or reversed a steady decline. This is anecdotal, but Gerber says, “We’ve seen that enough that it’s hard to chalk it up to coincidence.”
He agreed that studies in which half the patients receive placebos would not be able to enroll participants. “Nobody wants to be in the placebo arm in something that’s safe and might be helpful,” Gerber said. Instead of putting critically ill patients in that position, researchers “need to be smarter in how we design trials.”
“All the evidence suggests it has benefit,” Gerber added. “The question is what degree of benefit — when does it work and how best to utilize it.”
But Dr. Richard Kaufman, medical director of the Brigham and Women’s Hospital Transfusion Service, finds “a great deal of uncertainty as to whether this treatment is effective.”
Kaufman is leading a randomized controlled trial intended to assess the effectiveness of convalescent plasma, by giving half the patients plasma with antibodies against COVID-19, and half plasma from people who didn’t have COVID-19. That trial will still go forward, despite the announcement Sunday.
The Brigham is just starting to enroll patients, but the study will eventually involve other Boston hospitals and hospitals in Texas and Alabama, where COVID-19 infection rates are higher, Kaufman said.
He notes that the emergency use authorization — known as an EUA — contained language specifically saying that it does not establish a new standard of care and that randomized controlled trials are still needed. “My hope is that patients will still be willing to enroll in randomized studies like ours to try to determine if this works,” he said. “My fear is that it will make it more difficult to get the data.”
The evidence did not justify the FDA’s move, Kaufman said.
It was based on a Mayo Clinic study that compared patients who received plasma earlier with those who received it later in the course of their illness. Those who received it earlier were more likely to survive.
“I have concerns about the EUA making it harder to get definitive answers about whether this product works,” Kaufman said.
But UMass’s Gerber counters, “All of us would like a perfect trials but we live in an imperfect world. . . . We have other ways to answer these questions.”
He called the FDA’s decision “a reasonable move.”
“There is value to a patient population being able to access this,” Gerber said. “The data is fairly convincing that plasma has some benefit. We can argue about how much.”
In the end, said Dr. Jonathan Abraham, assistant professor of microbiology at Harvard Medical School, convalescent plasma is “not a sustainable long-term approach,” because the blood of recovered people is a scarce resource and varies from batch to batch in the amount and type of antibodies present. It’s extremely difficult to design studies that definitively show effectiveness, especially as the illness moves from one part of the country to another, he noted.
“Convalescent plasma is typically used as a bridge therapy until better options are available,” he said. With the Massachusetts Consortium on Pathogen Readiness, Abraham is working on identifying the specific antibodies that attack the coronavirus, testing them in animals and people, and then mass producing them.
Plasma looks “promising” but its benefits and harms remain unclear, Abraham said. “It just adds more confusion to clinical care,” he said.
Asked what he thought of the FDA’s decision, Abraham would only say, ”Science and medicine — given what’s at stake here in human lives — should be devoid of outside influences.”
This story has been updated to add Dr. Jonathan M. Gerber’s academic title and to clarify one of his comments.
Felice J. Freyer can be reached at firstname.lastname@example.org. Follow her on Twitter @felicejfreyer.