But even described correctly, you can’t claim that everyone who gets a transfusion of convalescent plasma could expect this result. That data applies only to the specific, somewhat non-representative sample of patients from the Mayo Clinic study and to transfusions of convalescent plasma that have a particular level of antibodies. What’s more, the number of people included in the study’s “lowest” and “highest” antibody groups was low—no more than 1,100 out of the total of 35,000. In the FDA’s supporting documentation, scientists reported that their reanalysis of the samples using a different antibody test got very different results. Change the threshold for what you call “high-,” and the numbers shift again. The point is, there are multiple levels of unreliability here.
Most of all, though, the finding isn’t solid enough to rely on, because this wasn’t a planned, randomized trial. The analysis raises an interesting hypothesis, but it can’t prove it. The Mayo study raises an equally likely, but worrying, possibility: that the vast majority of donated plasma—which doesn’t have the highest levels of antibodies and usually hasn’t been administered early in the course of illness—did not help at all, and maybe even did more harm than good to some people. Only randomized trials can disentangle this.
You don’t need to prove a treatment works, though, for the FDA to grant an Emergency Use Authorization—only that it may be effective. Still, I was shocked when I read the FDA’s documentation. It wasn’t just the weaknesses of the Mayo Clinic data. The other lines of evidence that Marks had mentioned at the seminar on August 1 were also deficient.
The authorization points to the historical studies on the use of convalescent plasma to support the “may be effective” case, for example. But these are taken almost entirely from a systematic review of research conducted up through July 2013, including no randomized trials. The conclusions of that paper are contested by a more recent systematic review from July 2020, not mentioned by the FDA, which included two randomized trials of the use of convalescent plasma for treating influenza. This one found “no convincing evidence” of benefit from those trials and expressed concern that convalescent plasma may end up doing little or nothing to help people recover from Covid-19. The new authorization also doesn’t mention a non-randomized trial of convalescent plasma during the 2016 Ebola outbreak that found no survival benefit.
Another basis for the new authorization was the handful of studies that compared similar people who did or did not receive convalescent plasma as a treatment for Covid-19. These showed fairly positive results, but they are small, and there’s a risk of publication bias this early in the pandemic. If everyone expects a treatment to work, then a hospital where people didn’t fare so well might not be in such a rush to broadcast those results. As if on cue, another study appeared a few days ago (too late to be in the FDA documentation), finding no reduction in mortality. The “totality of evidence” in patients is weak at every point.
Could convalescent plasma be an effective treatment for Covid-19? I don’t think we have a satisfying answer. It might work for some people, but even if it does, it might not help them very much. It doesn’t seem very likely to provide important benefits for most of the Americans who will end up getting the transfusions in the coming months.
This week’s announcements do teach us something else, however. The FDA’s Emergency Use Authorizations aren’t even a short arm’s length from politics, and its leader and spokespeople are spinning data to the American people. So is the US president. Similar decisions will be coming, at some point very soon, on vaccines. If we can’t trust how those are made and communicated, the consequences will be dire.
Photographs: Ivan Valencia/Bloomberg/Getty Images; Chip Somodevilla/Getty Images; Asaad Niazi/AFP/Getty Images
More From WIRED on Covid-19