Yesterday, President Trump, in an effort he described at one point as part of a plan to cut governmental red tape, announced that the FDA has given an emergency use authorization for convalescent plasma to treat COVID-19.
It was a controversial move because only a week earlier, the FDA had refused to grant such authorization. Its sudden about-face—especially since Trump himself was the one to announce it—makes it look as though the agency bowed to political pressure on the eve of the Republican National Convention.
Convalescent plasma treatment, as most people now know, uses blood plasma infusions from people who’ve recovered from a disease to treat people newly infected with it. The idea is that antibodies from the donor will help the recipient’s immune system get a head start on the disease, reducing its severity. It’s not a new concept: it was used as far back as 1918 to fight the Spanish Flu.
After initial promising results in small-scale trials, the FDA approved it for experimental use, setting up a program by which thousands of people were able to give it a try.
In the process, a team led by Michael Joyner of Mayo Clinic realized that they could collect data from 35,322 patients at 2,807 medical centers around the country to see just how well the treatment worked and, more importantly, how best to use it.
This is the study that drew all the attention. You can read it on medrxiv.
Even the most cursory glance reveals one important thing: it was never intended to be a definitive analysis of whether the treatment worked. The gold standard for such studies is the double-blind placebo-controlled trial, in which there is a large control group that gets a placebo instead of the treatment. In this case, everyone got the treatment and the Mayo team simply collected data.
What they found was interesting, though it was misreported by both the President and most of the press. What he said is that the treatment cut the fatality rate by 35 percent. That is not actually what the study found.
Instead, the researchers compared the fatality rates among different groups of people, all of whom got the treatment. It found two important things.
The first, widely touted, was that those who got the treatment early had a lower fatality rate than those who got it later—though not by 35 percent. Instead, it was a 27 percent lower 7-day fatality rate and a 19 percent lower 30-day fatality.
Based on this, I can see why the study gave the FDA scientists indigestion. This looks promising…but without a control group what exactly does it mean? In particular, why did some patients get the treatment earlier than others? Were they also getting other treatments earlier on—treatments that might have been more important to their survival?
The second finding is more technical…though possibly more significant.
It was that the level of antibodies in the donated plasma were strongly correlated with the survival rate, with those who received the most-antibody-rich plasma having a 35 percent lower fatality rate than those who got the least-antibody-rich plasma.
It is this 35 percent figure that Trump appears to have bootstrapped into an overall effectiveness of the treatment.
The study itself nowhere makes that claim. But this part of it actually has a hint of a double-blinded control group.
That’s because nobody knew, when the plasma was administered, what level of antibodies any given donor’s blood had. Instead, the doctors used whatever plasma was available—saving a small quantity for later testing.
And, as it turned out, the antibody levels ranged widely—over at least a factor of 10 from one donor to another. That means that those receiving the low-antibody plasma, simply by luck of the draw, may not have been all that different from a traditional placebo group.
Which brings me back to my deliberately provocative title to this post.
Science is not red tape. But it is complex. And the interpretation of its results can be open to dispute. Based on my read of the study, I come to two conclusions:
1. If I were hospitalized with access to convalescent plasma therapy, I would definitely want it.
2. We need to step back and let the scientists at the FDA continue to do their work. These are dedicated people who have spent their careers working to serve the public—not denizens of some deep-state conspiracy determined to use COVID-19 to unseat President Trump. I know the type of people who take these jobs. I knew them in grad school and I have watched them in action as a science writer. They have opinions—but their opinions are driven by the science…not the reverse. Yes, science can be slow and complex. But that’s not the same thing as “red tape.”