Enlarge /. Donald Trump shows Stephen Hahn, the head of the FDA, at an earlier press conference.
At a press conference today, President Trump announced that the FDA had granted emergency approval to treat COVID-19 cases with blood plasma from previously infected people. The move comes despite considerable uncertainty about how effective this treatment is and just days after Trump attacked the FDA for delaying its work as part of a conspiracy to sabotage his re-election.
Plasma is the liquid part of the blood that contains antibodies (among other things). It has been used to treat other infections because some antibodies may be able to neutralize the infecting agent – by attaching to bacteria or viruses in such a way that they cannot enter cells. Early studies have shown that it is relatively common for those who have had SARS-CoV-2 infection to produce antibodies that can neutralize the virus in laboratory tests, although the antibody response to the virus is also very different.
In the absence of effective treatments, people started testing this "convalescent plasma" as early as March, and the tests expanded as the pool of post-infected people continued to grow. So far, however, the evidence has been mixed. One of the largest studies conducted by researchers at the Mayo Clinic involving over 35,000 patients showed an effect, but it was very mild: mortality fell by 11.9 percent in people four days or more after starting treatment Plasma received. compared with 8.7 percent when treatment was started earlier. From a critical point of view, however, the study lacked a control group, so that the authors spoke of "signatures of effectiveness" rather than actual proof of effectiveness.
In support of the decision, only two small randomized clinical trials were cited in the FDA memo. No effect was seen while the second stopped prematurely and the researchers wrote, "Observations raised concerns about the potential benefits of convalescent plasma in the study population and after discussion with the Data Security Monitoring Panel, the study was discontinued."
Early data like this may still be enough to enable an emergency approval, which simply requires proof of effectiveness for a product that is known to be safe and for which there are no alternatives. However, the evidence was so weak that the New York Times reported last week that key health officials, including Anthony Fauci and Francis Collins, intervened when the FDA was considering giving plasma an earlier emergency approval. However, the Times reported that the permit has been put on hold, which is clearly not the case.
Politics, not evidence
What may have fueled the decision to move forward was a direct attack by President Trump on the FDA, who said the agency is delaying critical studies on COVID-19 therapies. Going on that the motivation was to undercut his re-election, he said, "Obviously they hope to postpone the answer until after November 3rd."
At the press conference at which the decision was announced, Trump described the convalescent plasma as "an effective therapy". He apparently used the difference between 11.9 percent and 8.7 percent seen in the Mayo study to say it "has been shown to reduce mortality by 35 percent". Trump called this "very effective".
The other speakers, FDA Commissioner Stephen Hahn and Alex Azar, Director of Health and Social Services, had to add reservations. That included starting treatment early, and the 35 percent figure only applied to those under 80 who did not need a ventilator. Hahn went on to say that the treatment was only really effective if the plasma had a high level of SARS-CoV-2-specific antibodies. When asked about the discrepancy between his claims and the more nuanced data, Trump passed it on to Hahn to respond. He gave an answer which basically said the plasma treatment met legal requirements for an emergency permit. That turned out to be the last question Trump asked (he only took three).
There are several reasons to believe that antibodies might be effective in limiting SARS-CoV-2 infections. A steady stream of neutralizing antibodies has been reported in the scientific literature, suggesting that these could be produced in large quantities, thereby avoiding the uncertain supply of regular plasma donation by those who cleared an infection. To date, however, there is no clear evidence that these antibodies have a clinical effect, or the course of treatment required to produce that effect (which is why we have not covered any of the antibody neutralization discoveries).
Given this uncertainty and the reports of disagreement among public health officials, the decision to postpone emergency approval now in the wake of Trump's attack on the FDA suggests that it was due more to political pressure than the best evidence available .