۲ݮƵ

Subscribe to the OSS Weekly Newsletter!

Statistical Acrobatics and Convalescent Plasma

Convalescent plasma is being touted as the latest treatment to COVID-19. Is there any science to back this up or is it all just hopeful hype?

It isn’t exactly “much ado about nothing,” but it is too much ado about something. That something is “convalescent plasma,” a much-talked-about treatment for COVID-19. Plasma is the yellowish liquid portion of blood in which blood cells are suspended and contain glucose, electrolytes, hormones, clotting factors and various proteins. Among the proteins are the “antibodies” produced by the body’s immune system in response to “antigens,” potentially harmful substances such as bacteria, fungi, parasites, and viruses. Antibodies combine chemically with antigens and neutralize their effect with each type of antibody being unique and able to defend against a specific antigen. Infection with SARS-CoV-2, the virus responsible for COVID-19, stimulates the formation of antibodies specific to this virus. These antibodies can be found in the plasma even after someone has recovered which is why transfusion of this “convalescent plasma” into an ailing patient can conceivably help in recovery.

The concept of using plasma to treat disease is by no means new, dating back to the late 19th century when German physiologist Emil von Behring cured animals of diphtheria with “antitoxins” isolated from the plasma of infected animals. These “antitoxins” were eventually identified as the proteins now known as antibodies by Paul Ehrlich who predicted that one day they could be used as “magic bullets” in medicine. For his pioneering work, Behring received the first-ever Nobel Prize in Physiology and Medicine in 1901, and Ehrlich was also awarded the Prize in 1909.

Behring and Ehrlich’s work quickly led to the investigation of the use of blood plasma to treat infectious diseases. Some types of pneumonia responded, and complications were reduced when serum, which is plasma with clotting factors removed, was used instead of plasma. The advent of antibiotics and the side effects such as fever and allergic reactions associated with serum treatment resulted in the abandonment of antibody therapy except in a few cases such as post-exposure to viral conditions such as rabies, measles, hepatitis A and B and the neutralization of toxins associated with diphtheria, tetanus, and botulism where it continued to be used.

In 1975 interest in antibodies was invigorated when Georges Kohler and Cesar Milstein found a way to produce specific individual antibodies in the lab. When an antigen is introduced into a mouse, immune B cells react by producing antibodies that bind to the antigen. These cells can be isolated from the animal’s spleen, and copied, or “cloned.” The antibodies they produce are specific to the antigen and can be isolated. Kohler and Milstein shared the 1984 Nobel Prize for Physiology and Medicine with Niels Kai Jerne for this discovery.

“Monoclonal antibodies” have revolutionized drug therapy with their application in major diseases such as cancer, inflammatory and autoimmune disorders, cardiovascular disease, infections, allergies, and osteoporosis. As soon as the SARS-CoV-2 virus was identified, researchers began to work on monoclonal antibodies that would bind to and neutralize the “spike” protein the virus uses to bind to a cell and initiate infection. Clinical trials are already underway and Dr. Anthony Fauci, head of the National Institute of Allergy and Infectious Diseases (NIAID), thinks that an effective use of monoclonal antibodies will come before a vaccine.

Until monoclonal antibodies that bind to SARS-CoV-2 are produced and shown to be effective, plasma from patients who have overcome a COVID-19 infection can be useful since it will contain a host of antibodies, including those for the virus in question. This is why the U.S. has just issued an Emergency Use Authorization (EUA) for convalescent plasma. This is not the same as an FDA approval which requires extensive safety and efficacy studies and allows a product to be generally prescribed.

An EUA is a temporary measure that allows for unapproved products to be used in an emergency situation when no approved products are available with the criterion that FDA has determined that the potential benefits outweigh the risks. In this case that was based on a trial coordinated by the Mayo Clinic involving over 35,000 hospitalized COVID-19 patients who were transfused at various times after diagnosis with convalescent plasma that had high, medium or low levels of antibodies.

Basically, the trial showed that patients who were transfused three or four days after diagnosis had a lower mortality rate after seven days had passed than patients who did not receive plasma that had been collected from people who had recovered from a COVID-19 infection. This was not a placebo-controlled trial so it isn’t clear whether there were other differences in treatment other than with plasma. In support of the EUA, FDA commissioner Stephen Hahn reported that mortality had been decreased by 35% a, meaning that in every 100 patients treated with convalescent plasma, 35 lives would be saved.

This number, also used by President Trump, is completely wrong, as Hahn later admitted. The 35% came from an analysis of a subgroup of 3000 patients who had received plasma either with high or low antibody levels. In the “high” group, 8.9% died after seven days, while 13.7% died in the “low” group. These numbers differ by 35%, which is termed the “relative risk.” The difference in “absolute risk,” a much more meaningful measure, is 13.7-8.9 which is 4.8. So, it isn’t 35, but 4.8 lives that would be saved for every one hundred patients. That is still significant, but much less than the highly-touted, and wrong, 35%. Convalescent plasma may help some patients, but it is not the “breakthrough” that was claimed. Monoclonal antibodies may, however, turn out to be Ehrlich’s “magic bullet” against SARS-CoV-2. We await with bated breath. Wearing a mask.


Back to top