Track the facts

Treating COVID with blood plasma? Uncertain

News - image_intro

Is treating people who have COVID-19 with the blood plasma from people who have recovered from COVID-19, the solution? Current research does not enable us to say with any certainty, in contrast to what we have been hearing since the end of August. 
The Facts 

Plasma is the liquid portion of blood that contains antibodies. Antibodies naturally develop following an infection. They are an integral part of our immune response to an enemy, evident when plasma is collected from an individual not long after the peak of his or her illness. This plasma is rich in antibodies. 

In the past, the injection of so-called « convalescent plasma » has been used to try and fight diseases such as the Spanish flu, Ebola, and SARS (Severe Acute Respiratory Syndrome). Although its effectiveness has varied from one epidemic to another, the original hypothesis makes sense. The injection would provide antibodies capable of neutralizing the virus for a patient who doesn’t have it.  

Another advantage of convalescent plasma is that it is both inexpensive and relatively easy to collect and administer. 

The origin of the excitement 

This is the reason hope is being placed on this treatment, in the context of the COVID-19 pandemic. Hope, which in in the U.S., the epicentre of the global pandemic, was rekindled on August 23, when Donald Trump announced that the FDA (Food and Drug Administration) was now approving it — just days after he criticized the agency for dragging its feet on its approval!  

In the scientific community, this decision is equated with political interference in the FDA’s activities. At issue were promising preliminary studies that did not lead to any conclusions about the efficacy of the treatment. 

The caveats 

In June, a U.S. study of 5,000 people with a severe form of COVID-19, concluded that there appear to be no negative side effects from convalescent plasma transfusion. The only problem… a lack of a control group so that results can be compared. Another study, published in China a week earlier, was discontinued along the way because the researchers found no significant benefit from blood plasma administration. But it involved a sample too small (103 people) to make a definitive judgement. 

Another example is this study (pre-published on August 12 but not yet peer-reviewed) that compares the course of the disease in 35,000 patients who received plasma with varying levels of antibodies. The results are encouraging but difficult to generalize.  The sample was not randomized. Variables other than those studied may have influenced the results. For example, some patients admitted to the ICU (intensive care unit) may have benefitted from supportive treatments such as oxygen.  

And, it should be remembered that even convalescent plasma studies conducted on viral diseases other than COVID-19, such as on EBOLA and SARS, were unconvincing. A meta-analysis on the subject published earlier this year was unable to definitively judge the efficacy and safety of this treatment due to the extremely low quality of the available scientific data. The authors of another meta-analysis published in 2015 arrived at a similar conclusion.  

The scarcity of randomized controlled trials where participants are randomly selected and can be compared with a similar group that did not receive the treatment can be explained in the context where each time convalescent plasma was used, it was for an emergency. Faced with an epidemic, or even a pandemic, researchers are caught short and cannot set up a well-meaning research budget. When they can, as in the case of the ongoing Canada-wide study on convalescent plasma, results are delayed. 

The rush in the U.S. to approve this antibody therapy can be contrasted with the controversy over hydroxychloroquine, a treatment that again was promoted by President Trump despite the lack of available data. It is now known that this anti-malaria drug does not appear to be able to reduce the mortality rate of COVID-19 in hospitalized patients with severe symptoms. A recent meta-analysis published on August 26 (reviewing 29 searches totalling 33,000 people) even suggests that for such patients the risk of mortality would be higher in groups where hydroxychloroquine was associated with the antibiotic, azithromycin. 

Moreover, in the case of plasma, a statistic used erroneously on August 23 at the White House, contributed to suspicions of political interference. It was said that blood plasma would have reduced the mortality rate by 35%. However, the verification was made, but the statistic was not included in the FDA approval document, nor in the (pre-published) study of the Mayo Clinic, which was cited as an example. The FDA director apologized for the error the next day in a tweet, without giving any details on the origins of the error. The FDA’s public relations officer, hired a week earlier and described as a right-wing conservative activist, was laid off on August 28. 

Read to complete: 

Evidence lags behind excitement over blood plasma as a coronavirus treatment, Nature, August 19. 

F.D.A. ‘Grossly Misrepresented’ Blood Plasma Data, Scientists Say, New York Times, August 24. 

This article was originally published on the website of L'Agence Science-Presse (French only).


Back to news list