August 17, 2020

expert reaction to unpublished, non-peer-reviewed preprint about convalescent plasma to treat hospitalised COVID patients

A preprint, uploaded to medRxiv*, has looked at usage of convalescent plasma to treat hospitalised COVID-19 patients in the US.

Prof Dave Roberts, Associate Director for Blood Donation at NHSBT, and one of the investigators for the NHSBT convalescent plasma trials, said:

“These are promising results that support the need for people to donate convalescent plasma in the UK.

“The results indicate preliminary and encouraging a significant reduction in mortality for people with COVID-19 treated with convalescent plasma.

“However, the research is not conclusive and cannot be easily correlated to our work or other trials. The combined data in the American study required used plasma from many sources, a great variety of different clinical situations and uses different testing methods to other studies. The results are also observational and not the fair comparison we would see using randomised control trial. We are carrying out two randomised control trials using convalescent plasma to treat intensive care patients (in the REMAP-CAP trial) and hospitalised patients (in the RECOVERY trial) with COVID-19 to answer these questions.  The American groups are now increasing their own efforts to increase participation and enrolment in their trials.

“We need people who have recovered from COVID-19, especially men, to offer to donate at nhsbt.nhs.uk. Men are particularly important donors because they are more likely to have the higher antibody levels which are observed to show more effect.” 

Prof Stephen Evans, Professor of Pharmacoepidemiology at the London School of Hygiene & Tropical Medicine, said:

“This study from the Mayo clinic is indicative of efficacy of convalescent plasma, but as the authors make clear, it is only a suggestion. It does seem to suggest that there are no major issues of safety.

“The study only looks at those who had received convalescent plasma (CoPl) and there was no control group for a comparison. The randomised trials underway at the moment will give much better answers on efficacy.

“What it finds is that those who had CoPl soon after diagnosis fared better. Those transfused within 3 days of Covid-19 diagnosis had about a 3% (11% down to 8%) reduction in absolute risk of death and this absolute risk was similarly reduced by about 5% at 30 days (27% vs 22%). Similar magnitudes of reduced mortality were seen in those who received plasma with a high level of antibodies.

“It looks as though this treatment is one of those that works early, rather than later in the course of disease. It is also a hint that vaccines that also produce antibodies may also be effective, but we again will have to wait for results of randomised trials to know whether any vaccine is effective.

“Mortality was not dramatically reduced to close to zero, so it is not a “cure”; but probably, (not yet certainly) it is a useful addition to the range of treatments that will be effective in early Covid-19 disease.”

* https://www.medrxiv.org/content/10.1101/2020.08.12.20169359v1

Declared interests

None received.