The Kiss of Death, Poblenou Cemetery, Barcelona, 2015

Convalescent plasma (CP) therapy gained FDA emergency use authorisation (EUA) in August. It appears that this is surrounded by some controversy and suspicion of politicisation.

Without getting anywhere near politics here, let’s not forget that health agencies’ interest in CP is nothing new. The FDA has previously encouraged survivors to donate plasma and supported developments in this space along with other international organisations (see also Blue Steens’ April newsletter). This is also not the first time that CP has been a hopeful for the treatment of viral disease.

What may be new in this context is heightened public scrutiny, unvarnished political commentary and wide-spread reporting. Below is an example of a debate on Twitter that was pointed out to me. Follow the thread if you’d like to understand some of the criticism relating to the use of published data. Note, though, that the FDA’s decision memo refers to several studies that have led to its conclusion. Thus, it is advisable to take the cited Twitter debate with a pinch of salt (like any social media discussion). As insightful as it is, Topol focuses on a statement made by FDA Commissioner Hahn rather than the full memo. Note as well that Hahn has since qualified his statement.

In general, it is of limited use to get too hung up on a single clinical study and very risky to base any far-reaching conclusions on it. This applies even more in the case of CP. Why? – Because we’re dealing with an inevitably extremely heterogeneous therapy. Not only is it applied to humans (as opposed to inbred lab mice in a controlled experiment say), but the therapeutic concoction itself is not of a prescribed standard composition because it is sourced from humans instead of chemical reactors. This is nothing like treating someone with dexamethasone for example. There, at least, we know and control the drug’s detailed features – its active compounds, additives, quantities etc. Sure, CP is to be prepared in compliance with regulatory standards. However, each donor will have mounted a different immune reaction to fight the virus. While one donor’s response has saved their own life, their individual plasma composition may or may not be helpful to another patient. Thus, getting clear-cut results even from several clinical studies is more than just a statistical challenge. Until we understand how to match donors and recipients at a more granular level, i.e. beyond simply matching blood types, treatment remains close to a therapeutic gamble. Possibly, artificial intelligence can improve therapeutic effectiveness one day. AI tools are already being explored to predict COVID-19 disease course and customise treatments to reduce mortality. All that said, as long as there is no adequate treatment for some patient groups, trialling an alternative therapy may indeed save lives. It is crucial, though, that the right influences lead to such decisions. Besides scientific scrutiny, medical ethics should equally lead the charge.

References