The FDA's convalescent plasma fallout continues
The National Institutes of Health on Tuesday released a statement undercutting the Food and Drug Administration's emergency authorization of convalescent plasma as a coronavirus treatment — an escalation of an extraordinary public disagreement between federal agencies.
Why it matters: Thankfully, the main question surrounding the treatment is whether it works, not whether it's safe. But this feud could erode public trust in any future coronavirus treatments and vaccines, potentially for good reason.
Driving the news: An NIH panel of experts reviewed the existing evidence on convalescent plasma, including the FDA's analysis, and determined that "there are currently no data from well-controlled, adequately powered randomized clinical trials that demonstrate the efficacy and safety of convalescent plasma for the treatment of COVID-19."
- Although it said that serious adverse reactions to convalescent plasma are rare, the panel wrote that it's still unknown whether the treatment makes patients more susceptible to reinfection.
What they're saying: "The public is best served when health agencies are aligned in the kind of advice they're giving to providers and patients. Different agencies can have different interpretations of data, but the actionable advice should reflect a consensus view so patients have clear guidance," former FDA commissioner Scott Gottlieb told Axios.
- Another former FDA commissioner, Robert Califf, told Bloomberg that "if you just took random data and divided it into enough subgroups just by chance alone you would find differences."
- The translation, via Bloomberg's Anna Edney: "It seems the agency did what it doesn't like pharma cos to do — analyze the data a bunch of different ways until something works."
The other side: "Surprised by media uproar on Treatment Guidelines on convalescent plasma for #COVID19. Guidelines mirror EUA: possible benefit, seems safe, randomized trials needed. ... No news here," NIH director Francis Collins tweeted last night.