Q&A: Is convalescent plasma effective for COVID-19?
Researchers reported recently in The BMJ that convalescent plasma was not associated with a reduction in progression to severe COVID-19 or all-cause mortality in adults with moderate disease.
The results were from a phase 2 randomized controlled trial conducted at 39 hospitals in India.
Healio spoke with Shmuel Shoham, MD, an associate professor of medicine at Johns Hopkins University School of Medicine, about the clinical implications of the new study, and how convalescent plasma has been used since receiving emergency use authorization (EUA) from the FDA in August.
Q: What should clinicians make of these results?
A: So, a few things about it. One is that it is helpful to have the additional information that inpatients with COVID-19 who are receiving a convalescent plasma product are probably not going to see a significant difference. That's helpful. It also gets us a little further down the road as to identifying who the patients are who would be best positioned to respond to passive immunization, whether it's plasma, or a monoclonal antibody, the patients who are less likely or unlikely to see a response.
What we know from work that has been done both in randomized clinical trials and in observational studies is that timing seems to matter. A passive immunization strategy is most likely to have a positive impact if done early in the course of infection and if done with enough firepower.
Look at table one [of the study], which looks at the baseline characteristics of the study participants. Of the patients in the intervention and in the control group, 80% in the control and 86% in the intervention had detectable neutralizing antibodies at baseline. And what this suggests is that they've had the infection for a number of days, where we are gaining an understanding of when the body starts making an antibody response to the virus. And it's certainly not in the very early days; maybe it's after a week or so, maybe after 10 days. So that suggests that many of the patients had been banging around with the infection whether they knew it or not for a few days before the time that they were randomly assigned in the study. I think that that's an important lesson in a group like that.
From the observational study, among the 30,000 or so patients from the FDA/Mayo group, if you got the drug to the patients within 72 hours of diagnosis, that's when you had an impact. If you waited longer, you didn't.
Now, the other thing that is important is that I believe it was about a third of the donors had low or no neutralizing antibody levels. I feel pretty good to say that if you give a product that has low antibody titers, it will be less effective than if you give a product that has high titer. In some ways, a third of the recipients got dud units. So, what does that mean? Does that mean that the calculations were maybe off when they looked at a subgroup of people that got the higher antibody concentration?
Basically, if you have convalescent plasma with low antibody titers that you give too late, it is probably not going to work. If you have high antibody titers and you give it late, it may still not work. If you have high antibody titers and you give it early, I think that is the best chance for it to work and I think that the nonrandomized trial data suggest that.
Q: Should the FDA withdraw its EUA for convalescent plasma?
A: I'll preface by saying I'm not a regulator, so I can't tell regulators how to decide because they're looking at the world through a very different prism. I would say EUA is the lowest level of nonapproval approval that there is. The first part I think that they must look at is safety. I don’t think that from this paper, or from our experience, that there's a scary safety signal. That's not an issue. The second thing is they look at efficacy. I don't think this paper knocks out the chance of efficacy. So, I think that doctors and patients and their families should be aware of this information with all of its limitations.
In my mind, the power of an EUA is that it sends a signal that this is the lowest level of authorization, this is not an approval and it also helps empower local doctors, patients and their families to make decisions about their health as opposed to having a product outlawed. If you withdraw the EUA, essentially, you're outlawing the product except in a clinical trial.
Q: Were hospitals using it?
A: Yes, people are using it all the time. Are they using it more or less or the same as when it was part of the expanded access protocol? I don't know. I don't have numbers. An informal survey that I've taken suggests that it's being used as much. But at the same time, there's been a big push toward trying to get people into randomized clinical trials in the inpatient arena and the outpatient arena to try to get more information and further nail this down.
Q: Are patients asking for it?
A: Some are it depends on their level of medical sophistication and comfort with a blood product. I think that is dependent on conversations with doctors and talking to them about the benefits of that therapy. People are asking for it. Sometimes people are asking for it in situations where it has, in my mind, the best chance of working early on, some people are asking for it in situations where I think it has a much lower chance of working in the ICU and among patients who are incredibly sick.
Agarwal A, et al. BMJ. 2020;doi:10.1136/bmj.m3939.