COVID: convalescent plasma not help

A recent study found that convalescent plasma from previously infected Covid-19 survivors was basically ineffective when transfused into sick new Covid patients (see covid convales plasma not help rct jama2020 in dropbox, or doi:10.1001/jama.2020.10044)

Details:

-- 103 patients with laboratory confirmed severe Covid-19 in an open-label, multicenter, RCT in 7 medical centers in Wuhan, China. Randomized to convalescent plasma in addition to standard treatment vs standard treatment alone, stratified by disease severity

-- mean age 70, more females in the transfusion group (48% vs 35%), comorbidities: hypertension 54%/cardiovasc dz 25%/diabetes 20%, body temp 36.5 deg C, resp rate 21 in transfusion group/14 control, lymphocyte 820, CRP 20 in transfusion vs 9, IL-6 17 in transfusion vs 22 control, D-dimer 2

-- convalescent plasma: they assessed the S-protein-RBD-specific (receptor binding domain) IgG antibody, which correlated strongly with the viral antibody neutralization titer (they comment that a serum neutralization titer of 1:80 was equivalent to a titer of 1:1280 for the S RBD, and all had S RBD titer of at least 1:640)

-- primary outcome: time to clinical improvement within 28 days, defined as patient discharge alive or reduction of 2 points in a 6 point disease severity scale, ranging from 1 (discharge) to 6 (death) [defined previously, but each step is a major clinical change (eg, going from needing noninvasive ventilation to mechanical ventilation or ECMO)

-- secondary outcomes included 28 day mortality, time to discharge, and the rate of viral PCR turning from positive at baseline to negative at up to 72 hours

--median interval between onset of symptoms and randomization was 30 days

--median plasma infusion was 200 ml, and 96% of patients had only 1 transfusion

 

Results:

-- clinical improvement within 28 days:

    -- convalescent plasma group: 52% (27 to 52 patients)

    -- control group: 43% (22 of 51 patients)

    -- difference 8.8%, HR 1.40 (0.279-2.49), p=0.26

-- in those with severe disease (respiratory distress with resting breathing >30 breaths per minute and/or hypoxemia with oxygen saturation 93% or less or PaO2/FiO2 <300):

    -- convalescent plasma group: 91% (21 of 23 patients)

    -- control group: 68% (15 of 22 patients)

    -- HR 2.15 (1.07-4.32),p=0.03

-- life-threatening disease (shock, organ failure, or requiring mechanical ventilation)

    -- convalescent plasma group: 21% (6 of 29 patients)

    -- control group: 24% (7 of 29 patients)

    -- HR 0.88 (0.30-2.63),p=0.83

-- 28 day mortality: no difference, 16% vs 24%, OR 0.65 (0.29-1.46), p=0.30

-- time from randomization to discharge: no difference, 51% vs 36% discharged by day 28, HR 1.61 (0.88-2.93), p=0.12

-- time from randomization to death: no difference, with HR 0.74 (0.30-1.82), p=0.52 (but too few people to be really significant: comparing transfusion to control, in the severe group, 0 vs 2 died; in the life-threatening group, 8 vs 10 died

-- PCR conversion to negative at 72 hours: 87% in convalescent group vs 38% in the control group, OR 11.39 (3.91- 33.18), p<0.001

    -- at 24 hours, 45% vs 15%, p=0.003; at 48 hours, 68% vs 33%, p=0.001

-- adverse events: 2 patients in the convalescent plasma group had adverse events within 2 hours after transfusion that improved with supportive care

Commentary:

--this trial was terminated early after only 103 of the plan 200 patients were enrolled, due to the containment of the Covid19 epidemic in Wuhan (good for Wuhan, not so good for the study)

--so, not much benefit from this convalescent plasma transfusion therapy.

    --there was an increased conversion rate from PCR positive to negative, but unclear what that means, and it did not appear to translate to much clinical improvement

    --there was  some apparent benefit in those with less severe disease (ie, not quite life-threatening), though statistically the interaction by disease severity was not significant. still, might have been significant if the study included more patients

    --there was a quite long delay (30 days) from symptom onset to initiation of therapy. maybe earlier would have been better? maybe by the time patients were this sick, the poor outcomes (and lack of neutralizing antibody benefit) was because other cycles of events had already been initiated beyond the direct viral effects (eg cytokine storm, ARDS, coagulopathy), and viral elimination didn't matter at this point??

--2 prior blogs of convalescent plasma, observational studies only:

    --5 people in china with severe and rapidly progressive pneumonia who received convalescent plasma, and 3 were discharged after >50 days

 http://gmodestmedblogs.blogspot.com/2020/03/covid-using-convalescent-serum-to-treat.html

    --10 patients on a variety of meds got the transfusion with improvement within a few days http://gmodestmedblogs.blogspot.com/2020/04/covid-convalescent-plasma-seems-to-help.html

    -- but, of course, these studies had very few patients, were not RCTs, and these patients might have gotten better on their own

--for coronaviruses (including SARS and MERS), convalescent plasma therapy has been used, though no definitive results; though there were observational studies suggesting benefit

--using high levels of antibodies to enhance passive immunity (vs the active immunity from vaccination) has been effective in many diseases: prevention of hepatitis B after exposure or with liver transplantation, rabies, vaccinia, botulinum, tetanus.... And, it does make mechanistic sense that giving lots of antibodies (probably best of the neutralizing sort, though not clear that this lab test translates into effective clinical response), might help eliminate infections prior to the body developing its own active immunologic response

--but,there are many concerns about how best to use convalescent plasma therapy:

    --limited understanding of mechanism and therapeutic components of the therapy

    --no standardized or evidence-based rationale for donor selection (eg: what cutpoint of neutralizing antibody or other antibody level should be required to be an effective donor, are there other non-immunologic components in transfused plasma that are either helpful or harmuful?)

    --how should the process be undertaken, from identifying potential donors and through transfusion in potential recipents, and with what quality control?

        --it is not so easy to get the appropriate convalescent plasma, and, especially if scaled up, there might also have significant harms (eg, also transfusing unwanted other blood products, such as other infections).

    --what are best times and indications for giving a convalescent plasma transfusion to a patient?

        --and, are their specific patient characteristics of the recipients that optimize or negate the benefit of the transfusion?

    --are their combination therapies that work better (eg: giving remdesivir along with transfusion therapy, especially earlier in the infection process, along with vitamin D, maybe a sprinkling of hydroxychloroquine.....)

 

Limitations:

-- this trial was terminated early, which may have led to its being underpowered to detect clinically important differences

-- they chose a specific antibody titer for the plasma donors. Was that appropriate? Should a higher titer cutpoint have been used? 

-- should they have evaluated other disease severity markers and stratified results by them as well (though too few patients in this study to have meaningful subgroup analyses, as above)?

so, as i wrote recently to a friend, this is all so confounding: we get bits and pieces of potentially useful interventions, and we get interesting and plausible mechanistic theories about why they might work, but then the better studies (usually quite flawed themselves) that undercut our hopes for an effective therapy. Again, and again. This study, not providing definitive answers about convalescent transfusion therapy, does suggest it is not likely to be a major important intervention on its own. though, i suppose, each time we go back to the drawing board, we are a bit more knowledgeable and more refined in our next approach....

geoff

 

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

 

to get access to all of the blogs:

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list