Convalescent plasma may not prevent progression to severe disease or reduce mortality risk in hospitalized patients with moderate COVID-19, based on a phase 2 trial involving more than 400 patients in India.

The PLACID trial offers real-world data with “high generalizability,” according to lead author Anup Agarwal, MD, of the Indian Council of Medical Research, New Delhi, and colleagues.

“Evidence suggests that convalescent plasma collected from survivors of COVID-19 contains receptor binding domain specific antibodies with potent antiviral activity,” the investigators wrote in the BMJ. “However, effective titers of antiviral neutralizing antibodies, optimal timing for convalescent plasma treatment, optimal timing for plasma donation, and the severity class of patients who are likely to benefit from convalescent plasma remain unclear.”

According to Dr. Agarwal and colleagues, case series and observational studies have suggested that convalescent plasma may reduce viral load, hospital stay, and mortality, but randomized controlled trials to date have ended prematurely because of issues with enrollment and design, making PLACID the first randomized controlled trial of its kind to reach completion.

The open-label, multicenter study involved 464 hospitalized adults who tested positive for SARS-CoV-2 via reverse transcription polymerase chain reaction (RT-PCR). Enrollment also required a respiratory rate of more than 24 breaths/min with an oxygen saturation (SpO₂) of 93% or less on room air, or a partial pressure of oxygen in arterial blood/fraction of inspired oxygen (PaO₂ /FiO₂ ) ratio between 200 and 300 mm Hg.

Patients were randomly assigned in a 1:1 ratio to receive either best standard of care (control), or best standard of care plus convalescent plasma, which was given in two doses of 200 mL, 24 hours apart. Patients were assessed via clinical examination, chest imaging, and serial laboratory testing, the latter of which included neutralizing antibody titers on days 0, 3, and 7.

The primary outcome was a 28-day composite of progression to severe disease (PaO₂/FiO₂ ratio < 100 mm Hg) and all-cause mortality. An array of secondary outcomes were also reported, including symptom resolution, total duration of respiratory support, change in oxygen requirement, and others.

In the convalescent plasma group, 19% of patients progressed to severe disease or died within 28 days, compared with 18% of those in the control group (risk ratio, 1.04; 95% confidence interval, 0.71-1.54), suggesting no statistically significant benefit from the intervention. This lack of benefit was also found in a subgroup analysis of patients with detectable titers of antibodies to SARS-CoV-2, and when progression to severe disease and all-cause mortality were analyzed independently across all patients.

Still, at day 7, patients treated with convalescent plasma were significantly more likely to have resolution of fatigue (RR, 1.21; 95% CI, 1.02-1.42) and shortness of breath (RR, 1.16; 95% CI, 1.02-1.32). And at the same time point, patients treated with convalescent plasma were 20% more likely to test negative for SARS-CoV-2 RNA (RR, 1.2; 95% CI, 1.04-1.5).

In an accompanying editorial, Elizabeth B. Pathak, PhD, of the Women’s Institute for Independent Social Enquiry, Olney, Md., suggested that the reported symptom improvements need to be viewed with skepticism.

“These results should be interpreted with caution, because the trial was not blinded, so knowledge of treatment status could have influenced the reporting of subjective symptoms by patients who survived to day 7,” Dr. Pathak wrote.

Dr. Pathak noted that convalescent plasma did appear to have an antiviral effect, based on the higher rate of negative RNA test results at day 7. She hypothesized that the lack of major corresponding clinical benefit could be explained by detrimental thrombotic processes.

“The net effect of plasma is prothrombotic,” Dr. Pathak wrote, which should raise safety concerns, since “COVID-19 is a life-threatening thrombotic disorder.”

According to Dr. Pathak, large-scale datasets may be giving a false sense of security. She cited a recent safety analysis of 20,000 U.S. patients who received convalescent plasma, in which the investigators excluded 88.2% of cardiac events and 66.3% of thrombotic events, as these were deemed unrelated to transfusion; but this decision was made by the treating physician, without independent review or a defined protocol.

Michael J. Joyner, MD, of the Mayo Clinic in Rochester, Minn., was the lead author of the above safety study, and is leading the Food and Drug Administration expanded access program for convalescent plasma in patients with COVID-19. He suggested that the study by Dr. Agarwal and colleagues was admirable, but flaws in the treatment protocol cast doubt upon the efficacy findings.

“It is very impressive that these investigators performed a large trial of convalescent plasma in the midst of a pandemic,” Dr. Joyner said. “Unfortunately it is unclear how generalizable the findings are because many of the units of plasma had either very low or no antibody titers and because the plasma was given late in the course of the disease. It has been known since at least the 1930s that antibody therapy works best when enough product is given either prophylactically or early in the course of disease.”

Dr. Joyner had a more positive interpretation of the reported symptom improvements.

“It is also interesting to note that while there was no mortality benefit, that – even with the limitations of the study – there was some evidence of improved patient physiology at 7 days,” he said. “So, at one level, [this is] a negative study, but at least [there are] some hints of efficacy given the suboptimal use case in the patients studied.”

The study was funded by the Indian Council of Medical Research, which employs several of the authors and PLACID Trial Collaborators. Dr. Pathak and Dr. Joyner reported no conflicts of interest.