Indian Study Shows No Survival Benefit of Plasma in COVID-19

The results, posted as a preprint on September 8 with a revision posted two days later, signal that the use of convalescent plasma in COVID-19 patients did not forestall progression to severe disease or mortality.

Study coauthor Pankaj Malhotra, a clinical hematologist at the Postgraduate Institute of Medical Education and Research in Chandigarh, India, says that these results won’t have much effect on clinical practice in India where convalescent plasma therapy is hugely popular and where the total number of COVID-19 cases has crossed 4.9 million. In June, the Indian health ministry recommended off-label use of plasma in COVID-19 patients who showed no improvement despite standard of care. Malhotra adds that the study could perhaps lead to a decline in the burgeoning illegal trade in plasma in the country. 

Arturo Casadevall, an immunologist at Johns Hopkins Bloomberg School of Public Health, tells The Scientist, “It’s the first RCT that was completed for convalescent plasma and I think that is a significant accomplishment.” An RCT in China was halted prematurely in late March due to dwindling cases and another in the Netherlands was discontinued as antibody titers in recipients were found to be comparable to that of donors, raising questions about the potential benefit of convalescent plasma. In the US currently, there are more than 20 RCTs that are actively recruiting patients, but none have been completed. A large Mayo Clinic study that was part of the US Expanded Access Program for COVID-19 convalescent plasma and that led to the FDA’s EUA was nonrandomized.

Despite the lack of survival benefit shown in the ICMR study, some positives gleaned from the trial include improved symptoms and oxygenation and faster viral clearance in patients in the intervention arm compared with the control arm.

Michael Joyner, the lead author of the Mayo Clinic study that prompted the FDA to grant EUA for convalescent plasma therapy to treat COVID-19 in the US, tells The Scientist that the authors, the Indian research council, and the country deserve a lot of credit for doing this trial under very difficult circumstances. “I think that’s most impressive. . . . So, high compliments there.”

“I see the cup being half full in terms of the viral load data and the improved oxygenation and so forth,” Joyner says. The half empty part, he adds, is that most of the plasma had low titers of antibodies and was given relatively late during the course of the disease—a median of eight days after onset of symptoms. “Those are the two main limitations of the study.”

See “The Search for Immune Responses that Stop COVID-19”

The Mayo Clinic investigators found that early treatment with plasma in hospitalized COVID-19 patients, within three days of diagnosis, with high levels of protective antibodies decreases mortality compared with later treatment using plasma with low titers of antibodies. Since this study did not include a control arm, however, the researchers can’t say if the reduced mortality was due to plasma treatment.

Casadevall, who is involved in two clinical trials at Hopkins to test whether plasma therapy is effective in COVID-19, finds the results of the Indian study encouraging, and contends that the low titers, likely because many of the donors were younger and fitter and had only mild disease, and the late use of plasma could perhaps account for the lack of survival benefit.

Adrian Newland, a hematologist at Barts Health NHS Trust, tells The Scientist in an email, “It has also been seen that the antibody titers often fall off quite quickly so the chances of only 200 ml of serum (given twice) of providing any significant protection is small.”

He adds that the standard of care is continuously improving and contributing to better survival. “Therefore, it is likely that if the benefit of convalescent serum is small, the impact is likely to be masked.”

Terry Gernsheimer, a hematologist at the University of Washington School of Medicine, writes in an email to The Scientist, “The numbers [in the ICMR study] are small and we don’t know how many patients it would take to treat before we would actually see a difference.”

She adds that randomization against standard of care is problematic as it is very difficult to keep either the patient or the staff blinded to the treatments that patients are receiving. In addition, the possible immunomodulatory effects of donor plasma, even plasma that is not convalescent, cannot be known. The best control arm, she says, would be non-convalescent plasma, which was not done. “Unfortunately that is true of all studies to date,” observes Gernsheimer.

See “First Antibody Trial Launched in COVID-19 Patients”

Thomas Marron, a cancer immunologist at the Icahn School of Medicine at Mount Sinai, tells The Scientist in an email, “I’m happy to see that there is finally some randomized data to report. That said, I still don’t think we are testing the appropriate population in many of these trials.”

He says that it is likely that viral control—which the antibodies in plasma are supposed to do—is more important during the early phase, while control of the hyperinflammatory response would be crucial in patients with more advanced disease. “I would worry that by the time you are having moderate respiratory distress, the horse has left the barn, and while we may give antivirals and convalescent plasma to stem any further viral spread, managing the inflammation should be the primary focus.”

There’s also the question of antibodies’ overall utility. “There is also the uncertainty of the importance of antibodies in fighting the [SARS-CoV-2] infection,” writes Newland. “There is increasing evidence that T lymphocytes are a more important source of immunity and this is one of the measures used in assessing the effectiveness of the vaccines now being developed.”

He says, “I think that, ultimately, convalescent serum has little to offer.”

Gernsheimer, who is involved in studies on convalescent plasma therapy in COVID-19 in the US, is more optimistic. “I think the best we can say here, which is what we have noted from the 70,000 patients who received convalescent plasma under a nonrandomized access to the plasma in the US, is that there is no apparent harm and further studies need to be done.”

A. Agarwal et al., “Convalescent plasma in the management of moderate COVID-19 in India: An open-label parallel-arm phase II multicentre randomized controlled trial (PLACID Trial),” medRxiv, doi:10.1101/2020.09.03.20187252, 2020.