Convalescent plasma could reduce death from COVID-19, early data suggests

COVID-19 patients treated with the blood of those who have recovered from the infection die at significantly lower rates than those given standard treatments alone, according to a preliminary analysis.

In their analysis, posted July 30 to the preprint database bioRxiv, the researchers looked at a dozen trials where hospitalized COVID-19 patients received convalescent plasma (CP) therapy — a treatment that involves drawing blood plasma from recovered patients and injecting the antibody-rich fluid into sick patients. The 12 trials, conducted at various sites around the world, included more than 800 participants in total, and when taken together, suggest that patients given plasma were less than half as likely to die as patients given other treatments, according to the report. 

Specifically, the mortality rate among patients given plasma was 13%, compared with 25% among patients given standard treatments. While that's trending in the right direction, the new analysis wasn't peer reviewed, and neither was some of the trial data the analysis looked at. What's more, only three of the 12 studies were randomized controlled trials (RCTs), where patients are randomly assigned to receive a treatment or standard of care, which is the gold standard for assessing medical treatments.

Related: Treatments for COVID-19: Drugs being tested against the coronavirus

"All studies come with limits, and basically what we're trying to do is provide a very high-level overview" of the data currently available, said author Dr. Michael Joyner, an anesthesiologist and physician-researcher at the Mayo Clinic in Rochester, Minnesota.

"The report provides a signal of hope that CP is beneficial, although unfortunately, it does not provide the confidence that is required to be able to responsibly recommend CP for the treatment of COVID-19," Dr. Mila Ortigoza, an instructor in the Departments of Medicine and Microbiology at NYU Langone Health, who was not involved in the research, told Live Science in an email. Ortigoza, who is currently co-leading a clinical trial of CP therapy for COVID-19, noted that none of the randomized controlled trials included in the analysis "recruited a sufficient number of participants to be able to make conclusions about efficacy" on their own.

"What the current study really highlights is the need to continue supporting ongoing RCTs of CP" to ensure that they enroll enough patients to provide "indisputable evidence" that the therapy really works, she said.

A positive signal 

As scientists design new medications for COVID-19, and clinicians repurpose existing drugs like remdesivir, doctors have also turned to CP therapy to treat the viral infection. 

"If you look at convalescent plasma, specifically … [it's] been applied to pandemics at least since the 1918 flu," Joyner said. CP therapies were subsequently used during the 2003 SARS outbreak, caused by a coronavirus related to the one that causes COVID-19, and the 2009 H1N1 pandemic, Ortigoza added. 

Because people who have recovered from a disease have mounted an effective immune response, CP therapy offers a way to treat infected patients by borrowing tools from the immune system itself — namely, antibodies that direct the immune system to attack a specific pathogen, or neutralize the bug directly, Live Science previously reported. 

Although promising on paper, CP has been difficult to study in practice. Trials of CP conducted during past pandemics often lacked control groups for comparison, meaning the effects of CP could not be weighed against those of an alternate therapy, or the standard of care, Ortigoza said. But in the context of a pandemic, well-controlled trials can be difficult to execute at the scale and speed required to draw clear-cut conclusions for people who may need treatment right away.

Related: 20 of the worst epidemics and pandemics in history

"In a pandemic, you can't always get a definitive 'ah-ha' study" that clearly demonstrates a therapy's efficacy, Joyner said. Plasma therapies pose a particular challenge, given that they rely on blood donations from donors who are both eligible to donate plasma and test positive for antibodies, he added. 

A typical donation yields about 20 to 27 ounces (600 to 800 milliliters) of plasma, which can then be used for several doses of CP at 6.7 to 10.1 oz (200 to 300 ml) each, Joyner said. Patients in the NYU trial receive one to two 8.4 oz (250 ml) doses of plasma, for example. Plasma can be stored for years at below-freezing temperatures, meaning hospitals could potentially build up a supply for COVID-19 patients. But given that hospitals only receive CP as the appropriate plasma donations become available, which may depend on both the rate of donations and the prevalence of COVID-19 in the area, organizing a trial where a significant number of patients randomly receive either plasma or a standard treatment becomes difficult. 

In addition, the number of COVID-19 patients at a given hospital fluctuates, making recruiting participants for RCTs even trickier.

But it's important, because "it is very challenging to draw conclusions from non-RCT studies with confidence because they lack the randomization process," and small RCTs, albeit randomized, don't include enough patients to yield reliable statistics or be generalized to larger populations, Ortigoza said. 

With these caveats in mind, Joyner said his team still felt it important to pool the available data and see what trends materialized; specifically, they wanted to know if COVID-19 patients who receive plasma die at lower rates than those who do not. In addition to three RCTs, the team analyzed four case series studies, which tracked the clinical outcomes of a small group of individuals given CP. The other five trials were matched-control studies, meaning each patient given CP was compared to a similar patient given a standard treatment, but these treatment assignments weren't randomized. 

Related: Why COVID-19 kills some people and spares others. Here's what scientists are finding.

In crunching all the numbers, "you start seeing this mortality benefit that's substantial," meaning CP patients do seem to die at significantly lower rates, Joyner said. However, more RCTs will be needed to nail down finer details, such as which patients benefit most from the treatment, or when the plasma should be given over the course of the infection to produce the best results, Joyner said. "You might see a more striking effect if [CP is] used optimally," he said. 

Joyner and his colleagues will  add more trials to their meta-analysis as they emerge, and will conduct similar analysis on how CP therapy impacts the length of a patient's hospital stay, their intensive care unit (ICU) status and symptom severity, such as whether or not they require supplemental oxygen. NYU is also leading an initiative to pool data from ongoing RCTs, called COMPILE, which could "provide a faster and reliable answer relating to the effectiveness of convalescent plasma, prior to the conclusion of many ongoing RCTs," Ortigoza said. The analysis is similar to Joyner's, but will only include RCTs that meet specific criteria.

"When evidence with a high degree of confidence emerges, the data and safety monitoring board will make a joint recommendation to the leadership of all trials," according to a statement on the COMPILE website.

Even if CP is proven effective one other obstacle stands in the way of its widespread use: the limited number of certified blood banks.

"Most hospitals across the United States are not equipped or certified to perform apheresis in-house," meaning they cannot separate plasma from red blood cells and other components in donated blood, Ortigoza said. "Providing support to the certified blood banks ... across the nation will be crucial for the success of this therapeutic strategy."

Originally published on Live Science.