Montefiore in the News
Please Reevaluate the Data on Convalescent Plasma for COVID-19
- August 28, 2021
— Early administration of convalescent plasma with high antibodies may offer hope to some patients
by Arturo Casadevall, MD, PhD, Brenda Grossman, MD, MPH, Jeffrey Henderson, MD, PhD, Michael Joyner, MD, Nigel Paneth, MD, MPH, Liise-anne Pirofski, MD, and Shmuel Shoham, MD
As the Delta variant surges, we have a therapy that is inexpensive, available, potentially effective when used correctly, and underused in the treatment of COVID-19. We are referring to COVID-19 convalescent plasma (CCP), a form of antibody therapy derived from donors who have recovered from COVID-19. When CCP was first deployed, many of us anticipated that it would be followed by powerful new antiviral drugs for patients hospitalized with COVID-19. Unfortunately, such drugs have not materialized. We also anticipated that with highly effective vaccines, the pandemic would have ebbed by now. That too has not happened. Instead, the country has again reported facing more than 1,000 deaths in a single day and has limited therapeutic options.
Recent History of COVID-19 Convalescent Plasma
To understand the situation with CCP we need to review some history. In March 2020, the U.S. became the first country to make CCP readily available. The FDA allowed CCP use first through case-by-case requests from doctors, and next through an expanded access program (EAP) coordinated by the Mayo Clinic, which carefully monitored CCP use. The rationale for CCP use was threefold: convalescent plasma worked in prior epidemics, antibody from recovered patients had strong antiviral activity, and plasma was known to be remarkably safe. Most importantly, no other specific therapy was available.
By mid-summer 2020, almost 100,000 patients had been treated with CCP under the aegis of the EAP, and strong evidence emerged that it was safe and effective when used early in hospitalization, leading the FDA to issue, in August 2020, an emergency use authorization. The Mayo Clinic and FDA, each independently analyzing the EAP data, found that mortality was lower in CCP-treated patients if units with high antibody content were administered before severe disease had developed.
The deployment of CCP -- beginning within a month of the first pandemic surge -- was an American success story. It was driven by a grassroots effort of physician networks, blood banks, community organizers, and individual donors who self-organized to study and promote CCP use under careful oversight by the FDA. In the months that followed, several observational studies and small randomized trials reported that CCP use was associated with reduced mortality, and by October 2020, CCP was used in approximately 40% of all hospital admissions for COVID-19. Tragically, as with so much in COVID-19, CCP was politicized, creating a cloud from which it has yet to fully emerge.
In the fall and winter 2020-21, three randomized controlled trials from India, Argentina, and the U.K. evaluated treatment in hospitalized patients with severe disease. They each reported that CCP had no effect on overall mortality, despite suggestions of efficacy in some subgroups. In contrast, another randomized trial from Argentina showed that early administration of CCP to elderly outpatients substantially prevented disease progression from mild to severe. Unfortunately, this result was less influential in the U.S., where the FDA had limited CCP use to hospitalized patients. Consequently, there was a dramatic decline in CCP use, and by March 2021, only about 10% of hospitalized COVID-19
patients received CCP. We have estimated that this retreat from CCP use was associated with 30,000 excess deaths from December 2020 to March 2021. In late spring of 2021, a double-blinded, placebo-controlled trial from a Columbia University-Brazil collaboration and a large analysis of real-world data from HCA Healthcare reported that CCP use was associated with statistically significant reductions in mortality. These results arrived at a time when vaccines seemed to have vanquished the epidemic and thus received little attention.
Efficacy of COVID-19 Convalescent Plasma
Why is the evidence on CCP efficacy so mixed? The answer lies in the complexity of antibody action and the nature of CCP itself. Antibody therapies are effective only when given early in the course of disease and when the infused plasma contains enough antibody to effectively contain the virus. In COVID-19, the initial viral phase can trigger an over-exuberant inflammatory response later in the course of illness and is often the process responsible for COVID-19-associated deaths. Early administration of CCP with sufficient antibody reduces the amount of virus in the patient, reducing the chance of progression to life-threatening disease. By the time the inflammatory response in COVID-19 has taken hold, antibody therapy is less likely to be helpful.
Tragically, but understandably, given the hectic early days of the pandemic, several randomized trials were designed without taking these principles into consideration. Most trials included substantial numbers of patients in advanced stages of disease when the inflammatory response was dominant. Some trials also used plasma with insufficient antibody content. These trials might have been anticipated to fail, and when they did, they created a circular logic that discouraged plasma use. Since then, we have learned that local plasma is most effective, providing another explanation for why some trials using nationally sourced plasma failed to show efficacy. The reliance on trials for decision-making in clinical medicine is now so powerful that when these trials, which tested CCP in suboptimal cases, delivered the message that CCP was ineffective, plasma use dropped substantially.
Take a Second Look at the Data
As our country confronts a rapidly accelerating Delta variant surge, there is no time to redo clinical trials. We ask that physicians reevaluate the available data, review the guidance for use, and consider using CCP early in hospitalization, ideally as soon as the decision to admit has been made. CCP lacks the toxicity of remdesivir (Veklury), or the risks of steroid use, which include superinfections.
Some physicians may respond by pointing out that the recently published C3PO trial of early CCP therapy did not show that CCP prevented disease progression. However, a close examination of the trial design and outcome does not support that conclusion. The trial design was flawed in that it included in the primary outcome patients admitted during the index visit and considered unrelated emergency room visits as outcome events. If these patients are removed from the analysis, the numbers show that the group receiving CCP had a 35% lower chance of progressing to hospitalization (see our critique of the study). The study also shows that two other important outcomes -- dyspnea and worsening of disease on a five-point scale after day 1 -- were significantly less frequent in the plasma arm.
In most hospitalized patients, the disease process may be too advanced for CCP to be effective. Some patients, particularly those who are delayed in producing their own antibody response, may still benefit late in illness. Among people with blunted antibody responses -- such as some hematologic malignancy patients, transplant recipients, and some being treated for auto-immune disorders -- CCP may be helpful at any point in the disease course. This is included in FDA guidance for CCP use.
Individuals at high risk for COVID-19 progression should be treated with monoclonal antibodies as outpatients (the FDA has only authorized outpatient use) and, if the disease progresses, should be seen in a hospital where CCP is available.
Overall, we estimate that CCP may have saved close to 100,000 lives during the pandemic in the U.S. alone. If physicians heed FDA recommendations -- use CCP with adequate antibody levels early in hospitalization – the toll of the Delta variant could be lessened.
The authors comprise the leadership group on the COVID-19 Convalescent Plasma Project. Arturo Casadevall, MD, PhD, is a professor of molecular microbiology and immunology at the Johns Hopkins School of Public Health. Brenda Grossman, MD, MPH, is a professor of pathology & immunology and medicine at Washington University in St. Louis. Jeffrey Henderson, MD, PhD, is an associate professor of medicine and Molecular Microbiology at Washington University in St. Louis. Michael Joyner, MD, is a professor of anesthesiology at the Mayo Clinic. Nigel Paneth, MD, MPH, is an emeritus university distinguished professor of epidemiology and biostatistics and pediatrics at Michigan State University. Liise-anne Pirofski, MD, is a professor at Albert Einstein College of Medicine & Montefiore Medical Center. Shmuel Shoham, MD, is associate professor of medicine at Johns Hopkins School of Medicine.
Casadevall disclosed participation on the scientific advisory board of SAb Biotherapeutics; receiving payments as a speaker for Ortho Diagnostics; and payments for data and safety monitoring board (DSMB) work from Bristol Myers Squibb. Shoham disclosed receiving payments for consulting from Adagio, Celltrion, and Immunome; payments for DSMB work from Adamis, Intermountain Health, and Karyopharm; and research grants from Ansun.