Over 120 years ago, a milk wagon horse named Jim was the United States’ most potent weapon against a raging diphtheria epidemic. During his lifetime, Jim—inoculated against the bacterial toxin that causes the disease—produced gallons of anti-diphtheria serum that, once extracted, could then be administered directly to patients. But tragic difficulties in making a safe and standard therapy from a single horse led Congress to pass the 1902 Biologics Control Act—the predecessor to the FDA’s current oversight over biologic products.
While convalescent sera have largely fallen out of favor since the development of modern vaccines, there is renewed hope in the space: the development of therapeutic sera from recovered COVID-19 patients. Encouraging the development of safe, pure, and potent COVID-19 convalescent serum has recently tasked policymakers with numerous challenges—some old, and some new. In this post, we explain the “manufacture” of COVID-19 convalescent sera and explore the regulatory and innovation policy difficulties in maintaining it.
What is convalescent serum therapy?
Recovered COVID-19 patients—including more than one million Americans—generally have antibodies to the SARS-CoV-2 virus, proteins in the blood to help their immune systems fight the infection. Some diagnostic tests look for these antibodies to determine whether someone has previously been infected; a successful COVID-19 vaccine would help patients actively produce their own antibodies before infection. These “neutralizing” antibodies are important to combat SARS-CoV-2 and are the source of significant promise for a robust treatment. One experimental intervention is to administer serum from recovered patients containing neutralizing antibodies to those who have not yet developed them—a process known as convalescent serum therapy, convalescent plasma therapy, or passive antibody therapy.
This labor-intensive process begins with collection of plasma (the liquid portion of blood with cells removed) from a patient with SARS-CoV-2 antibodies, a process that takes around 45 minutes as blood is removed, plasma is fractionated, and the remainder—mainly red blood cells—are returned to the patient. (The FDA has compiled information on donation options.) The antibody-rich plasma or serum (plasma with clotting factors removed) can then be administered to COVID-19 patients to help them fight the disease until their immune system develops its own active response.
Passive antibody therapy has a long history—including some improvements in survival during the 1918 flu pandemic and against the coronaviruses that caused SARS and MERS—and is expected to be most effective when administered before infection (when it can provide weeks to months of protection) or shortly after the onset of symptoms. Some researchers are studying the preventative effect of convalescent plasma in health care workers, but given the scarcity of plasma donors, this intervention mostly has been used for hospitalized COVID-19 patients.
The first randomized convalescent plasma trial for COVID-19 treatment, with 103 patients in China, found no statistically significant benefit among all patients but promising enough results among severely ill patients that a JAMA editorial found “optimism for the future of antibody therapy in this disease.” A matched control study in which 39 NYC patients with severe COVID-19 received convalescent plasma transfusions concluded that the treatment improved survival for non-intubated patients but not for intubated patients. But the lack of randomization makes interpretation more challenging. A report on 5,000 hospitalized COVID-19 patients as part of the FDA expanded access program for convalescent plasma concluded that the treatment appears safe—but the study did not have a control arm, leaving efficacy uncertain. A number of registered randomized trials are recruiting patients, although enrolling a sufficient number of participants can be challenging. As the New York Times reports, although the “only way to know for sure if the treatment works is to randomly assign patients to receive antibodies or a placebo … it can be impossible to find many patients who agree to have their treatment randomized to an unknown treatment,” especially because the product is already accessible outside the clinical trial setting.
How is convalescent serum regulated?
Convalescent serum is mainly regulated through the Public Health Services Act’s oversight of “biological products.” Section 262(i)(1) of the pertinent statute defines “biological products” subject to FDA approval, and specifically includes “therapeutic serum” in its list of covered products (a prophylactic serum would presumably be included in the term “blood component or blood derivative,” also in the list). Accordingly, any therapeutic serum must be approved by the FDA, something the Agency has not yet done for convalescent serum to treat COVID-19.
Despite the Act’s specific designation of therapeutic sera as biological products, historically, they have been quite rare. The use of serum was relatively popular in the 1920s and 1930s to treat polio, measles, and mumps, before the development of vaccines. The Public Health Services Act’s placement of therapeutic sera under the FDA’s jurisdiction wasn’t enacted until 1942. More recently, convalescent sera have been sparingly used in other pandemic contexts before other treatments were available; in 2009 it was used to treat patients in the H1N1 flu pandemic, and in 2013 in the West African Ebola outbreak.
Because it has not yet approved sera to treat COVID-19 patients, the FDA regulates convalescent serum as an Investigational New Drug (IND). In general, the FDA requires information to demonstrate that a product subject to an IND is adequately identified, pure, strong, and of sufficient quality (21 CFR 312.23(a)(7) and 21 CFR 312.305(b)(2)(vi)). This is easy enough to state. But regulation of serum is tremendously difficult. Every donor is slightly different, and the things the FDA normally measures—quality, purity, strength, and manufacturing chemistry processes—are not easy to measure for a product derived from individual donors’ blood. Serum is deeply unlike the manufacture of a standardized product produced at an identifiable facility. To the contrary, it has parallels with other difficult-to-regulate human donor products, such as stem cell therapy or fecal microbiota transplants. Regulating systems like these is substantially more complex than regulating standardized antibody products, which are complex enough on their own. The FDA’s May 1 guidance attempts to standardize the process of manufacturing convalescent serum, laying out requirements for patient and donor eligibility, the labeling of products, and recordkeeping. But substantial uncertainty—and significant variability, donor to donor, batch to batch—remains.
Normally, products being studied under an IND can only be obtained by participating in a clinical trial; the FDA first approved trials of the use of convalescent serum for COVID-19 in April. However, the FDA has created a nationwide expanded access (EA) program (not to be confused with its Emergency Use Authorization (EUA) program), within which patients can access the treatment without being part of a clinical trial. The COVID-19 therapeutic serum EA centers on clinical trials being run by the Mayo Clinic; over 2,600 sites are connected with the program, and over 40,000 patients have been infused. If patients cannot access convalescent serum through this EA, a physician can request an individual patient-level emergency IND; incredibly, the FDA normally responds within four hours. While this streamlined access helps patients get treatment, it also makes it difficult to enroll robust numbers of clinical trial participants.
How can innovation policy encourage convalescent serum?
Convalescent serum differs from standard pharmaceutical therapies in ways that limit the potential effectiveness of many typical approaches for encouraging innovation of a biologic product. As noted, because convalescent serum is typically obtained from donors rather than manufactured in large factories, one may worry that the lack of standardization and some randomness associated with finding a batch of effective plasma decreases companies’ incentives and ability to study it for its intended uses. In typical circumstances, this would suggest increasing the need for governmental funding in this space to overcome FDA regulatory challenges and to encourage standardization.
But there are more ways policymakers might encourage the development of convalescent sera at their source: encouraging sera donation from and sera administration to COVID-19 patients. First, there are a number of other therapeutic areas involving donation of human bodily materials (including blood, plasma, gametes like eggs and sperm, and fecal matter). The FDA and Congress in particular can learn from the ways both monetary and non-monetary incentives have been used to encourage donation, including the ways in which recovering COVID-19 patients might feel altruistically motivated to participate in a study that might help future patients. More directly, clinicians might seek to repurpose existing donation mechanisms for COVID-19 purposes. As one example, the Red Cross has begun testing all blood donations for COVID-19 antibodies. This provides a benefit to donors, who can learn whether or not they have been exposed to the virus, while holding themselves out to be asked whether they would like their serum to go to COVID-19 patients if they test positive.
Second, policymakers should prioritize federal funding not only for clinical trials evaluating the efficacy of convalescent serum in COVID-19 patients, but also in studying additional aspects of the use of convalescent serum. For instance, as has been documented in the fecal transplant space, it might be that material from particular donors is far more effective at treating the condition in question than material from others. It would be important to determine, for instance, whether donations from patients who had developed asymptomatic COVID-19 infections differed from patients who had developed symptomatic infections and displayed different levels of efficacy when later administered to symptomatic patients. Such information—which is comparatively nonexcludable and likely under-incentivized for larger-scale developers—should be a prime target for federal funding.
Third, the FDA might seek to set ex ante standards, as it has already begun to do in its May guidance document, to help scientists determine how to structure the collection process as well as clinical trials. Standardization in collection and processing might also help regulators compare the results of different clinical trials, help determine which convalescent sera have demonstrated efficacy, and encourage efforts to characterize the sera more clearly.
Policymakers also ought to pay attention to the ethical dimensions of the donation and use of convalescent plasma. A market has developed for blood donations from COVID-19 survivors, potentially raising concerns both about exploitation of donors and access to expensive treatments. More generally, the history of research involving human biological materials is filled with examples of human tissue sampling without subjects’ consent—especially from people who were disproportionately likely to be members of minority communities (although not exclusively so). Suffice it to say, the people from these examples did not share in the profits reaped by researchers and commercial developers. Convalescent serum, like other potential COVID-19 therapies, sheds yet more light on the racial disparities in COVID-19 cases. As policymakers work through ways to fund sera-based therapies, they should be attentive to equity in all parts of sera development—from donor to patient.
This post is part of a series on COVID-19 innovation law and policy. Author order is rotated with each post.
