Science’s COVID-19 reporting is supported by the Pulitzer Center.
We still don’t know how many people have been infected with the novel coronavirus, SARS-CoV-2. Not only have countries struggled to roll out wide-scale testing for the virus, those efforts inevitably will miss people who have recovered from an infection. The best way to figure out how far and wide the virus has spread in a population is to look at blood. Antibodies, blood proteins that the immune system produces to attack pathogens, are viral fingerprints that remain long after infections are cleared. Sensitive tests can detect them even in people who never felt a single symptom of COVID-19.
The World Health Organization has announced an ambitious global effort, called Solidarity II, of so-called serosurveys, studies that look for antibodies to SARS-CoV-2 in the population.
The United States has launched an unprecedented effort as well. One serosurvey is already underway in six metropolitan areas, including New York City, the hardest hit city in the United States. A second, even larger one, is on its heels, and together they should give a strong nationwide effort to track closely how many Americans have become infected as the pandemic unfolds. Serosurveys may also help efforts to develop vaccines, and, separately attempts to devise therapies to stop the virus from causing harm.
Science talked to Michael Busch, a transfusion specialist based at the University of California, San Francisco (UCSF), who is one of the leaders of these efforts. Busch has studied human blood infections caused by every imaginable virus. He directs the Vitalant Research Institute, a nonprofit that’s linked to 170 blood donation centers in the country and is world renowned for its infectious disease studies.
This interview had been edited for length and clarity.
Q: What’s your big picture view of serosurveys?
A: There’s a whole portfolio of “use cases” for serologic testing. You can use serology as an adjunct to the swab testing that detects the virus to help diagnose acute infections. It can identify donors of plasma from recovered patients that can be transfused into COVID-19 patients as a potential treatment. It can help estimate the timing of infection. And for vaccines, we need to have serologic tools that can discriminate between vaccine-induced antibodies and natural infection.
Q: Shortly after transmission occurs and people are acutely infected, you’re not going to have antibodies, are you?
A: The proportion of people who have recently acquired SARS-CoV-2 who would be positive with a single time point with nasal pharyngeal swab—the usual diagnostic sample, which uses the polymerase chain reaction to amplify tiny bits of viral nucleic acid so it can be detected—is probably 50%, or at best 70% to 80%. Antibody testing wouldn’t pick people up in the earliest stages of infection who had asymptomatic infections, but the data are becoming very solid that within 4 to 5 days of earliest disease onset, antibodies are detectable. So if you really want to pick up acute infections, you need to add serology to nucleic acid testing.
Q: What serosurveys are you doing in the United States?
A: We’re developing three large serosurvey studies. We need to do them at regular intervals to detect ongoing incidence, to determine if antibody responses are waning, and to assess herd immunity.
The first one, which will be funded by the National Institutes of Health, is already underway in six metropolitan regions in the U.S. It was started in Seattle when that outbreak happened, then New York City, then we quickly kicked in the San Francisco Bay area, and now we’ve added Los Angeles, Boston, and Minneapolis. Colleagues at regional blood centers are each saving 1000 samples from donors each month—often it’s just a few days each month—and they’re demographically defined so we know the age, the gender, and, most important, the zip code of the donor’s residence. Those 6000 samples, collected each month starting in March and for the next 5 months, will be assessed with an antibody testing algorithm, which we’re still finalizing, that will help us monitor how many people develop SARS-CoV-2 antibodies over time. That will show us when we’re going from, say, a half a percent to 2% of the donors having antibodies.
That will evolve into a national survey. With support from the U.S. Centers for Disease Control and Prevention [CDC], we’ll conduct three national, fully representative serosurveys of the U.S. population using the blood donors. That will be 50,000 donations in September and December of 2020 and November of 2021. We’re going to be estimating overall antibody prevalence to SARS-CoV-2 within each state, but also map it down within the states to regions and metropolitan urban areas, and look at the differences.
Q: Is this the largest serosurvey that’s planned in the United States right now?
A: To my knowledge, yes it is. It’s certainly the largest serosurvey I’ve ever been involved with.
Q: What’s the third serosurvey you’re doing?
A: We want to compare the results we’re getting from our blood donor serosurveys in several geographies with colleagues who are doing different types of populations surveys. We’re collaborating closely with colleagues at UCSF and the University of Washington, and they’re doing population serosurveys by neighborhood door knocking and capturing samples from the hematology lab.
Q: When do you think you’re going to have your first surveillance data that can answer the big questions about the percentage of the population that is asymptomatic or presymptomatic?
A: I can’t disclose the data, but we’ve got results for Seattle for March, and we’ll have results next week for New York City for the last week of March.
Q: Why don’t you reveal your data next week?
A: We’re cautious because blood donors are not a representative sample. They are asymptomatic, afebrile people [without a fever]. We have a “healthy donor effect.” The donor-based incidence data could lag behind population incidence by a month or 2 because of this bias.
Q: Infected people often do not have detectable SARS-CoV-2 in their blood, and so far, only viral nucleic acid—not infectious virus—has been found and there have been no reports of transfusion-related transmissions. Is SARS-CoV-2 contamination of the blood supply a concern?
A: At this point, it’s theoretical, and the Food and Drug Administration is recommending against screening either blood or tissue donors with laboratory testing for SARS-CoV-2 RNA. The FDA is very concerned that there’s not an overreaction to the blood safety risk. And I agree with them. But it’s enough of a concern that we are doing very large-scale studies to investigate the rate of detection of viral nucleic acids in donors, and also studies to address the potential for transfusion transmission in animal models including macaques and humanized mice.
Q: There are different types of antibodies that the immune system produces at different time points after an infection occurs. Can antibody tests act like clocks and reveal when someone was infected?
A: With a single sample from a patient, you can discriminate people who’ve been infected for 1 or 2 weeks as their immune response is maturing, versus longer periods of time following infection. There are many ways to do this—there’s a whole field that has evolved to detect new infections with HIV and hepatitis B and C viruses. These assays would not be very good at discriminating infections that occurred 3 or 4 days ago versus a week ago, but once you get out to 4 or 6 or 10 weeks following infection, the increasing signal intensity and pattern of antibodies allows you to impute the estimated date of infection with good precision.
Q: Why does it matter whether someone was infected 1 week ago or 1 month ago?
A: Dating when people were infected is important. When we get into the summer, fall, and winter, we will want to distinguish whether people were infected during this early outbreak period. You want to know whether this is a current seasonal infection or antibodies from the past outbreak season. That was the case with Zika. When a woman was pregnant and tested antibody positive, you wanted to tell her whether it was an infection acquired during this pregnancy—which meant her baby was at risk for microcephaly—or if she had been infected before the current pregnancy.
We’re also concerned about the related coronaviruses that cause the common cold in humans. These viruses induce antibody responses that, at least temporarily “neutralize” the virus, but they don’t appear to last. Because the neutralizing antibodies wane over the course of 1 to 2 years, people can be reinfected with the exact same coronavirus 1 or 2 years later. The simplistic idea is that people are going to get infected with SARS-CoV-2 and then they’ll be resistant to infection for the rest of their lives, and that herd immunity will accumulate over time, etc. But if this is anything like the more traditional coronaviruses that cause common colds, you’ll get robust neutralizing activity at first, but it will wane over time. And so our studies are really focused on assessing that issue of persistence of immunity from reinfection.
Q: What are the implications of waning immunity?
A: All this current effort to screen everybody and tell them, “Oh, you’re safe from future infection,” that may be true for now, but it may not be true a year or 2 from now. This also affects collecting plasma from recently recovered donors for transfusion to patients. A year from now, their antibodies may wane. You could potentially use SARS-CoV-2 vaccines to boost their immunity and also to harvest their neutralizing antibodies as therapeutics or for prophylaxis.
Q: What happens when SARS-CoV-2 infects a person who has antibodies to the other four coronaviruses that infect humans and cause the common cold?
A: We were on a call today with the CDC about this. If we look at people who just went through a SARS-CoV-2 infection and have a burst of antibodies against the virus, they’ve also boosted their pre-existing antibodies against the classic cold coronavirus. And the earliest antibody responses that CDC researchers have seen in careful longitudinal studies to SARS-CoV-2 are actually those cross-reactive memory responses to the classic cold coronaviruses.
Q: How might these cross-reactive antibody responses matter?
A: The immune memory to previous infections may help control infection with those cold viruses and even ameliorate symptoms of SARS-CoV-2 infection. But it can cause problems with the accuracy of SARS-CoV-2 diagnostics, as people reinfected with common cold coronaviruses could score as false positive with some SARS-CoV-2 serological assays.