Predicting SARS-CoV-2 antibody titers

From the Vaccine and Infectious Disease Division and the University of Washington

Two effective vaccines for Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) are now in circulation, but worldwide vaccine rollout will still lag behind ongoing infections, necessitating urgent treatments for those yet-unvaccinated individuals with severe COVID-19 symptoms. Passive immunotherapy treatment – where SARS-CoV-2-neutralizing antibodies (nAbs) from the plasma of convalescing patients are administered to acutely sick patients – is a promising strategy for COVID-19 treatment in severe cases. For example a recent phase II clinical trial in Argentina showed that convalescent plasma with high levels of nAb, specifically when given early in COVID-19 symptoms to older persons who did not yet require hospitalization, had a significant beneficial health impact as measured by keeping people out of the hospital.  However, not all SARS-CoV-2-infected people produce a robust neutralizing antibody response, so donor plasma must be screened for SARS-CoV-2-neutralizing antibody activity to determine which convalescent patients are suitable nAb donors. These neutralization assays are challenging and require a high level of biosafety containment. Alternatively, the identification of correlates of high nAb titers would eliminate the need for slow, time-consuming screening procedures and streamline plasma donor selection, but the clinical predictors for high SARS-CoV-2 nAb titer are yet unknown.

Scientists at the University of Washington and the Fred Hutch are studying the nAbs from patients recovering from COVID-19 to investigate which clinical factors predict good passive immunotherapy donors. In a recent publication in the Journal of Clinical Investigation, Dr. Jim Boonyaratanakornkit, a researcher and physician from the Vaccine and Infectious Disease Division), along with the Koelle Lab (University of Washington), measured SARS-CoV-2-nAb titers in the plasma of 250 people with PCR-confirmed SARS-CoV-2. Team members at the University of Washington Department of Laboratory Medicine and Pathology studied fast, high throughput commercial SARS-CoV-2 antibody tests in parallel with the nAb assay, and identified tests with good correlation characteristics that make them reasonable surrogates for the cumbersome nAb assays.   They then correlated nAb titer with clinical or demographic data such as age, sex, co-morbidities, or severity of COVID-19 disease. This comparison demonstrated that increased age, male sex, and severe COVID-19 symptoms such as breathing difficulty and fever correlated with higher nAb titers. Additionally, a longer period between SARS-CoV-2 infection and antibody screening correlated with decreased nAb titers. Together, these results suggest that severe COVID-19 disease generates higher levels of nAbs than less severe disease, and that anti-SARS-CoV-2 antibody titers in the blood may wane over time.

Graphical abstract of study workflow.
Graphical abstract of study workflow. Image provided by Dr. Boonyaratanakornkit

Since the onset of the SARS-CoV-2 pandemic, understanding the longevity of antibody-mediated immunity generated by natural infection has been a top priority. To further investigate their initial findings that the amount of time since COVID-19 symptoms correlated with decreased nAb titers, Dr. Boonyaratanakornkit and colleagues collected follow-up plasma samples from 41 of the original study participants. Further supporting their earlier data, all but four patients had decreased nAb titers by the second visit, with an average nAb half-life of 66 days. This suggests that SARS-CoV-2 nAbs may not provide long-term sterilizing immunity and that there may be a limited window after SARS-CoV-2 infection in which an individual’s plasma contains sufficient nAbs for clinically effective transfer.

Interestingly, seven, or approximately three percent of study participants with PCR-confirmed SARS-CoV-2 infection did not have detectable nAbs in their plasma. To further investigate the immune responses in these seronegative individuals, T cells from the blood of these participants were cultured with SARS-CoV-2 peptides that cover the major structural proteins of the virus. However, the T cells from seronegative individuals behaved similarly to T cells from healthy donors and did not produce proinflammatory cytokines, while seropositive donor T cells did, suggesting that a lack of seroconversion correlates with T cells that do not respond to SARS-CoV-2. Dr. Koelle commented on this finding: “A small percent[age] of people with documented COVID-19 are both seronegative and lacked T cell responses after recovery. This highlights the puzzle that some people seem to be able to clear the infection without elicitation of acquired immunity.”

This work demonstrates that factors such as age and severity of disease could be used to predict which convalescing COVID-19 patients might be good candidates to donate plasma for passive immunotherapy. These correlates could replace the need for traditional nAb assays. “COVID-19 seems to [be] one of a group of infections where the sicker one is, and presumably the more virus and therefore the more antigen that is around, the higher the levels of antibody,” Dr. Koelle said. However, this finding seems paradoxical, as a stronger immune response should limit SARS-CoV-2 replication and COVID-19 disease. Dr. Koelle explained that the plausible cause for this discrepancy is that the “immune response [in those with both severe disease and high nAb titers] obviously was not ‘effective’ in time to stop the illness. This points to the probable temporal ‘race’ between viral replication and host inflammation, and adaptive immunity. The virus and inflammation get out front, provoking an immune response that ends up looking stronger in retrospect.”

Going forward, the Koelle Lab hopes to understand why some people fail to seroconvert after SARS-CoV-2 infection and whether a lack of seroconversion is related to either a low viral inoculum or a successful innate immune defense at the onset of infection. Additionally, the lab is interested in understanding how antibody levels at the site of infection in the respiratory mucosa compare to plasma nAb kinetics. The lab is also curious if “there is a certain level of neutralizing antibody after natural infection that correlates with protection from re-infection,” Dr. Koelle said, but he added that “now with vaccines, we may never know!”

This work was supported by the Frederick National Laboratory for Cancer Research, the National Institute for Allergies and Infectious Disease, the Fred Hutchinson Joel Meyer’s Endowment, and the American Society for Transplantation and Cell Therapy.

UW/Fred Hutch Cancer Consortium members Keith Jerome and David Koelle contributed to this work.

Boonyaratanakornkit J, Morishima C, Selke S, Zamora D, McGuffin SA, Shapiro AE, Campbell VL, McClurkan CL, Jing L, Gross R, Liang J, Postnikova E, Mazur S, Lukin VV, Chaudhary A, Das MK, Fink SL, Bryan A, Greninger AL, Jerome KR, Holbrook MR, Gernsheimer TB, Wener MH, Wald A, Koelle DM. Clinical, laboratory, and temporal predictors of neutralizing antibodies to SARS-CoV-2 among COVID-19 convalescent plasma donor candidates. J Clin Invest. 2020 Dec 15;144930. doi: 10.1172/JCI144930. Online ahead of print.