News Release 

New longitudinal study documents antibody responses to COVID-19 5 months after symptom onset

American Association for the Advancement of Science

Research News

A new 5-month longitudinal analysis of 254 COVID-19 patients who displayed a wide range of disease severity - from asymptomatic to deadly illness - suggests that IgA and IgM antibodies to the SARS-CoV-2 virus disappear quickly during convalescence. IgG antibodies likely persist for longer, but Katharina Röltgen and colleagues nonetheless documented a slow, inexorable decline in this antibody class as well, even in severely ill patients who mounted very strong initial antibody responses. Studying 983 plasma samples collected up to 5 months post-infection from 79 hospitalized COVID-19 patients and 175 outpatients and asymptomatic people infected with SARS-CoV-2, Röltgen et al. found a higher ratio of antibodies that react to the viral spike protein complex (versus those that target the nucleocapsid protein) in patients with mild illness compared with the severely ill patients, suggesting that this preponderance of spike-reactive antibodies may have helped to blunt the severity of symptoms. To facilitate their work, the researchers also developed an immunoassay that can detect successful blocking of the viral receptor binding domain's interaction with the human ACE2 receptor in human samples, providing an effective surrogate for detecting neutralizing antibodies. The authors say their findings may raise important questions about the reliability of seroprevalence studies, since the rapid waning of antibody titers may lead to an underestimation of how many people may have been previously infected in a given population. "The decrease in antibodies after infection also raises the question of how long antibodies elicited by vaccination will last, and whether frequent boosting will be needed to maintain protection," the authors further note.


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.