Respiratory infections are a common cause of illness in older adults, potentially resulting in severe morbidity or mortality. While up to 10% of respiratory infections in this population are caused by one of the four human coronaviruses (hCoVs), OC43, HKU1, NL63, and 229E, data on hCoV epidemiological and immunological responses are limited in communitydwelling older adults. In addition, it is often difficult to distinguish and identify distinct hCoV infections. Therefore, both clinical characteristics and the possibility of using serology to identify recent infections were investigated.
Clinical characteristics and humoral immune responses were studied in community-dwelling older adults who presented with hCoV-related symptomatic influenza-like illness (ILI). Serum antibodies specific for each hCoV were identified by protein microarray using recombinant spike proteins.
The symptoms of participants with molecular confirmation of hCoV infection were difficult to distinguish from symptoms of other viral pathogens causing ILI. Overall, severity based on a cumulative symptom score was less for hCoV than the other ILI-causing infections present in the study. Furthermore, symptom score did not correlate with changes in antibody levels. Using single serum samples to identify recent infections resulted in limited distinction among infections with receiver operating characteristic (ROC) area under the curve (AUC) values between 0.5 and 0.7, depending on the hCoV. However, paired serology samples collected at acute and recovery timepoints with an 8-week interval show an increase in type-specific antibodies with ROC AUC values between 0.78 and 0.96, depending on the hCoV.
Although clinical characteristics are comparable between hCoVs, the analysis of antibody kinetics may provide an alternative method for identifying recent hCoV infections.