RESUMO
BACKGROUND: Biomarker guided therapy could improve management of COVID-19 inpatients. Although some results indicate that antibody tests are prognostic, little is known about patient management using point-of-care (POC) antibody tests. METHODS: COVID-19 inpatients were recruited to evaluate 2 POC tests: LumiraDX and RightSign. Ease of use data was collected. Blood was also collected for centralized testing using established antibody assays (GenScript cPass). A nested case-control study assessed if POC tests conducted on stored specimens were predictive of time to sustained recovery, mortality, and a composite safety outcome. RESULTS: While both POC tests exhibited moderate agreement with the GenScript assay (both agreeing with 89% of antibody determinations), they were significantly different from the GenScript assay. Treating the GenScript assay as the gold standard, the LumiraDX assay had 99.5% sensitivity and 58.1% specificity while the RightSign assay had 89.5% sensitivity and 84.0% specificity. The LumiraDX assay frequently gave indeterminant results. Both tests were significantly associated with clinical outcomes. CONCLUSIONS: Although both POC tests deviated moderately from the GenScript assay, they predicted outcomes of interest. The RightSign test was easier to use and was more likely to detect those lacking antibody compared to the LumiraDX test treating GenScript as the gold standard.