Clinical Performance of Two SARS-CoV-2 Serologic Assays

Tang, Mei San; Hock, Karl G; Logsdon, Nicole M; Hayes, Jennifer E; Gronowski, Ann M; Anderson, Neil W; Farnsworth, Christopher W

BACKGROUND: The recent emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has resulted in a rapid proliferation of serologic assays. However, little is known about their clinical performance. Here, we compared two commercial SARS-CoV-2 IgG assays. METHODS: 103 specimens from 48 patients with PCR confirmed SARS-CoV-2 infections and 153 control specimens were analyzed using SARS-CoV-2 serologic assays by Abbott and EUROIMMUN (EI). Duration from symptom onset was determined by medical record review. Diagnostic sensitivity, specificity, and concordance were calculated. RESULTS: The Abbott SARS-CoV-2 assay had a diagnostic specificity of 99.4% (95% CI; 96.41-99.98%), and sensitivity of 0.0% (95% CI; 0.00-26.47%) at <3 days post symptom onset, 30.0% (95% CI; 11.89-54.28) at 3-7d, 47.8% (95% CI; 26.82-69.41) at 8-13d and 93.8% (95% CI; 82.80-98.69) at ≥14d. Diagnostic specificity on the EI assay was 94.8% (95% CI; 89.96-97.72) if borderline results were considered positive and 96.7% (95% CI; 92.54-98.93) if borderline results were considered negative. The diagnostic sensitivity was 0.0% (95% CI; 0.00-26.47%) at <3d, 25.0% (95% CI; 8.66-49.10) at 3-7d, 56.5% (95% CI; 34.49-76.81) at 3-7d and 85.4% (95% CI; 72.24-93.93) at ≥14d if borderline results were considered positive. The qualitative concordance between the assays was 0.83 (95% CI; 0.75-0.91). CONCLUSIONS: The Abbott SARS-CoV-2 assay had fewer false positive and false negative results than the EI assay. However, diagnostic sensitivity was poor in both assays during the first 14 days of symptoms.