We evaluated two SARS-CoV-2 antibody point-of-care tests (POCTs) (Abbott Panbio COVID-19 IgG/IgM and Roche SARS-CoV-2 Rapid Antibody tests) and compared the results to their respective chemiluminescent immunoassays (CLIAs) (Abbott Architect IgM, Architect IgG, Roche Cobas total antibody assays). 200 pre-pandemic sera and 48 samples positive for various conditions (18 viral hepatitis, 18 dengue, 11 ANA and 1 dsDNA) were used as controls and to assess for cross-reactivity. Anonymised residual leftover sera positive for SARS-CoV-2 on RT-PCR were recruited as cases (n = 133). The sensitivity/specificity/cross-reactivity/positive predictive value (PPV)/negative predictive value (NPV) of the POCTs were assessed. Concordance between the POCTs and chemiluminescent immunoassays (CLIAs) were analysed. Abbott/Roche POCT specificity was 98.7%/100% (95% CI 96.5–99.8/98.5–100) and sensitivity was 97.2%/97.2% (95% CI 85.5–99.9/85.5–99.9) in cases ≥14 days post-first positive RT-PCR (POS), PPV was 68.7%/100% (95% CI 41.3–87.2/94.7–100.0), and NPV was 97.4%/97.6% (95% CI 97.0–97.8/97.2–98.0). In cases ≥14 days POS, concordance of Abbott/Roche POCT and CLIAs was 97.2%/100% (35/36 and 36/36 results). The sensitivity of individual IgM-band results on both POCTs did not increase >95% even after 14 days POS (Abbott 2.78%, Roche 44.4%). Both POCTs have good specificity, little cross-reactivity with other antibodies, and sensitivity >95% when used in subjects ≥14 days POS. Analysis of individual POCT IgG/IgM-bands did not provide any additional information. POCTs can substitute for CLIAs in cases ≥14 days POS. In low prevalence areas, POCTs would be especially useful when combined with antigen testing in an orthogonal format to increase the PPV of COVID-19 results.
Introduction: SARS-CoV-2 antigen tests can complement and substitute for RT-PCR tests. Centralized laboratory automated SARS-CoV-2 antigen tests that can be scaled to process a large number COVID-19 cases simultaneously are now available. We have evaluated the new Roche Elecsys SARS-CoV-2 antigen electro-chemiluminescent immunoassay. Methods: The Roche SARS-CoV-2 antigen assay is a double-antibody sandwich electro-chemiluminescent immunoassay, which reports a cut-off index (COI) (COI ≥ 1.0 considered positive). We assessed assay precision and linearity, and confirmed the reactivity limit. We determined the assay sensitivity and specificity with a verification group (289 controls and 61 RT-PCR positive COVID-19 cases). Assay performance was also validated against the consecutive samples we received (7657 controls and 17 cases) for SARS-CoV-2 antigen testing from June to October 2021. Result: The assay had a within-run precision CV of 3.0% at COI 0.68, and a CV of 1.5% at COI 3.49. Between-run precision was 3.0% at COI 0.68 and 1.8% at COI 3.49. The assay was linear from COI 0.65 to 7.84. All 35 C50 ± 20% test results performed over 7 days were positive/negative, respectively. In the verification group, overall sensitivity was 42.6% (26/61 positive, 95% CI 30.0–55.9), and specificity was 99.7% (1/289 positive, 95% CI 98.1–100). The agreement between the SARS-CoV-2 antigen and the RT-PCR cycle threshold (Ct) count was good (r = 0.90). In cases with Ct counts ≤ 30, the antigen assay sensitivity improved to 94.7% (18/19 positive, 95% CI 74.0–99.9). In our validation group, antigen sensitivity was 62.5% (5/8 antigen positive, 95% CI 24.5–91.5) within the first week of disease onset, but no cases were reactive after the first week of disease onset. Conclusion: The Elecsys SARS-CoV-2 antigen assay has good performance within manufacturer specifications. The sensitivity of the Roche antigen assay was greatest when used in patients with lower RT-PCR Ct values (≤30) and within the first week of disease onset.