Handling and accuracy of four rapid antigen tests for the diagnosis of SARS-CoV-2 compared to RT-qPCR

BACKGROUND: SARS-CoV-2 molecular diagnostics is facing material shortages and long turnaround times due to exponential increase of testing demand.

OBJECTIVE: We evaluated the analytic performance and handling of four rapid Antigen Point of Care Tests (AgPOCTs) I-IV (Distributors: (I) Roche, (II) Abbott, (III) MEDsan and (IV) Siemens).

METHODS: 100 RT-PCR negative and 84 RT-PCR positive oropharyngeal swabs were prospectively collected and used to determine performance and accuracy of these AgPOCTs. Handling was evaluated by 10 healthcare workers/users through a questionnaire.

RESULTS: The median duration from symptom onset to sampling was 6 days (IQR 2-12 days). The overall respective sensitivity were 49.4 % (CI95 %: 38.9-59.9), 44.6 % (CI95 %: 34.3-55.3), 45.8 % (CI95 %: 35.5-56.5) and 54.9 % (CI95 %: 43.4-65.9) for tests I, II, III and IV, respectively. In the high viral load subgroup (containing >106 copies of SARS-CoV-2 /swab, n = 26), AgPOCTs reached sensitivities of 92.3 % or more (range 92.3 %-100 %). Specificity was 100 % for tests I, II (CI95 %: 96.3-100 for both tests) and IV (CI95 %: 96.3-100) and 97 % (CI95 %: 91.5-98.9) for test III. Regarding handling, test I obtained the overall highest scores, while test II was considered to have the most convenient components. Of note, users considered all assays, with the exception of test I, to pose a significant risk for contamination by drips or spills.

DISCUSSION: Besides some differences in sensitivity and handling, all four AgPOCTs showed acceptable performance in high viral load samples. However, due to the significantly lower sensitivity compared to RT-qPCR, a careful consideration of pro and cons of AgPOCT has to be taken into account before clinical implementation.