Clinical evaluation of the Roche distributed SD Biosensor SARS-CoV-2 & Flu A/B Rapid Antigen Test amongst mild symptomatic people during the 2022/2023 winter season

Both influenza and SARS-CoV-2 are seasonal respiratory illnesses with similar symptoms, however distinguishing one from the other can have benefits for the patient and have different implications in various settings.

In this study we have evaluated the clinical performance of the Roche distributed SD Biosensor SARS-CoV-2 & Flu A/B Rapid Antigen Test during the 2022/2023 winter season, in a non-hospitalized, mild symptomatic population, comparing results with reverse transcription quantitative polymerase chain reaction (RT-qPCR). Participants also filled in a short questionnaire about their symptom onset, symptoms, vaccination status for both influenza and SARS-CoV-2.

We could include 290 people with complete records with female majority (72%, 209/290). Age ranged from 18 years old (minimum age for inclusion) to 71 years (mean age was 40.4 years). From the 290 inclusions 93 tested positive with SARS-CoV-2 PCR, 12 by influenza A and 6 by influenza B PCR. For SARS-CoV-2 overall sensitivity was 72.0% (confidence interval, CI 61.8-80.9%) and specificity 99.5% (CI 97.2-99.9%). SARS-CoV-2 RDT performed best up to and including PCR ct value of 25 (sensitivity 96% CI 85.8-99.5%), but could also detect samples less or equal to PCR ct 33, however with lower sensitivity (sensitivity 80.0% CI 69.6-88.1%). For influenza limited amount of samples were available; the RDT detected influenza A with 58.3% sensitivity (CI 27.7-84.8) and 100% specificity (CI 98.6-100.0%). In case of influenza B the inclusions were too low to calculate sensitivity reliably (2/6, 33.3% CI 4.3-77.7%); specificity was 98.2% (5/274, CI 95.8-99.4%). No cross reaction between SARS-CoV-2 and Flu A/B was experienced.

As was shown before, SARS-CoV-2 could be determined with high sensitivity in recent onset and lower than ct 25 samples. In spite of performing the study throughout the influenza season, we had sub optimal inclusions for determining RDT clinical performance; further studies are needed.