Background: RT-PCR is the gold standard for COVID-19 diagnosis, but the lack of standardization of assays, whose diagnostic performance may widely vary, complicates the interpretation of the discrepancies that may be encountered. Study design. We conducted a retrospective study over a ten-month period at the Central Laboratory of Virology of Ibn Sina University Hospital of Rabat. We included nasopharyngeal swabs, positive and negative for SARS-CoV-2 on FilmArray BioFire® Respiratory Panel 2.1 Plus, which were subjected to our laboratory's reference test, MAScIR SARS-CoV-2 M kit 2.0, initially or after a freeze-thaw cycle. The results were compared, and each discrepant sample with sufficient volume underwent the third test, using ARGENE® SARS-CoV-2 R-GENE kit. Results: Of 80 SARS-CoV-2 negative samples on FilmArray, there were no discordant results, whereas of 80 SARS-CoV-2 positive samples on FilmArray, 21 had discordant results on MAScIR, and only 11 could be tested on ARGENE, revealing positive results in 6 cases. 12.7% and 76.5% correspond to the discordance rates for MAScIR (with one or both targets detected on FilmArray), while 14.3% and 100% correspond to those of ARGENE. As the estimated sensitivity and specificity of FilmArray, compared with MAScIR, were 100% and 79.2%, respectively, its lower limit of detection, and ARGENE assay results, made it difficult to distinguish between false positives on FilmArray and false negatives on MAScIR without further investigations. Conclusion: The implementation of a new assay in our laboratory revealed discrepancies suggesting a lack of sensitivity of our laboratory's reference test, leading us consequently to retain the SARS-CoV-2 positive result of these discordant samples on FilmArray, regardless of the detection of one or both targets. Our study, which is, to our knowledge, the first comparing FilmArray RP2.1 and MAScIR 2.0 assays for SARS-CoV-2 detection, highlights the urgent need to standardize RT-PCR assays for COVID-19 diagnosis.