Performance evaluation of the BioFire respiratory panel 2.1 plus and exploratory analysis of Ct values for respiratory pathogen quantification

BACKGROUNDS: The BioFire FilmArray RP2.1 plus is a syndromic multiplex PCR panel for qualitative detection of 20 respiratory pathogens. BioMérieux recently introduced FIREWORKS, a software platform providing access to amplification data, including cycle threshold (Ct) values. OBJECTIVES: This study evaluated the diagnostic performance of the BioFire RP2.1 plus and explored the potential utility of Ct values for semi-quantitative assessment of pathogen load. METHODS: This retrospectively study included 72 specimens. Some were serially diluted for quantitative analyses, yielding 130 PCR tests. Samples were tested on the BioFire RP2.1 plus, and compared with Xpert, Xpress Cov-2/Flu/RSV plus (Cepheid) or in-house real-time PCR assays QuantStudio, resulting in 708 PCR results. Agreement was assessed qualitatively using the kappa coefficient, and exploratory analyses were performed using Passing-Bablok regression and Spearman correlation. RESULTS: The BioFire RP2.1 plus demonstrated excellent qualitative agreement with comparator methods (kappa of 0.963). Among 130 tests, 117 were concordant. Nine discordant negatives involved Bordetella pertussis (n = 4), Coronavirus NL63 (n = 1), Influenza A (n = 2), and Influenza B (n = 2) predominantly in specimens with high Ct values (low pathogen load). Four samples were negative by QuantStudio but detected by BioFire RP2.1 plus. Exploratory quantitative analyses demonstrated strong correlations between BioFire Ct values and reference assays, with Spearman coefficients of 0.71 (QuantStudio) and 0.93 (Xpert). CONCLUSIONS: Overall the BioFire RP2.1 plus showed high qualitative agreement with routine assays for respiratory pathogens detection. Quantitative analyses suggested a reproducible relationship between BioFire-derived Ct values and comparator methods, although these findings should be interpreted cautiously given the exploratory design and methodological limitations of cross-platform comparisons.