Robust Automation and Point of care IDentification of COVID
As a rapid and definitive diagnosis of SARS-CoV-2 is essential, it is also key to consider that a recent data review has shown that 19% of patients with COVID-19 have co-infections and that these show an association with poor outcomes (JS Musuuza et al. 2021). Due to the benefits in the management of treatment and the timely isolation of infected patients with overlapping clinical symptoms, this highlights the need to identify other common viral and bacterial pathogens. The objective of this study was to evaluate the performance of the Respiratory 1 Panel (RESP003) CE marked kits, developed by GeneFirst (United Kingdom), for multiplex respiratory pathogen testing including 17 common causes of upper-respiratory tract infections and 11 common causes of pneumonia on biobanked clinical samples on the ACSIA - a High Throughput robotic system provided by PrimaDiag (France). The results from this system were assessed against BioFire® FilmArray®, a comparable testing platform used as the gold standard comparator, to determine the clinical sensitivity and specificity of the tests. Results from conventional microbiology techniques were also considered as necessary or required. APHP established a clinical study protocol and has been granted with the necessary approvals for the study. Samples have been formerly screened with FilmArray® or GeneXpert® in order to include the most adequate number of positives and negative cases. Overall, the evaluation criteria for the respiratory panels was: - Clinical sensitivity of the RESP003 test coupled with the ACSIA platform versus the comparator (target: 90%) - Clinical specificity of the RESP003 test coupled with the ACSIA platform in relation to the comparator (target: 98%) - Background noise (white limit) will be also evaluated. The study allowed the constitution of a biobank encompassing all major variants of SARS-CoV-2 samples, along with most of the other respiratory pathogens commonly reported in community acquired respiratory infections. The assay does not yet meet the required sensitivity specification using the automated result call, but the specificity is within range when the automated calling is verified. The reason for the discrepancy between the reference result and the post-biobanking results will be explored to curate and validate the biobank.
Steps to reproduce
Samples were collected between August 01st 2020 and April 01st 2022. The assay was registered with the French National Ethical Review Board under number ID 11032 N° 21.01.14.34949 ID-RCB number: 2020-A03542-37 under paragraph MR-002. All rhino-pharyngeal samples were collected with a swab, discharged into a transport medium, analysed and biobanked at -30°C. Lower respiratory samples, with remaining fluid compatible with the molecular analysis, were biobanked when positive for the bacterial pathogens targeted in the bacterial panel for testing using the reference assay and the GeneFirst RAPID COVID assay. The collected samples were divided across a “learning cohort” during which the assay design will remain open. While the chemistry tested on mock samples was frozen at this stage, the implementation on the automated system remained flexible in order to adjust the extraction, the program and the software to the variability observed in clinical samples. Nucleic Acid Extraction The RESP003 assay is an evolution of the GeneFirst SARS-CoV-2 assay that was performed on the upper respiratory samples. The performance of the PCR kit downstream of the Ademtech Viral-Prep DNA Extraction kit was validated in April 2020. This extraction kit was therefore adapted from the Kingfisher (ThermoFisher Scientific) instrument for use with the RESP003 assay for viral targets on the ACSIA platform. For the bacterial panel (RESP005) a custom protocol was used which included an in-house bacterial DNA extraction protocol and the Ademtech Viral Prep kit, already validated on the ACSIA platform. The in-house bacterial DNA extraction protocol is based on the Invitrogen MagMax™ Pathogen Extraction kit. The sample is suspended in lysis buffer, and a 1:1 volume of 100µm zirconium beads is added to the sample in a 2ml screwcap tube. The tube is placed on a Retsch MM400 bead mill for 2:30 min and centrifuged for 90 secs at 4000g. The sample is then processed according to the manufacturer’s instruction on a KingFisher magnetic bead handling automate. Automated Extraction, PCR Setup and Amplification The ACSIA instrument, allows for the extraction and setup of these 48 samples along with positive and negative controls. The current version of the platform receives the samples manually distributed in a 96 well plate. The plate setup is manually entered into the thermal cycler software, 10 runs on the instrument were performed. Each run took 2 hours and yielded a plate with extracted nucleic acid leftovers and a PCR plate ready to be run on the Bio-Rad CFX96. The PCR cycle takes 2 hours 30 min to complete. PCR Cycling The plate setup by the ACSIA platform was transferred to the Bio-Rad CFX96 instrument for amplification and the results exported to the controlling PC. Automated analysis using the Biosistemika software was performed, and a manual interpretation was carried out by the GeneFirst MPA development team for the validation of the automated analysis.