key: cord-0993604-c5znj8dv authors: Litchfield, Mark; Brookes, Paul; Ojrzynska, Agnieszka; Kavi, Janki; Dawood, Richard title: Comparison of the clinical sensitivity and specificity of two commercial RNA SARS-CoV-2 assays date: 2022-02-23 journal: Int J Infect Dis DOI: 10.1016/j.ijid.2022.02.032 sha: 22c03008a20f7a5425aab119aa10a995a0515c4e doc_id: 993604 cord_uid: c5znj8dv Objectives This study aimed to compare the performance of the NeuMoDx™ SARS-CoV-2 Assay, implemented on the NeuMoDx 96 Molecular System, with the ThermoFisher TaqPath™ COVID-19 CE-IVD RT-PCR Kit (reference method). Methods Overall, 450 nasopharyngeal swab samples, previously tested using the reference method, were tested by the NeuMoDx Assay, and the clinical sensitivity and specificity of the assay was analysed. Results By retrospective statistical analysis of all valid results, the NeuMoDx Assay had a clinical specificity of 100% (95% confidence interval [CI]: 98.65–100.00) and a clinical sensitivity of 98.73% (95% CI: 95.47–99.85). Conclusions The NeuMoDx SARS-CoV-2 Assay demonstrated comparable analytical and clinical performance to the ThermoFisher TaqPath COVID-19 CE-IVD RT-PCR Kit. The NeuMoDx 96 Molecular System is well suited for automating medium-throughput routine SARS-CoV-2 testing or as an addition to high-throughput systems to allow fast-tracking for highly urgent clinical samples. Responding to the COVID-19 pandemic, researchers have developed several diagnostic assays for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Reverse transcriptase polymerase chain reaction (RT-PCR) is the gold standard for diagnosis of active SARS-CoV-2 infections due to its high sensitivity and specificity (Park et al., 2020) . Automation in molecular diagnostics enables scaling of testing capacity, which is critical for enabling a large number of tests (Eigner et Table S2 ). Samples were collected and processed in April 2020 (I week apart between sites), coinciding with period of low SARS-CoV-2 community prevalence. Of the 450 samples, three were excluded due to invalid results per the NeuMoDx manufacturer's specifications (Table S3) ; 447 were therefore included in the primary analysis (Table S4) . Of these, 12 had discordant results between assays, related to high cycle threshold (Ct) values using the reference method (Table S5a and S5b). Low level positives were excluded in a secondary analysis (Supplementary Methods); these results are presented here. In this secondary analysis, only two samples had discordant results. The clinical specificity of the NeuMoDx Assay was 100% (n/N=272/272; The LoD for the NeuMoDx Assay was 150 copies/mL (Table S1 ), in agreement with the manufacturer's evaluation (NeuMoDx Molecular, 2021). This exceeds the DHSC threshold (Department of Health and Social Care, 2021). In our study, the clinical specificity and sensitivity of the NeuMoDx Assay demonstrated similar analytical and clinical performance to the reference method and met the acceptance criteria for all the DHSC standards ( Table S6) . The observed shift in Ct values between assays may be partially explained by sample transport and freeze-thaw. Unfortunately, due to the number of samples needed, it was impossible to mitigate this. Freeze-thaw can have marked effects: one study reported a 0.41 increase in Ct and a change in 10.2% of sample results after just one freeze-thaw cycle (Li et al., 2020) . The two discordant results could also be accounted for by freeze-thaw or transcription errors. It was not possible for either laboratory to repeat test the discordant samples; this represents a study limitation. Other factors potentially impacting results are the use of off-label collection devices (see Supplementary Methods), samples not being tested in parallel, and a modest sample size. Furthermore, this study assumed the reference method is 100% sensitive and specific; however, samples were tested when prevalence was low and therefore the probability of false positives was higher. Importantly, the criteria for reference method positivity and low-level positivity were internally determined. Diagnostic laboratories frequently select cut-offs for assays, above which Ct values are considered negative (Sule and Oluwayelu, 2020); in this case, Ct ≥30 was applied. Furthermore, most diagnostic laboratories will verify the assay's LoD before implementation into routine testing (Burd, 2010) : here, the internally determined reference method LoD was 100 copies/mL compared with the manufacturer-stated 250 copies/mL (ThermoFisher Scientific, 2021), which may be due to lack of standardisation of SARS-CoV-2 reference materials. The recent availability of an International Standard (IS) for SARS-CoV-2 RNA may help alleviate these challenges (World Health Organization, 2020). As routine SARS-CoV-2 testing remains critical, the NeuMoDx Assay has demonstrated good clinical sensitivity and specificity on a platform well suited for automated clinical testing in our laboratory. Validation of laboratory-developed molecular assays for infectious diseases Guidance: Day 2 and day 8 PCR testing for international arrivals: minimum standards for providers Clinical evaluation of multiplex RT-PCR assays for the detection of influenza A/B and respiratory syncytial virus using a high throughput system Influence of Storage Conditions on SARS-CoV-2 Nucleic Acid Detection in Throat Swabs NeuMoDx Molecular I. NeuMoDx™ SARS-CoV-2 Assay Instructions For Use Development of Reverse Transcription Loop-Mediated Isothermal Amplification Assays Targeting Severe Acute Respiratory Syndrome Coronavirus Real-time RT-PCR for COVID-19 diagnosis: challenges and prospects Thermo Fisher Scientific's TaqPath COVID-19 CE-IVD RT-PCR Kit Now Available for the Incoming International Travel Quarantine Protocol Testing in the United Kingdom CE-IVD-RT-PCR-Kit-Now-Available-for-the-Incoming-International-Travel-Quarantine-Protocol-Testing-in-the Considerations for diagnostic COVID-19 tests 2402 Collaborative Study for the Establishment of a WHO International Standard for SARS-CoV-2 RNA The authors would like to thank staff at the UK Biocentre (Milton Keynes, UK) for the sourcing and provision of clinical samples and performance of reference method