key: cord-0953569-bbiqvodi authors: Perchetti, Garrett A.; Pepper, Gregory; Shrestha, Lasata; LaTurner, Katrina; Yae Kim, Da; Huang, Meei-Li; Jerome, Keith R.; Greninger, Alexander L. title: Performance Characteristics of the Abbott Alinity m SARS-CoV-2 Assay date: 2021-05-14 journal: J Clin Virol DOI: 10.1016/j.jcv.2021.104869 sha: 30388cda9887369d6e250586875c8a01634897c5 doc_id: 953569 cord_uid: bbiqvodi Mass molecular diagnostic testing for the SARS-CoV-2 pandemic has drawn on laboratory developed tests, commercial assays, and fully-automated platforms to accommodate widespread demand. The Alinity m instrument by Abbott is capable of detecting several clinically relevant pathogens and has recently received FDA emergency use authorization for SARS-CoV-2 molecular testing. The Alinity m performs automatic sample preparation, RT-PCR assembly, amplification, detection, and result calculation in under two hours. Here, we validate the performance characteristics of the Alinity m SARS-CoV-2 assay in comparison with the Roche cobas 6800 and Hologic Panther Fusion platforms. Across 178 positive and 195 negative nasopharyngeal swab specimens (C(T) range 14.30-38.84), the Alinity m detected one additional positive specimen that was found to be negative on the Roche cobas 6800 (PPA 100%, NPA 99.5%). Across a separate set of 30 positive and 174 negative nasopharyngeal swab specimens (C(T) range 14.1-38.5), the Alinity m had 100% positive and negative agreement with the Hologic Panther Fusion. Using SeraCare SARS-CoV-2 RNA standards, the assay limit of detection was verified to be two-fold more sensitive than the parameters stated by the SARS-CoV-2 AMP kit package insert, at 50 virus copies/mL. Assay specificity was 100% over 20 specimens positive for other respiratory viruses and intraday precision was 100% concordant with <2% CV. These data indicate the Abbott Alinity m system's high concordance with reference assays and exquisite sensitivity distinguishes the platform's suitability for SARS-CoV-2 molecular detection. As SARS-CoV-2 continues to spread, multiple automated molecular diagnostic platforms have received emergency use authorization (EUA) from the US Food and Drug Administration (FDA) in 2020. The benefits of assay automation and standardization are many, such as improved reproducibility, efficiency, and safety (1) (2) (3) (4) (5) (6) (7) . However, automated platforms can take considerable time and resources to implement into clinical workflows (8) (9) (10) . The Alinity m SARS-CoV-2 assay is intended for the qualitative detection of nucleic acid from nasal swabs, nasopharyngeal (NP) and oropharyngeal (OP) swabs, or bronchoalveolar lavage fluid (BAL). The assay detects two targets, located in the RdRp and N genes of SARS-CoV-2. A sequence unrelated to SARS-CoV-2 (RNA from the hydroxypyruvate reductase gene of Cucurbita pepo, a pumpkin plant) is introduced into each specimen at the beginning of sample preparation. This unrelated RNA sequence is simultaneously amplified by RT-PCR and serves as an internal control (IC) to demonstrate that the process has proceeded correctly for each sample. Here, we evaluate the clinical performance of the Alinity m SARS-CoV-2 assay (Abbott Laboratories, Chicago, IL, USA) using clinical NP specimens in comparison with reference assays on the cobas 6800 (Roche, Basel, Switzerland) and Panther Fusion SARS-CoV-2 (Hologic, Marlborough, MA) real-time RT-PCR assays (11) (12) (13) . To determine the analytical sensitivity for this assay, we ran initial and confirmatory dilutions of previously quantified SARS-CoV-2 RNA standards. To evaluate specificity, a panel of non-SARS-CoV-2 laboratory-confirmed respiratory infections was run on the Alinity m. Assay precision was evaluated utilizing RNA standards in triplicate with appropriate controls over 72 hours for each target concentration. Although the Alinity m SARS-CoV-2 assay is qualitative, here we also assess the linearity and PCR efficiency by serial tenfold dilutions of a remnant clinical NP specimen. Finally, we compare two Alinity m instruments side-by-side using positive clinical specimens over a range of SARS-CoV-2 viral loads. NP specimens in viral transport media (VTM) were sent to the University of Washington (UW) Clinical Virology Laboratory for SARS-CoV-2 RT-PCR testing, and deidentified excess sample was used to evaluate platform clinical concordance (14, 15 For analytical sensitivity, recombinant virus containing SARS-CoV-2 RNA (SeraCare Life Sciences, Milford, MA, USA) was used as a positive control at a stock concentration of 1,000 copies/mL. Assay calibrators at 1,000 copies/mL, 200 copies/mL, 100 copies/mL, 50 copies/mL, and 25 copies/mL were prepared based on dilutions of the positive control in dH2O and tested in triplicate for initial analytical sensitivity. Confirmatory limit of detection experiments were performed using 20 replicates at concentrations of 100 copies/mL, 50 copies/mL, and 25 copies/mL. Twenty non-SARS-CoV-2 respiratory NP specimens with CTs ranging 17.5-37.6 from an in-house respiratory panel were used to evaluate assay specificity. Using 373 remnant clinical NP specimens (178 positive and 195 negative) assayed on the Roche cobas 6800 platform, the clinical performance of the Alinity m SARS-CoV-2 assay was evaluated ( Table 1 , Supp. Data 1). The accuracy and percent agreement between the candidate (Alinity m) and reference method (cobas 6800) were calculated as follows: overall percent agreement = 99.7%, PPV = 99%, NPV = 100%, PPA = 100%, and NPA = 99.5%. Individual CTs for all 178 positive specimens run on the cobas 6800 were plotted against mean CTs from the two Alinity m instruments (Fig. 1) . The clinical performance of the Alinity m SARS-CoV-2 assay was also compared to the Hologic Panther Fusion SARS-CoV-2 assay. 204 remnant clinical NP specimens (30 positive and 174 negative) were assayed (Table 1) . Overall percent agreement, PPV, NPV, PPA, and NPA were all calculated at 100% accuracy for the candidate method. Individual CTs from the 30 positive specimens assayed on the Panther Fusion were plotted against mean CTs from both Alinity m platforms (Fig. 2 ). The initial estimation of LoD was performed in triplicate at 1,000 copies/mL, 200 copies/mL, 100 copies/mL, 50 copies/mL, and 25 copies/mL using SeraCare RNA standards of recombinant virus containing SARS-CoV-2 ( Table 2 , Supp. Data 2). Confirmation of the LoD was performed at 100 copies/mL and 50 copies/mL with 20 replicates for each instrument. A total of 110 contrived samples ranging from 50-1,000 virus copies/mL were run to evaluate the analytical sensitivity of the assay on two separate Alinity m instruments. Further assay sensitivity was interrogated after the Alinity platform detected all contrived samples below the manufacturer's claimed LoD at 100 virus copies/mL. Twenty-three additional samples, at 25 virus copies/mL, were assayed to confirm an empirical limit of detection on a single Alinity m due to reagent and consumable constrictions. The mean CN for the manufacturer declared 100 copies/mL concentration was determined at 36.09. However, the Alinity m was able to detect >95% of samples below this threshold, with a confirmed empirical limit of detection of 50 virus copies/mL, with a mean CN of 37.02. Non-SARS-CoV-2 infections from patient NP swabs (n = 20), laboratory-confirmed by UW Virology's laboratory-developed respiratory panel test, were evaluated for assay specificity (16) (17) (18) . Specimens included infection with adenovirus (n = 2), bocavirus (n = 2), non-SARS-CoV-2 coronavirus (n = 5), influenza (n = 3), human metapneumovirus (n = 1), parainfluenza (n = 3), respiratory syncytial virus (n = 2), and rhinovirus (n=2). Respiratory panel CT ranges for laboratory-confirmed respiratory infections spanned from 17.5 to 37.6. All respiratory infection samples were negative for SARS-CoV-2 when tested on the Alinity m. RNA standards were run daily at 1,000 copies/mL, 200 copies/mL, and with a negative control -He-La cellsover the course of 3 days. The mean CN, SD, and %CV were calculated at each target concentration (Table 3A) . All positive dilutions were detected with <2% CV and 100% concordance based on qualitative interpretation for both Alinity m instruments. All negative controls did not amplify. The Abbott Alinity m SARS-CoV-2 assay precision was then compared to UW (Table 3B ). All other positive dilutions were detected with <3% CV. Remnant clinical NP swabs were serially diluted 10-fold to evaluate assay linearity and PCR efficiency (Supp. Data 3). The most diluted samples (1:10x10 5 ) were omitted from the PCR efficiency calculation. The assay is linear from a CN/CT of approximately 23 -37 with an R 2 value > 0.99. PCR efficiency was calculated at 87.4%, respectively. Notably, the Alinity m assay is qualitative so does not require a quantitative calculation of linearity and PCR efficiency. To Here, the Alinity m SARS-CoV-2 assay by Abbott has been validated to be accurate, specific, and more sensitive than the AMP kit package insert declared, with an empirical limit of detection determined at 50 virus copies/mL. The instrument's sensitivity permits detection of very low viral load specimens that may be missed by other platforms. The clinical concordance to reference assays was greater than 95% NPA and PPA when compared to the cobas 6800 and Panther Fusion. In summary, the Abbott Alinity m SARS-CoV-2 assay performance was superior to the parameters stated in the SARS-CoV-2 AMP kit package insert and equivalent to other FDA EUA molecular assays. Advantages and limitations of total laboratory automation: a personal overview Total Automation for the Core Laboratory: Improving the Turnaround Time Helps to Reduce the Volume of Ordered STAT Tests Total laboratory automation: Do stat tests still matter The "hospital central laboratory": automation, integration and clinical usefulness A one-step, one-tube realtime RT-PCR based assay with an automated analysis for detection of SARS-CoV-2 Lab Automation in the Microbiology Lab: An Ongoing Journey, not a Tale? Benefits Derived from Full Laboratory Automation in Microbiology: A Tale of Four Laboratories Understanding, Verifying, and Implementing Emergency Use Authorization Molecular Diagnostics for the Detection of SARS-CoV-2 RNA Molecular Diagnosis of COVID-19: Challenges and Research Needs Clinical laboratory automation: a case study Evaluation of a quantitative RT-PCR assay for the detection of the emerging coronavirus SARS-CoV-2 using a high throughput system Comparison of Commercially Available and Laboratory-Developed Assays for In Vitro Detection of SARS-CoV-2 in Clinical Laboratories Detection of SARS-CoV-2 by Use of the Cepheid Xpert Xpress SARS-CoV-2 and Roche cobas SARS-CoV-2 Assays Comparative Performance of SARS-CoV-2 Detection Assays using Seven Different Primer/Probe Sets and One Assay Kit Validation of SARS-CoV-2 detection across multiple specimen types Detection and quantification of human metapneumovirus in pediatric specimens by real-time RT-PCR Comparison of Real-Time PCR Assays with Fluorescent-Antibody Assays for Diagnosis of Respiratory Virus Infections in Children Clinical Disease in Children Associated With Newly Described Coronavirus Subtypes Evaluation of the Alinity m HIV-1 assay for the quantification of HIV-1 RNA plasma viral load in a high-throughput molecular laboratory in South Africa Multicenter clinical comparative evaluation of Alinity m HIV-1 assay performance Multicenter clinical evaluation of alinity m HCV assay performance We would like to thank Jared Castor and Aria Bovell for thoughtful discussion and assistance with clinical data. This research was funded from departmental funds from the Department of Laboratory Medicine and Pathology. ALG reports contract funding from Abbott and Gilead for testing,research funding from Merck, all outside of the submitted work. Abbott had no role in the design or execution of the study.