Abstract
Background
The coronavirus disease 2019 pandemic has accelerated the need for rapid validation and implementation of assays for detection of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) in diagnostic specimens. Multiple molecular methods have received emergency use authorization by the U.S. Food and Drug Administration for detection of SARS-CoV-2 in upper respiratory specimens, with testing of nasopharyngeal (NP) specimens serving as the foundation for these assays. However, supply chain constraints and the need for improved ease and safety of collection have prompted consideration of other specimen types as alternatives to NP specimens for detection of SARS-CoV-2. Here, we compared two methods for SARS-CoV-2 detection in saliva: the Roche cobas® 6800 SARS-CoV-2 real-time RT-PCR Test (“Roche”), which tests for viral ORF1ab (target 1, T1) and envelope E genes (target 2, T2); and the Agena Biosciences MassARRAY® SARS-CoV-2 Panel/MassARRAY® System (“Agena”), which tests for targets in the ORF1ab gene (ORF1, Orf1ab) and nucleocapsid N gene (N1, N2, N3).
Methods
Sixty saliva specimens collected within 48 hours of SARS-CoV-2 detection in an upper respiratory (anterior nares or NP) specimen from the same individual were tested in both the Roche and Agena platforms. Each system was evaluated for overall detection results and agreement with results of matched upper respiratory specimens. In addition, we determined the limit of detection (LoD) for each system and its component targets using an in-house SARS-CoV-2 standard generated from pooled positive saliva specimens quantitated against a commercially available standard (ZeptoMetrix NATSARS(COV2)-ERC).
Results
Both platforms demonstrated a similarly high sensitivity (97%) and specificity (100%) when compared to matched patient upper respiratory specimens and had high agreement with one another (Cohen’s κ = 0.9321, p = 2.6x10-13). Overall, the LoD (copies/mL) for the Roche assay was four times lower than that of Agena for saliva specimens (390.6 v. 1562.5). Furthermore, we determined that the LoD differed among the target components of each assay. The experimental LoD was comparable across Roche targets, but probit analyses indicate T2 has greater sensitivity (LoD: 228.6), Of the five Agena targets, the N2 target had the lowest LoD (1562.5).
Conclusions
In sum, we demonstrate that saliva is an acceptable specimen for testing in both the Roche cobas® 6800 SARS-CoV-2 real-time RT-PCR Test and the Agena Biosciences MassARRAY® SARS-CoV-2 Panel/MassARRAY® System, and both provide sensitive and specific detection of SARS-CoV-2 in saliva specimens. Although there was a high level of agreement between platforms, the LoD was lower for the Roche compared to the Agena assay with T2 and N2 being the most sensitive targets on each platform, respectively. The addition of saliva as an acceptable specimen and understanding the sensitivity for testing on these platforms can further inform public health measures for screening and detection to combat the pandemic.
Evaluation of the Cepheid Xpert Xpress SARS-CoV-2 test for bronchoalveolar lavage
