Clinical Performance of Direct RT-PCR Testing of Raw Saliva for Detection of SARS-CoV-2 in Symptomatic and Asymptomatic Individuals.

Abstract

RT-PCR tests based on RNA extraction from nasopharyngeal swabs (NPS) are promoted as the "gold standard" for SARS-CoV-2 detection. However, the use of saliva samples offers noninvasive self-collection more suitable for high-throughput testing. This study evaluated performance of the TaqPath COVID-19 Fast PCR Combo kit 2.0 assay for detection of SARS-CoV-2 in raw saliva relative to a lab-developed direct RT-PCR test (SalivaDirect-based PCR, SDB-PCR) and an RT-PCR test based on RNA extraction from NPS. Saliva and NPS samples were collected from symptomatic and asymptomatic individuals (N = 615). Saliva samples were tested for SARS-CoV-2 using the TaqPath COVID-19 Fast PCR Combo kit 2.0 and the SDB-PCR, while NPS samples were tested by RT-PCR in RNA extracts according to the Irish national testing system. TaqPath COVID-19 Fast PCR Combo kit 2.0 detected SARS-CoV-2 in 52 saliva samples, of which 51 were also positive with the SDB-PCR. Compared to the NPS "gold standard" biospecimen method, 49 samples displayed concordant results, while three samples (35<Ct<37) were positive on raw saliva. Among the negative samples, 10 discordant cases were found with the TaqPath COVID-19 Fast PCR Combo kit 2.0 (PPA-83.05%; NPA-99.44%), compared to the RNA extraction-based NPS method, performing similarly to the SDB-PCR (PPA-84.75%; NPA-99.63%). The direct RT-PCR testing of saliva samples shows high concordance with the NPS extraction-based method for SARS-CoV-2 detection, and therefore provides a cost-effective and highly scalable system for high-throughput COVID-19 rapid-testing. IMPORTANCE The scale of the COVID-19 pandemic highlighted the need for viral diagnostic systems that are accurate and could be deployed at large population scales. Large-scale diagnostic or surveillance testing of large numbers of people requires collection of infected biological samples that is easy and rapid. Here, we demonstrate that raw saliva samples can be easily collected and tested by RT-PCR assays. Indeed, we find that direct testing of raw saliva by two different RT-PCR assays is as accurate (if not more accurate) than nasal swab-based RT-PCR testing. We present a cost-effective and highly scalable system for high-throughput COVID-19 rapid-testing.