Direct RT-qPCR detection of SARS-CoV-2 RNA from patient nasopharyngeal swabs without an RNA extraction step

Abstract

The ongoing COVID-19 pandemic has created an unprecedented need for rapid diagnostic testing. The World Health Organization (WHO) recommends a standard assay that includes an RNA extraction step from a nasopharyngeal (NP) swab followed by reverse transcription-quantitative polymerase chain reaction (RT-qPCR) to detect the purified SARS-CoV-2 RNA. The current global shortage of RNA extraction kits has caused a severe bottleneck to COVID-19 testing. The goal of this study was to determine whether SARS-CoV-2 RNA could be detected from NP samples via a direct RT-qPCR assay that omits the RNA extraction step altogether. The direct RT-qPCR approach correctly identified 92% of a reference set of blinded NP samples (n = 155) demonstrated to be positive for SARS-CoV-2 RNA by traditional clinical diagnostic RT-qPCR that included an RNA extraction. Importantly, the direct method had sufficient sensitivity to reliably detect those patients with viral loads that correlate with the presence of infectious virus. Thus, this strategy has the potential to ease supply choke points to substantially expand COVID-19 testing and screening capacity and should be applicable throughout the world.

Fig 1. SARS-CoV-2 RNA can be detected from COVID-19 patient NP swabs by RT-qPCR without an RNA extraction step.

(A) NP swab diluents from 2 confirmed COVID-19 patients were pooled, and using the 2019-nCoV_N3 primer/probe set, the mixture was either (i) subjected to RNA extraction using the Qiagen QIAamp Viral RNA Mini Kit followed by subsequent testing by RT-qPCR (using the equivalent of 11.3 μl of swab diluent) or (ii) directly added to the RT-qPCR reaction, with or without a preheating step (5 minutes at 70°C, “NP sample + heat”). As a control, the indicated quantities of the CDC 2019-nCoV Positive Control SARS-CoV-2 synthetic RNA were spiked into M6 transport medium, purified using the QIAamp Viral RNA Mini Kit, and screened by RT-qPCR. NP swab samples from 7 additional donors were screened by direct RT-qPCR for SARS-CoV-2 RNA using the 2019-nCoV_N1 primer/probe set (B) or the 2019-nCoV_N2 primer/probe set (C), or for human RNase P RNA using the RNase P primer/probe set (D). NP swab samples from donors 1–4 were previously shown to contain SARS-CoV-2 RNA by standard clinical RT-qPCR, while donors 5–7 were negative. For each primer/probe set, 7 μl (A) or 3 μl (B–D) of NP swab diluent was tested in the RT-qPCR reaction per donor. For the N1 and N2 primer/probe sets, the fully synthetic SARS-CoV-2 RNA Control 2 from Twist Bioscience was loaded at serial 10-fold dilutions (A, 3 × 10⁶ copies; B, 3 × 10⁵ copies; C, 3 × 10⁴ copies; D, 3 × 10³ copies; E, 3 × 10² copies; F, 3 × 10¹ copies) as indicated in (B) and (C). NTC wells were included for each primer/probe set, and each was negative. For (B) and (C), the correlation coefficients (R²) of the standard curves were 0.999 and 0.995, respectively. The dashed line at cycle 40 in each graph indicates the limit of detection. CDC, Centers for Disease Control and Prevention; CT, cycle threshold; NP, nasopharyngeal; NTC, no template control; RT-qPCR, reverse transcription–quantitative polymerase chain reaction.

Fig 2. Distribution of CT values from COVID-19 patient NP swabs following direct RT-qPCR versus standard RT-qPCR that included RNA extraction.

A total of 150 NP swab samples representing high (CT values less than 20), intermediate (CT values of 20–30), or low (CT values of more than 30) SARS-CoV-2 RNA loads as determined by standard clinical RT-qPCR at the University of Washington in Seattle (aqua circles) were analyzed by the indicated method. All assays used the 2019-nCoV_N2 primer/probe set. Direct RT-qPCR was performed on 3 μl of NP swab diluent after heating for 10 minutes at 95°C (green circles). In parallel, RNA was extracted from 30 μl of NP swab diluent that had been previously heated at 95°C for 10 minutes, and RNA representing 3 μl of the original diluent was used in RT-qPCR (purple circles) to allow a head-to-head comparison with direct RT-qPCR on the same quantity of NP swab diluent. The limit of detection (CT of 40) is denoted with a dashed line. Samples with CT values above this cutoff were considered negative for SARS-CoV-2 RNA. The fitted curves are LOESS (locally estimated scatterplot smoothing)–smoothed CT values, with 95% confidence intervals in gray, against the mean of CT values detected in the clinical RT-qPCR assay with primer sets N1 and N2. Samples are ordered by the latter mean. The full dataset for this experiment and controls are provided in S1 Table. CT, cycle threshold; NP, nasopharyngeal; RT-qPCR, reverse transcription–quantitative polymerase chain reaction.

Fig 3. Limit of detection of direct RT-qPCR approach.

A total of 60 NP swab samples representing low loads (CT of 27–36) of SARS-CoV-2 RNA as determined by standard clinical RT-qPCR at UW in Seattle (purple circles) were analyzed by the indicated method. All assays used the 2019-nCoV_N2 primer/probe set. Direct RT-qPCR was performed on 3 μl of NP swab diluent after heating for 10 minutes at 95°C (green circles). In parallel, RNA was newly extracted from 200 μl of NP swab diluent (aqua circles) and processed with the UW LDT to control for the effect of freeze/thaw cycles. The limit of detection (CT of 40) is denoted with a red dashed line. Samples with CT values above this cutoff were considered negative for SARS-CoV-2 RNA. The fitted curves are LOESS (locally estimated scatterplot smoothing)–smoothed CT values, with 95% confidence intervals in gray, against the CT values detected in the 200-μl freshly extracted RT-qPCR assay with primer set N2. Samples are ordered by the CT value of freshly extracted 200-μl LDT samples. The full dataset for this experiment and controls are provided in S2 Table. CT, cycle threshold; LDT, laboratory developed test; NP, nasopharyngeal; RT-qPCR, reverse transcription–quantitative polymerase chain reaction; UW, University of Washington.