Overcoming Supply Shortage for SARS-CoV-2 Detection by RT-qPCR

Abstract

In February 2020, our laboratory started to offer a RT-qPCR assay for the qualitative detection of severe acute respiratory syndrome coronavirus 2. A few months after the assay was released to our patients, some materials, reagents, and equipment became in short supply. Alternative protocols were necessary in order to avoid stopping testing to the population. However, the suitability of these alternatives needs to be validated before their use. Here, we investigated if saliva is a reliable alternative specimen to nasopharyngeal swabs; if 0.45% saline is a reliable alternative to guanidine hydrochloride as a collection viral transport media; the stability of SARS-COV-2 in guanidine hydrochloride and in 0.45% saline for 10 and 50 days at room temperature; and if the primers/probe concentration and thermocycling times could be reduced so as to overcome the short supply of these reagents and equipment, without a significant loss of the assay performance. We found that saliva is not an appropriated specimen for our method-nasopharyngeal swabs perform better. Saline (0.45%) and guanidine hydrochloride have a similar SARS-CoV-2 diagnostic capability as tube additives. Reliable SARS-CoV-2 RNA detection can be performed after sample storage for 10 days at room temperature (18-23 °C) in both 0.45% saline and guanidine hydrochloride. Using synthetic RNA, and decreasing the concentration of primers by five-fold and probes by 2.5-fold, changed the assay limit of detection (LOD) from 7.2 copies/reaction to 23.7 copies/reaction and the subsequent reducing of thermocycling times changed the assay LOD from 23.7 copies/reaction to 44.2 copies/reaction. However, using real clinical samples with Cq values ranging from ~12.15 to ~36.46, the results of the three tested conditions were almost identical. These alterations will not affect the vast majority of diagnostics and increase the daily testing capability in 30% and increase primers and probe stocks in 500% and 250%, respectively. Taken together, the alternative protocols described here overcome the short supply of tubes, reagents and equipment during the SARS-CoV-2 pandemic, avoiding the collapse of test offering for the population: 105,757 samples were processed, and 25,156 SARS-CoV-2 diagnostics were performed from 9 May 2020 to 30 June 2020.

Keywords: RT-qPCR; SARS-CoV-2; validation.

Figure 1

Evaluation of whether saliva (green) is a reliable alternative specimen to nasopharyngeal swabs (blue). The N1 Cq values were consistently higher in saliva compared with the paired nasopharyngeal swab sample, suggesting a decreased diagnostic capability of SARS-CoV-2 RNA in saliva.

Figure 2

Evaluation of whether 0.45% saline (blue) is a reliable alternative viral transport media to guanidine hydrochloride (red). The N1 Cq values were analogous or slightly decrease in 0.45% saline compared with the paired guanidine hydrochloride sample, suggesting an acceptable diagnostic capability of SARS-CoV-2 in 0.45% saline.

Figure 3

Evaluation of the SARS-CoV-2 detection capability in guanidine hydrochloride (red) (a) and in 0.45% saline (blue) (b) after 10 and 50 days of incubation at room temperature (Nasopharyngeal samples). The N1 Cq values were slightly increased on day 10 versus day 0 for both tubes additives suggesting that a reliable SARS-CoV-2 detection can be performed for 10 days at room temperature after sample collection. However, the N1 Cq values were higher on day 50 versus day 0 for both tubes’ additives, indicating that after 50 days at room temperature, the diagnostic capability is affected, especially in guanidine hydrochloride. In summary, SARS-CoV-2 diagnostics can reliably be performed in both 0.45% saline and guanidine hydrochloride after sample storage for 10 days at room temperature, but for extended periods (e.g., 50 days), the detection seems to be more reliable in 0.45% saline. The dashed line is the Cq value cut-off value and a Cq of 45 was arbitrarily attributed to samples where SARS-CoV-2 was not detected.

Figure 4

(a) Amplification efficiencies and (b) Cq values observed for 105 SARS-CoV-2 positive samples. The values for both parameters for [+] (blue), [−] (green), and [−] fast (red) were almost identical, suggesting that the three tested conditions did not affect the diagnostic capability of the N1 assay. A three-fold decrease in the limit of detection (LOD) was observed for [−] compared with [+] (7.3 versus 23.7 copies/reaction), and a six-fold decrease in the LOD was observed for [−] compared with [−] fast (7.3 versus 44.2 copies/reaction) using synthetic SARS-CoC-2 RNA and probit regression, suggesting that the differences between the tested conditions will be noted only on samples with a very low viral load.