Comparison of 4 molecular assays for detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)

Severe acute respiratory syndrome coronavirus 2 (SARS- CoV-2), the causative agent of 2019 coronavirus disease (COVID-19), has ubiquitously spread across the world, infecting over 2 million people and causing over 150,000 deaths globally, creating a catastrophic global health emergency (Food and Drug Administration, 2020a). Limited access to laboratory tests early in the course of its global spread has further fueled the COVI-19 pandemic.

In the United States, the Centers for Disease Control and Prevention's (CDC's) 2019-Novel Coronavirus (2019-nCoV) Real-Time RT-PCR Diagnostic Panel was the first assay to receive emergency use authorization (EUA) from the Food and Drug Administration (FDA) on February 4, 2020 (Food and Drug Administration, 2020b, Food and Drug Administration, 2020c). Therefore, many public health and clinical laboratories in the United States have implemented this assay. Numerous commercial device manufacturers subsequently obtained EUAs for the molecular detection of SARS-CoV-2. However, reagents are still being allocated by manufacturers. Hence, most clinical laboratories, including ours, have resorted to using multiple platforms to meet testing demands.

In this report, we describe the analytical performance of 4 platforms that are currently being run at Robert Wood Johnson University Hospital (RWJUH) in New Brunswick, NJ: CDC 2019-nCoV Real-Time RT-PCR Diagnostic Panel (Centers for Disease Control and Prevention, Atlanta, GA), Simplexa COVID-19 Direct assay (DiaSorin Molecular LLC, Cypress, CA), Xpert Xpress SARS-CoV-2 test (Cepheid Inc., Sunnyvale, CA), and Cobas SARS-CoV-2 (Roche Molecular Systems, Inc., Branchburg, NJ) (Food and Drug Administration, 2020c, Food and Drug Administration, 2020d, Food and Drug Administration, 2020e; World Health Organization (WHO), 2020). Currently, there are little published data comparing the analytical sensitivity of molecular assays that have received FDA EUA for SARS-CoV-2. Per package insert, the claimed limit of detection (LoD) for the CDC 2019-nCoV assay on 2 separate extraction platforms is 10^0.5 (Qiagen EZ1 Advanced XL) and 10⁰ (Qiagen DSP Viral RNA Mini Kit), Simplexa COVID-19 Direct assay is 500 copies/mL, Xpert Xpress SARS-CoV-2 test is 250 copies/mL, and the Cobas SARS-CoV-2 assay is 0.007 TCID₅₀/mL (ORF1 gene) and 0.004 TCID₅₀/mL (E gene) (Food and Drug Administration, 2020c, Food and Drug Administration, 2020d, Food and Drug Administration, 2020e; World Health Organization (WHO), 2020). A major limitation of this study is the lack of LoD studies to directly compare the analytical sensitivity of the all 4 molecular assays for the detection of SARS-CoV-2. Such studies could not be done due to the shortage of reagents and the need to conserve the test supplies to meet increasing clinical demands. Nevertheless, it is important to note that Ct values should not serve as comparators for sensitivity of a molecular assay as the results can vary depending on testing platform, amplification chemistry, and sample input. There is also an issue with the interpretation of the “inconclusive” result obtained from the CDC assay as, per the SOP provided with the EUA, all controls (positive, negative, and extraction) need to be passed and a valid signal (<40Ct) needs to be present for both viral targets, N1 and N2, for a result to be valid on the CDC assay. If only 1 of the viral targets is positive, the extracted RNA is retested and the final result reported. If the sample repeatedly gives an inconclusive result, it is forwarded to the state public health lab or retested on another platform for the detection of SARS-CoV-2. There is a concern that these “inconclusive” results on the CDC assay may reflect the presence of low level viral RNA that was not consistently detected for all viral targets.

Herein, we show the data on nasopharyngeal specimens from 10 positive and 10 negative patients that were tested on 4 different testing platforms (CDC, Simplexa, Xpert, Cobas assays) that received EUA from the FDA. These results provide some data on the comparative analytical sensitivity of the 4 testing platforms and also highlight the issues with the current interpretation of the “inconclusive” results from the CDC assay.

The CDC assay was introduced to test for SARS-CoV-2 at RWJUH on March 12, 2020, and used to test nasopharyngeal (NP) and oropharyngeal (OP) specimens per protocol (Food and Drug Administration, 2020b). The PCR amplification kit (CDC 2019-nCoV Real-Time RT-PCR Diagnostic Panel, lot #0000505284) and positive control (lot #000505848) were purchased from Integrated DNA Technologies (Coralville, IA). Total nucleic acid extraction and purification from clinical specimens were achieved using Roche Magna Pure LC 2.0 instrument and total Nucleic Acid Isolation kit (Roche Diagnostics, Indianapolis, IN). In order to determine the relative analytical sensitivity of the 4 molecular assays for SARS-CoV-2, NP specimens from 10 positive patients and 10 negative patients were run on the 4 platforms in parallel. The Simplexa assay had the lowest average Ct (20.2 for S, 21.1 for ORF1ab) and hence was the most sensitive assay, whereas CDC assay had the highest average Ct (24.7 for N1 and 25.7 for N2) and appeared to be the least sensitive assay. The sensitivities of Xpert (22.4 for E, 24.8 for N2) and Cobas assays (22.3 for ORF1, 22.8 for E) were intermediate between the Simplexa and CDC assays. There were 2 negative specimens (sample 16, 17) that had a high Ct (37.5 and 36.2) for only 1 of the 2 targets (E) for the Cobas SARS-CoV-2 assay, suggesting that this may have represented contamination introduced during pipetting while aliquoting the specimens or carryover contamination in the instrument during the run. Inadequate volume of remaining sample precluded repeat testing of samples #16 and #17 on the Cobas assay.

Between 12 March 2020 and 31 March 2020, a total of 538 NP or OP specimens were run on the CDC with 234 positive (43.5%), 277 negative (51.5%), and 22 inconclusive results (4.1%). Results of testing in each platform are shown in Table 1 . Among 22 samples with inconclusive result on the first test, 18 were rerun on the CDC assay from the extracted nucleic acid, of which 8 (44.4%) were positive and 10 (55.6%) were negative. Once the Simplexa assay was verified at RWJUH, the specimens with an inconclusive result from the first run on the CDC assay (n = 4) were automatically reflexed to the Simplexa assay due to its potential superior sensitivity, ease of specimen setup, and rapid turnaround time (90 min). All 4 reflexed samples tested positive (Table 2 ). There was a concern that the 10 samples that tested negative on the second run on the CDC assay may have represented positive samples with a low viral load. Only 5 specimens were available for retrospective testing, of which 4 samples (80%) tested positive on the Simplexa assay with average Ct of 31.8 for S gene and 35.3 for ORF1ab gene (Table 2). In summary, if all specimens with an initial inconclusive results on the CDC assay were reflexed to a second, potentially more sensitive Simplexa assay, the positivity on the repeat run would increase from 44.4% (CDC assay) to 70.6% (Simplexa assay). These results suggest that the specimens with an “inconclusive” result on the CDC assay could represent positive samples with a low viral load; hence, these specimens need to be reflexed to a more sensitive molecular platform for the detection of SARS-CoV-2.

Table 1.

SARS-CoV-2 results from nasopharyngeal samples obtained from 10 negative and 10 positive patients run on 4 testing platforms.

Samples	CDC SARS-CoV-2, Ct		Simplexa COVID-19 Direct, Ct		Xpert® Xpress SARS-CoV-2, Ct		cobas® SARS-CoV-2, Ct
	N1	N2	S	ORF1ab	E	N2	ORF1	E
1	26.1	28.3	22.8	24.2	23.7	25.6	24.1	24.8
2	30.2	32.8	27.9	29.7	27.4	29.9	27.4	28.5
3	18.0	18.9	9.9	10.6	13.1	15.5	13.2	13.3
4	25.0	26.6	19.1	19.9	21.5	23.9	22.1	22.1
5	20.8	22.6	17.0	18.0	18.3	20.6	18.2	18.3
6	22.7	22.8	19.5	20.5	21.2	23.2	21.5	21.7
7	27.3	27.2	21.2	21.4	24.0	26.6	24.7	25.0
8	22.8	24.0	20.6	21.4	24.1	26.4	23.4	23.8
9	22.0	22.2	17.2	17.8	21.0	23.2	20.0	20.6
10	32.4	31.9	26.7	27.2	30.0	32.6	28.8	30.2
11	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
12	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
13	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
14	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
15	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
16	0.0	0.0	0.0	0.0	0.0	0.0	0.0	37.5
17	0.0	0.0	0.0	0.0	0.0	0.0	0.0	36.2
18	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
19	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
20	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
Average positive samples Ct	24.7	25.7	20.2	21.1	22.4	24.8	22.3	22.8

Table 2.

Resolution of inconclusive results from the CDC SARS-CoV-2 assay on Simplexa COVID-19 Direct.

	CDC SARS-CoV-2, Ct				Simplexa COVID-19 Direct, Ct
Samples	N1	N2	Rnase P	Interpretation	S	ORF1ab	IC	Interpretation
1	UND	38.1	28.5	INC	32.5	32.8	30.2	Positive
1-repeat	UND	UND	28.2	Negative
2	UND	38.3	27.5	INC	0.0	39.1	30.9	Positive
2-repeat	UND	UND	27.4	Negative
3	UND	35.9	30.4	INC	30.1	31.2	31.2	Positive
3-repeat	UND	UND	30.7	Negative
4	UND	36.4	29.8	INC	32.9	37.6	31.3	Positive
4-repeat	UND	UND	29.7	Negative
5	UND	29.6	27.4	INC	0.0	0.0	30.5	Negative
5-repeat	UND	UND	28.1	Negative
6	UND	37.8	29.4	INC	32.2	32.6	29.8	Positive
6-repeat	N/A	N/A	N/A	N/A
7	UND	38.4	32.4	INC	0.0	0.0	30.5	Positive
7-repeat	N/A	N/A	N/A	N/A
8	37.0	UND	23.6	INC	32.5	32.1	29.8	Positive
8-repeat	N/A	N/A	N/A	N/A
9	UND	35.9	27.3	INC	34.5	37.4	27.8	Positive
9-repeat	N/A	N/A	N/A	N/A

UND = undetected; INC = inconclusive; N/A = not applicable; Rnase P = ribonuclease P; ORF = open reading frame target; IC = internal control.

Once Simplexa COVID-19 Direct assay was validated for clinical use, initial inconclusive results from the CDC assay (n = 4) were automatically reflexed to the Simplexa assay (patients 6, 7, 8, 9).

In conclusion, based on the differences in the average Ct values obtained on specimens from 10 patients who tested positive for SARS-CoV-2, the Simplexa assay had the lowest Ct, suggesting that it was the most sensitive of the assays tested, and CDC assay had the highest Ct, suggesting that it was the least sensitive assay. These results also suggest that the “inconclusive” result from the CDC assay may represent positive samples with a low viral load, and therefore, these samples need be reflexed to another molecular platform that is more sensitive.