key: cord-1054825-ahip00v5
authors: Antonara, Stella; Ozbolt, Patrick; Landon, Lorie; Fatica, Lisa; Pleasant, Tamra; Swickard, Juanita; Drury, Andrew; Wongchaoart, Nicholas; Cradic, Kendall W
title: Detection of SARS-CoV-2 infection in asymptomatic populations using the DiaSorin Molecular Simplexa and Roche Cobas EUA assays
date: 2021-07-31
journal: Diagn Microbiol Infect Dis
DOI: 10.1016/j.diagmicrobio.2021.115513
sha: a2d86126adc077f3801a7b330e4772037442998a
doc_id: 1054825
cord_uid: ahip00v5

Identification of asymptomatic patients is necessary to control the COVID-19 pandemic and testing is one of the measures to detect this population. We evaluated the clinical correlation of the DiaSorin Molecular Simplexa COVID-19 Direct (DiaSorin Molecular) and Roche Cobas 6800 SARS-CoV-2 (Roche) assays using 253 oropharyngeal (OP) swab specimens collected from asymptomatic patients. Agreement between DiaSorin Molecular and Roche was 97% (95% CI, 0.94 to 0.99), with a κ statistic of 0.90 (95% CI, 0.83 to 0.97) and a PPA of 89% (95% CI, 0.76 to 0.96) and NPA of 99% (95% CI, 0.97 to 0.99). Simple regression analysis of Ct values revealed a regression line of y = 1.065*X - 5.537 with a Pearson's r of 0.8542, indicating a good correlation between both platforms. The DiaSorin Molecular assay demonstrates clinical performance comparable to that of Roche in this population.

positive for SARS-CoV-2 onboard the Diamond Princess cruise ship (7) . Asymptomatic transmission has been described previously for coronaviruses and other respiratory viruses such as Middle East respiratory syndrome coronavirus (MERS-CoV), human rhinoviruses (HRVs), influenza virus, and severe acute respiratory syndrome coronavirus (SARS-CoV) (8) (9) (10) (11) (12) . While the prevalence of asymptomatic individuals with SARS-CoV-2 is not yet fully characterized, it is estimated that approximately 45% of those infected with SARS-CoV-2 will remain asymptomatic (13) . It has been challenging to accurately quantify the prevalence of asymptomatic individuals including limitations of previous study designs such as population size, longitudinal data, and distinguishing truly asymptomatic cases from those who are merely pre-symptomatic. Hence, the occurrence of asymptomatic patients with SARS-CoV-2 infections poses a significant threat to COVID-19 control efforts and may have been a critical factor in community spread (14) (15) (16) . In fact, recent studies have shown that in some cases, the detection pattern and viral load of infected asymptomatic individuals have been equal to that of symptomatic individuals, suggesting a similar likelihood of viral transmission (17) (18) (19) (20) .

Molecular diagnostic testing has been critical in identifying infected symptomatic cases during the current pandemic. However, as lockdown measures have started to ease, Emergency Use Authorization (EUA) in vitro diagnostic tests for SARS-CoV-2 are being used to assess symptomatic patients and asymptomatic individuals to facilitate a return to normalcy. Recent guidelines from the Centers for Disease Control and Prevention (CDC) recommend using available EUA in vitro diagnostic tests for SARS-Cov-2 to diagnose acute infection of both symptomatic and asymptomatic individuals, to guide contact tracing and treatment options, pre-operative testing, and isolation requirements (21) .

From a healthcare setting perspective, it is imperative that testing programs include asymptomatic testing since identifying individuals with suspected COVID-19 is critical for real-time cohorting decisions and infection control measures.

Accurate and reliable results are necessary for correctly identifying asymptomatic patients since molecular assays with low clinical sensitivity and specificity for asymptomatic populations may inevitably lead to more exposures and have profound safety implications for the public health and healthcare systems. Simultaneously, to meet the exponential demand in testing, there has been accelerated development of both molecular and serological assays across a plethora of platforms (22, 23) . The selection from over 200 molecular diagnostic assays and counting requires understanding their limitations in diagnostic performance and further data-based clarification of their testing utility with different COVID-19 populations. Therefore, in this study, we evaluated the clinical performance of the DiaSorin Molecular Simplexa COVID-19 Direct and Roche Cobas 6800 SARS-CoV-2 EUA assays using oropharyngeal swabs from asymptomatic patients. This information may provide significant insights to support population screening strategies and the use of these two assays for the detection of SARS-CoV-2 in asymptomatic patients, providing the rapid detection needed for diagnosis, isolation, and contact tracing of COVID-19 cases.

Specimen collection and storage. Flocked swabs were used to collect oropharyngeal specimens from asymptomatic patients with known exposures to suspected COVID-19 infections or for pre-operative testing for surgical patients at OhioHealth Riverside Methodist Hospital (Columbus, OH). After collection, the swabs were placed into 3 ml of sterile Universal Viral Transport (UVT; BD). Specimens were tested as soon as possible after collection, or if testing was delayed, were stored for up to 72 h at 2-8°C. Following routine testing, samples were stored frozen (≤-80°C) until comparator testing could be completed.

Study design. Assay performance was evaluated using a total of 253 oropharyngeal specimens originally submitted for routine COVID-19 testing at OhioHealth Riverside Methodist Hospital on the Roche Cobas SARS-CoV-2 assay. Positive and negative specimens were selected during the submission process for EUA approval of the Roche assay for use in asymptomatic patients. Samples were thawed and subsequently tested on the DiaSorin Molecular Simplexa COVID-19 Direct assay. The study population included patients 4-85 years of age and both genders and were selected based on the patient's status (asymptomatic individual reported on the order entry questions). 

Following testing of 253 specimens, DiaSorin Molecular and Roche's total percent agreement was 97% (95% CI, 0.94 to 0.99), with a κ statistic of 0.90 (95% CI, 0.83 to 0.97), demonstrating strong agreement between assays. Overall, a positive percent agreement of 89% (95% CI, 0.76 to 0.96) and a negative percent agreement of 99% (95% CI, 0.97 to 0.99) were observed when OP swabs from asymptomatic patients were tested using the DiaSorin Molecular assay ( Table 1) . Table 2 . An overall discordance rate of 2.8% was found between the two systems. were considered presumptive positives on initial testing by the Roche Cobas assay since they were negative for the ORF1ab target. Repeat testing by DiaSorin Molecular was also negative for all samples.

Additionally, review of records revealed that two of these samples were from patients with previous SARS-CoV-2 infections which had resolved. Therefore, these samples were categorized as true negatives during discordant resolution. Two additional discordant samples (OH-39 & OH-41) were identified as positive for SARS-CoV-2 by the DiaSorin Molecular assay with Ct values of 36.3 and 36.5 for the ORF1ab target but were negative by the Roche Cobas. Following discordant analysis and retrospective chart review, the DiaSorin Molecular SARS-CoV-2 assay showed an improvement of PPA to 95% (95% CI, 0.84 to 0.99). Furthermore, the total percent agreement of DiaSorin Molecular and Roche increased to 98% (95% CI, 0.95 to 1.0), with a κ statistic of 0.94 (95% CI, 0.88 to 1.0) ( Table 2) .

distribution obtained by DiaSorin Molecular Simplexa analysis from oropharyngeal-positive patients was compared with that obtained by Roche Cobas. The corresponding R 2 values indicated a good correlation (R 2 = 0.7297), with 95% confidence intervals of the slope (0.85 to 1.04) and intercept (-12.01 to 0.93) between DiaSorin Molecular Simplexa and the Roche Cobas assays ( Figure 1A) . Next, the differences in Ct values were plotted against the average values to generate a Bland-Altman plot. The mean difference was 3.6 Cts, with 95% limits of agreement (LoA) of -2.8 to 10.0 ( Figure 1B) . Two samples showed a notable discrepancy in Ct values for both targets suggesting an analytical or pipetting error in one assay.

Unfortunately, the specimens was not available for retesting. The overall Ct values of both DiaSorin Molecular Simplexa assay targets (S gene and ORF1ab) were significantly lower than that those from Roche Cobas (P = 0.0001), showing an average of 3.6 ± 1.0 Ct difference for the tested asymptomatic population ( Figure 1C) . Further breakdown analysis per target on each assay shows an average 0.8 ± 1.3

Ct difference between S gene and ORF1ab targets in the DiaSorin Molecular Assay and 0.8 ± 1.6 Ct difference between the ORF1ab and Pan-SARS for Roche Cobas (Figure 1D ).

Numerous molecular assays have been granted EUA by the Food and Drug Administration (FDA), however, all of them have been focused on symptomatic individuals. Testing of the symptomatic population was critical early in the pandemic and it was the core of medical decision-making during that time, especially for the population deemed high-risk. Now that we have moved past this initial stage of the pandemic, we have had to adapt to testing of both symptomatic and asymptomatic individuals, and our testing algorithms and criteria for detecting the novel coronavirus have also shifted. Testing individuals without symptoms is now part of the concentrated effort to curb SARS-CoV-2 transmission within the community; From active infection-finding through mandated population testing based on symptoms to contact tracing after possible exposure, pre-requirement testing before being admitted to the hospital for a procedure (e.g., pregnant women admitted for labor and delivery, patients admitted for elective surgery), or screening to safely return to schools or workplaces. Hence, the use of EUA approved assays should not be limited to the symptomatic population, and it is imperative to adopt new testing practices to curb SARS-CoV-2 transmission. While there are performance-driven publications of many EUA assays for symptomatic testing, data on the accuracy and performance of the existing molecular assays for asymptomatic populations is lacking.

Previously, we were able to compare the Simplexa COVID-19 Direct assay clinical performance to Roche Cobas SARS-CoV-2 using samples from symptomatic patients and made several observations, including limit of detection (LOD) and how each test performed in a head-to-head clinical comparison (24) . With the new measures to prevent the spread of COVID-19 infection, we expanded the comparison to the asymptomatic population. This evaluation is critical as numerous assays are used for testing asymptomatic individuals, and the clinical sensitivity of the assays has not been established thoroughly due to the limited availability of testing resources and the constraints on testing guidelines and clinical recommendations for this population.

Our data suggest that both DiaSorin Molecular and Roche real time-PCR methods yielded comparable results (κ = 0.90), with an overall agreement of 97% (95% CI, 0.94 to 0.99) ( Table 1 ). Of the samples tested, there were a total of seven discordant results between the two assays showing an overall discordance rate of 2.8% between the two systems ( Table 2) . These discordances may be due to differences in primer sequences, assay limits of detection, or other factors, highlighting the importance of comparing the performance of different testing platforms. Upon further review of the medical chart of those patients where discordance was observed, it was shown that three patients (OH-11, OH-22, OH-27) were previously positive patients who returned for care to our institution seeking a negative test result for medicosocial reasons. These patients were reported as presumptive positive due to the positivity of only the Pan-SARS target by Roche which is not specific for SARS-CoV-2. In addition, these patients reported no symptoms during the time of collection, were at least three weeks from the time of the initial diagnosis, and may not have been infectious at that time during testing. Although amplification only occurred in the PAN-SARS target, it is unlikely that the samples were SARS-CoV 1 since the virus is not in circulation. Hence, detection of this target is indicative of the current SARS-CoV-2.

There is a possibility with these high Ct specimens, that patients are shedding inactive fragments of viral RNA and that the positive test result is not indicative of active viral replication. This has been previously East respiratory syndrome coronavirus, and influenza virus), it is well known that after the immune system neutralizes these viruses, inactivated viral RNA degrades slowly over time, and may still be detected by RT-PCR for months after infection had resolved (28) (29) (30) (31) . This highlights the challenges that many labs may have with highly sensitive molecular diagnostic platforms and high Ct value cut-offs that Observed Ct values in asymptomatic positive patients covered a wide range indicating that the viral load in these patients is variable (Figure 1A-1B) . Similar ranges of Ct values have been reported for symptomatic patients (24, (32) (33) (34) (35) (36) . Hence, when performing tests on asymptomatic patients, both clinical and analytical sensitivity is critical as the most sensitive assay would identify individuals who carry and transmit the virus unknowingly.

In the Bland Altman analysis, our data showed the DiaSorin Molecular Simplexa COVID-19 assay Ct values were, on average, 3.6 Cts lower than those of the Roche Cobas SARS CoV-2. All samples, except two, were within the acceptable upper and lower 95% confidence interval. These two samples were not available for repeat testing and investigation of the discrepant Ct values. The mean difference of 3.6 Cts between the assays (Figure 1B) may be due to the number of variables that can affect assay performance. Assay Ct values depend on multiple factors including sample volume, the gene targets used for the virus detection, assay parameters, fluorescence thresholds, and the cut-off for each assay.

For example, the DiaSorin Molecular Simplexa assay requires 50 µl of a non-extracted sample with 10 µl used directly in the amplification reaction, while Roche Cobas utilizes 600 µl of sample volume that is subsequently extracted and used in the amplification reaction. Additionally, each assay has its specific gene targets; DiaSorin Molecular uses the two SARS-CoV-2 specific targets S gene and ORF1ab, while the Roche Cobas design includes a non-specific PAN-SARS and a specific ORF1ab target. The assay Ct cut-off for each assay also differs. DiaSorin Molecular has an assay cut-off of 40 cycles, compared to 45 cycles for Roche Cobas. These data reinforce the idea previously reported by Rhoads et al. (37) that Ct values from different assays cannot be compared or correlated as significant differences in absolute Ct values can be demonstrated in the same sample with different assays. Also, Ct values can vary significantly between and within methods by as much as 14 cycles, including tests that assess different gene targets for SARS-CoV-2 (37). These considerations of each platform's unique capabilities and limitations are essential when laboratories examine the possibility of implementing existing assays for asymptomatic testing, and the study design requirements during method comparisons.

For this study, we examined the correlation of the Ct values between assay targets within both assays to evaluate the efficiency of the assay design when used in the asymptomatic population ( Figure   1C-1D) . The primer and probe target design must allow for amplification efficiency and specificity, reducing reagent competition, reducing preferential amplification, and allowing for a similar LOD between the two targets. Additionally, with the new SARS-CoV-2 variants circulating within the population, primer and probe target design must be within conservative regions where viral diversity will have the least potential impact (38, 39). The potential impact of missing positive SARS-CoV-2 samples amongst the rise of SARS-CoV-2 variants is reduced by designing assays with multiple specific targets. As part of the DiaSorin Molecular's Simplexa assay algorithm, if one of the specific SARS-CoV-2 targets is detected, the sample is reported as positive. For Roche, both ORF1a and PAN-SARS are required to be detected to obtain positive results. If PAN-SARS target only is detected, the sample is reported as presumptive positive. We observed similar Ct values between the two targets in both assays and confirmed how they complement each other as part of the assay designed (Figure 1C-1D) .

This study has several limitations. Oropharyngeal swabs (OPS) from asymptomatic individuals were collected and may raise the question if OPS provide accurate detection of SARS-CoV-2 in asymptomatic carriers. To date, there is a lack of published data on the accuracy of alternative sample types such as OPS (17) . It should be noted, CDC guidelines recommend collecting and testing an upper respiratory specimen, with a nasopharyngeal swab (NPS) being the preferred choice for swab-based SARS-CoV-2 testing. When the collection of an NPS is not possible, an oropharyngeal specimen is acceptable. However, these guidelines are for symptomatic patients and do not comment on what sample type is preferred for asymptomatic testing. Previously, we have shown that there is concordance between an OP and NPS specimen for both Roche and DiaSorin Molecular (24) .

Additionally, a CDC group showed that NPS and OP swabs did not show meaningful differences in SARS-CoV-2 RNA detection in patients with < 7 days post symptom onset supporting OP swabs use when supply chain issues persist (40) . A second limitation is that this study included samples from patients from a single geographic region, Central Ohio. However, the patient samples spanned the entire range of clinical positives and reflected our overall true positivity rate between 4-7% during July 15-August 30, 2020 of the COVID-19 outbreak (41). Thirdly, the present assay comparison and characterization were made using assays authorized only for symptomatic patients. Although this study has a small sample size, it is clear that patients reported as asymptomatic can have low viral loads that can be detected by assays proven to have high clinical sensitivities. The testing of asymptomatic individuals can be performed after the evaluation of assays currently approved only for symptomatic individuals. Lastly, it is hard to assess if the current asymptomatic samples used in the study are genuinely asymptomatic or merely pre-symptomatic. With the vast range of symptoms across COVID-19 patients, mild symptoms can be missed as some patients may confuse a sore throat or a runny nose for allergies.

In summary, we have evaluated two molecular in vitro diagnostic assays for the qualitative detection of SARS-CoV-2 using OP swabs. Our evaluation data suggest that the DiaSorin Molecular and Roche Cobas assays have comparable results and that both can reliably detect SARS-CoV-2 in an asymptomatic population. As we continue in this evolving pandemic and the available vaccines reach the population, testing efforts will continue shifting in the United States. The focus on testing of symptomatic individuals will decrease and move towards asymptomatic testing as a crucial part of the process of monitoring for infection. The lab testing choices will become increasingly complex as many platforms with widely disparate performance characteristics may not be appropriate for all testing populations. The development of guidance and recommendations for asymptomatic screening by datadriven clinical studies is warranted for public health or infection control initiatives to successfully curb viral spread in future epidemics.

The authors confirm contribution to the paper as follows: study conception and design: SA, KWC; data collection: PO, LF, LL, TP, JS, AD; analysis and interpretation of results: SA, KWC, NW; draft manuscript preparation: SA. All authors reviewed the results and approved the final version of the manuscript. 

COVID-19) outbreak. World Health Organization

Coronavirus COVID-19 Global Cases by the Center for Systems Science and Engineering

Presumed Asymptomatic Carrier Transmission of COVID-19

A familial cluster of pneumonia associated with the 2019 novel coronavirus indicating person-to-person transmission: a study of a family cluster

Transmission of 2019-nCoV infection from an asymptomatic contact in Germany

The relative transmissibility of asymptomatic COVID-19 infections among close contacts

Estimating the asymptomatic proportion of coronavirus disease 2019 (COVID-19) cases on board the Diamond Princess cruise ship

Asymptomatic Middle East Respiratory Syndrome Coronavirus (MERS-CoV) infection: extent and implications for infection control: a systematic review

Comparison of asymptomatic and symptomatic rhinovirus infections in university students: incidence, species diversity, and viral load

Heterogeneous and dynamic prevalence of asymptomatic influenza virus infections

Asymptomatic SARS coronavirus infection among healthcare workers

SARS-CoV-2 and COVID-19: The most important research questions

Prevalence of Asymptomatic SARS-CoV-2 Infection: A Narrative Review

Chinese Pediatric Novel Coronavirus Study Team. SARS-CoV-2 Infection in Children

Asymptomatic coronavirus infection: MERS-CoV and SARS-CoV-2 (COVID-19)

Review of Burden, Clinical Definitions, and Management of COVID-19 Cases

SARS-CoV-2 Viral Load in Upper Respiratory Specimens of Infected Patients

Public Health -Seattle & King County; CDC COVID-19 Investigation Team. Asymptomatic and Presymptomatic SARS-CoV-2 Infections in Residents of a Long-Term Care Skilled Nursing Facility

Clinical Course and Molecular Viral Shedding Among Asymptomatic and Symptomatic Patients With SARS-CoV-2 Infection in a Community Treatment Center in the Republic of Korea

SARS-CoV-2 detection, viral load and infectivity over the course of an infection

Overview of Testing for SARS-CoV-2 (COVID-19)

Testing for SARS-CoV-2 (COVID-19): a systematic review and clinical guide to molecular and serological in-vitro diagnostic assays

FDA

Clinical Evaluation and Utilization of Multiple Molecular In Vitro Diagnostic Assays for the Detection of SARS-CoV-2

The measurement of observer agreement for categorical data

Interrater reliability: the kappa statistic

Household Transmission Study Team, Persistent SARS-CoV-2 RNA Shedding without Evidence of Infectiousness: A Cohort Study of Individuals with COVID-19

Clinical progression and viral load in a community outbreak of coronavirus-associated SARS pneumonia: a prospective study

Detection of SARS coronavirus in patients with suspected SARS. Emerg Infect Dis

Viral Load Kinetics of MERS Coronavirus Infection

CAP-China Network. Factors Associated With Prolonged Viral Shedding in Patients With Avian Influenza A(H7N9) Virus Infection

et alUpper respiratory viral load in asymptomatic individuals and mildly symptomatic patients with SARS-CoV

Clinical Course and Molecular Viral Shedding Among Asymptomatic and Symptomatic Patients With SARS-CoV-2 Infection in a Community Treatment Center in the Republic of Korea

Viral load of SARS-CoV-2 in clinical samples

SARS-CoV-2 viral load in upper respiratory specimens of infected patients

College of American Pathologists (CAP) Microbiology Committee Perspective: Caution Must Be Used in Interpreting the Cycle Threshold (Ct) Value

Genetic Variants of SARS-CoV-2-What Do They Mean?

Performance of oropharyngeal swab testing compared to nasopharyngeal swab testing for diagnosis of COVID-19 -United States