key: cord-1018186-eygkkcro
authors: Varadhan, Hemalatha; Ahuja, Vishal; Pitman, Catherine; Dwyer, Dominic E.
title: Weak positive SARS-CoV-2 N2 gene results using the Xpress Xpert assay: the need for an alternate interpretative criteria in a low prevalence setting
date: 2021-11-23
journal: Pathology
DOI: 10.1016/j.pathol.2021.10.001
sha: 1e15fc9d71de30f49b6b7d69ea734a004f3700b6
doc_id: 1018186
cord_uid: eygkkcro

nan

The Xpert Xpress SARS-CoV-2 (Xpert) test (Cepheid, USA) is the most common Australian rapid nucleic acid test (NAT) used for detection of SARS-CoV-2 in nasal and nasopharyngeal specimens. This rapid RT-PCR assay detects two targets: the E (envelope) and N2 (nucleocapsid) SARS-CoV-2 coding regions. The assay cartridge is a closed system that includes sample processing, probe check internal controls which ensure adequate sample processing, and checks for sample-associated inhibition and reagents functionality. 1 Following the manufacturer's instructions 1 the results are interpreted automatically by the assay software into four categories: 'positive', 'presumptive positive', 'negative' and 'error'. Specimens positive for both E and N2 targets, or single N2 target, are reported as 'positive', while a single E target positive result is reported as 'presumptive positive'.

The Xpert assay is currently performed at 37 NSW Health Pathology (NSWHP) laboratories, and by 10 May 2021, 55,411 tests had been performed on nasal or nasopharyngeal swabs. A total of 317 samples tested positive for both E and N2 targets. There were 42 notified instances where only N2 target positive results were obtained and two where only the E target was detected. All of these 44 tests detected the target at a high cycle threshold (Ct) number (range 36.8-44.6, average 42.0). Of the 42 N2 target only positive tests, 16 (38.1%) were deemed as true infection as confirmed by alternate NAT, SARS-CoV-2-specific serology, and/or consistent epidemiological exposure ( Table 1 ). The remaining 26 (61.9 %) individuals with only N2 target positive results had neither clinical nor epidemiological clues suggestive of COVID-19 (Table 2) , and all had Ct values over 40 (range 40.6-44.9, average 42.5). The test was repeated on the same sample and Xpert assay for 15/26 (57.7%) specimens and all were negative. Nineteen of the 26 (73.1%) N2 target positive individuals had a repeat specimen collected and all were negative using the Xpert assay and/or alternate SARS-CoV-2 NAT. SARS-CoV-2-specific serology was performed in 8/26 (30.8%) with one patient returning a positive result (SARS-CoV-2-specific IgG at a titre of 640 using a reference SARS-CoV-2 immunofluorescence assay 2 and Xpress Xpert N2 target positive at Ct 41.3 collected in May 2021). This patient was a known COVID-19 NAT positive case from March 2020 and had received their first dose of the AstraZeneca COVID-19 vaccine a few weeks prior to the Xpert test, a possible explanation of the positive SARS-CoV-2-specific IgG.

What do single target N2 results mean? In a high prevalence setting (where the assay was developed) a single target positive result may indicate early infection, late infection, or reflect a false-positive finding. One USA study published in October 2020 where there was a high prevalence of COVID-19 reported 4% E negative N2 positive results among 1123 positive Xpert results. 3 Approximately one-third (29.5%) were symptomatic at clinical presentation, and they concluded that the late single N2 positive results suggested prolonged subclinical or resolving infections. An Italian study reported that all 12/1639 samples tested with N2 only positive results with a Ct >39 were established to be false-positives. 4 In this study, repeat testing on the same or alternative platform was performed after a centrifugation step. It should be noted that the manufacturer does not recommend repeat testing for E negative N2 positive results, regardless of Ct values, and instructs these samples to be reported as 'positive'. In our low prevalence setting in Australia, all N2 only positive samples with a Ct greater than 40 were found to be false-positives once a clinical history had been obtained excluding prior COVID-19 infection or exposure to high-risk settings (either contact with a known COVID-19 case or residing in a hotel quarantine facility). These findings were supported by further testing on the same or a freshly collected sample. The samples in our study that were deemed to be 'true-positive' findings were generally observed during the initial stages of the pandemic of 2020 when disease prevalence was higher.

Factors such as the site and quality of sampling, swab type and viral transport media, timing of collection, and sensitivity of the platforms employed need to be considered when interpreting single target high Ct positive results. An inter-laboratory quality assurance survey performed across all NSWHP laboratories in November 2020 revealed no major differences in Xpert Ct values across a range of different viral transport media at a concentration of as low as 10 3 copies per mL (NSWHP SARS-CoV-2 Internal Quality Assurance Survey; data not shown). It has been reported that prolonged virus shedding is observed in lower respiratory tract compared to upper respiratory tract specimens (median duration 34 vs 19 days post-infection) 5 and whether the former might be considered to confirm the initial positive results needs assessment. A study reported increased analytical sensitivity of the Xpert assay over the BD Max assay (BD Diagnostics, Canada) which could explain the variation in the results on the same specimen across NAT platforms. 6 One of the limitations in the study was that two of 26 recollected specimens were tested solely by BD Max. The differential performance in sensitivity across platforms could have resulted in a 'false-negative' result. While it is plausible that the E negative, N2 positive results in our setting may have represented early infection before other markers became available, the absence or progression of transmission events and lack of clusters associated with the index case affirms a 'false-positive' finding.

The evaluations of many 'Emergency Use Only' authorised NAT were mostly based on the analytical performance in terms of sensitivity and specificity: they lacked data on actual diagnostic performance based on prospective clinical evaluation due to early urgent deployment of these assays during the initial pandemic stages. This resulted in limited knowledge of predictive values prior to laboratory adoption of assays. With the low incidence of COVID-19 in Australia (less than 1% daily positivity rate as at 31 May 2021, www.health.gov.au), one should expect a lower positive predictive value (including highly specific assays) and a two-step testing algorithm may be required to resolve the 'weak' J o u r n a l P r e -p r o o f positive results. These weak positive (Ct >40), E negative, N2 only positive results pose several challenges for contact tracers in that the initial 'positive' finding triggers significant public health measures. This is especially true when these results stem from smaller laboratories that have no alternate testing platforms available for immediate confirmation. We calculated that the confirmatory steps such as repeat specimen collection and re-testing could result in an additional cost. The financial and social impact costs of triggering COVID-19 testing in a region, on the basis of a later confirmed false-positive result, cannot be underestimated. There is an urgent need for a standardised reporting pathway for such 'weak' positive single target results. A study in a low endemic setting suggests that isolated N2 positive results with a Ct value of >40 can be reported as 'not detected'. 7 We suggest a flowchart (Fig. 1 ) that may assist with initial interpretation and further confirmation of weak positive N2 results. This flowchart is contingent upon a low prevalence environment, appropriate clinical history taking, and re-testing and re-sampling based on the clinical scenario, and should be reviewed as the situation changes.

To conclude, we report 42 cases of E negative, N2 positive results on the Cepheid rapid Xpert Xpress SARS-CoV-2 assay. In this study we found that if the flowchart of Fig. 1 is used along with appropriate clinical and epidemiological history, no false-positive results would have been reported. To our knowledge, this is the first report from Australia and a large-scale study from a low prevalence setting. Given the downstream public health and clinical implications, we suggest a flowchart that laboratories in a low prevalence setting might be considered when reporting such results. Interpretation of such results requires caution and a close liaison with the medical microbiologist and public health physician to ensure an accurate clinical history is obtained. 

Xpert Xpress SARS-CoV-2: Instructions for Use

The antibody response to SARS-CoV-2 infection

A needle in the haystack? Assessing the significance of envelope (E) gene-negative, nucleocapsid (N2) genepositive SARS-CoV-2 detection by the Cepheid Xpert Xpress SARS-COV-2 assay

Detection of SARS-COV N2 gene: very low amounts of viral RNA or false positive?

SARS CoV-2 detection from upper and lower respiratory tract specimens: diagnostic and infection control implications

Comparison of the analytical sensitivity of seven commonly used commercial SARS-CoV-2 automated molecular assays

GeneXpert for the diagnosis of COVID-19 in LMICs

The authors acknowledge the following NSWHP laboratories that