Prospective and Retrospective Evaluation of the Performance of the FDA-Approved Cepheid Xpert Flu/RSV XC Assay

Abstract

Rapid and accurate detection of respiratory viruses is important in patient care and in guiding therapy and infection prevention policy. Rapid viral antigen assays are simple to perform and provide results within 15 to 30 minutes but are limited by their modest-to-moderate sensitivity. Molecular assays are more sensitive and specific but require more technical time and expertise and are more expensive. We verified the performance of the Xpert Flu/RSV XC assay prospectively, using patient respiratory samples from the 2014–2015 respiratory season, and, retrospectively, with frozen patient samples from the previous respiratory season. A total of 60 specimens were assayed on the Xpert Flu/RSV XC assay and by the GenMark Diagnostics eSensor Respiratory Viral Panel. The sensitivity of the Xpert Flu/RSV XC for Flu A was 100% (23/23), for Flu B, 80% (8/10), and for respiratory syncytial virus (RSV), 94.1% (16/17), compared to the reference assay (GenMark). The specificity was 100%. Eight specimens were positive for viruses other than Flu A/B or RSV, and this did not interfere with detection of targets in the Xpert assay. We demonstrated that the performance of the Xpert Flu/RSV XC was comparable to the more comprehensive molecular respiratory assay.

Introduction

The clinical diagnosis of influenza can be challenging, as many of the symptoms are nonspecific to influenza and overlap with other respiratory pathogens. Rapid identification of influenza as the etiology of the respiratory tract infection is important, as it guides therapy and permits the rapid institution of appropriate infection control precautions.¹ Furthermore, the antiviral therapy available for the treatment of influenza has to be given within the first 48 hours after onset of influenza symptoms to be effective.² Influenza B is inherently resistant to the M2 ion channel inhibitors, such as amantadine, and, therefore, there is a need to know the serological type of influenza causing the infection.³ Respiratory syncytial virus causes acute respiratory tract illness in persons of all ages. Premature infants, children younger than 2 years old with congenital heart or chronic lung disease, children or adults with compromised immune systems, and people 65 years and older are at risk of severe disease and often require hospitalization for the management of the infection.⁴ Almost all children have been infected with RSV by the age of 2 years, and reinfection is common throughout life.⁵ Many caregivers use rapid antigen testing systems for the diagnosis of influenza and respiratory syncytial virus infection because of their simplicity of use and their rapid turnaround time. It is well known, however, that their sensitivity and specificity are limited and fluctuate depending on the prevalence of the virus in the tested population and on the strains of the virus circulating, like the novel H1N1 influenza A 2009 strain.^6,7 The performance of the assays also depends on the operator performing the testing; rapid antigen testing systems have a tendency to perform worse in the hands of non-laboratorians.⁸ These limitations have driven commercial vendors to develop a second generation of rapid antigen assays to try to improve their performance and to make them as sensitive as nucleic acid‒based testing assays. Evaluations of these newer rapid antigen assays showed that they are still lacking in high sensitivity and specificity, leaving nucleic acid‒based testing as the only alternate choice of testing.⁹ Molecular assays are more sensitive and specific but require technical expertise and time to generate results. They can be expensive, as most commercial respiratory molecular assays offer comprehensive expanded panels capable of detecting 15 to 20 respiratory pathogen targets.

In November 2014, Cepheid received FDA clearance for its new Xpert Flu/RSV XC, a real-time multiplex PCR-based assay capable of simultaneously detecting influenza A viruses (Flu A), inclusive of both human and avian strains; influenza B viruses (Flu B); and RSV directly from respiratory specimens in little over an hour. The previous generation assay, the Xpert Flu from Cepheid, detected only Flu A and Flu B.

We verified the performance of the Xpert Flu/RSV XC assay prospectively using patient respiratory samples from the 2014–2015 respiratory seasons and retrospectively with frozen patient samples from previous respiratory seasons.

Methods

To assess the accuracy of the Xpert Flu/RSV XC to detect and identify correctly influenza A, influenza B, and RSV in upper respiratory samples, 60 specimens (19 specimens from the 2014–2015 respiratory season and 41 specimens frozen at -80 °C from previous respiratory seasons) were tested on the Xpert Flu/RSV XC assay, and also assayed by the GenMark Diagnostics eSensor Respiratory Viral Panel (RVP), for which the performance characteristics had been validated in our laboratory.

Results

The Xpert Flu/RSV XC results were as follows (Table 1): 23 specimens positive for Flu A, 8 positive for Flu B, 16 positive for RSV, and 13 negative for all three viruses. The sensitivity of the Xpert Flu/RSV XC was 100% (23/23) for Flu A, 80% (8/10) for Flu B, and 94.1% (16/17) for RSV compared to the reference assay (GenMark).

The specificity of the Xpert Flu/RSV XC assay was 100%; it correctly identified as negative the 11 specimens that were negative by the GenMark assay. Two specimens were positive for Flu B by the eSensor Respiratory Viral Panel but negative for Flu B by the Xpert Flu/RSV XC assay (specimens 24 and 30 in Table 1). Specimen 24 was a “low-positive” with a signal of 12.2 nanoamperes on the GenMark eSensor instrument. Specimen 30 had a high signal of 188.1 nanoamperes on the GenMark eSensor instrument, and on the Cepheid Xpert Flu/RSV XC assay, a small curve was seen at a high cycle threshold but called negative by the Xpert Flu/RSV XC assay software. One specimen was positive for Flu B by the Xpert Flu/RSV XC Assay, but positive for Flu B and RSV by GenMark eSensor.

Eight specimens were positive by the GenMark eSensor for viruses other than Flu A/B or RSV, and this did not interfere with detection of targets by the Cepheid Xpert Flu/RSV XC assay.

Conclusion

Three expanded direct multiplex respiratory panels (FilmArray Respiratory Panel, GenMark’s ePlex Respiratory Pathogen Panel, and VERIGENE Respiratory Pathogens Flex Test) are available to clinical laboratories. The disadvantage of these multiplex assays is that they are too expansive—some detecting up to 20 targets, including viruses and bacteria—for the general population that present to primary care providers or emergency departments with typical influenza symptoms. The other drawback is the high cost associated with these expanded multiplex panels and their associated testing platforms. The Centers for Medicare and Medicaid Services (CMS) list a price difference of almost $400.00 between a panel of 3 to 5 respiratory targets and a panel of 12 to 25 respiratory targets. Clinicians are increasingly conscious of the rise in healthcare costs and are ordering tests targeted to the specific infectious agent they are expecting to be the cause of the infection. The advantage of this newer Xpert platform is that it is a rapid, direct, limited molecular respiratory assay with a better performance than rapid antigen tests, and the predictive values of the assay are not affected by the prevalence of influenza in the patient population being tested.¹⁰ The assay detects simultaneously Flu A, Flu B, and RSV, in contrast to most rapid antigen testing, which needs two assays, one for Flu A/B and one for RSV. The Xpert Flu/RSV XC assay was able to detect accurately the influenza A H1N1 2009 strain, unlike rapid antigen assays that have a poor performance in detecting that strain.⁹ We also demonstrated that the performance of the Xpert Flu/RSV XC was comparable to the expanded molecular respiratory panel.

The presence of viruses other than the three targets in the new Xpert Flu/RSV XC assay did not interfere with its performance.

The capacity of the FDA cleared Xpert Flu/RSV XC to detect simultaneously Flu A, Flu B and RSV made it more practical and efficient than its predecessor (the Xpert Flu), which detected only Flu A and Flu B and had issues with sensitivity (ranging from 65.% to 95.3% depending on the study) of detection.^11-13

The Xpert Flu/RSV XC could replace the use of rapid antigen assays performed on patients from emergency departments and in hospitalized patients for whom a rapid and accurate result for Flu A, Flu B, or RSV is needed for clinical management and infection control containment. LM

Declaration of Interest

None of the authors has any financial disclosure or conflicts of interest to disclose.

Abbreviations

 
• Flu

  influenza

 
• Flu A

  influenza A

 
• Flu B

  influenza B

 
• RSV

  respiratory syncytial virus

 
• RSV A

  respiratory syncytial virus subtype A

 
• RSV B

  respiratory syncytial virus subtype B

 
• PIV 2

  parainfluenza virus type 2

 
• PIV 3

  parainfluenza virus type 3

 
• HRV

  human rhinovirus

 
• HMPV

  human metapneumovirus

 
• Adeno B/E

  adenovirus species B/E

 
• Adeno C

  adenovirus species C

Acknowledgments

We are grateful for the excellent technical expertise of our staff in the Clinical Microbiology Laboratory. Also, the contents of this article were presented in part at the 50th Annual Meeting, Academy of Clinical Laboratory Physicians, Minneapolis, Minnesota, May 2015.

References