Background The FilmArray Respiratory Panel (RP) detects multiple pathogens, including Bordetella pertussis. The multiplex PCR system is appropriate for a core laboratory or point of care due to ease of use. The purpose of this study is to compare the analytical sensitivity of the FilmArray RP, which targets the promoter region of the B. pertussis toxin gene, with the Focus real-time PCR assay, which targets the insertion sequence IS481.
Methods Seventy-one specimens from patients aged 1 month to 18 years, which had tested positive for B. pertussis using the Focus assay, were analysed using the FilmArray RP.
Results Forty-six specimens were positive for B. pertussis by both the Focus and the FilmArray RP assays. Twenty-five specimens were negative for B. pertussis using the FilmArray RP assay, but positive using the Focus assay.
Conclusions The FilmArray RP assays will detect approximately 1/3 less cases of B. pertussis than the Focus assay.
- BORDETELLA PERTUSSIS
- MOLECULAR PATHOLOGY
This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Statistics from Altmetric.com
The target with the greatest analytical sensitivity to detect Bordetella pertussis is the insertion sequence, IS481.1–3 IS481 is internally repeated between 50 and >200 times in each B. pertussis genome, resulting in a PCR assay with high analytical sensitivity. However, studies have identified the presence of IS481 in Bordetella holmesii, an organism that also has been associated with cough-like illness.1 ,4–6 When compared with B. pertussis, IS481 is internally repeated 8–27 times within the B. holmesii genome.5 ,7 The conundrum related to PCR testing for B. pertussis revolves around whether the assay implemented in a clinical laboratory should be highly sensitive (IS481 target) or highly specific (pertussis toxin gene target).
The FilmArray Respiratory Panel (RP) (BioFire Diagnostics, Salt Lake City, Utah, USA) uses PCR to detect 20 respiratory pathogens, including B. pertussis.8 The closed system makes testing easy, resulting in the system being amenable to less complex laboratories or point of care.9 The genetic target for identification of B. pertussis is the promoter region of the pertussis toxin gene,8 which is present in only one copy per B. pertussis genome. While this target is specific, with lack of cross-reactivity with B. holmesii, it should have lower analytical sensitivity compared with detection of the IS481 locus. Previous analyses of the FilmArray RP reported high agreement between the results of FilmArray RP and other PCR assays for B. pertussis, suggesting that the analytical sensitivity of the FilmArray RP for B. pertussis may be equivalent to assays that target the insertion sequence IS481.9 ,10
The purpose of this study is to use clinical samples to compare the sensitivity of the FilmArray RP test, a closed system with on-board extraction coupled with the toxin gene target, to the Focus PCR assay (Focus Diagnostics, Cypress, California, USA), which amplifies insertion sequence IS481, following offline extraction.
Reserve samples, held in −80°C in universal transport medium in the microbiology laboratory at Children's Healthcare of Atlanta (Atlanta, Georgia, USA), were used for testing. All samples were initially collected using flocked swabs (Copan Diagnostics, Murrieta, California, USA) between September 2012 and June 2013. Flocked swabs are approved for collection of both viruses and bacteria, and have been verified for collection and performance of PCR by our laboratory. The samples were originally tested between September 2012 and June 2013 using the Focus assay that targets the insertion sequence IS481. This is a laboratory-developed test using analyte-specific reagents, with a limit of detection of five cells at a CT of 40. Each sample was positive for B. pertussis. Residual specimen in universal transport medium, which was not used for the initial testing, was frozen within 48–72 h at −80°C.
For the comparison between the FilmArray RP assay and the Focus assay, samples were thawed and retested using the Focus IS481-based assay and FilmArray RP assay. Testing was performed the same day the specimen was thawed for the repeat Focus assay and within three days of thawing for the FilmArray RP assay.
Specimens were tested using the FilmArray RP per the manufacturer's instructions. At the time of testing, the FilmArray RP, including the test for B. pertussis, had been cleared by the US Food and Drug Administration. The lower limit of detection for B. pertussis was stated to be 4000 copies/reaction by the manufacturer. Briefly, for each sample, 1.0 mL of hydration solution (molecular reagent grade water) was added to the pouch to rehydrate the reagents. In total, 300 μL of the reserve sample was added to 500 μL of sample buffer mix and thoroughly combined. Also, 300 μL of sample/sample buffer mix was added to the pouch, which was then loaded onto the instrument. Each run contained internal process controls for extraction, dilution and PCR.
Focus PCR-based IS481 assay
DNA from 200 µL of the reserve sample was extracted using the NucliSENS easyMAG system according to the recommendations of the manufacturer (bioMérieux Diagnostics, Marcy l'Etoile, France). Also, 5 µL of extracted nucleic acid was combined with the Focus Diagnostics B. pertussis and Bordetella parapertussis analyte-specific reagent reaction matrix. Forty cycles of PCR were accomplished using a 3M Integrated Cycler (3M Health Care, St. Paul, Minnesota, USA).
Seventy-four samples originally positive for B. pertussis between September 2012 and June 2013 were retested using the Focus assay. Seventy-one were positive on repeat testing. The specimens that did not repeat as positive were associated with original CT values between 37.3 and 38. These 71 samples were used for the analysis with the FilmArray RP.
Forty-six samples were positive for B. pertussis by both the Focus assay and the FilmArray RP assay. Twenty-five samples were negative for B. pertussis using the FilmArray RP assay, but positive using the Focus assay. The range of cycle thresholds at which all samples were positive using both the FilmArray RP assay and the Focus assay was 12.4–26.2. At CT values between 26.9 and 32.9, there was variability in the correlation with 14 samples positive by both FilmArray RP and Focus assays, and 16 negative using the FilmArray RP assay and positive using the Focus assay. At CT values from 34.8 to 39.9, the FilmArray RP assay was negative and the Focus assay was positive.
The 25 discrepant samples were submitted to the Bordetella laboratory at the Centers for Disease Control and Prevention (CDC, Atlanta, Georgia USA) and tested for B. pertussis and B. holmesii according to their published assay.2 Two samples were negative for B. pertussis. Twenty-three samples were positive for the IS481 gene target. Of these, 10 samples were also positive for the toxin gene and deemed positive for B. pertussis. Thirteen were negative for the toxin gene and deemed intermediate for B. pertussis. B. holmesii was not detected in any of the 25 samples.
The rationale for this study is straightforward. We wanted to evaluate the analytical sensitivity of the FilmArray RP compared with a real-time PCR assay for the detection of B. pertussis using a large number of clinical samples. We hypothesised that the FilmArray RP, which targets the pertussis toxin gene, would have less analytical sensitivity than the Focus assay, which targets the insertion sequence IS481. Two studies did not find a loss of analytical sensitivity for B. pertussis when using the FilmArray RP. 9 ,10 However, the studies were designed to evaluate the entire panel, and so B. pertussis was not a primary focus of the studies. One study evaluated nine clinical samples, comparing the FilmArray RP with a laboratory-developed test targeting the insertion sequence IS481, and there was 100% correlation.10 From evaluation of the CT values, all CT values were 30 or lower, suggesting that the burden of organisms was high in all the samples.
We did identify a loss of analytical sensitivity for B. pertussis using the FilmArray RP, which detected B. pertussis in 65% (46/71) of samples that tested positive using the Focus assay. FilmArray-negative samples occurred only at Focus CT values ≥26.9. The reason for the difference in analytical sensitivity could be
detection of B. holmesii by the Focus assay instead of B. pertussis (ie, false-positive tests);
differences in the extraction methods;
differences in the gene target.
We did not identify B. holmesii using the reference method performed at the CDC, nor would it be logical that there would be such a high prevalence of B. holmesii in our population. While we did not conduct a detailed evaluation of the extraction methodologies, the FilmArray RP uses a greater sample volume than the easyMag, which was the nucleic acid purification method used prior to amplification with the Focus assay. It is likely, therefore, that the difference lies in the PCR target, as IS481 is known to be present in multiple copies, resulting in increased analytical sensitivity.2 ,11 Our data differ from those from Qin et al,12 who demonstrated little difference between their limit of detection for the toxin promoter region and IS481. The performance specifications of their assay were reported to detect a similar limit of detection for each gene, indicating that their assay was not optimised for maximal sensitivity for the IS481 target. We acknowledge the need for multitarget analysis to sort out the optimal diagnostic regimen for diagnosis of pertussis, but focus this study on two commercially available systems (table 1).
FilmArray RP is less sensitive than the Focus assay targeting the insertion sequence IS481 for B. pertussis detection.
Caution should be executed when examining kits for diagnosis of B. pertussis as they may differ.
The authors would like to acknowledge Brunilís Burgos-Rivera, PhD, Katherine E. Bowden, PhD, and Maria L. Tondella, PhD, Centers for Disease Control, for providing sample analysis for Bordetella holmesii. The authors also acknowledge Megan Hamling for assistance with manuscript preparation.
Handling editor Runjan Chetty
Contributors RCJ, SRW and BBR: planning, conduct and reporting. HJM and DRI: conduct and reporting. LFW: reporting. Brunilís Burgos-Rivera, Katherine E. Bowden and Maria L. Tondella—conduct (performed confirmatory testing at the CDC reference laboratory).
Funding BioFire Diagnostics supplied the reagents to perform testing on the FilmArray instrument.
Competing interests RCJ and BBR served on advisory boards for BioFire and also perform research and development with the FilmArray instrument.
Ethics approval Institutional Review Board.
Provenance and peer review Not commissioned; externally peer reviewed.