Comparison of three molecular assays for detection of enteric viruses in stool samples

Accepted Manuscript Title: Comparison of three molecular assays for detection of enteric viruses in stool samples Authors: Haciba Moudjahed, Claire Pinc¸on, Kazali Alidjinou, Anny Dewilde, Anne Goffard PII: DOI: Reference:

S0166-0934(17)30278-1 https://doi.org/10.1016/j.jviromet.2017.09.026 VIRMET 13352

To appear in:

Journal of Virological Methods

Received date: Revised date: Accepted date:

4-5-2017 15-9-2017 27-9-2017

Please cite this article as: Moudjahed, Haciba, Pinc¸on, Claire, Alidjinou, Kazali, Dewilde, Anny, Goffard, Anne, Comparison of three molecular assays for detection of enteric viruses in stool samples.Journal of Virological Methods https://doi.org/10.1016/j.jviromet.2017.09.026 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Comparison of three molecular assays for detection of enteric viruses in stool samples. Haciba Moudjahed 1, Claire Pinçon 2, Kazali Alidjinou 3, Anny Dewilde 3, Anne Goffard4* 1

CH Valenciennes, Central Laboratory, F-59322 Valenciennes, France.

2

Univ. Lille, CHU Lille, EA 2694 - Santé publique : épidémiologie et qualité des soins, F59000 Lille, France 3

Univ. Lille, CHU Lille, Laboratoire de Virologie EA3610, F-59000 Lille, France

4

Univ. Lille, CNRS, INSERM CHU Lille, Institut Pasteur de Lille, U1019 - UMR 8204 - CIIL - Centre d'Infection et d'Immunité de Lille, F-59000 Lille, France Corresponding author: *Anne Goffard: Molecular & Cellular Virology, Univ. Lille, CNRS, Inserm, CHU Lille, Institut Pasteur de Lille, U1019 - UMR 8204 - CIIL - Centre d'Infection et d'Immunité de Lille, F-59000 Lille, France. Tel: +33-320871162; Fax: +33-320871201, E-mail address: [email protected]

Abstract: Three molecular assays (FTD® Viral GE from Fast-track diagnostics, RIDA®GENE VSP1 from R-Biopharm, and Xpert Norovirus from Cepheid) were compared for virus detection in acute diarrhea samples. RIDA®GENE and FTD® Viral GE showed perfect/almost perfect agreement for Rotavirus, Sapovirus and Norovirus, substantial agreement for Adenovirus, and moderate agreement for Astrovirus. Keywords: Norovirus; Adenovirus; Astrovirus; Sapovirus; Rotavirus; molecular assay

In hospital, viruses frequently cause acute gastroenteritis outbreaks. The associated mortality and morbidity rates are especially high in pediatric, elderly and immunocompromised patients. The cost of controlling acute outbreaks of viral gastroenteritis is related to the duration of the outbreak, the number of infected cases, and the type of measures taken to control the infection (e.g. the closure of care units) (Danial et al., 2011; Navas et al., 2015; Sadique et al., 2016). The viruses responsible for these acute gastroenteritis outbreaks are typically members of Caliciviridae (such as Norovirus and Sapovirus), Rotaviridae, Adenoviridae and Astroviridae families. The Norovirus genus is the most common cause of acute viral gastroenteritis worldwide, and affects all age groups. The genus comprises six genogroups (Vinjé, 2015). Although both genogroups I and II are frequently detected in human Norovirus infections, genogroup II is particularly implicated in hospital and food-borne outbreaks (Iturriza-Gómara and Lopman, 2014; Navarro et al., 2005; Sheahan et al., 2015; Sommer et al., 2009). In France (where the vaccination against Rotavirus is not recommended), Rotavirus is the leading cause of acute gastroenteritis in children (Parez et al., 2007). Of the 57 identified types of Adenovirus, types 40 and 41 are mainly associated with diarrhea (primarily acute in children under the age of 4) whereas the serotypes 31, 12, 18, 1, 2, 5 and 6 are less often associated with gastroenteritis (Oude Munnink and van der Hoek, 2016). Sapovirus causes acute diarrhea (especially in infants and young children) and is frequently associated with food-borne outbreaks (Kobayashi et al., 2012; Usuku et al., 2008). Astrovirus predominantly affects

1

children under the age of 2 years, elderly people, and immunocompromised individuals, and may be responsible for systemic infections (De Benedictis et al., 2011; Quan et al., 2010). Several laboratory methods are now available for the detection of gastroenteritis virus in stool samples: enzyme immunoassays, immunochromatography (ICT) assays, real-time RT-PCR assays, and electron microscopy (although the latter is not used in clinical laboratories). Hospital microbiology laboratories require sensitive, easy-to-use, efficient diagnostic assays if they are to provide physicians with accurate, timely information and thus help to control the spread of gastrointestinal viral infections. With a view to implementing a molecular assay for the diagnosis of viral gastroenteritis, we compare the performance of molecular assays for the detection of Rotavirus, Adenovirus, Astrovirus and Sapovirus using the FTD® Viral GE kit from Fast-Track diagnostics Ltd, (Sliema, Malta) and RIDA®GENE Viral Stool Panel 1 (VSP1) from r-Biopharm AG, (Darmstadt, Germany). For the detection of Norovirus, we compare the performance of the RIDA®GENE (VSP1) kit with the FTD® Viral gastroenteritis (GE) kit and with the Xpert Norovirus kit from Cepheid Inc., (Sunnyvale, CA, USA). Samples were collected from patients admitted to the Lille University Medical Center (Lille, France) for suspected infectious diarrhea (defined as three or more loose stools within a 24h period) between August, 2014 and February, 2015. The samples were tested using an ICT assays, and then stored at -20°C for subsequent evaluation. All biological samples and the associated data were obtained using standard procedures following a physician’s request; there was no additional or modified sampling. Given that the data were de-identified, French legislation did not require the provision of informed consent. A total of 45 samples met the following criteria and were included in the study: a positive test (using the laboratory’s standard ICT method) for a gastroenteritis-causing virus, and a negative test for any diarrhea-causing bacterial, fungal or parasitic pathogen. ImmunoCard® STAT Norovirus (Meridian, Paris, France), Diarlex® MB Rotavirus and Diarlex® MB Adenovirus (Orion Diagnostica, Levallois Perret, France) were used for ICT detections of virus in stool samples. Stool samples were variously collected from the adult hematology department (n=13), the geriatric medicine unit (n=9), the pediatric hematology department (n=4), the pediatric medicine unit (n=10) and other clinical units (n=9) (Figure 1). Frozen stool samples were thawed, brought to room temperature and then homogenized. Prior to extraction, a 200 mg aliquot was suspended in 1 ml of nuclease-free water or 1 ml of stool transport and recovery buffer (Roche Diagnostics, Meylan, France) for the RIDA®GENE VSP1 and the FTD® Viral GE kit, respectively. For the FTD® Viral GE kit, the suspension was immediately clarified by centrifugation at 11,000 g for 5 min. For the RIDA®GENE VSP1, the suspension was incubated at room temperature for 3 min and then clarified by centrifugation at 11,000 g for 1 min. For the Cepheid Xpert Norovirus kit only, a portion of the stool sample was collected with a swab, placed into Elution reagent viral and vortexed. The sample solution was dispensed into the kit’s dedicated cartridge and processed according to the manufacturer’s instructions. After preparation of a supernatant (as described above), 210 µl were used for RNA extraction using MagDEA® Viral DNA/RNA 200 (GC) reagent (Precision System Science, Tokyo, Japan) in an automated Magtration® Systems 12GC extractor (Precision System Science, Tokyo, Japan). Nucleic acids were eluted into 50 or 100 µl of elution buffer, depending on the PCR amplification kit used.

2

RT-PCRs were performed according to the manufacturer’s instructions: 10 µl of nucleic acid and the LightCycler 480 real-time PCR thermocycler (Roche, Meylan, France) were used for the FTD® Viral GE kit, and 5 µl of nucleic acid and the ABI 7500 real-time PCR thermocycler (Applied Biosystems, Thermo Fisher, Illkirch, France) were used for the R-Biopharm RIDA®GENE VSP1. Data are presented as the number (proportion) of positive tests. The level of diagnostic agreement between the assays was quantified by calculating Cohen’s kappa and its 95% confidence interval (CI), and applying McNemar’s test when appropriate. All statistical tests were two-tailed, and the threshold for statistical significance was set to p<0.05. All analyses were performed with SAS software (version 9.3, SAS Institute Inc., Cary, NC, USA). We performed a retrospective analysis of 45 stool samples collected from 45 patients admitted for acute diarrhea (Figure 1). The patients’ median age was 38.5 years (range: 0–101; interquartile range: 3–64). The 45 stool samples were tested for the presence of Rotavirus, Astrovirus and Sapovirus RNA and Adenovirus DNA using the R-Biopharm RIDA®GENE VSP1 and FTD® Viral GE assays. Seven of the 45 samples (15.5%) were positive for Rotavirus RNA (figure 2A) in both assays (full agreement: κ=1.00). For the molecular detection of Adenovirus DNA, 4 of the 45 (8.9%) samples were positive in both assays, and another 4 samples (8.9%) were positive in the FTD® Viral GE assay only (figure 2B). The level of agreement was substantial ( [95%CI]=0.62 [0.25; 0.95] p=0.13). For the detection of Astrovirus RNA, 2 of the 45 (4.4%) samples were positive in both assays, and an additional 4 samples (8.9%) were positive in the R-Biopharm RIDA®GENE assay only (figure 2C). The level of agreement was moderate (=0.46 [0.04; 0.89]; p=0.13). For the detection of Sapovirus RNA, only 1 of the 45 (2.2%) samples (figure 2D) was positive in both assays (full agreement: κ=1.00). Because we could not test the 45 stool samples with Cepheid assay, 39 samples were selected and also tested for Norovirus RNA with all three assays. Eleven (28.2%) were positive in all assays, one additional sample (2.5%) was positive in the FTD® Viral GE assay and 2 additional samples (5.1%) were positive in the Cepheid Xpert assay (figure 2E). The level of agreement was very strong for all comparisons: =0.94 [0.82; 1.00] for R-Biopharm RIDA®GENE vs. FTD® Viral GE (p=1.00), =0.82 [0.64; 1.00] for R-Biopharm RIDA®GENE vs. Cepheid Xpert (p=0.25), and =0.89 [0.73; 1.00] for FTD® Viral GE vs. Cepheid Xpert (p=0.50). The FTD® Viral GE and Cepheid Xpert assays both specify the Norovirus genotype; type II was detected in all cases. Coinfections were detected in 7 of the 45 (15.5%) stool samples using R-Biopharm RIDA®GENE and FTD® Viral GE. Three of the 45 samples (6.7%) evidenced Rotavirus and Adenovirus coinfections in children, 2 (4.4%) evidenced with Norovirus and Astrovirus coinfections in elderly patients, 1 (2.2%) evidenced a Rotavirus and Norovirus coinfection in an elderly patient, and 1 (2.2%) evidenced Adenovirus and Sapovirus coinfection in an adult patient. Seventeen of the 45 stool samples (37.8%) were negative for all viruses, according to the three molecular assays. Our study revealed 11 discordances between the R-Biopharm RIDA®GENE and FTD® Viral GE assays. In some cases, two viruses were detected in discordant samples. Firstly, FTD® Viral GE assay apparently detected Adenovirus but not the R-Biopharm RIDA®GENE assay in four cases. Co-infections were observed in all four of the discordant Adenovirus samples (Norovirus: n=2; Rotavirus: n=1; Sapovirus: n=1). Secondly, the R-Biopharm RIDA®GENE assay apparently detected Astrovirus but not the FTD® Viral GE assay in four cases. Coinfection with Norovirus was detected in a sample collected from a child but not in the other 3

three samples. Additionally, the Cepheid Xpert assay detected Norovirus but not the FTD® Viral GE and R-Biopharm RIDA®GENE assays in two samples, and in one sample, the Cepheid Xpert and FTD® Viral GE assays detected it but not the R-Biopharm RIDA®GENE assay. In both the samples collected from children, co-infections either with Astrovirus (n=1) or Rotavirus (n=1) were observed. For the last sample, no co-infection was detected. Overall, 11 of the 45 stool samples gave discordant results. The detection of co-infection in seven samples suggests the presence of competition between PCR targets. The discrepancies in the other four cases may have been due to poor storage conditions (e.g. storage at room temperature for too long before initial freezing, or repeated freeze-thaw cycles). The present study compares the performance of commercial multiplex molecular assays for the detection of viruses in stools. Our statistical analyses showed perfect or almost perfect agreement between the R-Biopharm RIDA®GENE and FTD® Viral GE assays for Rotavirus, Sapovirus and Norovirus detection, substantial agreement for Adenovirus detection and moderate agreement for Astrovirus detection. However, the small number of samples included in our study prevents us from concluding that all three molecular assays have equivalent levels of performance. Although ICT assay have long been used to detect viruses in the stools of patients with diarrhea, molecular methods are now more efficient to detect virus responsible for acute outbreaks of gastroenteritis (Farkas et al., 2015; Gonzalez et al., 2016; Jiang et al., 2014; Mo et al., 2015; Rovida et al., 2016). McAuliffe et al. recently suggested that multiplex PCR should be systematically applied to detect pathogens in stool samples as early as possible (McAuliffe et al., 2013). Our present results suggest that R-Biopharm RIDA®GENE and FTD® Viral GE assays could be used to investigate outbreaks of acute gastroenteritis while the Cepheid Xpert test seems appropriate to the specific research of Norovirus. Acknowledgments Cohen’s Kappa estimates and p-values of McNemar’s tests are depicted. Results obtained for A. Rotavirus. B. Adenovirus. C. Astrovirus. D. Sapovirus. E. Norovirus. NA*: non-applicable.

4

Bibliography Danial, J., Cepeda, J.A., Cameron, F., Cloy, K., Wishart, D., Templeton, K.E., 2011. Epidemiology and costs associated with norovirus outbreaks in NHS Lothian, Scotland 20072009. J. Hosp. Infect. 79, 354–358. doi:10.1016/j.jhin.2011.06.018 De Benedictis, P., Schultz-Cherry, S., Burnham, A., Cattoli, G., 2011. Astrovirus infections in humans and animals - molecular biology, genetic diversity, and interspecies transmissions. Infect. Genet. Evol. J. Mol. Epidemiol. Evol. Genet. Infect. Dis. 11, 1529–1544. doi:10.1016/j.meegid.2011.07.024 Farkas, T., Singh, A., Le Guyader, F.S., La Rosa, G., Saif, L., McNeal, M., 2015. Multiplex real-time RT-PCR for the simultaneous detection and quantification of GI, GII and GIV noroviruses. J. Virol. Methods 223, 109–114. doi:10.1016/j.jviromet.2015.07.020 Gonzalez, M.D., Langley, L.C., Buchan, B.W., Faron, M.L., Maier, M., Templeton, K., Walker, K., Popowitch, E.B., Miller, M.B., Rao, A., Liebert, U.G., Ledeboer, N.A., Vinjé, J., Burnham, C.A., 2016. Multicenter Evaluation of the Xpert Norovirus Assay for Detection of Norovirus Genogroups I and II in Fecal Specimens. J. Clin. Microbiol. 54, 142–147. doi:10.1128/JCM.02361-15 Iturriza-Gómara, M., Lopman, B., 2014. Norovirus in healthcare settings. Curr. Opin. Infect. Dis. 27, 437–443. doi:10.1097/QCO.0000000000000094 Jiang, Y., Fang, L., Shi, X., Zhang, H., Li, Y., Lin, Y., Qiu, Y., Chen, Q., Li, H., Zhou, L., Hu, Q., 2014. Simultaneous detection of five enteric viruses associated with gastroenteritis by use of a PCR assay: a single real-time multiplex reaction and its clinical application. J. Clin. Microbiol. 52, 1266–1268. doi:10.1128/JCM.00245-14 Kobayashi, S., Fujiwara, N., Yasui, Y., Yamashita, T., Hiramatsu, R., Minagawa, H., 2012. A foodborne outbreak of sapovirus linked to catered box lunches in Japan. Arch. Virol. 157, 1995–1997. doi:10.1007/s00705-012-1394-8 McAuliffe, G.N., Anderson, T.P., Stevens, M., Adams, J., Coleman, R., Mahagamasekera, P., Young, S., Henderson, T., Hofmann, M., Jennings, L.C., Murdoch, D.R., 2013. Systematic application of multiplex PCR enhances the detection of bacteria, parasites, and viruses in stool samples. J. Infect. 67, 122–129. doi:10.1016/j.jinf.2013.04.009 Mo, Q.-H., Wang, H.-B., Tan, H., Wu, B.-M., Feng, Z.-L., Wang, Q., Lin, J.-C., Yang, Z., 2015. Comparative detection of rotavirus RNA by conventional RT-PCR, TaqMan RT-PCR and real-time nucleic acid sequence-based amplification. J. Virol. Methods 213, 1–4. doi:10.1016/j.jviromet.2014.11.006 Navarro, G., Sala, R.M., Segura, F., Arias, C., Anton, E., Varela, P., Peña, P., Llovet, T., Sanfeliu, I., Canals, M., Serrate, G., Nogueras, A., 2005. An outbreak of norovirus infection in a long-term-care unit in Spain. Infect. Control Hosp. Epidemiol. 26, 259–262. doi:10.1086/502536 Navas, E., Torner, N., Broner, S., Godoy, P., Martínez, A., Bartolomé, R., Domínguez, A., Working Group for the Study of Outbreaks of Acute Gastroenteritis in Catalonia, 2015. Economic costs of outbreaks of acute viral gastroenteritis due to norovirus in Catalonia (Spain), 2010-2011. BMC Public Health 15, 999. doi:10.1186/s12889-015-2289-x Oude Munnink, B.B., van der Hoek, L., 2016. Viruses Causing Gastroenteritis: The Known, The New and Those Beyond. Viruses 8. doi:10.3390/v8020042

5

Parez, N., Allaert, F.-A., Derrough, T., Caulin, E., et le groupe d’investigateurs, 2007. [Frequency and clinical characteristics of rotavirus acute gastroenteritis in children under 5 years of age treated in community practice in France]. Pathol. Biol. (Paris) 55, 453–459. doi:10.1016/j.patbio.2007.07.007 Quan, P.L., Wagner, T.A., Briese, T., Torgerson, T.R., Hornig, M., Tashmukhamedova, A., Firth, C., Palacios, G., Baisre-De-Leon, A., Paddock, C.D., Hutchison, S.K., Egholm, M., Zaki, S.R., Goldman, J.E., Ochs, H.D., Lipkin, W.I., 2010. Astrovirus encephalitis in boy with X-linked agammaglobulinemia. Emerg. Infect. Dis. 16, 918–925. doi:10.3201/eid1606.091536 Rovida, F., Premoli, M., Campanini, G., Sarasini, A., Baldanti, F., 2016. Evaluation of Xpert® Norovirus Assay performance in comparison with real-time RT-PCR in hospitalized adult patients with acute gastroenteritis. Diagn. Microbiol. Infect. Dis. 85, 426–427. doi:10.1016/j.diagmicrobio.2016.05.002 Sadique, Z., Lopman, B., Cooper, B.S., Edmunds, W.J., 2016. Cost-effectiveness of Ward Closure to Control Outbreaks of Norovirus Infection in United Kingdom National Health Service Hospitals. J. Infect. Dis. 213 Suppl 1, S19-26. doi:10.1093/infdis/jiv410 Sheahan, A., Copeland, G., Richardson, L., McKay, S., Chou, A., Babady, N.E., Tang, Y.-W., Boulad, F., Eagan, J., Sepkowitz, K., Kamboj, M., 2015. Control of norovirus outbreak on a pediatric oncology unit. Am. J. Infect. Control 43, 1066–1069. doi:10.1016/j.ajic.2015.05.032 Sommer, C., Mueller, W., Resch, B., 2009. Two nosocomial norovirus outbreaks in the neonatal intensive and intermediate care unit. Eur. J. Clin. Microbiol. Infect. Dis. Off. Publ. Eur. Soc. Clin. Microbiol. 28, 1133–1136. doi:10.1007/s10096-009-0735-3 Usuku, S., Kumazaki, M., Kitamura, K., Tochikubo, O., Noguchi, Y., 2008. An outbreak of food-borne gastroenteritis due to sapovirus among junior high school students. Jpn. J. Infect. Dis. 61, 438–441. Vinjé, J., 2015. Advances in Laboratory Methods for Detection and Typing of Norovirus. J. Clin. Microbiol. 53, 373–381. doi:10.1128/JCM.01535-14 Figure Caption

6

The authors thank Dr. David Fraser for editorial assistance. Figures

Figure 1. The workflow diagram for stool samples.

A. Rotavirus

FTD Viral GE (Fast-track diagnosis)

RIDA®GENE VSP1 (r-Biopharm) Negative Positive Negative 37 / Positive

/ =1.00, p=NA*

7

B. Adenovirus

RIDA®GENE VSP1

FTD Viral GE

Negative 37

(r-Biopharm)

Negative

Positive /

(Fast-track diagnosis)

Positive

4 4 =0.62 [0.29 ; 0.95] p=0.13

Astrovirus

RIDA®GENE VSP1

FTD Viral GE

Negative 39

(r-Biopharm)

Negative

Positive 4 7

C.

(Fast-track diagnosis)

Positive

/ 2 =0.46 [0.04 ; 0.89] p=0.13

D. Sapovirus

RIDA®GENE VSP1 (r-Biopharm)

FTD Viral GE

Negative

Negative 43

Positive /

(Fast-track diagnosis)

Positive

/ 1 =1.00 p=NA*

8

E. Norovirus

FTD Viral GE

Negative

RIDA®GENE VSP1

FTD Viral GE

(r-Biopharm)

(Fast-track diagnosis)

Negative 27

Positive /

Negative

Positive

Negative 25

Positive /

(Fast-track diagnosis)

Positive

Cepheid Xpert

Negative Positive

1 11 =0.94 [0.82 ; 1.00] p=1.00 Negative Positive 25 / 3 11 =0.82 [0.64 ; 1.00] p=0.25

2 12 =0.89 [0.73 ; 1.00] p=0.50

Figure 2. Contingency tables of the agreement between tested methods for each virus.

9

Table 1

RIDA®GENE VSP1 (rBiopharm) n NOROVIRUS Positive Negative Total ROTAVIRUS Positive Negative Total ADENOVIRUS Positive Negative Total ASTROVIRUS Positive Negative Total SAPOVIRUS Positive Negative Total

%

FTD Viral GE (Fast-track diagnosis) n

%

Xpert Norovirus (Cepheid) n 14 25 39

Concordant results with all methods tested % 35.9 64.1

11

11 28 39

28.2 71.8

12 27 39

30.8 69.2

7 38 45

15.5 84.4

7 38

15.5 84.4

7

4 41 45

8.9 91.1

17.4 82.6

4

6 39 45

13.3 86.6

1 44 45

2.1 97.8

45

8 37 45

2 43

% 4.4 95.5

2

45

1 44

2.1 97.8

1

45

10