Journal of Clinical Virology - Volume 82 - Supplement 1 - September 2016

38 / 152

S34 Abstracts / Journal of Clinical Virology 82S (2016) S1–S142
workflow simplicity, and less chance for user error as compared to
the GeneXpert system.
Abstract no: 196
Presentation at ESCV 2016: Poster 26
Performance of a molecular diagnostic,
multicode based, sample-to-answer assay for
the simultaneous detection of Influenza A, B
and Respiratory Syncytial Viruses
J. Voermans ∗, S. Deniz, M. Koopmans,
A. van der Eijk, S. Pas
Erasmus MC, The Netherlands
Introduction: Rapid diagnostics is required in cases with respi-
ratory failure for clinical decision making regarding isolation and
antiviral therapy. Techniques like immune-chromatographic test
(ICT) and direct immunofluorescence assay (DFA) have lower sensi-
tivities and specificities than molecular diagnostic assays, but have
the advantage of quick turnaround times and ease-of-use. Here, we
evaluated the performance of an automated, easy to use, sample-
to-answer system, which performs an Influenza A/B virus (fluA/B),
respiratory syncytial virus (RSV) and internal control multiplex RT-
PCR of 1–12 samples within 2 h.
Methods: The analytical performance of the FluA/B/RSV assay
on the ARIES (Luminex), a system using multicode technology
(a probe-free real-time RT-PCR method with melting curve con-
firmation), was evaluated using published laboratory developed
automated real-time RT-PCR assays (LDA) for fluA, fluB, RSV-A and
RSV-B. Genotype inclusivity of the ARIES was tested using 16 avian
(H1–H16) and 33 human fluA strains, 3 fluB strains and the two
RSV (A/B) strains. Specificity was assessed using 40 high positive
non-fluA/fluB/RSV-viruses and analytical sensitivity was compared
to LDA assays by testing 0.5 log dilution series. The clinical per-
formance was compared to both LDA + ICT (BinaxNOW influenza
A/B and RSV test) + DFA using selected (pretreated), −80 ◦ C stored,
respiratory tract samples from 2006 until 2015 (retrospective)
and prospective testing of original respiratory tract samples from
December 2015 onwards.
Results: All fluA, fluB and RSVA/B strains tested for analytical
performance evaluation were detected and no aspecific reactions
were identified. ARIES FluA/B/RSV assay was 0.5 log less sensitive
for fluA, 1 log for RSV-A, 2 logs for RSV-B and 2.5 logs for fluB com-
pared to LDA. In total, 447 samples were included in the clinical
performance evaluation, of which 15.4% tested positive for fluA,
9.2% for fluB and 26.0% for RSV, (RSV-A, 13.2% and RSV-B 12.9%) in
both LDA and ARIES. Confirmed discrepant results were found in 11
samples (1 fluA, 4 fluB and 6 RSV-A), which tested positive in LDA
and negative in ARIES (2%, LDA Ct values 28.8–36.0), resulting in
an overall clinical sensitivity and specificity of 98.6% and 100% for
fluA, 91.1% and 100% for fluB and 95.1% and 100% for RSV, respec-
tively. If compared to the DFA (n = 217) and ICT (n = 119), ARIES
detected 38 (17.5%; 4 fluA, 23 fluB, 11 RSV) and 32 (26.9%; 7 fluA, 3
fluB, 22 RSV) more samples respectively, all confirmed by LDA (Ct
range 14.9–35.0). In terms of robustness, 2.2% cassettes failed dur-
ing operation in clinical specimen, of which 90% was an undiluted
bronchio-alveolar lavage, nose wash or sputum.
Conclusion: The ARIES influenza A/B/RSV assay is a specific and
rapidmolecular assay. Although analytically the ARIES is less sensi-
tive for fluB, RSV-A and RSV-B than the LDA assays, the performance
in clinical samples is comparable to LDA and better than those of
the established rapid assays. Other respiratory samples than throat
swabs can be analyzed by the ARIES, but need to be diluted prior
analysis.
Abstract no: 200
Presentation at ESCV 2016: Poster 27
Fully automated diagnosis of MERS-CoV
infection in respiratory specimen on the
IdyllaTM MDx Platform
N. Trost ∗, A. Gilles , I. Erquiaga , H. Kenes ,
D. Nauwelaers Steimer Carman
E. Sablon
1 Fast-track Diagnostics, Esch-sur-Alzette,
Luxembourg
2 Biocartis NV, Mechelen, Belgium
Background: Rapid diagnosis of MERS-CoV infection is essen-
tial for the successful clinical management and isolation of MERS
patients. The prototype Idylla TM MERS assay is a RT-PCRbased assay
which generates highly sensitive, specific results with a minimal
turn-around time. Two independent PCR assays, targeting differ-
ent regions in the MERS-CoV genome, are combined to detect and
at the same time confirm infection with MERS-CoV. The proto-
type IdyllaTM MERS assay is a single-use cartridge that will be run
in the fully automated IdyllaTM MDx Platform. The cartridge con-
tains all reagents and is capable of processing samples without any
user manipulation, minimizing the possibility of errors in setup
and decreasing the risk of infection or contamination. The aim of
this work was to demonstrate the performance of the prototype
IdyllaTM MERS assay on the IdyllaTM Platform.
Methods: Performance of the prototype IdyllaTM MERS assay
was assessed using serial dilutions of viral culture spiked in MERS-
CoV negative clinical material. The performance of the prototype
IdyllaTM MERS assay was compared to a conventional RT-PCR kit in
combination with extraction by the NucliSENS® easyMag®. In vitro
transcribed MERS-CoV RNA was used to determine the LoD of the
assay and to show reproducibility. Cross-reactivity was analysed
using culture and clinical specimen positive of other respiratory
pathogens. Additionally, an in-silico analysis was performed to
prove the reactivity with all available MERS-CoV sequences and to
exclude any cross-reactivity with organisms present in respiratory
specimen or with the human genome.
Results: The prototype IdyllaTM MERS assay demonstrated high
sensitivity and specificity. The analysis of MERS-CoV viral culture
showed the same sensitivity with the IdyllaTM MERS assay as a
conventional MERS RT-PCR kit in combination with NucliSENS®
easyMag® extraction. No cross-reactivity with other pathogens or
the human genome was observed in-vitro or in-silico. The in-silico
reactivity analysis showed 100% identity in 98.31% of the avail-
able sequences for the MERS screening assay and 97.07% for the
confirmatory assay. The remaining sequences only showed minor
mismatches and we confirmed the binding capability of our assay
by using plasmids containing the mismatches.
Conclusions: The fully automated prototype IdyllaTM MERS
assay requires less than 2min of hands-on time for sample handling
and provides results in less than 90min without need for expe-
rienced staff or extensive training. Due to the integrated sample
preparation the handling of the infectious material is reduced to an
absolute minimum. The automated sample processing and RT-PCR
and data analysis will lead to a sensitive and accurate calling of any
MERS-CoV positive sample. The sample-to-result format of the pro-