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

30 / 152

S26 Abstracts / Journal of Clinical Virology 82S (2016) S1–S142
(bioMérieux) followed by amplification on 7500 Fast Real-Time PCR
System Dx (Applied Biosystems® ).
Results of analytical sensitivity, exclusivity and inclusivity stud-
ies are presented below.
Analytical sensitivity was determined on the whole system
using in vitro transcript spiked in respiratory samples. This study
showed a 95% limit of detection at 2.89 log10 cp/mL of sample [IC
95%: 2.65–3.29] i.e. 780 cp/mL of sample [IC 95%: 450–1950].
Exclusivity was confirmed with the major human respiratory
viruses including other human coronaviruses. No cross-reaction
was observed. The QCMD Panel MERS-CoV 2015 was tested
and results are as expected, Core and educational samples were
detected. Among the 19 other commercial kits, all but one gave
also the correct identification. The bioMerieux’s solution targeting
S gene gave equivalent results than the kits targeting upE or N gene.
The combination of RNA internal control r-gene® ready-to-use
premix with the MERS-CoV primers r-gene®, MERS-CoV probe r-
gene® and MERS-CoV transcript r-gene® (research use Only) is a
good candidate solution for the detection of MERS-CoV virus and
demonstrate the interest and reliability of the RNA internal control
for the rapid development of detection tool, in outbreak situation.
Abstract no: 131
Presentation at ESCV 2016: Poster 10
Multicentre evaluation of the variability of
adenovirus quantification by PCR
Jacqueline F. Fryer ∗ son G. Hockley
Clare L. Morris
1 Division of Virology, NIBSC, UK
2 Biostatistics, NIBSC, UK
Background: Viral load measurements using nucleic acid
amplification techniques (NAT) are critical for the diagnosis and
management of human adenovirus (HAdV) infections. A variety of
laboratory-developed tests (LDT) and commercial assays are used.
The aim of this study was to evaluate variability in the quantifi-
cation of HAdV by NAT, and the effectiveness of candidate HAdV
reference materials to harmonise viral load measurements.
Methods: HAdV positive patient samples, including; whole
blood, plasma, serum, urine, stool, eye swab, nasal lavage and
sputum, were sourced from clinical laboratories and typed by
sequencing. Virus stocks representing the 9 HAdV types identified
in the clinical samples (Types 1, 2, 4, 5, 7, 14, 31, 40, 41), were grown
in Hep2C cells. The HAdV DNA concentration in the clinical and cul-
tured virus samples was determined at NIBSC using a commercial
and LDT.
Study samples comprised cultured virus representing 9 HAdV
types in 10mM Tris–HCl (pH7.4) containing 0.5% human serum
albumin (TCS1-9), clinical samples diluted in HAdV-negative sam-
ple matrix, and a dilution series of cultured virus prepared in
HAdV-negative sample matrix (TCS-matrix samples, these repre-
sented the same HAdV type and matrix as the clinical samples).
Twelve laboratories from 6 European countries took part in the
study. Each laboratory tested TCS1-9, and the clinical and TCS-
matrix samples relevant to their quantitative HAdV NAT assay.
Results: In total, 16 datasets were received. The SD of the over-
all laboratory mean HAdV concentrations for TCS1-9 ranged from
0.40 to 1.03 Log10 copies/mL. The SDs were highest for Types 7,
31 and 41. The SD of the overall laboratory means for clinical
samples (representing the same HAdV types) ranged from 0.33
to 1.09 Log10 copies/mL. For all TCS and clinical samples inter-
laboratory variation in HAdV quantification was higher than the
intra-laboratory variation.
The effectiveness of candidate HAdV reference materials to har-
monise viral load measurements by NAT was evaluated by ‘relative
potency’. For all clinical samples, apart from one of the stool sam-
ples, the SD of the overall laboratory mean was reduced when the
results were expressed relative to the corresponding TCS-matrix
sample. The effect of diluting the cultured virus in different sam-
ple matrices was determined by plotting the individual laboratory
results for the TCS-matrix dilution series. The mean slope for each
TCS-matrix sample ranged from 0.76 to 1.19.
Conclusions: The results suggest that there is variability in
the quantification of different HAdV types by pan-HAdV NAT, but
this would reduce through standardisation to a common reference
material. A proposal to develop the 1stWHO International Standard
for HAdV for NAT has been endorsed. The results of this study will
be used to determine the most appropriate source material and
formulation for the candidate standard. The availability of a WHO
International Standard for HAdV for NAT will help to standardise
these assays and enable comparison of measurements within and
between different laboratories, thereby improving patientmanage-
ment.
Abstract no: 132
Presentation at ESCV 2016: Poster 11
Evaluation of two algorithms for diagnosis of
Epstein-Barr virus infection
L.C. Rodrigues ∗, P. Silva, C. Cardoso, J.M. Figueira
Patologia Clínica, Centro Hospitalar de Lisboa
Ocidental, Portugal
Introduction: Epstein-Barr virus (EBV) serology is mainly used
to identify primary infections. Two algorithms have been proposed
for diagnostic approach: the anti-VCA approach, in which the study
continues if at least one of the markers, IgG or IgM, is reactive; and
the anti-EBNA approach, in which the study continues only if the
anti-EBNA-1 marker is negative. In our hospital we have used the
full panel until now.
Since the EBV seroprevalence rate is as high as 95% amongst
adults and themajority is reactive to anti-VCA IgG and anti-EBNA-1
antibodies, it is expected that the anti-VCA approach would trigger
more sequels than the anti-EBNA approach, with more costs.
Objective: To evaluate if the use of the anti-EBNA and anti-VCA
approaches prevent the identification of important clinical situa-
tions.
Material and methods: We have retrospectively applied both
screening algorithms to all EBV serology results done in our hospital
between January 2013 and December 2015 ( n = 3090). Diagnoses
obtainedwith both algorithms were comparedwith those obtained
with the full panel. We have analysed the distribution of serological
patterns and the clinical relevance of those patterns that would be
lost with the algorithms’ implementation.
Results: The anti-VCA approach would prevent the use of anti-
EBNA-1 test in 3.9% of all cases. The anti-EBNA approach would
prevent the use of anti-VCA IgG and IgM tests in 95.7% of all cases.
Regarding the anti-VCA approach, when we have an isolated
anti-EBNA-1 pattern (suggesting a past infection), it would be clas-
sified as negative (2.94% of all cases). It will not be a problem in
the investigation of a primary infection but it may be relevant in
the study of diseases that we know can be associated with EBV
infection (autoimmune diseases, post-transplantation lymphopro-
liferative disease, transplant rejection, lymphomas in the context