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

27 / 152

Abstracts / Journal of Clinical Virology 82S (2016) S1–S142 S23
ation of results from the negative samples was <0.1 SD across all
three analytes.
Conclusions: The data demonstrate that the BioPlex 2200 ToRC
IgM results are comparable to other commercially available assays.
Furthermore, the simultaneous detection and identification of anti-
bodies to T. gondii, Rubella, and CMV allows laboratories to increase
throughput and improve overall workflow.
Abstract no: 109
Presentation at ESCV 2016: Poster 3
Comparison of the Beckman Coulter DxN VERIS
and Abbott RealTime assays in analyzing HCV
positive plasma samples
G. Naeth, P. Braun, F. Wiesmann, B. Haase,
H. Knechten
PZB Aachen, Germany
Background: The accurate determination of viral load in Hep-
atitis C infected patients is very important for treatmentmonitoring
and efficacy. The DxN VERIS Molecular Diagnostics System has
become commercially available more recently and the VERIS HCV
assay is performed on this fully-automated, random-access system.
This study was conducted to assess the precision, linearity and sen-
sitivity of the VERIS HCV assay. A direct comparison between the
VERIS HCV assay and the Abbott RealTime HCV assay was also per-
formed which included frozen plasma samples from individuals on
HCV treatment.
Methods: Four HCV quality controls, diluted to nominal con-
centrations of 1.56, 3.38, 5.0 and 7.9 log IU/mL, and one negative
control were tested on DxN VERIS in duplicate for 20 days. For lin-
earity analysis, one high viremic sample (>10,000,000 IU/mL) was
diluted to several concentrations to demonstrate the linearity of
the VERIS HCV assay. Analytical sensitivity was also determined
for the VERIS HCV assay using 4th WHO traceable material and the
result calculated using Probit analysis (95% hit rate). For method
comparison 100 HCV-positive clinical specimens with viral loads
ranging from 12 to 22million IU/mL were tested on both systems.
Additionally, 80 frozen EDTA-plasma specimens derived from 20
individuals undergoing HCV treatment on both systems and viral
load profiles generated.
Results: Coefficients of variation calculated from the DxN VERIS
results ranged from 1.55% for the highest concentration to 9.4% for
the lowest nominal concentration in precision analysis. All HCV
negative samples were confirmed to be undetectable. The linear-
ity was validated for range of 1.7–6.7 log IU/mL and showed a good
correlation (R2 = 0.988). With an analytical sensitivity of 6.2 IU/mL
(CI95%: 5.1–7.9 IU/mL) VERIS showed similar sensitivity to Abbott.
In Bland-Altman analysis both assays showed an overall mean dif-
ference of 0.245 log IU/mL (VERIS HCV – RealTime HCV, n = 91). The
correlation co-efficient based on 91 qualified results (9 results were
excluded from analysis as they were outside the linear range of
either assay) was 0.982 (slope 1.11, intercept −0.19).
The results of 80 frozen EDTA-plasma specimens, covering 4
different blood collection time points, showed a good overall agree-
ment in the viral load profiles obtained for genotype 1–3. For
genotype 4 (n = 4) the mean difference between both assays results
were −0.96 log IU/ml (Veris minus RealTime) in this analysis.
Conclusions: The VERIS HCV assay demonstrated a good cor-
relation over the linear range, particularly at the lower end of the
linear range. The limit of detectionwas comparable to that reported
for the Abbott RealTime assay. Sensitivity and precision of both
assayswere comparable on a high level in general analysis. Discrep-
ancies between both assays were measureable higher at viral load
above 3 log IU/ml. Results received from the VERIS assay tended to
be higher quantified than results received from the RealTime assay.
However, genotype 4 isolates in clinical samples were lower quan-
tified as compared to RealTime. With random access and time to
first result of about 105min the VERIS system is faster and less
time-consuming than the Abbott RealTime System.
Abstract no: 118
Presentation at ESCV 2016: Poster 4
Characterisation and standardisation of
material in QCMD EQA programmes in the
absence of higher order standards
Elaine McCulloch ∗ ubert Niesters
Alastair Ricketts , Shubana Kazi , Paul Wallace
1 QCMD, United Kingdom
2 University Medical Center Groningen, United
Kingdom
3 Qnostics, United Kingdom
Background: Diagnosis and subsequent monitoring of infec-
tious disease levels after treatment using quantitative molecular
assays is well established in clinical practical for a range of differ-
ent viral pathogens including HIV, HCV, HBV, CMV, EBV, and BK.
Within EQA programmes for these pathogens, quality assessment
requires assessing laboratory performance in both the detection
and quantitation of the target pathogens relative to their respective
peer group and where an International Standard is available in the
same common units. For these type of programmes EQA has always
calibrated the material it uses against the International Standard.
However for the material of infectious diseases there are currently
no international standard or certified reference material hence
QCMD develops and uses internal reference materials to charac-
terise the materials used within the EQA regardless of whether the
EQA programme is ‘qualitative’ or ‘quantitative’. By doing this we
ensure consistency within and across EQA programmes over time.
In this study the use of digital PCR (dPCR) to quantify materials
for two EQA programmes that are currently primarily qualitative
(herpes simplex virus 1 (HSV1) and Varicella Zoster virus (VZV))
was examined and compared to reference quantitative real-time
PCR (qPCR) methods.
Methods: Pathogen specific materials were prepared for use
within each respective EQA programme. Each target pathogen
HSV1 and VZV was prepared transport media. The samples were
characterised using both qPCR based methods and digital PCR (Bio-
RadQX200 droplet dPCR) by selected reference testing laboratories.
The samples were then included in the EQA panels and distributed
to laboratories in two challenges across 2015 to registered partici-
pating laboratories. Laboratories were asked to treat the materials
as they would a clinical sample. Results were returned to QCMD via
a dedicated online system where sample results including quanti-
tative data along with information on the assay workflow were
collected.
Results: Data analysis was performed on all quantitative
datasets HSV1 (106), VZV (86). Datasets were categorised based
on the amplification assay manufacturer and method. Outliers
were assessed through the application of Grubbs’ analysis to each
assessment group. Datasets were considered suitable assessment
if more than five laboratories reported data using the same assay
manufacturer after removal of outliers. This allowed a consensus
concentration to be derived. Each assessment groups datasets were