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

56 / 152

S52 Abstracts / Journal of Clinical Virology 82S (2016) S1–S142
Abstract no: 91
Presentation at ESCV 2016: Poster 62
Comparison of the performances of the three
nucleic acid extraction systems: The Roche
Magna Pure LCTM System, the Biomerieux
easyMAG® system, and the new Biomerieux
eMAGTM System
Julien Lupo ∗, Laurence Grossi ,
Gwladys Perrin-Confort , Maud Lemoine ,
Touyana Semenova , Olivier Epaulard ,
Patrice Morand , Raphaële Germi
1 Structural Biology Institute, UMR 5075
CEA/CNRS/UGA – Virology Laboratory, Grenoble
University Hospital, Grenoble, France
2 Structural Biology Institute, UMR 5075
CEA/CNRS/UGA, France
3 Virology Laboratory, Grenoble University Hospital,
France
4 BioMérieux, Centre Christophe Mérieux, 5 rue des
Berges, Grenoble, France
5 Structural Biology Institute, UMR 5075
CEA/CNRS/UGA – Department of Infectious Diseases,
Grenoble University Hospital, France
Quantification by qPCR of Epstein-Barr DNA load in whole
blood (EBV-L) is required for the monitoring of post-transplant
lymphoproliferative disorder risk or adjustment of the immuno-
suppressive regimen. It may also be useful in other EBV-associated
diseases. EBV-L measurement is not currently standardized but
automated system for DNA extraction and commercial assays could
help to obtain more reliable results in the routine setting of non-
specialized laboratories. Here we compared the performances of
three automated extraction systems: NucliSENS easyMAG® and
eMAGTM (bioMérieux) and MagNA Pure LC (Roche). eMAG is the
new generation of easyMAG: eMAG provides full automation while
keeping high flexibility in term of sample management.
86 samples were collected from the routine of the labo-
ratory before any congelation and DNA was extracted using
the three automated systems. EBV-L was measured on a LC480
(Roche applied science) by using the EBV R-gene quantification kit
(bioMérieux).
Qualitative analysis showed that the overall concordance was
78% between the 3 extraction methods. 11 results of 86 (13%) were
discordant between MagNA Pure (MP) and eMAG (8 positives with
eMAG/negatives with MP and 3 positives with MP/negatives with
eMAG); 22 results of 86 (25%) were discordant between MP and
easyMAG (EZM) (11 positives with EZM/negatives with MP and
11 positives with MP/negatives with EZM); and 23 results of 86
(27%) were discordant between eMAG and EZM (9 positives with
EZM/negatives with eMAG and 14 positives with eMAG/negatives
with EZM. All discordant results were below 2000 copies/mL.
Blant–Altman analysis showed that few samples were quantita-
tively discordant (variability > [mean difference]/2 ± 1.96 standard
deviation): 1 of 29 (3.4%) for eMAG/MP comparison; 1 of
27 (3.7%) for EZM/MP; and 1 of 35 (2.8%) for eMAG/EZM.
Blant–Altman analysis also showed that EBV-L obtained after
MP extraction were higher than after eMAG extraction (mean
differences = 0.2 log copies/mL) or EZM extraction (mean differ-
ences = 0.29 log copies/mL). Nevertheless, the final sensitivity is not
impacted and this would not change the therapeutic manage-
ment of patients: 22% of EBV-L obtained after MP extraction were
0.5 log copies/mL higher than after eMAG or EZM extractions.
In conclusion, the new eMAG extraction system fulfills the
extraction performances forwhole blood samples. The 3 automated
systems allowed a reliable extraction of EBV DNA in whole blood
with similar qualitative and quantitative performances. Discrep-
ancies were mostly observed for low viral load. This emphasizes
the current difficulty to compare EBV-L measured with different
technologies, even statistically well correlated and suggest that the
monitoring of EBV-L should be manage with a single method.
Abstract no: 94
Presentation at ESCV 2016: Poster 63
Comparison of the Hologic Aptima HCV Quant
Dx assay to the Roche COBAS Ampliprep/COBAS
TaqMan HCV Test v2.0 for the quantification of
HCV-RNA in plasma samples
K. Schønning ∗, K. Johansen, B. Landt, H. Westh
Department of Clinical Microbiology, Hvidovre
University Hospital, Hvidovre, Denmark
Background: Monitoring of HCV RNA levels remain useful in
evaluating antiviral treatment in chronic HCV infection. Variations
in performance of different tests may impact clinical decisions
and knowledge of the analytical performance of tests is therefore
important for clinical care.
Objectives: To compare the analytical performance of the
APTIMA HCV Quant Dx Assay (APTIMA) and the COBAS
Ampliprep/COBAS TaqMan HCV Test v2.0 (CAPCTM) for the quan-
tification of HCV RNA in plasma samples.
Study design: The performance of the two tests was compared
on 125 archived clinical plasma samples of known genotypes, and
on dilutions series in six replicates of clinical samples of genotype
1a, 2b, 3a and 4a.
Results: Mean bias in quantification between the two test
(APTIMA–CAPCTM) was 0.13 Log IU/mL (SE: 0.32 Log IU/mL) and
changed little when results were stratified on genotypes (1a
(N = 35): 0.26 Log IU/mL (SE: 0.29); 1b (N = 27): 0.23 Log IU/mL
(SE: 0.25); 2b (N = 12): 0.31 log IU/mL (SE: 0.30); 3a (N = 36):
−0.12 Log IU/mL (SE: 0.27). Although the two tests were highly
correlated (R = 0.977), Deming regression showed that APTIMA
quantified higher than CAPCTM for high viral loads. In the dilu-
tions series the APTIMA test was linear with slopes very close to
the expected (1a: 1.01; 2b: 1.02; 3a: 1.02; 4a: 1.00). For all four
genotypes tested CAPCTM yielded slopes less than one (1a: 0.96;
2b: 0.94; 3a: 0.87; 4a: 0.93). The APTIMA assay appeared at least as
sensitive as the CAPCTM test detecting for all four genotypes more
replicates than the CAPCTM test. Precision of both tests was com-
parable with %CV less than 5% for HCV RNA levels above 100 IU/mL.
Conclusion: The APTIMA assay and the CAPCTM test are highly
correlated. Both tests are sensitive and precise. Linearity of the
APTIMA test is excellent and the test will be useful to monitor ther-
apy responses during antiviral treatment of chronic HCV infection.