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

35 / 152

Abstracts / Journal of Clinical Virology 82S (2016) S1–S142 S31
Abstract no: 168
Presentation at ESCV 2016: Poster 20
Comparative evaluation of the Aptima HSV 1&2
assay and a lab developed real-time PCR test for
detection of HSV-1 and HSV-2 viruses
A. Jassem ∗, M. Krajden , D. Getman ,
P. Hovey , C. Hentzen , N. Barakat , A. Jiang
1 British Columbia Centre for Disease Control Public
Health Laboratory, Vancouver, BC, Canada
2 Hologic Inc., San Diego, CA USA
Introduction: Herpes simplex virus 1 and 2 (HSV-1/HSV-2)
cause significant morbidity in humans. Accurate diagnosis of HSV
infections is important for treatment as well as counselling to
reduce transmission. The Aptima Herpes Simplex Virus 1&2 Assay
(AHSV, Hologic, Inc.) is an in vitro real-time nucleic acid amplifi-
cation test (NAAT) for the qualitative detection and differentiation
of UL42 mRNA from HSV-1 and HSV-2 on the automated Panther
System. This study evaluated the clinical performance characteris-
tics of AHSV in comparison to a lab developed real-time PCR test
(LDT) targeting glycoproteins genes D and G for HSV-1 and HSV-2,
respectively.
Materials and methods: Swab specimens (n = 1249) in viral
transport medium from ocular, genital, skin and mouth lesions
were submitted to the British Columbia Centre for Disease Control
Public Health Laboratory for testing with the lab developed, plate-
based real-time PCR test for HSV-1, HSV-2, and VZV DNA. Remnant
samples were testedwith the AHSV assay. Samples with discordant
results were tested with a validated DNA PCR sequencing assay
specific for HSV-1 and 2.
Results: A consensus result was established with both AHSV
and the LDT real-time PCR assay. A total of 221 positives (17.5%;
103HSV-1 and 118HSV-2) were identified. For all specimens, AHSV
and LDT positive agreement for HSV-1was 82.4% and for HSV-2was
94.4%, while negative agreement was 99.8% for HSV-1 and 99.7%
for HSV-2. The kappa value for HSV-1 was 0.885 and for HSV-2 was
0.955. In 555 anogenital samples, positive agreement for HSV-1
was 89.4% and for HSV-2 was 94.1%, and negative agreement was
99.8% for HSV-1 and 99.6% for HSV-2. The kappa value for HSV-1
was 0.927 and for HSV-2 was 0.958. Most of the discordant results
(29/34, 85%) were positive results obtained only by the LDT. For
these, the average cycle threshold (Ct) values were 30.6 for HSV-1
and 30.7 for HSV-2, while positive samples that were in agreement
had average Ct values of 20.5 for HSV-1 and 22.4 for HSV-2 (P < 0.001
for both targets). For samples with Ct values <30, sequencing anal-
ysis confirmed 9/11 (82%) discordant results as true positive, while
for samples with Ct values >30, sequencing analysis confirmed 6/23
(26%) discordant positive results. Of the LDT positive, AHSV nega-
tive samples, 21/29 (72%) were directly adjacent to HSV positive
samples with low HSV Ct values.
Conclusions: Overall, detection of HSV-1 and -2 viral mRNAs in
clinical specimens using the Aptima HSV 1&2 assay showed good
agreement with amplified molecular tests for HSV genomic DNA.
Although the AHSV failed to detect several low positive samples
identified by the LDT, some of those positives may be the result of
cross contamination. The LDT used a plate-based PCR assay which
can be prone to sample cross contamination. The Aptima HSV 1&2
assay is a single tube assay with very low risk of sample cross
contamination that is a good alternative to LDT.
Abstract no: 170
Presentation at ESCV 2016: Poster 21
Plasma and serum are suitable specimen types
for quantitation of HCV RNA using real-time
transcription mediated amplification or PCR
S. Hau ∗, A.M. Geretti , M. Hopkins
1 Liverpool Clinical Laboratories, United Kingdom
2 University of Liverpool, United Kingdom
Aims: Hepatitis C virus (HCV) RNA assays form an integral part
of patient management, from confirmation of antibody results at
diagnosis through to monitoring treatment efficacy. This dual pur-
pose often means both serum and plasma samples are tested by
diagnostic laboratories. The aim of this study was to evaluate per-
formance of the new Aptima HCV Quant DX Assay (Aptima) in
comparison to the Roche COBAS Ampliprep/COBAS Taqman HCV
test v2.0 (CAP/CTM) using both serumand plasma samplematrices.
Method: A total of 319 surplus clinical samples were tested:
141 plasma (27 negative), 178 serum (40 negative) in parallel with
Aptima and CAP/CTM assays in accordance with manufacturers’
guidelines. The sample set included paired serum and plasma sam-
ples collected at the same time point from 20 patients. All samples
were obtained from 225 patients received as part of routine care
between July 2011 and March 2016.
Paired plasma and serum from three patients infected with HCV
genotypes 1a, 3a and 4d were also used to assess assay precision.
Aliquots of the plasma and serum were tested in replicates (n = 7)
over a period of 3 days.
Three standard panels were also used: Qnostic performance
panel for concordance of results; Acrometrix HCV RNA standard
panel tested in triplicate to assess linearity in serum and plasma;
SeraCare HCV RNA genotype performance panel was used to assess
quantitation across genotypes in serum and plasma.
Results: Aptima and CAP/CTM showed excellent accuracy and
linear correlation with expected results across the Acrometrix HCV
RNA panel range (1.5–6.3 log10 IU/ml) in serum and plasma sam-
ples. In terms of precision at a nominal value of 1000 IU/ml, Aptima
showed coefficients of variation up to 3.5% and 4.5% log10 IU/ml in
plasma and serum respectively. Corresponding values for CAP/CTM
were 2.1% and 2.2%. CAP/CTM reported significantly higher quan-
tification in serum replicates but this effect was not observed
in the paired plasma and serum samples of 20 patients. Within
the tested clinical sample set the prevalence of genotypes 1, 2,
3 and 4 were as follows: 35%; 1%; 35%; 3% respectively and 26%
were unknown. Aptima gave good quantitation across all geno-
types tested, although results were generally lower than CAP/CTM
in the SeraCare and Qnostics panels and closer to the Abbott
Real-Time HCV results reported by SeraCare. Analysis of clini-
cal samples showed excellent correlation (95.0%, Kappa = 0.852)
between the two assays for detection of HCV RNA, and 87.5%
agreement for quantitation at 15 IU/ml. Regression and Bland-
Altman analysis showed a proportional bias: Aptima quantified
lower than CAP/CTM closer to the lower limit of quantitation,
whilst the converse was observed towards the upper limit of
quantitation. The effect was similar in plasma (mean differ-
ence 0.07 log10 IU/ml, R 2 = 0.9665) and serum (mean difference
0.08 log10 IU/ml, R2 = 0.9774).
Conclusion: CAP/CTM and the new Aptima HCV assay show
excellent analytical performance in both serum and plasma. Both
tests are highly accurate with equivalent sensitivity. Comparative
quantitation between the two assays deviated most when close
to the assay limits of quantitation. Overall, this evaluation demon-