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

10 / 152

S6 Abstracts / Journal of Clinical Virology 82S (2016) S1–S142
(MR) was introduced in order to analyse the qPCR data without
input from the operator, accounting at the same time for sub-
optimal reactions. In the present work, wemodifiedMR to filter out
results inconsistent with positive reactions using an assumption-
free approach. We applied this novel algorithm of qPCR analysis to
several primer sets targeting a plethora of viruses in order to assess
its effectiveness with respect to the CT.
Methods: Clinical samples (n = 328) were obtained from resid-
ual faecal specimens processed by the Clinical Microbiology and
Public Health Laboratory at Addenbrooke’s Hospital (Cambridge,
UK). The samples were extracted by QIAsymphony SP and amplified
on Custom TaqMan Array 384-well Card by TaqMan Fast Virus 1-Step
Master Mix 2× on Viia7 thermalcyclers. The results were issued as
either positive or negative by three operators and a consensus clas-
sification was generated. A training dataset of 1920 reactions was
obtained from a pool of 54 primer sets performed over 50 plates.
The resulting MR data were analysed by EM algorithm to obtain a
cut-off for the positive/negative results. This filtered MR was then
applied to 23 primer sets targeting different viruses for a total of
6038 reactions. MR values below the empirical cut-off were consid-
ered negative and the consensus classification was used to assess
the accuracy of detection in comparison to CT.
Results: Five of the 23 primer set analysed (21.74%) showed
a better accuracy and negative predictive value using the MR
rather than the CT method, both being in average 0.987 ± 0.013
and 0.996 ± 0.006 for the CT and MR, respectively. The clinical sen-
sitivity for four of these primer sets was in average 0.500 ± 0.136
and 0.764 ± 0.274 for the CT and MR, respectively; for the single
primer set where this parameter could not be computed, CT and
MR showed ten and none false negative reactions, respectively. In
all other instances, CT and MR performed equally.
Discussion: The data gathered suggested that MR is a com-
petitive analytical algorithm for qPCR analysis, providing higher
accuracy than CT in one fifth of the targets tested while being com-
parable to CT in all other cases. MR also had the advantage over CT
of (a) being assumption-free and (b) taking into account primer
specific inhibitions. The use of MR can be beneficial for several
qPCR applications by increasing the effectiveness and reproducibil-
ity of the assay. MR can also assist the operators during the visual
inspection of the individual reactions by highlighting problems in
the amplification.
Abstract no: 202
Presentation at ESCV 2016: Oral 11
Anti-BK virus neutralizing antibody titers
before transplantation predict BK virus
replication in kidney transplant recipients after
transplantation
M. Solis ∗, A. Velay , R. Porcher ,
P. Domingo-Calap , E. Soulier , M. Joly ,
M. Meddeb , W. Kack-Kack , B. Moulin ,
S. Bahram , F. Stoll-Keller , H. Barth ,
S. Caillard , S. Fafi-Kremer
1 Laboratoire de Virologie, Hôpitaux Universitaires
de Strasbourg, Strasbourg, France
2 Inserm UMR S1109, LabEx Transplantex, Fédération
de Médecine Translationnelle de Strasbourg (FMTS),
Université de Strasbourg, Strasbourg, France
3 Centre d’Epidémiologie Clinique (CRESS),
UMR1153, Université Paris Descartes, Paris, France
4 Département de Néphrologie – Transplantation,
Hôpitaux Universitaires de Strasbourg, Strasbourg,
France
BK virus-associated nephropathy (BKVN) is the most fre-
quent BKV-associated disease after renal transplantation, with BKV
reactivation occurring in up to 80% of kidney transplant recipi-
ents (KTR). Virological diagnosis of BKVN relies on the detection
and quantification of viral load in urine and blood by real-time
PCR techniques, allowing preemptive immunosuppressive therapy
adaptation. However, the delayed nature and incomplete success of
this preemptive strategy underscore the need for prognostic mark-
ers of BKV reactivation. Neutralizing antibodies (Nabs) against BKV
genotypes were analyzed in a prospective KTR cohort to investi-
gate whether Nabs titers may predict BKV replication. Blood and
urine samples were prospectively collected from 168 KTR the day
of transplantation, weekly the first month post-transplantation
then monthly during 96 weeks. Using the BKV pseudovirus system
(Pastrana et al., J Virol 2013), anti-BKV Nabs titers were mea-
sured on the day of transplantation and at additional time points
post-transplantation. BKV DNA load was quantified in urine and
blood samples using a commercial qPCR kit (BK virus R-gene® ,
Biomérieux, France). BKV strains of KTR displaying viruria and/or
viremia were genotyped as previously described (Solis et al, JCM
2016). Anti-BKV Nabs were positive in 164 (97.6%) patients before
transplantation. Hundredten (67.1%) KTR harbored higher Nabs
titers against genotype I, while 16 (9.8%) and 7 (4.3%) KTR showed
higher Nabs titers against genotype II and genotype IV, respectively.
Twenty eight KTR harbored higher titers for two genotypes (17
for genotype I and II, 5 for genotype I and IV and 6 for genotype
II and IV). Three harbored similar Nabs titers against the 3 geno-
types. BKV viruria was detected in 52 (31%) patients 1 to 78 weeks
(median 5 weeks) after transplantation. BKV viremia was observed
in 28 (16.7%) patients 5–75 weeks (median 18 weeks) after trans-
plantation, among them 13 (7.7%) developed BKVN 10–76 weeks
(median 17 weeks) after transplantation. In BKV-replicating KTR,
BKV genotype I, genotype II and genotype IV were identified in
45 (86.5%), 1 (1.9%) and 6 (11.5%) patients, respectively. The risk
of developing viruria was higher for patients with lower Nabs
titers before transplantation against their subsequently-replicating
genotype (HR (95% CI) = 0.44 (0.25–0.76; p = 0.003). The replicating
BKV is acknowledged to be of donor origin. Indeed, donor/recipient
mismatches in regard to genotypic neutralization profiles and
replicating strains were found to be greater in BKV-replicating
KTR (p < 0.05). Anti-BKV Nabs titer before transplantation may
represent a valuable prognostic marker of BKV replication after