Potential uses of EBV and CMV
viral load assays
In solid organ and hematopoietic stem cell transplantation
 As triggers for pre-emptive therapy for disease
prevention
 For disease diagnosis
 For treatment monitoring
 As surrogate markers of anti-viral resistance
 For safety monitoring in clinical trials (new
immunosuppressive agents)
Other
 Disease diagnosis and treatment monitoring other EBVrelated disease- nasopharyngeal carcinoma ,NK and
Hodgkin’s lymphoma
 Population based screening- congenital CMV disease
CMV and EBV Viral Load Assays
Current Problems
 Many “In-house” ; not standardized or
cross referenced
 Optimal sampling site uncertain - serum,
Leukocytes/lymphocytes, whole blood
 Optimal sampling frequency uncertain
 Natural history studies are scarce so that
“trigger points” for intervention have not
been clearly defined
Development of an International
Standard for EBV and CMV Viral
Load Assessment
Dr Jutta K Preiksaitis
Provincial Public Health Laboratory (Alberta)
University of Alberta
Edmonton and Calgary, Alberta Canada
On behalf of the American Society of Transplantation
Infectious Diseases Community of Practice and the
Canadian Society of Transplantation
Objective of Study
 To examine the inter-laboratory variability in
qualitative and quantitative CMV and EBV viral
load assessments


Funded by the American Society of Transplantation
and the Canadian Society of Transplantation ( armslength educational grant Roche Canada)
Coordinated through the American Society of
Transplantation Infectious Diseases Community of
Practice
CMV Viral Load Assays
Establishing the
“expected result”
 Viral stock (purified nucleocapsids of Merlin, a
clinical isolate in human in CMV seronegative
human plasma)


Quantified by nucleocapsid count using electron
microscopy log 10 copies/ml =4.52
Calculation of a mean of replicate viral load results
from seven reference laboratories (included use of all
available commercial assays) log 10 copies/ml =5.0
Panel Design
 12 samples
 2 negatives (CMV seronegative plasma)
 7 samples -dilutions of purified viral stock;
replicates of two dilutions were included
 3 clinical samples (1:30 dilution in CMV
seronegative plasma)



UL54 mutation (not ganciclovir resistant)
UL97mutation (ganciclovir resistant) and gB mutation
No mutation
CMV PCR Methods Utilized
n=35 panels (33 labs)
19 US, 12 Canada , 2 EU
Other
commercially
available kits
(ASRs)
23%
Roche
Amplicor
26%
Inhouse
51%
Results Summary
35 panels / 33 laboratories
CMV DNA Copies/ml (log10)
7.00
6.50
6.00
5.50
5.00
4.50
4.00
3.50
3.00
2.50
2.00
1.50
1.00
0.50
0.00
02
09
Clinical sample
07
08
04
11
03
12
06
CMV Sample Number
CMV viral panel sample
Expected result based on stock quantified by reference laboratories
Positive but not quantifiable (assigned lowest detectable value)
10
05
01
Summary of CMV Qualitative Results
(constructed samples)
35 panels / 33 labs
Sample
No
EM –based
expected
result
copies/ml
(log10)
Reference
lab
expected
result
copies/ml
(log10)
02
0.0
†09
Number of panels
Negative
(%)
Positive-NQ
(%)
Positive-Q
(%)
0.0
34 (97)
0
1 (3)
0.0
0.0
33 (94)
0
1 (3)
07
1.5
2.0
26 (74)
6 (17)
3 (9)
08
2.5
3.0
4 (11)
4 (11)
27 (77)
04
3.5
4.0
0
1 (3)
34 (97)
11
3.5
4.0
0
2 (6)
33 (94)
03
4.5
5.0
0
0
35 (100)
12
4.5
5.0
0
0
35 (100)
06
5.5
6.0
0
0
35 (100)
† One test was invalid
Pos-NQ: positive but not quantifiable
Pos-Q: positive with quantifiable results
Summary of CMV Quantitative results
(constructed samples)
35 panels / 33 laboratories
Sample
No
EM –based
expected
result
copies/ml
(log10)
Reference lab
expected
result
copies/ml
(log10)
07
1.5
2.0
08
2.5
3.0
04
3.5
4.0
11
3.5
4.0
03
4.5
5.0
12
4.5
06
5.5
Number
positive
†GM SD
copies/ml
(log10)
Median (range)
copies/ml
(log10)
9
2.2  0.44
0 (0-2.78)
31
3.1  0.58
3.01 (0-4.32)
35
3.89  0.52
4.02 (2.33-5.08)
35
3.84  0.52
3.95 (2.62-5.01)
35
4.83  0.44
4.89 (3.42-5.89)
5.0
35
4.80  0.49
4.90 (3.68-5.91)
6.0
35
5.59  0.52
5.51 (4.65-6.73)
† Geometric mean; negative results were excluded
CMV quantitative results relative to expected result
[reference labs as “gold” standard]
§Number of panel results falling within specified
parameter relative to expected result [reference labs]
(copies/ml, log10)
Sample
No
#
positive
log±0.2
(%)
log±0.5
(%)
log±1
(%)
> log±1
(%)
07
9
2 (22)
7 (78)
9 (100)
0
08
31
8 (26)
21 (68)
27 (87)
4 (13)
04
35
17 (49)
26 (74)
33 (94)
2 (6)
11
35
16 (46)
25 (71)
32 (91)
3 (9)
03
35
19 (54)
25 (71)
34 (97)
1 (3)
12
35
16 (46)
25 (71)
32 (91)
2 (6)
06
35
7 (20)
15 (43)
32 (91)
3 (9)
§negative results were excluded
CMV Qualitative and Quantitative results
(clinical samples)
35 panels / 33 laboratories
Clinical Sample Number
Qualitative
Result
Negative (%)
Pos-NQ (%)
Pos-Q (%)
Quantitative
Result
copies/ml
(log10)
†GMSD
Median (range)
#10
#05
#01
13 (37)
0
0
9 (26)
1 (3)
0
13 (37)
34 (97)
35 (100)
2.78  0.72
3.89  0.53
3.97  0.47
2.24
(0-4.18)
3.87
(2.73-4.89)
3.99
(3.08-5.05)
† GM=Geometric mean; negative results were excluded
Result linearity over dynamic range
(each line represents results from one lab)
CMV copies/ml (log10) using PCR
Commercial assays (Lab =17)
In-house assays (Lab =18)
7.00
7.00
6.00
6.00
5.00
5.00
4.00
4.00
3.00
3.00
2.00
2.00
1.00
1.00
0.00
0
0.00
0
1
2
3
4
5
6
7
1
2
CMV Copies/ml (log10) Expected result
3
4
5
6
7
Comparison of Intra and Inter laboratory
variation in CMV vial load assessments in
duplicate specimens
mean coefficient of variation (CV), %
Duplicate samples
(sample 04 and 11)
35 panels
Duplicate samples
(sample 03 and 12)
35 panels
p value*
Intra-Lab
21.48
17.62
0.720
Inter-Lab
149.23
139.15
0.316
p value*
< 0.0001
< 0.0001
* Fisher Exact Test (two tailed)
CMV Conclusions
 Significant variation exists in reported results. The greatest
variation was observed in clinical samples and constructed samples
with low viral load. As viral load increased, there was less variation
independent of the assay platforms used (commercial versus inhouse)
 False negative results were not observed in samples with viral load
greater than 3.0 log copies/ml (expected result) even when lower
limit of detection reported was higher than this value
 Variation is lower limits of detection may have significant impact on
duration of treatment based on recommendation of treatment until
viral load is non-detectable
 If ± 0.5 log10 is considered “acceptable” assay variation, only 62.5
% of the results observed fell within this range
CMV Conclusions
 As a group, commercial assays demonstrated
overall less variability compared to all “in house”
developed assays, but some of the former have
limitations related to lower sensitivity and limited
dynamic range
 Inter-laboratory variability was significantly greater
than intra-laboratory variability, highlighting the need
for an international reference standard for assay
calibration
EBV Viral Load Assays
Establishing the “expected result”
 EBV viral stock (Namalwa cell line in EBV
seronegative plasma)


Quantified by Namalwa cell count using
assumption of 2 EBV genome copies per cell
Calculation of a geometric mean of replicate
viral load results from seven reference
laboratories ( included use of all available
commercial assays)
Panel Design
12 samples
Constructed samples-(total cell count in each sample fixed
to mimic total white cell count in normal peripheral
blood)
 2 negatives ( EBV-negative Molt-3 cells in EBV
seronegative plasma)
 7 samples -dilutions of EBV-positive Namalwa cells and
EBV-negative Molt-3 cells ; two dilutions were replicated
 3 clinical plasma samples (diluted in EBV seronegative
plasma)
 Two patients had EBV-positive B cell post-transplant
lymphoproliferative disorder
EBV PCR Methods Utilized
n=30 panels (28 labs)
16 US, 11 Canada, 2 EU
Other
commercially
available kits
(ASRs)
40%
Inhouse
60%
Results Summary
30 panels / 28 laboratories
7.00
EBV DNA copies/ml (log10)
6.00
5.00
4.00
3.00
2.00
1.00
0.00
01
08
09
03
05
Clinical Samples
EBV Viral Panel Samples
10
02
11
06
EBV Sample Number
Gold Standard Expected Result Based on Cell Count
Positive but not quantifiable (assigned lowest detectable value)
07
04
12
Summary of EBV Qualitative Results
(constructed samples)
30 panels reported from 28 laboratories
Sample
No.
Expected result
based on
Namalwa cell
count copies/ml
(log10)
Number of panels
Negative
(%)
Positive-NQ
(%)
Positive-Q
(%)
01
0.0
30 (100)
0
0
08
0.0
28 (93)
0
2 (7)
09
1.3
27 (90)
1 (3)
2 (7)
03
2.3
16 (53)
3 (10)
11 (37)
05
3.3
3 (10)
2 (7)
25 (83)
10
3.3
3 (10)
1 (3)
26 (87)
02
4.3
0
1 (3)
29 (97)
11
4.3
0
2 (7)
28 (93)
06
5.3
0
1 (3)
29 (97)
 Quantitation based on cell count
Pos-NQ: positive, not quantifiable
Pos-Q: positive, quantifiable
Summary of EBV Quantitative results (Constructed Samples)
30 panels reported from 28 labs
Sample
No.
09
Expected result
based on Namalwa
cell count
copies/ml
(log10)
1.3
Number
of
positive
results
† GM SD
copies/ml
(log10)
Median (range)
copies/ml
(log10)
3
1.890.93
0.00 (0.002.74)
14
2.480.59
0.00 (0.003.78)
03
2.3
05
3.3
27
2.970.52
2.92 (0.00-4.14)
10
3.3
27
3.020.61
2.92 (0.00-4.12)
02
4.3
30
3.920.59
4.03 (2.76-5.04)
11
4.3
30
3.880.66
06
5.3
30
4.810.81
†Geometric mean; negative results were excluded
3.97 (2.18-5.00)
4.96 (2.15-6.09)
EBV quantitative results (constructed samples) relative to
expected result [Namalwa cell count as “gold” standard]
§Number of panel results falling within specified parameter
relative to expected result [Namalwa cell count] (copies/ml,
log10)
Sample
No
Number
positive
results
log±0.2
(%)
log±0.5
(%)
log±1
(%)
> log±1
(%)
09
3
0
1 (33)
1 (33)
2 (67)
03
14
3 (21)
10 (71)
12 (86)
2 (14)
05
27
5 (19)
16 (59)
26 (96)
1 (4)
10
27
6 (22)
14 (52)
25 (93)
2 (7)
02
30
10 (33)
17 (63)
25 (83)
5 (17)
11
30
8 (27)
15 (50)
25 (83)
5 (17)
06
30
4 (13)
17 (57)
25 (83)
5 (17)
§ negative results were excluded
EBV Qualitative and Quantitative results
(clinical samples)
30 panels reported from 28 labs
Clinical Sample Number
Qualitative
Result
#07
#04
#12
Negative (%)
0
0
0
Pos-NQ (%)
0
0
0
30 (100)
30 (100)
30 (100)
†GMSD
4.08  0.60
3.95  0.56
4.21  0.61
Median (range)
4.09
(3.09- 5.12)
3.96
(3.10 – 5.31)
4.36
(3.08 – 5.12)
Pos-Q (%)
Quantitative
Result
copies/ml,
log10
† GM= Geometric mean
Result linearity over dynamic range
(each line represents results from one lab)
EBV copies/ml (log10) using PCR
Commercial assays
(Lab =12)
In-house assays
(Lab = 18)
7.00
7.00
6.00
6.00
5.00
5.00
4.00
4.00
3.00
3.00
2.00
2.00
1.00
1.00
0.00
0.00
1.00
2.00
3.00
4.00
5.00
0.00
6.00
0.00
1.00
2.00
3.00
4.00
EBV copies/ml expected quantification based on cell count
5.00
6.00
Comparison of Intra and Inter laboratory variation
in EBV vial load assessments in duplicate
specimens
Mean coefficient of variation (CV), %
Duplicate (sample
05 and 10)
25 panels
Duplicate (sample
02 and 11)
30 panels
p value*
Intra-Lab
39.01
30.48
0.234
Inter-Lab
135.56
135.26
1.0
p value*
< 0.0001
< 0.0001
* Fisher Exact Test (two tailed)
Conclusions
 Significant variation in reported results exists in all
samples independent of viral load and of assay platforms
used (commercial versus in-house)
 If ± 0.5 log10 is considered “acceptable” variation in a Q
NAT assay, our results indicate that only 56 % of all
results fell within that parameter
 Greater QNAT variations were observed in cellular
constructed samples when compared to acellular
plasma samples indicating that DNA extraction in cellular
samples may need further improvement
 Inter-laboratory variability was significantly greater than
intra-laboratory variability, highlighting the need for an
international reference standard for assay calibration
Next Steps
Highest Priority
Establishment of an international reference
standard for EBV and CMV qualitative and
quantitative assay calibration
Acknowledgments
Technical Committee
Dr
Dr
Dr
Dr
Xiao-Li Pang
Julie Fox
Geraldine Miller
Angie Caliendo
Technical and other support
Jayne Fenton
Sandra Shokopoles
Kim Martin
Ana Shynader
AST ID Community of Practice
Dr John Saldanha
Dr Alan Heath
Participating Laboratories
Canada
 Children’s Hospital of British Columbia,
Vancouver
 St. Paul’s Hospital, Vancouver
 Provincial Laboratory for Public Health
Alberta, Edmonton & Calgary
 National Microbiology Laboratory, Winnipeg
 St. Joseph’s Health Care, Hamilton
 Hospital for Sick Children, Toronto
 Mt. Sinai Hospital, Toronto
 Children’s Hospital of Eastern Ontario,
Ottawa
 London Laboratory Services, London
 St. Justine Hospital, Montreal
 Centre hospitalier de l'Université Laval,
Quebec City
 QE II Health Sciences Centre, Halifax
 Newfoundland Public Health Laboratory, St.
John’s
Europe
 Erasmus MC, University Medical Center
Rotterdam, The Netherlands
 Institute for Medical Microbiology, Basel,
Switzerland
USA
 UCLA Healthcare Clinical Labs, Los Angeles
 Stanford Hospital and Clinics, Stanford
 Yale-New Haven Hospital, New Haven
 Emory Hospital, Atlanta
 University of Iowa, Iowa City
 University of Chicago Hospitals, Chicago
 Johns Hopkins Hospital, Baltimore
 University of Michigan Medical Center, Ann
Arbor
 Warde Medical Laboratory, Ann Arbor
 Mayo Clinic, Rochester
 St. Louis Children’s Hospital, St. Louis
 Viracor Laboratories, Lee’s Summit
 University of North Carolina Hospital, Chapel
Hill
 Mt. Sinai Hospital, New York
 Cleveland Clinic, Cleveland
 Oregon Health & Science University,
Portland
 Children’s Hospital of Pittsburgh, Pittsburgh
 Vanderbilt University Medical Center,
Nashville
 Seattle Cancer Care Alliance, University of
Washington, Seattle
 Children’s Hospital, Birmingham