As presented at the Annual Meeting of the Association for Molecular Pathology, November 2009
EVALUATION OF THE EFFECT OF SPECIMEN HANDLING CONDITIONS IN BD VACUT
Evaluation of the Effect of Specimen Handling Conditions
VIRAL LOAD USING ROCHE COBAS® AMPLIPREP/COBAS® TAQMAN® HIV-1 T
®
in
BD
Vacutainer
PPT on the Stability of HIV-1 Viral Load
H Fernandes, M Ramanathan, S Morosyuk, T Do, L Rainen
®
®
®
AmpliPrep/COBAS
TaqMan
HIV-1Systems,
TestPleasanton, CA
using
COBAS
University ofRoche
Medicine and Dentistry
of New Jersey,
Newark, NJ; BD Diagnostics, Franklin
Lakes, NJ; Roche Molecular
1
1
2
3
2
1
2
3
H. FERNANDES, 1M. RAMANATHAN, 2S. MOROSYUK, 3T. DO, 2L. RAINEN
1
University of Medicine and Dentistry of New Jersey, Newark, NJ;
BD Diagnostics, Franklin Lakes, NJ; 3Roche Molecular Systems, Pleasanton, CA
1
2
Introduction
Accurate quantification of HIV-1 RNA is essential to the management of HIV-1 infected
patients. The aim of this study was to validate the performance of the BD Vacutainer® Plasma
INTRODUCTION
Preparation Tube (PPT™) with the Roche
COBAS® AmpliPrep/COBAS® TaqMan® HIV-1 Test and
evaluate the effect of the following parameters on HIV-1 viral load:
Table 1: Available Results per Tube Type and Handling Condition
MATERIALS AND METHODS
Tube/Condition
Accurate
quantification of HIV-1 RNA is essential to the
1) Storage of whole blood in PPT for six hours prior to centrifugation.
management
of HIV-1 infected patients. The aim of this study
2) Storage of plasma in situ in the PPT at room temperature (RT) or 4ºC for up to 5 days.
®
was
to
validate
thetransport)
performance
the BD
Plasma
3) Agitation (to simulate
of the tubeof
followed
by aVacutainer
re-centrifugation.
™
®
Preparation Tube (PPT ) with the Roche COBAS
AmpliPrep/COBAS® TaqMan® HIV-1 Test and evaluate the
effect of the following parameters on HIV-1 viral load:
PPT 2H (B)
PPT invert (F)
11
30
55
14
1
9
31
55
4
1
6
44
Collection Tubes:
55
15
2
6
• Tube A:
BD Vacutainer® K2EDTA Tube, Ref.#367861
32
*ND= Not Detected
®
™
• Tubes
B-G:2:BD
Vacutainer
Plasma
Tube
Figure
Correlation
plot
of LogPreparation
HIV-1 viral
load(PPT
in ),
10
Ref.
#362788
PPT 2H (B) to EDTA
Assay:
5.0
Roche COBAS® AmpliPrep/COBAS® TaqMan® HIV-1 Test (CAP/CTM)
2
BD Vacutainer® Plasma Preparation Tube (PPT™), Ref. #362788
3)
Agitation
(to simulate
transport)
theTest
tube
followed
Assay:
Roche COBAS
AmpliPrep/COBAS
TaqManofHIV-1
(CAP/CTM)
Statistical
by a Analysis:
re-centrifugation.
F
4.5
®
Intercept = 0.213
Statistical Analysis:
Slope = 0.947
4.0
• Cochran’s Q
test
to
compare
the
proportion
subjects with
R squaredof
= 92.1%
3.5
undetectable viral load count (ND) and results under the limit 3.0
of detection (LOD) of <48 copies/mL between tubes.
log10 PPT 2H (copies/mL)
®
>48 Copies/mL
1
PPT invert/spin (G)
Collection
Tubes:
2)
Storage
of plasma in situ in the PPT at room temperature A: or
BD Vacutainer
K EDTA
Ref.# 367861
• Tube
(RT)
4˚C for up
to Tube,
5 days.
®
Data not available <48 Copies/mL
Patient Population:
55
18
0
9
28
Blood
was collected
from 55
HIV+
PPT
1D RT (D)
55
18 adult, consented
0
5 subjects32
PPT
5D 4ºCprevious
(E)
55 load (VL)14test results0 of <1000 13
with
viral
copies/mL.28
1) Storage of whole blood in PPT for six hours prior
Patient
Population:
Blood was collected from 55 adult, consented HIV+ subjects with previous
viral
to
centrifugation.
load (VL) test results of <1000 copies/mL.
®
*ND Count
PPT 6H (C)
Materials and Methods
• Tubes B-G:
Total Count
Study Design: 55See Figure 1.
13
EDTA (A)
F
• Cochran’s Q test to compare the proportion of subjects with undetectable viral load count
(ND) and results under the limit of detection (LOD) of <48 copies/mL between tubes.
• McNemar's Test for paired samples.
• Analysis of concordance of dichotomized CAP/CTM HIV-1 Test, v1.0 results (<50 copies/mL,
≥50 copies/mL) using 2x2 tables for comparison between results from all PPT tube handling
conditions to EDTA results.
2.5
• McNemar’s Test for paired samples.
2.0
Regression Line
Ideal Line
• Analysis of concordance of dichotomized CAP/CTM HIV-1 Test,
2.0
2.5
3.0
3.5
4.0
4.5
5.0
v1.0 results (<50 copies/mL,
>50 copies/mL) using 2x2 tables
log EDTA (copies/mL)
for comparison between results from all PPT tube handling conditions
Figure 1: Study
Design to EDTA results.
10
Figure 1: Study Design
F
55 consented adult HIV+ Subjects
Collect 7 Tubes
EDTA(A)
PPT(B)
PPT(C)
PPT(D)
PPT(E)
PPT(F)
PPT(G)
2 hrs. after collection, spin
tube at 1100xg for 20 min.
6 hrs. after collection, spin tubes at 1100xg for 20 min.
“EDTA”
“PPT 2H”
“PPT 6H”
Store at room temp.
for 24 hours
Store at 4˚C
for 5 days
Invert tube
20 times
Invert tube
20 times.
Re-spin at 600xg
for 5 min.
“PPT 5D 4˚C”
“PPT Invert”
“PPT Invert/spin”
Aliquot plasma and store at -70˚C
“PPT1D RT”
Aliquot plasma and store at -70˚C
On day of analysis, thaw aliquots and test in Roche COBAS® AmpliPrep/COBAS® TaqMan® HIV-1 Test
F
RESULTS
Table 1: Available Results per Tube Type and Handling Condition
Tube/Condition
Total Count
*ND Count
Data Not Available
<48 Copies/mL
>48 Copies/mL
EDTA (A)
55
13
1
11
30
PPT 2H (B)
55
14
1
9
31
PPT 6H (C)
55
18
0
9
28
PPT 1D RT (D)
55
18
0
5
32
PPT 5D 4˚C (E)
55
14
0
13
28
PPT invert (F)
55
4
1
6
44
PPT invert/spin (G)
55
15
2
6
32
*ND=Not Detected
RESULTS
Figure 2: Correlation plot of Log10 HIV-1
viral load in PPT 2H (B) to EDTA
Figure 3: Correlation plot of Log10 HIV-1
viral load in PPT 6H (C) to EDTA
Figure 4: Correlation plot of Log10 HIV-1
viral load in PPT 1D RT (D) to EDTA
Figure 5: Correlation plot of Log10 HIV-1
viral load in PPT 5D 4ºC (E) to EDTA
Figure 6: Correlation plot of Log10 HIV-1
viral load in PPT invert (F) to EDTA
Figure 7: Correlation plot of Log10 HIV-1
viral load in PPT invert/spin (G) to EDTA
Table 2: Summary of Results per PPT Tube Handling Condition vs. EDTA
Eval. Tube
Total
Subjects
Both Eval.
and EDTA
<50cp/mL
(Concordant)
Eval.
<50cp/mL EDTA
>50cp/mL
(Discordant)
Eval.
>50cp/mL EDTA
<50cp/mL
(Discordant)
Both Eval.
and EDTA
>50cp/mL
(Concordant)
Overall
Agreement1
vs. EDTA
P-value for
McNemar’s
Test2
PPT 2H (B)
53
22
1
1
29
96.2%
1.0000
PPT 6H (C)
54
23
4
1
26
90.7%
0.1797
PPT 1D RT (D)
54
20
3
4
27
87.0%
0.7055
PPT 5D 4˚C (E)
54
24
4
0
26
92.6%
0.0455
PPT invert (F)
53
10
0
14
29
73.6%
0.0002
PPT invert/spin (G)
52
19
2
4
27
88.5%
0.4142
1
Percent overall agreement = Number of subjects with concordant results/Total number of subjects x 100.
2
P-value is for McNemar’s Test for paired samples with two binary outcomes (<50 copies/mL; >50 copies/mL).
A P-value of <0.05 may indicate a statistically significant difference between the evaluation and control tubes being compared.
No adjustment for multiple comparisons was performed.
Summary OF RESULTS
1. Table 1 shows that the number of specimens with quantifiable HIV-1 VL in Tubes A, B, C, D, E, and G were comparable (28–32).
The tube that was inverted to simulate transportation had more specimens (44) with quantifiable VL.
2. Figures 2–6 show that R2 values in correlation plots of evaluation tubes A, B, C, D, and G for comparisons to EDTA control (Tube A)
ranged from 91.7% to 93.7% while R2 values for comparisons of Tubes E and F to Tube A were 87.2% and 88.6% respectively.
3.
Table 2 shows that the PPT invert (Tube F) was significantly different to the EDTA control (Tube A). The PPT 5D at 4˚C (Tube E) may
also differ from the EDTA control. The p-value for McNemar’s Test was <0.05 for both comparisons, though no adjustment for multiple comparisons was made. The overall agreement, however, for Tube E was 92.6% while that for Tube F was 73.6% indicating that inverting the tube without re-centrifugation may have a greater impact on VL results <50 copies/mL as compared to EDTA than storing the plasma for 5D at 4˚C.
Summary
This study shows that with the Roche TaqMan HIV-1 viral load assay (CAP/CTM), HIV VL overall agreement with EDTA is unaffected by:
• Storage of whole blood in PPT for 6h at room temperature before centrifugation.
• Storage of plasma overnight in situ in PPT at room temperature.
• Storage in PPT at 4˚C for up to 5 days.
Additionally:
• As compared to EDTA plasma, VLs are increased in a significant number of samples in PPT near or below the LOQ when PPT was inverted and not re-centrifuged.
CONCLUSIONS
We conclude that the BD Vacutainer® PPT™ is equivalent to EDTA
plasma for HIV-1 viral load as measured by the Roche TaqMan
HIV-1 viral load assay (CAP/CTM) if:
• Whole blood is stored in PPT for no longer than 6h at ambient
temperature.
• Plasma is stored in PPT for no longer than 1 day at ambient temperature or 5 days at 4˚C.
• PPT is re-centrifuged at 600 x g for 5 minutes in the receiving laboratory prior to aliquoting, testing, or further storage.
Acknowledgements: We would like to thank Karen Byron and Valentin Parvu (BD) and John Duncan (Roche Molecular Systems) for statistical analyses.
Roche COBAS AmpliPrep/COBAS TaqMan HIV-1 is a trademark of Roche Molecular Systems, Inc.
BD, BD Logo and all other trademarks are property of Becton, Dickinson and Company ©2010 BD VS8188