Validation of HCV-RNA detection in small test pools on cadaveric samples

1
Validation of HCV-RNA detection in small test
pools on cadaveric samples

• Marco Koppelman, Theo Cuypers, Mirjam de Waal,
  
• Maarten Koot, Arlinke Bokhorst, Noor Holsboer.
• Sanquin Diagnostic Services, AmsterdamBIS-
• Foundation, Leiden, The Netherlands

2
Tests on blood borne infections for tissue release

• Bone, ossicle
• Anti-HIV-1/2
• HIV p24 antigen
• HIV-1 DNA
• Anti-HTLV-I/II
• HBsAg
• Anti-HBc total
• Anti-HCV
• HCV-RNA
• IgG anti-CMV
• TPPA

• Tissue cornea, skin, cardiac valve
• Anti-HIV-1/2
• HIV p24 antigen
• Anti-HTLV-I/II
• HBsAg
• Anti-HBc total
• Anti-HCV (pos. HCV-RNA)
• TPPA

3
HCV-RNA Bone-tissue and confirmation samples
positive in serology

• NucliSens extractor (BioMerieux), Amplicor HCV
  test vs 2.0
• 1 ml plasma input 95 hit-rate 60 geq/ml (8.7
  IU/ml) 15 invalid results with neg. IC signal.
• 0.1 ml plasma input with 0.9 ml negative human
  plasma 95 hit-rate 600 geq/ml (87
  IU/ml)

total number of tissues 5188
4
Tests on blood borne infections for tissue release

• Bone, ossicle
• Anti-HIV-1/2
• HIV p24 antigen
• HIV-1 DNA
• Anti-HTLV-I/II
• HBsAg
• Anti-HBc total
• Anti-HCV
• HCV-RNA
• IgG anti-CMV
• TPPA

• Tissue cornea, skin, cardiac valve
• Anti-HIV-1/2
• HIV p24 antigen
• Anti-HTLV-I/II
• HBsAg
• Anti-HBc total (pos. anti-HBs)
• Anti-HCV (pos. HCV-RNA)
• HCV-RNA test pools
• TPPA

5
Testing tissue donors in test pools of 4 for
HCV-RNA NAT

• NucliSens extractor (BioMerieux), AmpliScreen HCV
  test vs 2.0. NAT Blood screening for HCV-RNA and
  HIV-RNA.
• Tissues testing in test pools of 4 samples.
  Input per donor sample 50 ul EDTA plasma.
  Filling up to 1 ml with 0.8 ml negative human
  plasma.
• Sensitivity per donation
• 99 hit rate 1600 geq/ml (880-4320 95 CI)
  313 IU/ml
• 95 hit rate 840 geq/ml (520-1760 95 CI)
  165 IU/ml
• 50 hit rate 160 geq/ml (120-240 95 CI)
  31 IU/ml
• Calculated on dilution factor based on
  analytical validation Jongerius et al. 42,
  792-797 (2002).

6
Validation of testing tissue donors in test pools
of 4 for HCV-RNA NAT.

• Robustness.320 non-selected samples from tissue
  donors of period 6 weeksComposition of 80 test
  pools - invalid results, traceability sample(s)
  responsible for invalid result - pools with
  inhibited IC-signal, distribution of IC
  signals
• Sensitivity. -Spiking of 26 test pools with 300
  geq/ml HCV-RNA and analyses test pools with and
  without the HCV- RNA spike.
• 300 geq per extraction load run control in
  standard HCV-RNA blood screening

7
Acceptance criteria testing tissue donors in test
pools of 4 for HCV-RNA NAT.

• Robustness
• Percentage invalid results in test pools lt 5 .
• Invalid test pools breakdown into one invalid
  donation and three donations with valid result.
• Total invalid results on tissue level lt 2 .
• gt 90 test pools OD/CO gt 12.5, saturated signal.
• Sensitivity
• Non-spiked test pools negative for HCV-RNA,
  spiked test pools (300 geq per pool) positive.

8
Results test pools of 4 tissue donors in HCV-RNA
NAT

• 78 test pools negative for HCV-RNA.
• 2 test pools invalid results, with negative
  IC-signal
• 2 test pools with inhibited IC-signal, one OD/CO
  5, one 12.

9
Retest and breakdown invalid test pools.

• Two invalid test pools composed once more.
  Re-test one test pool invalid for HCV-RNA the
  other one inhibited.
• Breakdown of the two invalid test pools in
  individual donations, one donor responsible for
  invalid result.

10
Wild type and IC signal (OD/CO) test pools spiked
with 300 geq/ml HCV-RNA
11
Conclusions

• Testing in pools of 4 with 50 ul of plasma per
  tissue donor yields an acceptable level of
  invalid results of 2/320 (0.6 ) for the HCV-RNA
  NAT test in an unselected sample cohort.
• Test pools of 4 composed in this way yield a
  robust test, 95 of the pools have a saturated
  signal for the IC.
• Based on the analytical validation, this test has
  a 95 success rate for a positive result of 840
  geq/ml (165 IU/ml) for samples tested in this
  dilution.
• Spiking experiment show that the level of 6000
  geq/ml (1175 IU/ml) for the individual sample is
  guaranteed. Nearly all spiked samples have
  saturated wild-type signals.