61eca454c045b82f6d6d1db4836b3670.ppt
- Количество слайдов: 19
Evaluation of Commercially Available HIV Assays to Address Alternative Screening/Diagnostic Algorithms S. Michele Owen, Ph. D. Laboratory Branch Division of HIV AIDS Prevention Centers for Disease Control and Prevention
Background • Why Evaluate Tests and Consider Alternative Algorithms? – Abundance: Multiple new FDA approved testing methods- NAT, Rapid Tests, New EIAs – Ambiguity: reduce or eliminate WB indeterminate results – Cost and Efficiency: sequential EIAs have been effectively used internationally (WHO/UNAIDS)
Objective Compare performance of commercially available HIV tests as a basis for evaluating alternative algorithms for HIV diagnostics or surveillance.
Methods- Samples • 1002 specimens from the U. S. (Boston Biomedica Inc. ) Ø Plasma centers or blood banks. • 62 non-U. S. samples (Boston Biomedica Inc) Ø World Wide Performance Panels • 205 non-U. S. samples (Cameroon Blood Bank Study) Ø Units that were reactive in one or more screening tests for HIV, HBV, HCV or Syphilis • All samples were collected, processed and stored using standard diagnostic protocols
Methods-Testing • Plasma samples: Ø randomized and blinded Ø tested by 6 EIAs, 4 rapid tests and 3* NAT- based tests. • Any sample found to be reactive by any of the above tests was subjected to Western Blot • Sample was considered to be: Ø Positive if the Western Blot was positive (current “gold standard”) Ø Negative if any sample was either § negative by all 13 tests described above or § Western Blot negative. • All tests were performed by trained laboratory personnel (Gen-Probe and Ampli. Screen NAT testing was done by individuals certified by the company to run the test)
EIAs Evaluated EIA Bio. Merieux Vironostika HIV-1 (2 nd) Components HIV-1 viral lysate Bio. Merieux Vironostika HIV-1 Plus O (2 nd) HIV-1 viral lysate, purified viral env proteins, and synthetic peptide from transmembrane epitope of HIV-1 Group O BIO-RAD Genetic Systems r. LAV (2 nd) LAV lysate and recombinant gp 41 BIO-RAD Genetic Systems 1/ 2 Peptide (2 nd) Synthetic peptides from env and pol regions of both HIV-1 and HIV– 2 BIO-RAD Genetic Systems HIV 1/2 Plus (3 rd) HIV-1 recombinant gp 160 and p 24, HIVgp 36 synthetic peptide, and HIV-1 group O synthetic oligopeptide Abbott HIVAB HIV-1/HIV-2 (r. DNA) (3 rd) Recombinant HIV-1 core and env proteins and HIV-2 env protein
Rapid, NAAT, WB Tests Rapid Components Med. Mira Reveal conserved immunodominant peptides Ora. Sure Ora. Quick peptides, gp 41, gp 36 Trinity Biotech Uni-Gold Recombigen recombinant immunodominant proteins BIO-RAD Multispot HIV-2 gp 36 peptide, HIV-1 gp 41 peptide, recombinant gp 41 Western Blot Calypte Biomedical Cambridge Biotech HIV-1 H 9/HTLV-IIIB Lysate BIO-RAD Genetic Systems HIV-1 CEM/HIV LAV Lysate NAAT Gen-Probe Procleix LTR and Pol Roche Ampliscreen Gag In house LTR
EIA Sensitivity and Specificity Relative to WB Test Sensitivity Specificity Bio-Rad Genetic Systems r. LAV n=1264 96. 9 98. 6 Bio-Rad Genetic Systems 1/2 Peptide n=1264 98. 1 98. 6 Biomeriuex Vironostika n=1264 98. 4 96. 5 Biomerieux Vironostika Plus O n=1264 98. 9 96. 6 Bio-Rad Genetic Systems 1/2 Plus O n=1264 99. 4 95. 8 Abbott HIV 1/2 r. DNA n=1264 98. 8 96. 3 Sensitivity range 96. 9 -99. 4 % Specificity range 95. 8 -98. 6%
Rapid Test Sensitivity and Specificity Relative to WB Test Sensitivity Specificity Ora. Sure Ora. Quick n=1264 98. 0 98. 9 Med. Mira Reveal G-1 n=1264 98. 4 Trinity Biotech 98. 5 Uni-Gold Recombigen HIV n=1197 99. 4 BIO-RAD Multi-Spot n=83 97. 8 97. 4 Sensitivity range 97. 4 -98. 5% Specificity range 97. 8 -99. 4%
NAT Sensitivity and Specificity Relative to WB Test Sensitivity Specificity Gen-Probe Procleix n=1264 96. 7 98. 7 Roche Ampliscreen n=1171 92. 6 96. 8 CDC In- House NAT n=1264 94. 9 98. 8 Sensitivity range 92. 6 -96. 7% Specificity range 96. 8 -98. 8%
Indeterminate Characteristics • 58 indeterminate samples Ø 5 U. S. plasma donors Ø 52 Cameroon samples Ø 1 BBI non-U. S. performance panel sample • Most had 3 or fewer EIA/Rapid positive results Ø low S/CO values on EIA Ø One or few bands on WB § p 24 >> p 66 > p 55 • 4 samples positive on multiple EIAs Ø 2/4 positive by NAT Ø 1/4 almost complete WB pattern (known O from Spain) Ø 3/4 p 24 Ag positive (BBI)
Current Diagnostic Algorithm Screening EIA Non-Reactive Repeat EIA (duplicate) +/- +/+ Pos Neg -/- WB WB Pos Ind* Negative Ind* * follow-up sample, HIV-2
Potential Simple Algorithms • • EIA Screen /NAAT Confirmation EIA Screen/Rapid Confirmation EIA Screen/Alternate EIA Confirmation Rapid Screen/Alternate Rapid Confirmation
Summary Potential Algorithms Relative to Current EIA/WB False Negative False Positive EIA/NAAT 3. 3% (25) 0% EIA/Rapid 1. 3% (9) 0% EIA/EIA 0. 7% (5) 0% Rapid/Rapid 1. 7% (12) 0%
Proposed Blood Bank Algorithm HIV EIA Repeat Reactive (BIO-RAD Plus) 713 Gen-Probe NAT 38 675 Reactive No WB or Alternate EIA Required (optional) Non-Reactive Alternate EIA 26 12 Reactive 675 HIV-1 Positive 17 Reactive Indeterminates from 58 to 9 WB 9 Vironostika Plus O 0 Non-reactive 12 Negative 2/12 False Negative IND Non-Reactive True answer for indeterminates? ? ?
Important Caveats • No follow-up samples available for discordant samples (true answer unknown) • Limited demographic or epidemiological data available • Collection, processing, and storage of samples was conducted using routine diagnostic procedures. Ø 1 freeze/thaw of specimen prior to NAT testing Ø 1 -2 freeze/thaws prior to Serological testing
Summary • Range of sensitivity observed for all tests was 92. 1% - 99. 4% • Range of specificity observed for all tests was 95. 8% - 98. 8% • Discordant results between serological and NATbased tests were observed. – True answer unknown • Most indeterminate samples – Non-U. S. – Few WB Bands – Low EIA S/CO values • 4 Indeterminate samples likely or known positive
Conclusions • All FDA approved HIV detection assays have comparable Sensitivity and Specificity Ø Lower values may be due to the stringent testing methods employed in the study • NAAT alone can not replace WB for confirmation • EIA/EIA, EIA/Rapid or Rapid/Rapid algorithms yielded better sensitivity than EIA combined with NAT • Proposed Blood Bank algorithm would likely reduce indeterminate WB results • Much work left to do to establish “best algorithm” – Seroconversion samples – Discordant/Indeterminate samples with follow-up
Acknowledgements CDC Chunfu Yang Wei Luo Chou Pau Nick Delatorre Chin-Yih Ou Tom Spira Bharat Parekh Faye Cowart Susan Kennedy Debbie Kuehl Debra Candal Donna Rudolph Tammy Barnett Silvina Masciotra Marcia Kalish Steve Mc. Dougal Industry Bio. Merieux BIO-RAD Med. Mira Trinity Biotech Gen-Probe Roche l
61eca454c045b82f6d6d1db4836b3670.ppt