Model Performance Evaluation Program
Human Immunodeficiency Virus Type 1 (HIV-1) Antibody Testing
Report of Results
for the Performance Evaluation Survey
Conducted During January 2006
U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES
Centers for Disease Control and Prevention
Division of Laboratory Systems
Atlanta, Georgia 30333
Use of trade names and commercial sources is for identification only and does not imply endorsement by the
Centers for Disease Control and Prevention, or
U.S. Department of Health and Human Services.
Report of the January 2006 Human Immunodeficiency Virus Type 1 (HIV-1)
Antibody Performance Evaluation Sample Testing Results Provided by Participant Laboratories in the Model Performance Evaluation Program,
Centers for Disease Control and Prevention (CDC)
The production of this report was coordinated in CDC by:
Division of Laboratory Systems…………………………….Robert Martin, Dr.P.H., Director
Laboratory Practice Evaluation and Genomics Branch.…….Devery Howerton, Ph.D., Branch Chief
Model Performance Evaluation Program (MPEP)……….....G. David Cross, M.S., Co-Manager
Laurina O. Williams, Ph.D., M.P.H., Co-Manager
The material in this report was developed and prepared by:
Model Performance Evaluation Program (MPEP)………….Sandra Neal, B.S., MT(ASCP), M.S.
HIV-1 Project Coordinator
Inquiries about this report should be directed to the Model Performance Evaluation Program by
calling (404) 718-1004 or (404) 718-1006.
Table of Contents
Laboratory Demographics and Methods 12
Test methods by laboratory type 12
EIA antibody test kit manufacturers 16
EIA Ag/Ab test kit manufacturers 17
EIA false-positive and false-negative results 18
Western Blot Methods and Results 19
WB false-positive and false-negative results by test kits 21
IFA test kits, by manufacturer 22
IFA false-negative and indeterminate results 22
“Other” Test Methods and Results 24
List of Tables
Table 2: CDC Western blot (WB) testing results for the January 2006 shipment 10
Table 4: Location of MPEP participants reporting HIV-1 Ab results 14
Table 8: False-positive and indeterminate determinations for "Other" test kits 25
List of Figures
Figure 3: The combination of HIV-1 antibody tests reported by participant laboratories 15
Figure 4: Percentage and number of laboratories using EIA test kits, by manufacturer 16
Figure 5: Percentage and number of laboratories using Ag/Ab test kits, by manufacturer 17
Figure 6: Percentage and number of laboratories using WB test kits, by manufacturer 19
Figure 7: Percentage and number of laboratories using IFA test kits, by manufacturer 22
Purpose |
The purpose of this report is to present the analysis of results provided to the CDC by laboratories participating in the Model Performance Evaluation Program (MPEP) after they tested the human plasma samples shipped to them in January 2006. |
Response |
Of the 757 laboratories that were sent performance evaluation panels,
|
Contents |
This report contains the analysis of results for
|
Laboratory demographics |
Six hundred and sixty-three laboratories reported results for the January 2006 samples panel shipment. In this shipment
The 663 laboratories identified themselves as
|
EIA |
The percentage of false-negatives increased to 1.0% (35/3659) in this shipment, from 0.7% (20/2811) in the July 2005 shipment. The percentage of false-positives remained the same, 0.8%, in this shipment (6/727) as in the 2005 July shipment (11/1394). However, the overall performance is nearly the same, with 99.1% correct responses for this shipment and 99.3% in the July 2005 shipment. |
WB |
The overall performance for laboratories reporting WB results was 99.6% correct results compared to 99.3% in the July 2005 shipment.
There were 53 interpretations reported for Donor 4, the negative donor, although most laboratories do not normally perform WB testing of EIA nonreactive specimens as part of their routine algorithm for HIV antibody testing. Four laboratories that reported reactive WB reported EIA nonreactive results. |
IFA |
The overall performance for IFA in this shipment was 97.8% (due primarily to false-negative results) compared with 96.0% in the July 2005 shipment. |
Other Tests |
The overall performance of the laboratories using tests in the “other” category decreased slightly compared to the previous survey: In this shipment, laboratories reported 97.2% correct results, compared to 99.2% in the July 2005 shipment. |
Survey Samples
|
The survey samples are undiluted, defibrinated plasma obtained from individual donors who are either
HIV-1 infected (HIV-1 antibody positive):
These samples were heat-treated at 56º C for 60 minutes to inactivate blood-borne viruses including HIV-1, human T-lymphotropic virus types I and II (HTLV-I/II), and hepatitis B and C viruses.
HIV-1 uninfected (HIV-1 antibody-negative):
These samples were not heat-treated. |
Donor testing |
Before shipment, each donor sample was tested with the following:
|
Donor status |
Donors 1 (single samples) and 2 (duplicate samples) are HIV-1 antibody positive donors demonstrating factors consistent with seroconversion, including:
Donor 3: strong-positive HIV-1 (duplicate samples)
Donor 4: HIV-1 negative sample. |
Continued on next page
Challenge Samples, Continued
The CDC results were obtained after composite testing with all commercially available HIV-1 and HIV-1/HIV-2, HIV-1 WB, and IFA licensed by the Food and Drug Administration (FDA). The CDC WB interpretation is consistent with Association of Public Health Laboratories/Centers for Disease Control and Prevention (APHL/CDC) interpretations which are identical to the manufacturers’ interpretive criteria for WB results.
Laboratory Interpretation space is to be completed by participant laboratories to facilitate comparison of their results with CDC results.
Continued on next page
Challenge Samples, Continued
Western blot (WB) results are based on the band intensity of ≥1+ staining.
The CDC interpretation is consistent with APHL/CDC and the manufacturers’ criteria for the interpretation of WB results.
Cambridge Biotech/Calypte Biomedical.
Overall results |
Table 3 summarizes the results grouped by test type: EIA, WB, IFA, and Other.
Table 3:
Results Summary
Positive
Donors
Negative
Donor
Method
Total
# of laboratories
Total
# of results
Positive
I*
False-
negative (% false negative)
Negative
I
False-positive
(% false positive)
Overall
Performance (TP+TN/total # results)†
EIA
656
4386
3624
nv‡
35
(1.0%)
721
nv
6
(0.8%)
99.1%
WB
216
1132
1027
51
1
(0.09%)
49
2
2
(3.8%)
99.6%§
IFA
32
179
142
11
4
(2.5%)
22
0
0
97.8%§
Other¶
31
182
150
2
5
(3.2%)
25
0
0
97.3% † TP, true positives; TN, true negatives. ‡ nv, not valid. Indeterminate is not a valid interpretation for reporting final EIA results. § When calculating overall performance, indeterminate interpretations are considered to be correct for HIV-1 antibody-positive donors, and incorrect for HIV-1 antibody-negative donors. ¶ “Other” test methods refer to test types other than EIA, WB or IFA, such as line or strip assays.
|
Test methods by laboratory type |
Figure 1 shows laboratory types and the test methods used. Some laboratories reported using more than one method. Therefore, the sum is greater than the total number of laboratories.
The “n” value in all figures refers to the number of laboratories reporting results, not the number of methods or tests kits used.
Figure 1: Number of HIV-1 participants reporting EIA, WB, IFA, and "Other" results, by laboratory type
|
*“Other” laboratories include university-associated research centers, university clinics, Federal government facilities, STD clinics, etc.
Continued on next page
Laboratory Demographics and Methods, Continued
U. S. laboratories |
Figure 2 shows the number and location of MPEP laboratories in the U.S. and U.S. Territories submitting results for the January 2006 shipment.
Figure 2: Geographic distribution of MPEP participant laboratories in the United States and U.S. Territories
|
Continued on next page
Laboratory Demographics and Methods, Continued
All MPEP laboratories |
Including the United States and U. S. territories, MPEP participants are located in 73 countries. Table 4: Location of MPEP participants reporting HIV-1 Ab results N=663 |
||||||
|
Country |
Number of Laboratories |
Country |
Number of Laboratories |
Country |
Number of Laboratories |
|
|
Algeria |
1 |
Guyana |
1 |
Slovenia |
2 |
|
|
Argentina |
1 |
Honduras |
2 |
South Africa |
3 |
|
|
Australia |
6 |
Hong Kong |
2 |
South Korea |
2 |
|
|
Austria |
3 |
Hungary |
1 |
Spain |
3 |
|
|
Bahamas |
1 |
India |
6 |
Sri Lanka |
5 |
|
|
Barbados |
1 |
Ireland |
1 |
St. Kitts/Nevis |
1 |
|
|
Belgium |
2 |
Israel |
3 |
Suriname |
3 |
|
|
Bolivia |
1 |
Italy |
1 |
Switzerland |
1 |
|
|
Botswana |
3 |
Jamaica |
1 |
Taiwan |
2 |
|
|
Brazil |
3 |
Japan |
1 |
Tanzania |
3 |
|
|
Cameroon |
1 |
Kazakhstan |
6 |
Thailand |
5 |
|
|
Canada |
15 |
Kenya |
2 |
Trinidad |
2 |
|
|
Chile |
1 |
Kyrgyzstan |
3 |
Turkey |
1 |
|
|
Columbia |
2 |
Malaysia |
2 |
US Territory |
16 |
|
|
Costa Rica |
1 |
Mali |
1 |
Uganda, East Africa |
3 |
|
|
Cote d'Ivoire |
2 |
Malta |
1 |
United Arab Emirates |
1 |
|
|
Croatia |
1 |
Mexico |
1 |
United Kingdom |
1 |
|
|
Denmark |
3 |
Myanmar |
1 |
United States |
489 |
|
|
Dominican Republic |
2 |
Peru |
2 |
Uruguay |
1 |
|
|
Ecuador |
1 |
Philippines |
2 |
Uzbekistan |
10 |
|
|
England |
1 |
Portugal |
1 |
Venezuela |
3 |
|
|
Eritrea |
1 |
Republic of Singapore |
1 |
Vietnam |
1 |
|
|
Ethiopia |
1 |
Romania |
1 |
Zimbabwe |
1 |
|
|
Germany |
2 |
Saudi Arabia |
1 |
|
|
|
|
Ghana |
3 |
Scotland |
1 |
|
|
Continued on next page
Laboratory Demographics and Methods, Continued
Test methods |
The test combinations used by the MPEP laboratories are shown in Figure 3.
Figure 3: The combination of HIV-1 antibody tests reported by participant laboratories |
Test Summary |
Of the 663 laboratories reporting results;
|
Introduction |
Of the 663 laboratories reporting results, 656 (98.9%) performed EIA testing using test kits that are designed to detect the presence of HIV-1 and/or HIV-2 antibodies (2nd and 3rd generation tests). Many laboratories outside the U.S. use test kits that detect the presence of these antibodies to HIV and the presence of p24 antigen (HIV Ag/Ab test kits [4th generation assays]). These test kits are not yet FDA approved.
Participant laboratories located in the U.S. reported using seven different EIA test kits for detection of antibodies in plasma and serum. These test kits are listed in Figure 4 below, and include:
MPEP participant laboratories outside the U.S. reported using 30 different EIA test kits for detection of antibodies to the HIV-1 and/or HIV-2 virus and HIV p24 antigen. |
EIA antibody test kit manufacturers |
Figure 4 shows the percentage and number of laboratories using a particular EIA antibody-only test kit. The numbers at the end of the bars show the number of laboratories using each test kit.
Figure 4: Percentage and number of laboratories using EIA test kits, by manufacturer |
* FDA approved EIA test kits.
† Other EIA test kits for which no manufacturers’ code is provided in the result booklet.
Continued on next page
EIA Methods and Results, Continued
EIA Ag/Ab test kit manufacturers |
Figure 5 shows the percentage and number of laboratories that used antigen-antibody test kits. These test kits are not FDA approved, and therefore are not used by U. S. participants.
Figure 5: Percentage and number of laboratories using Ag/Ab test kits, by manufacturer
|
Other EIA test kits |
Laboratories reported using 15 other EIA kits for which no manufacturers’ codes are listed in the results booklet or online. Some of these EIA test kit manufacturers are listed below. The number in parenthesis is the number of laboratories that reported using these kits.
|
Continued on next page
EIA Methods and Results, Continued
EIA false-positive and false-negative results |
Table 5: False-positive and false-negative EIA results, reported by participant laboratories, by kit manufacturer§
* Bio-Rad Genetic Systems HIV-1 Antigen and Vironostika HIV-1 Antigen* tests detect the presence of HIV antigen. The survey samples do not contain HIV antigens. † Bio-Rad Genetic Systems HIV-2 EIA detects the presence of HIV-2 antibody. The challenge samples in this survey only contain HIV-1 antibodies.
§Note: Only false-positive and false-negative results are contained in this table. Those test kits for which false-positive and/or false-negative results were not reported are not included. |
EIA results by donor |
Incorrect results for donors for all reported EIA tests are as follows;
|
EIA comments |
The percentage of false-negatives increased to 1.0% (35/3659) in this shipment, from 0.7% (20/2811) in the July 2005 shipment. The percentage of false-positives remained the same, 0.8%, in this shipment (6/727) as in the 2005 July shipment (11/1394). However, the overall performance is nearly the same, with 99.1% correct responses for this shipment and 99.3% in the July 2005 shipment. |
Continued on next page
Introduction |
Of the 663 laboratories reporting test results in this survey, 216 (32.6%) performed WB testing using 7 different commercially manufactured WB test kits and one in-house preparation. Five laboratories use OraSure HIV-1 to test these survey samples, however OraSure HIV-1 Western Blot is not FDA approved for testing serum or plasma.
In the U.S., two FDA approved WB kits are available for testing serum or plasma. |
WB test kits |
The WB test kits used by MPEP participant laboratories are shown below. The numbers at the end of the bars show the number of laboratories using that test kit.
Figure 6: Percentage and number of laboratories using WB test kits, by manufacturer
|
* FDA approved WB test kits.
† Orasure HIV-1 Western Blot Kit is only FDA approved for oral fluid.
‡ Other, WB tests for which no manufacturers’ codes are included in the result booklet.
Continued on next page
Western Blot Methods and Results, Continued
WB interpretative criteria |
Of the 216 laboratories reporting WB test results, 214 (99.1%) indicated which WB criteria they used to interpret tests results. Most laboratories used the Association of Public Health Laboratories/Centers for Disease Control and Prevention (APHL/CDC) WB interpretive criteria.
The number of laboratories using specific criteria are as follows:
|
WB interpretive guidelines |
The WB interpretive guidelines recommended by the two FDA-licensed WB kit manufacturers are identical to the APHL/CDC HIV-1 WB interpretive criteria. According to these guidelines:
|
WB band patterns |
The WB bands for the donor samples in this survey, as determined in pre-shipment testing with two FDA-licensed WB test kits, are shown in Table 2, page 10. |
WB results by donor |
The results by donor are
|
Continued on next page
Western Blot Methods and Results, Continued
WB false-positive and false-negative results by test kits |
Table 6: False-positive, false-negative, and indeterminate interpretations for Western blot test, by manufacturer§
|
||||||||||
|
Manufacturer |
|
Negative Donor |
Positive Donors |
|
||||||
|
Total # of Results |
Negative |
False-positive |
I* |
Positive |
False-negative |
I |
|
|||
|
Bio-Rad Genetic Systems HIV-1 |
716 |
27 |
1 |
0 |
649 |
1 |
38 |
|
||
|
Bio-Rad New LAV Blot I |
75 |
5 |
0 |
1 |
67 |
0 |
2 |
|
||
|
Genelabs HIV Blot 2.2 |
126 |
5 |
1 |
1 |
114 |
0 |
5 |
|
||
|
*I, Indeterminates
§Note: Only false-positive and false-negative results are contained in this table. Those test kits for which false-positive and/or false-negative results were not reported are not included. |
|
|
WB comments |
There were 53 interpretations reported for Donor 4, the negative donor, although most laboratories do not normally perform WB testing of EIA nonreactive specimens as part of there routine algorithm for HIV antibody testing. Four laboratories reported reactive WB for the HIV-negative donor although the EIA results were reported as nonreactive. Laboratories are reminded to handle and report the MPEP challenge samples in the same manner as patient specimens.
Several laboratories reported indeterminant results for the HIV-1 positive donors although the correct bands were present and of sufficient intensity to have been reported reactive.
The overall performance for laboratories reporting WB results was 99.6% correct answers compared to 99.3% in the July 2005 shipment. |
IFA Introduction |
Of the 663 laboratories reporting results, 32 (4.8%) performed IFA tests. Most (90.6%) used the only commercially available IFA test kit, Sanochemia Fluorognost IFA. However, a few laboratories used in-house or other methods. (Figure 7). |
IFA test kits, by manufacturer |
The IFA test kits reported are shown in Figure 7. The numbers at the end of the bars are the number of laboratories using that test kit.
Figure 7: Percentage and number of laboratories using IFA test kits, by manufacturer |
*Other-IFA: Manufacturers for which no codes are included in the result booklet.
IFA false-negative and indeterminate results |
Table 7: False-negative and indeterminate IFA results reported by participant laboratories, by kit manufacturer
|
|||||||||
|
|
|
Negative Donors |
Positive Donors |
|
|||||
|
Methods/Manufacturer |
Total # of Results |
Negative |
False-positive |
I* |
Positive |
False-negative |
I |
|
|
|
In House |
17 |
2 |
0 |
0 |
12 |
0 |
3 |
|
|
|
Other† |
6 |
1 |
0 |
0 |
4 |
0 |
1 |
|
|
|
Sanochemia Fluorognost |
156 |
19 |
0 |
0 |
126 |
4 |
7 |
|
*I, indeterminates
†Other, Manufacturer for which no code is included in the result booklet.
Continued on next page
IFA Methods and Results, Continued
IFA results by donor |
Incorrect results for each donor are as follows;
|
Comments |
The overall performance for IFA in this is 97.8%, due primarily to false-negative results. |
Introduction |
Thirty-one (4.7%) of the 663 laboratories reported using “Other” tests. Tests in this category are based on line immunoblot assay technology or particle agglutination. Laboratories reported their results in the “Other” test type section of the result form since the form is not designed to capture these types of results. |
“Other” test kits, by manufacturer |
Figure 8 shows the “Other” test kits used by laboratories participating in this survey. The numbers at the end of the bars show the number of laboratories using that test kit.
Figure 8: Percentage and number of "Other" HIV-1 antibody test kits reported by participants, by manufacturer
|
*Other includes tests for which no manufacturers’ codes are included in the result booklet.
Other “other” test kits |
Test kits for which no manufacturers’ code is included in the result booklet or the test kits were too new to be included are listed below. The number in parentheses is the number of laboratories using that test kit.
|
Continued on next page
“Other” Test Methods and Results, Continued
Results by donor |
The results by donor are as follows;
|
“Other” results |
Table 8: False-positive and indeterminate determinations for "Other" test kits* |
||||||||
|
|
|
Negative Donor |
Positive Donors |
|||||
|
Methods/Manufacturer |
Total # of Results |
Negative |
False-positive |
I* |
Positive |
False- negative |
I |
|
|
In-House |
12 |
2 |
0 |
0 |
5 |
5 |
0 |
|
|
Innogenetics INNO-LIA |
119 |
14 |
0 |
0 |
103 |
0 |
2 |
*I, Indeterminates
*Note: Only false-positive and false-negative results are contained in this table. Those test kits for which false-positive and/or false-negative results were not reported are not included.
Comments |
The overall performance of the laboratories using tests in the “other” category decreased slightly compared to the previous survey: In this shipment, laboratories reported 97.2% correct results, compared to 99.2% in the July 2005 shipment. |
Glossary of Terms |
EIA: Enzyme immunoassay, sometimes referred to as ELISA (enzyme-linked immunosorbent assay), is a screening test to detect antibodies to HIV and other viruses and some bacteria.
Evaluation: A process for determining how well health systems, either public or private, deliver or improve services and for demonstrating the results of resource investments.
False-negative: A negative test result for a sample that is actually positive.
False-positive: A positive test result for a sample that is actually negative.
HIV Antibody: Specific immunoglobulin produced the body’s immune system in response to the HIV virus.
HIV Antigen: Specific immune-recognizable proteins, such as p24, which cause the production of antibodies.
HIV test: More correctly referred to as an HIV antibody test, this test detects antibodies to HIV, rather than detecting the virus itself.
IFA test: Indirect immunofluorescence assay, a confirmatory test for the detection of antibodies to Human Immunodeficiency Virus Type I (HIV-1) in human serum or plasma.
Indeterminate test result: A possible result for IFA, WB or “Other” test that might represent a recent HIV infection, but does not meet the criteria for positive.
Positive test: For HIV, a specimen that is reactive on a screening test such as an EIA test and confirmed positive on Western blot or other confirmatory test indicating that the donor is infected with HIV.
Seroconversion: Initial development of detectable antibodies specific to a particular antigen; the change of a serologic test result from negative to positive as a result of antibodies induced by the introduction of antigens or microorganisms into the host.
Western blot: For HIV, a laboratory test that detects antibodies specific for components of the HIV virus. It is chiefly used to confirm the presence of HIV antibodies in specimens found reactive using a screening test such as the EIA test.
|
HIV-1 Antibody Testing for January 2006
File Type | application/msword |
File Title | Overview |
Author | swn0 |
Last Modified By | bmm1 |
File Modified | 2007-01-10 |
File Created | 2006-11-17 |