Interpretation and Use of the Western Blot Assay for Serodiagnosis of Human Immunodeficiency Virus Type 1 Infections (2024)

Please note: This guideline document is obsolete and may not reflect current evidence or best practice and likely contains out-of-date information.

Reported by: Association of State and Territorial Public HealthLaboratory Directors and AIDS Program, Center for Infectious Diseases,Public Health Practice Program Office, Centers for Disease Control*The Association of State and Territorial Public Health LaboratoryDirectors (ASTPHLD) and CDC have collaborated in preparing this report.It includes a description of various interpretive criteria associatedwith the Western blot test for HIV-1, evaluates the sensitivity andspecificity of these criteria as tools for public health practice, andprovides recommendations for use of the Western blot and the manner inwhich to report results in order to provide clinicians and publichealth policy officials with useful information in their efforts toreach an accurate diagnosis for persons tested for HIV-1 infection.INTRODUCTION

The development of sensitive and specific tests for antibody to humanimmunodeficiency virus type 1 (HIV-1) progressed rapidly after thisretrovirus was identified as the cause of acquired immunodeficiencysyndrome (AIDS). These tests have been used for various purposes,including clinical diagnosis of HIV-1 infection--for symptomatic andasymptomatic patients in counseling and testing programs--forseroprevalence surveys, and for blood-donor screening.

Enzyme immunoassay (EIA) is the most widely used serologic test fordetecting antibody to HIV-1. Serum samples that are repeatedly reactivein the EIA for HIV-1 antibody are then retested with a supplemental andmore specific test, the most common of which is the Western blot (1-3).To date, only one commercial Western blot test (Du PontœPr) has beenlicensed by the Food and Drug Administration (FDA). The purpose of thisreport is to provide guidance for interpreting Western blot testresults and their use in diagnosing HIV-1 infection.THE WESTERN BLOT ASSAY

The Western blot assay is a method in which individual proteins of anHIV-1 lysate are separated according to size by polyacrylamide gelelectrophoresis. The viral proteins are then transferred ontonitrocellulose paper and reacted with the patient's serum. Any HIVantibody from the patient's serum is detected by an antihumanimmunoglobulin G (IgG) antibody conjugated with an enzyme that in thepresence of substrate will produce a colored band. Positive andnegative control serum specimens are run simultaneously to allowidentification of viral proteins.

Table 1 lists the major structural proteins coded for by the HIVgenome. Antibodies to the HIV-1 major group-specific antigen (GAG)protein p24, and its precursor p55, are the earliest detected afterinfection by Western blot and tend to decrease or become undetectablewith onset or progression of clinical symptoms (4-9). In contrast,antibodies to the envelope (ENV) precursor protein gp160 and the finalENV proteins (gp120 and gp41) can be detected in specimens fromvirtually all HIV-infected persons regardless of clinical stage (4-9).Antibodies to the polymerase (POL) gene products (p31, p51, and p66)are also commonly detected if these antigens are present on the Westernblot strips. However, in a recent study, the protein with a mobility of160 kilodaltons (kd) present in commercially available Western blotsand in viral lysate antigen preparations was identified as a multimerof the gp41 protein (10,11). Furthermore, this study presented evidencethat the reaction observed against the gp120 on certain Western blotsmay have resulted in part from a reaction with a multimeric form of thegp41. In fact, the true gp120 was shown to be absent from somecommercial Western blot antigens. When these reagents were used, serumspecimens with only gp41 antibodies produced bands at the 41-, 120-,and 160-kd positions.Interpretive Criteria

Although the overall sensitivity and specificity of the Western blotfor detection of antibodies to the various viral proteins are high,there has been substantial debate regarding the interpretive criteria.The currently licensed Du Pont Western blot test specifies that thetest result should be interpreted as positive only when the detectedbands include p24 and p31, and gp41 or gp120/160 (12) (see Table 2).Conversely, a negative Du Pont Western blot test result requires theabsence of any and all bands--not just viral-bands. All other patternsare regarded as indeterminate. This interpretation scheme maximizes thespecificity of the assay and is mainly intended for use with samplesfrom persons, such as blood donors, for whom there is usually littleclinical or virologic information available. (Donated units of bloodthat are repeatedly reactive by EIA are discarded; Western blot resultsare used to guide donor notification and deferral.) These criteria arenot ideal for all situations, especially the testing of persons atincreased risk for HIV infection, or with symptoms suggestive of thisinfection.

Alternative criteria have been proposed by various groups. ASTPHLD hasproposed that a positive test result be defined by the presence of anytwo of the following bands: p24, gp41, and gp120/160 (13). TheConsortium for Retrovirus Serology Standardization (CRSS) has defined apositive test result as the presence of either p24 or p31, plus adiffuse envelope band (i.e., gp41 or gp120/160) (14). The American RedCross has defined a positive test result as greater than or equal to 1band from each of the GAG, POL, and ENV gene-product groups (15). Thesethree groups and DuPont all agree that an indeterminate result is thepresence of any other band or bands that fail to meet the positivecriteria, and that a negative result is the absence of all bands.

The criteria for a negative Western blot interpretation specify "nobands." This interpretation is essential because some observed bandsmay reflect the presence of antibodies to HIV regulatory proteins ormay indicate partially processed or degraded viral structural proteins.Furthermore, different Western blots (commercial, as well as "in-house"preparations) and different virus-antigen preparations used to prepareWestern blots may contain different numbers and concentrations of bothviral-specific and contaminating cellular proteins that may haveunpredictable molecular weights.Evaluation of Criteria

To compare the four sets of criteria for Western blot interpretation,CDC selected 424 serum samples on the basis of the patients' clinicalstatus and EIA results only, and analyzed them using the licensed DuPont Western blot test (CDC unpublished data). The samples werescored according to each of the criteria (Table 3). For all threecategories with repeatedly reactive EIA test results, the Western blotresults demonstrate that the ASTPHLD definition gives the highestpercentage of positive and the lowest percentage of indeterminateresults. The interpretive standards that require the identification ofbands from each of the three groups of gene products tend to haveindeterminate results for some AIDS and other symptomatic patients dueto absence of antibodies to p24 (n=5) or to p31 (n=14) or absence ofboth types of antibodies (n=2). Since these patients clearly areinfected with HIV, the three-gene-product approach to Western blotinterpretation is not sensitive enough for public health or clinicalpractice.

The ASTPHLD/CDC criteria for a positive Western blot differ from theCRSS criteria in two ways: first, ASTPHLD/CDC deletes p31, a changethat does not affect the sensitivity or specificity of the criteria(Table 3), and second, ASTPHLD/CDC adds "gp41 and gp120/160," acombination not interpreted as positive with the CRSS criteria. Thislatter combination of bands represents antibody to envelopeglycoproteins only. In practice, this is a rare finding forasymptomatically infected persons, but it has been reported to bespecific for HIV-infected persons and should be included in thepositive criteria (9). However, when a Western blot test has only themultimeric form of gp41 and no true gp120 present, a serum sample wouldbe scored as positive on the basis of the presence of antibody to asingle envelope glycoprotein, gp41. HIV-1-infected persons with thisprofile have lost their antibodies to the GAG proteins and are usuallysymptomatic and do not present a diagnostic problem.

The ASTPHLD/CDC interpretive criteria for a negative result areidentical to the FDA recommendation for blood-donor reentry or theWestern blot interpretive criteria that are specified in the licensedWestern blot kit package insert.RECOMMENDATIONS

On the basis of the results described above, CDC concurred with theASTPHLD criteria and recommends their use in public health and clinicalpractice.

Laboratories should report test results as positive, indeterminate, ornegative. The Public Health Service recommends that no positive testresults be given to clients/patients until a screening test has beenrepeatedly reactive (i.e., greater than or equal to two tests) on thesame specimen and a supplemental, more specific test such as theWestern blot has been used to validate those results (3). Upon request,laboratory reports may also contain a list of the bands detected andreference to the interpretive criteria the laboratory uses. Because ofthe variability of unlicensed reagents, laboratories usingnon-FDA-licensed Western blots should compare, on a routine basis,their tests with the FDA-licensed Western blot kit usingwell-characterized serum specimens.

Clinical diagnosis and follow-up of patients is the responsibility ofthe clinical practitioner. Serologic test results are but onecontribution to a patient's data base, which contains medical history(including high-risk behavior or exposure to HIV), results of physicalexamination, and other clinical findings. Clinicians must consider thetotal profile for a client when attempting to make a diagnosis afterindeterminate Western blot results have been obtained. Accuratediagnosis for such persons can be challenging--and the challenge can becomplicated by the tendency of some clients to become distressed by theapparent "uncertainty" of their test results.

Clinical follow-up of patients with indeterminate Western blot resultsmay require many months of observation, interviewing, and testing. Mostindeterminate patterns involve p18 (also referred to as p17), p24, orp55, or any combination of these three proteins (16-18). In one studyof 390 "atypical" or indeterminate samples, 53% reacted against p24,with or without p18 or p55; 47% reacted against p18 (but not p24), withor without p55 (18).

Some indeterminate results may be obtained with serum samples frompersons who are in the process of seroconverting. A compilation of 209volunteer blood donors with GAG-only indeterminate Western blot resultswere followed for as long as 2 years (17-21). During that time, onlyfive of 134 persons who had initially reacted to p24 developedadditional bands on the Western blot test. None of the 75 persons whoinitially reacted against p18 (but not p24) developed additional bands.The five persons who did seroconvert had positive results when theirfirst follow-up samples were tested. The intervals between initial andfollow-up tests were 8 weeks (two persons), 20 weeks (two persons), and32 weeks (one person). The three longest intervals reflected delays infollow-up testing and not the actual time to seroconversion. Theseresults do not refute earlier findings that seroconversion typicallyoccurs within 3 months of infection (5,22). The importance of carefulrisk assessment for persons with indeterminate Western blot patternswas reemphasized when in one study (18) two of three people whoinitially had indeterminate results (but later seroconverted) disclosedhistories of risk behavior when they were reinterviewed duringfollow-up.

A person whose Western blot test results continue to be consistentlyindeterminate for at least 6 months--in the absence of any known riskfactors, clinical symptoms, or other findings--may be considered to benegative for antibodies to HIV-1. Such persons should be reassured thatthey are almost certainly not infected with HIV-1. However, nolarge-scale studies have been done to provide virologic data to confirmindependently the serologic findings from the studies of clients whoseWestern blot test results are consistently indeterminate. In contrast,an asymptomatic person who has an indeterminate Western blot testresult and a history of possible exposure to or symptoms compatiblewith HIV infection requires additional diagnostic follow-up. Thisshould include conducting serial Western blot testing, assessing thefunction of the individual's immune system, and eliciting thecooperation of the person's sexual and needle-sharing partners todetermine whether they are infected. Individuals with a pattern ofindeterminate Western blot test results should not donate blood orplasma for either transfusion or use in manufactured blood products.

As the HIV/AIDS epidemic continues, additional tests of higherspecificity will be needed to decrease the number of false-positivereactions and to permit correct diagnosis of HIV infection in a largerspectrum of clinical situations in which an indeterminate antibodyprofile exists. The use of new antibody tests based on antigens derivedby recombinant deoxyribonucleic acid (DNA) technology or theapplication of DNA probe technology--particularly DNA amplification bythe polymerase chain reaction (PCR)--already shows promise in this area(23).

References

1. Centers for Disease Control. Provisional public health service

inter-agency recommendations for screening donated blood and plasma forantibody to the virus causing acquired immunodeficiency syndrome. MMWR1985;34:1-5.

2. Centers for Disease Control. Public health service guidelines forcounseling and antibody testing to prevent HIV infection and AIDS. MMWR1987;36:509-15.

3. Centers for Disease Control. Update: Serologic testing for antibodyto human immunodeficiency virus. MMWR 1988;36:833-40, 845.

4. Lange JMA, Coutinho RA, Krone WJA, et al. Distinct IgG recognitionpatterns during progression of subclinical and clinical infection withlymphadenopathy associated virus/human T lymphotropic virus. Brit Med J1986;292:228-30.

5. Esteban JI, Shih JW, Tai CC, et al. Importance of Western blotanalysis in predicting infectivity of anti-HTLV-III/LAV positive blood.Lancet 1985;2:1083-6.

6. Goudsmit J, Lange JMA, Paul DA, et al. 1987. Antigenemia andantibody titers to core and envelope antigens in AIDS, AIDS-relatedcomplex, and subclinical human immunodeficiency virus infection. JInfect Dis 1987;155:558-60.

7. Lange JDA, Paul DA, Huisman HG, et al. Persistent HIV antigenemiaand decline of HIV core antibodies associated with transition to AIDS.Brit Med J 1986;293:1459-62.

8. Weber JN, Clapham PR, Weiss RA, et al. Human immunodeficiency virusinfection in two cohorts of hom*osexual men: neutralizing sera andassociation of anti-gag antibody with prognosis. Lancet 1987;i:119-21.

9. McDougal JS, Kennedy MS, Nicholson JKA, et al. Antibody response tohuman immunodeficiency virus in hom*osexual men: Relation of antibodyspecificity, titer, and isotype to clinical status, severity ofimmunodeficiency and disease progression. J Clin Invest 1987;80:316-24.10. Zolla-Pazner S, Gorney MK, Honnen WJ, et al. Reinterpretation ofHIV Western blot patterns. NEJM 1989;320:1280-1.11. Pinter A, Honnen WJ, Tilley SA, et al. Oligomeric structure ofgp41, the transmembrane protein of human immunodeficiency virus type 1.J Virology 1989;63:2674-9.12. Du Pont Diagnostics. Human immunodeficiency virus (HIV): Biotech/DuPont HIV western blot kit for detection of antibodies to HIV.Wilmington, Delaware:Du Pont Diagnostics, 1987.13. Committee on Human Retrovirus Testing. Report of Fourth ConsensusConference on Testing for Human Retroviruses. Association of State andTerritorial Public Health Laboratory Directors. Infection Control andHospital Epidemiology 1989; in press.14. The Consortium for Retrovirus Serology Standardization. Serologicaldiagnosis of human immunodeficiency virus infection by Western blottesting. JAMA 1988;260:674-9.15. Sandler SG, Dodd RY, Fang CT. Diagnostic tests for HIV infection:Serology. In: DeVita VT, Hellman S, Rosenberg SA, eds. AIDS: Etiology,Treatment, and Prevention, Second Edition. Philadelphia: J.B.Lippincott, 1988:121-6.16. Fang C, Le P, Mallory D, et al. Western blot patterns of antibodiesto human immunodeficiency virus (HIV). Transfusion 1988;27:539.17. Dock NL, Lamberson HV, O'Brien TA, et al. Evaluation of atypicalhuman immunodeficiency virus immunoblot reactivity in blood donors.Transfusion 1988;28:412-8.18. Kleinman S, Fitzpatrick L, Secord K, et al. Follow-up testing andnotification of anti-HIV Western blot atypical (indeterminant) donors.Transfusion 1988;28:280-2.19. Biberfeld G, Bredberg-Raden U, Bottiger B, et al. Blood donor serawith false-positive Western blot reactions to human immunodeficiencyvirus (letter). Lancet 1986;2:289-90.20. Courouce A, Muller J, Richard D. False-positive Western blotreactions to human immunodeficiency virus in blood donors (letter).Lancet 1986;2:921-2.21. van der Poel CL, Reesink HW, Tersmette T, et al. Blood donationsreactive for HIV in Western blot, but non-reactive in culture andrecipients of blood. Lancet 1986;2:752-3.22. Groopman JE, Caiazzo T, Thomas MA, et al. Lack of evidence ofprolonged human immunodeficiency virus infection before antibodyseroconversion. Blood 1988;71:1752-4.23. Schochetman G, Ou C-Y, Jones WK. Polymerase chain reaction. JInfect Dis 1988;158: 1154-7.Use of trade names is for identificationonly and does not imply endorsem*nt by DHHS or PHS or byASTPHLD.backœs#U.S. Government Printing Office: 1989-631-108/02016Region IV*List of participants: ASTPHLD Committee on Testing for Retroviruses:William J. Hausler, Jr., Ph.D. (Chairman), J. Mehsen Joseph, Ph.D.,Arthur F. DiSalvo, M.D., Mahadeo P. Verma, Ph.D. (President). ProgramCommittee ASTPHLD Consensus Conference: Jane P. Getchell, Dr.P.H. CDC:D. Peter Drotman, M.D., M.P.H., J. Richard George, Ph.D., GeraldSchochetman, Ph.D., Charles A. Schable, M.S., Wanda K. Jones, Dr.P.H.,Thomas L. Hearn, M.S., William Schalla, M.S.

Disclaimer All MMWR HTML documents published before January 1993 are electronic conversions from ASCII text into HTML. This conversion may have resulted in character translation or format errors in the HTML version. Users should not rely on this HTML document, but are referred to the original MMWR paper copy for the official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.

Page converted: 08/05/98