Journal of Clinical Virology 52S (2011) S23S27

Contents lists available at SciVerse ScienceDirect

Journal of Clinical Virology

journal homepage: www.elsevier.com/locate/jcv

Comparison of alternative interpretive criteria for the HIV-1 Western blot and results of the Multispot HIV-1/HIV-2 Rapid Test for classifying HIV-1 and HIV-2 infections
Muazzam Nasrullah a, , Steven F. Ethridge a , Kevin P. Delaney a , Laura G. Wesolowski a , Timothy C. Granade a , Joe Schwendemann b , Robert D. Boromisa b , James D. Heffelnger a , S. Michele Owen a , Bernard M. Branson a
a b

Division of HIV/AIDS Prevention, National Center for HIV, Hepatitis, STD and TB Prevention, Centers for Disease Control and Prevention, Mailstop E46, Atlanta, GA 30333, USA Wadworth Center, New York State Department of Health, Albany, NY, USA

a r t i c l e

i n f o

a b s t r a c t
Background: HIV-1 Western blot (WB) may be positive in specimens from persons with HIV-2 infection due to cross-reactive antibodies. HIV-1 and HIV-2 infections may be identied using assays designed to differentiate HIV-1 and HIV-2 antibody reactivity. Objectives: To evaluate the ability of the current CDC WB criteria, alternative more stringent HIV-1 WB criteria (2 env plus one gag or pol band) and the Multispot HIV-1/HIV-2 Rapid Test to accurately differentiate HIV-1 and HIV-2 infections. Study design: Two panels were used to determine the ability of each method to properly classify HIV-1 and HIV-2 infections: an HIV-2 panel (n = 114) determined to be HIV-2 antibody-positive by both Multispot and by a validated HIV-2 WB, and 2135 HIV-1/HIV-2 immunoassay repeatedly reactive (IA-RR) specimens from the New York State Department of Health Laboratory (NYS). Results: By CDC WB criteria, 53 (46.5%) HIV-2 panel specimens were HIV-1 WB positive, 60 (52.6%) were indeterminate, and 1 (0.9%) was negative; the alternative WB criteria re-classied 75.5% of the positives as indeterminate. Among 2135 NYS IA-RR specimens, the alternative WB criteria increased the proportion of indeterminates by 0.8%. Only 6 (0.3%) of the NYS specimens were determined to be HIV-2 infections; all 6 were classied either as HIV-1 positive or indeterminate by both WB criteria, but were classied as HIV-2 (n = 4) or HIV-1/2 undifferentiated (n = 2) by Multispot. Conclusions: The alternative WB criteria classied most of the HIV-2 specimens that were HIV-1 positive by CDC criteria as indeterminate, but also slightly increased the proportion of HIV-1 specimens classied as indeterminate. The WB indeterminate specimens would require further testing or follow-up to resolve the infection status, whereas Multispot directly distinguished HIV-1 from HIV-2. Published by Elsevier B.V.

Keywords: HIV-1 HIV-2 HIV subtype differentiation HIV-1/HIV-2 Rapid Test Supplemental HIV testing HIV-1 Western blot criteria

1. Background Since 1989,1 laboratories in the United States have followed the HIV-1 Western blot (WB) interpretive criteria established by the Centers for Disease Control (CDC) and the Association of State and Territorial Public Health Laboratory Directors. This standard requires the presence of at least two of the three major viral protein bands, p24, gp41, or gp120/160, to be considered positive for HIV-1 antibodies. However, several other interpretative criteria for the HIV-1WB have been promulgated. The French Autorit de Sant requires both the gp120 and gp160 bands (but not gp41) and the

Corresponding author. Tel.: +1 404 639 3271; fax: +1 404 639 8640. E-mail address: snasrullah@cdc.gov (M. Nasrullah). 1386-6532/$ see front matter. Published by Elsevier B.V. doi:10.1016/j.jcv.2011.09.020

presence of one gag (p55, p24, p17) or pol (p65, p51 and p31) band.1 The World Health Organization requires at least two env bands, but does not require a p24 band.2 Although HIV-2 infections are rare in the United States,36 WB results for some persons with HIV-2 infection may be classied as positive for HIV-1 by the CDC interpretive criteria due to antibody cross-reactivity to the proteins of the two viruses, particularly those of the gag region.3,7,8 Correct identication of HIV-2 infection is important because therapeutic monitoring and treatment regimens differ for HIV-1 and HIV-2 infected individuals. Quantitative viral load assays detect HIV-1 RNA but not that of HIV-2, and some antiretrovirals, including non-nucleoside reverse transcriptase inhibitors and some protease inhibitors, are not effective against HIV-2.9 Traditionally, HIV-2 testing has been recommended for persons with risk factors for HIV-2, such as a link to West Africa;4 a clinical suspicion of AIDS

S24

M. Nasrullah et al. / Journal of Clinical Virology 52S (2011) S23S27

in the absence of a positive test for antibodies to HIV-1; an HIV-1 WB with unusual indeterminate band patterns, such as gag p55, p24 or p17 plus pol bands p66, p51 reverse transcriptase or p31/32 (integrase) without env bands.1012 However, an assay has been approved by the Food and Drug Administration (FDA)13 that is capable of differentiating HIV-1 from HIV-2 infections. This assay, the Multispot HIV-1/HIV-2 Rapid Test (Bio-Rad Laboratories, Redmond, WA),14 differentiates HIV-1 from HIV-2 antibody reactivity by using a synthetic gp36 peptide glycoprotein to detect HIV-2 and a synthetic gp41peptide and a recombinant gp41 glycoprotein to detect HIV-1. This test has been shown to perform with high sensitivity and specicity15 and correctly identies HIV-2 specimens that were initially HIV-1 positive by WB.3,16 2. Objectives The current study was conducted to evaluate the ability of the current CDC WB criteria, alternative more stringent HIV-1 WB criteria, and the Multispot HIV-1/HIV-2 Rapid Test to accurately differentiate HIV-1 and HIV-2 infections. 3. Study design HIV-2 specimens from HIV-2 infected persons (n = 114) were obtained from BocaBiolistics (Coconut Creek, FL) (n = 91) and from a CDC eld study conducted in the Republic of Cte dIvoire (n = 23). These specimens were classied as HIV-2 antibody-positive using a validated research use only (RUO) HIV-2 WB (MP Biomedicals Asia Pacic Ptv Ltd, Singapore) and all were positive for HIV-2 by Multispot. The HIV-2 panel was tested with the GS HIV-1 WB (BioRad Laboratories, Redmond, WA). The Diagnostic HIV Laboratory at the Wadsworth Center, New York State Department of Health (NYS) provided CDC with electronic datasets and completed laboratory forms that contained Multispot and GS HIV-1 WB results for 2135 immunoassay repeatedly reactive (IA-RR) specimens from persons who were tested for HIV in NYS publicly funded HIV test sites during 20022006. From 2002 to 2005, specimens were considered repeatedly reactive if reactive by the two following IAs which were run in parallel: Vironostika HIV-1 Microelisa System (BioMeriux, Durham NC) and the GS rLAV IA (Bio-Rad Laboratories, Redmond, WA). In the case of discordant results, the Vironostika was run again and served as the tie breaker. From November 2005 to 2006, specimens were screened with only the GS HIV 1/2 Plus O IA (Bio-Rad Laboratories). All data for these specimens were routinely collected as part of NYS HIV diagnostic testing. Neither the electronic data sets nor the laboratory forms used for this study contained any patient identiers. If results were discordant between the Multispot and WB, the stored specimens at NYS were retrieved for retesting by an HIV1 nucleic acid test (NAT), if sufcient volume was available or a data search was performed to determine whether there was a follow-up specimen on which testing had been performed. The single-use cartridge Multispot HIV-1/HIV-2 Rapid Test detects and differentiates circulating antibodies to HIV-1 and HIV-2 in serum and plasma employing recombinant HIV-1 env glycoproteins, HIV-1 peptides, and HIV-2 peptides.14 Test samples are diluted in specimen diluent, and dispensed onto the prelter followed by addition of the conjugate. The conjugate binds to the human antibodyantigen complexes that are immobilized in the spots on the cartridge membrane and any unbound conjugate is removed by a wash step. A purple color develops on the test spots in proportion to the amount of antibodies against HIV-1 and/or HIV2 that have been bound to the antigen-coated microparticles and detected by the conjugate. Color development is stopped by the addition of stop solution and a purple color will develop on the

procedural control spot when the test has been performed correctly. The cartridge membrane has 4 spots for interpretation: 2 for gp41 peptides, 1 for gp36, and 1 for procedural control that are examined visually after 15 min. In NYS specimens, the Multispot results that were positive for both HIV-1 and HIV-2 were classied as undifferentiated because results of the dilution protocol recommended by the manufacturer to resolve undifferentiated specimens were not available.14 Specimens with HIV-2 or undifferentiated HIV-1/HIV-2 results using Multispot were tested with a RUO HIV-2 WB and/or a laboratorydeveloped RUO HIV-2 DNA NAT. For both specimen sets, the presence of each HIV-1 WB band was determined according to the manufacturers instructions.12 Alternative WB criteria for a positive result (which required the presence of 2 of 3 env bands [gp120, gp160 or gp41] plus one gag [p55, p24, p17] or pol [p65, p51, p31] band) were applied to the HIV1 WB results. The number and proportion of specimens classied as positive, indeterminate, and negative for HIV-1were calculated for both specimen sets using the CDC and the alternative WB criteria. Multispot results were compared to both WB criteria. Because all specimens used in this study were unlinked from personal identiers, this study was determined to be research not involving identiable human subjects by the CDC. 4. Results 4.1. HIV-2 panel The proportion of HIV-2 positive panel specimens for which each HIV-1 WB band was positive or weakly reactive is presented in Table 1. The p24 band was present in the largest proportion of specimens (93.9%), followed by p40 (88.6%) and p31 (83.3%). The env bands were present less frequently: gp160 (48.3%); gp120 (10.5%) and gp41 (1.8%). Based on the CDC WB criteria, 53 (46.5%) of the HIV-2 specimens were classied as HIV-1 positive, 60 (52.6%) as indeterminate, and 1 (0.9%) as negative (Table 2). The alternative WB criteria classied 13 specimens (11.4%) as HIV-1 positive, 100 (87.7%) as indeterminate, including 40 of the 53 (75.5%) HIV-2 specimens that were HIV-1 WB positive by the CDC criteria. Both the p24 and the gp160 bands were present in all of the 53 HIV-2 specimens that were classied as HIV-1 positive by the CDC WB criteria. Of the 13 specimens that remained HIV-1 WB positive by both criteria, 11 had the p24, gp120 and gp160 bands present, and 2 had p24, gp41 and gp160 bands. 4.2. Specimens from New York State Using the CDC WB criteria, 1790 (83.8%) of the 2135 IA-RR specimens from NYS were positive for HIV-1, 96 (4.5%) were indeterminate, and 249 (11.7%) were negative (Table 3). The alternative WB criteria classied 1773 (83.0%) specimens as HIV-1 positive (including 5 specimens with gp120, gp160 and at least one gag or pol band that were classied as indeterminate by the CDC criteria), 113 (5.3%) as indeterminate, including 17 specimens that were positive by the CDC criteria, and 249 (11.7%) as negative. Of the 6 (0.3%) specimens that were conrmed HIV-2 by HIV-2 WB and/or HIV-2 DNA NAT testing, 2 were HIV-1 WB positive by both criteria, 3 were indeterminate by both, and 1 was HIV-1 positive by the CDC criteria but indeterminate by the alternative criteria. Of 1790 specimens that were WB positive by the CDC criteria, Multispot was positive (for HIV-1, HIV-2, or undifferentiated) in 1788 specimens (Table 3). Of 96 specimens that were indeterminate by the CDC WB criteria, Multispot was positive in 36 and negative in 60 specimens. Of 249 WB-negatives, Multispot was positive in 20 and negative in 229 specimens. Multispot correctly

M. Nasrullah et al. / Journal of Clinical Virology 52S (2011) S23S27

S25

Table 1 Percentage of specimens with each HIV-1 Western blot band in 114 specimens collected from persons infected with HIV-2 and 1761 specimens positive for HIV-1 by Western blot and Multispot HIV-1/HIV-2 differentiation assay. p17 HIV-2 (n = 114) Present Present but weak Absent HIV-1 (n = 1761) Present Present but weak Absent 18.4 14.9 66.7 78.8 6.3 14.9 p24 93.9 4.4 1.8 91.4 7.3 1.4 p31 83.3 7.0 9.7 95.2 2.0 2.8 p40 88.6 9.7 1.8 gp41 1.8 0.9 97.4 97.4 1.7 0.9 p51 74.6 17.5 7.9 97.2 1.4 1.4 p55 73.7 17.5 8.8 93.3 1.3 5.4 p66 29.8 10.5 59.7 95.0 2.8 2.2 gp120 10.5 10.5 79.0 98.6 0.6 0.8 gp160 48.3 22.8 29.0 99.9 0.1 0.0

Table 2 Comparison of two HIV-1 Western blot interpretive criteria applied to specimens collected from 114 persons known to be infected with HIV-2.a Current CDC HIV-1 WB criteriaa Alternative HIV-1 WB criteriab , n (%) Negative Negative Indeterminate Positive Total 1 (0.9) 0 (0.0) 0 (0.0) 1 (0.9) Indeterminate 0 (0.0) 60 (52.6) 40 (35.1) 100 (87.7) Positive 0 (0.0) 0 (0.0) 13 (11.4) 13 (11.4) Total 1 (0.9) 60 (52.6) 53 (46.5) 114 (100.0)

a CDC HIV-1 WB interpretive criteria: Positive dened as at least 2 of p24, gp41 or gp120/gp160 bands; indeterminate dened as presence of any other viral bands not meeting the criteria for positive; negative dened as absence of any bands. b Alternate HIV-1 WB interpretive criteria: Positive requires presence of at least 2 of the 3 env bands (gp41, gp120, and/or gp160) and one gag or pol band; indeterminate dened as presence of any other viral bands not meeting criteria for positive; negative dened as absence of any bands.

identied as HIV-1 17 specimens with indeterminate (n = 15) or negative (n = 2) WB results by both WB criteria that were determined to be HIV-1 positive based on follow-up or HIV-1 NAT. Another 34 (1.6%) specimens that were HIV-1 positive on Multispot but indeterminate (n = 16) or negative (n = 18) by both WB criteria had no HIV-1 NAT or follow-up results. Two specimens that were positive by both WB criteria were negative on the Multispot test. Of 6 specimens conrmed as HIV-2 by HIV-2 WB and/or HIV-2 DNA NAT, 4 were positive only for HIV-2 by Multispot and 2 were undifferentiated. The proportion of the 1761 IA-RR, Multispot HIV-1 positive specimens for which each HIV-1 WB band was present is shown in Table 1. All bands except p17 were detected in over 91% of specimens. Of the env bands, the lowest rate of detection was observed for gp41, which was detected in 1715 (97.4%) specimens. However, the percentage increased to 99.1% when weak gp41 bands were included.

5. Discussion In this study, more stringent alternative WB interpretive criteria that required the detection of two env bands and one pol or gag band were investigated. The alternative criteria were anticipated to reduce misclassication of HIV-2 infections since serologic crossreactivity occurs less frequently for the viral env glycoproteins than for the other viral proteins. The alternative criteria classied 75.5% fewer known HIV-2 specimens as HIV-1 positive than did the CDC criteria, but still misclassied 11.4% as HIV-1. Applying these alternative criteria to a sample of 2135 IA-RR specimens that were routinely submitted to a large US public health laboratory led to a small increase (0.8%) in specimens that were classied as indeterminate, including 16 specimens that were HIV-1 positive by CDC criteria and one specimen that was known to represent an HIV-2 infection. However, use of the alternative criteria in this laboratory would still misclassify 2 of 6 HIV-2 specimens as HIV-1.

Table 3 Number and proportion of 2135 IA-repeatedly reactive specimens from New York State Department of Health, classied as positive, negative or indeterminate by two HIV-1 Western blot (WB) interpretive criteria, stratied by Multispot HIV-1/HIV-2 rapid test results. WB criteria Multispot HIV-1/HIV-2 results HIV-1 only (n = 1812), n (%) Current CDC criteria Positive 1761 (97.2) Indeterminate 31d (1.7) 20g (1.1) Negative Alternative interpretive criteria 1745 (96.3) Positive 47 (2.6) Indeterminate Negative 20 (1.1)
a b c

Total, n (%) HIV-2 only (n = 4), n (%) 1 (25.0)b 3 (75.0)b 0 (0.0) 1 (25.0) 3 (75.0) 0 (0.0) HIV-1 and HIV-2 (n = 28), n (%) 26c (92.9) 2e (7.1) 0 (0.0) 25 (89.3) 3 (10.7) 0 (0.0)
a

HIV-negative (n = 291), n (%) 2 (0.7) 60f (20.6) 229 (78.7) 2 (0.7) 60 (20.6) 229 (78.7) 1790 (83.8) 96 (4.5) 249 (11.7) 1773 (83.0) 113 (5.3) 249 (11.7)

No dilution testing was done for these Multispot undifferentiated specimens. Specimens determined to be HIV-2 infections by HIV-2 WB and/or HIV-2 DNA nucleic acid test (NAT). Includes 2 specimens determined to be HIV-2 infections by HIV-2 WB and/or HIV-2 DNA nucleic acid test (NAT). d 15 specimens with positive Multispot and indeterminate WB results were determined to be HIV-1-positive based on positive NAT or follow-up. Denitive results for remaining 16 specimens were not available. e Includes 1 specimen determined to be HIV negative. f 1 of 60 specimen was conrmed HIV-1-positive on follow-up. g 2 specimens with positive Multispot and negative WB results were conrmed HIV-1 positive on follow-up. Denitive results for remaining 18 specimens were not available.

S26

M. Nasrullah et al. / Journal of Clinical Virology 52S (2011) S23S27

The Multispot HIV-1/HIV-2 differentiation assay correctly classied four of the 6 HIV-2 specimens from NYS, and was reactive for both HIV-1 and HIV-2 antibodies in the other two. In this study, 28 of 2135 IA-RR specimens had Multispot results that were reactive for both HIV-1 and HIV-2. Most such undifferentiated results are due to cross-reactivity and can be resolved by a dilution protocol to distinguish whether the specimen is positive for HIV-1 or HIV214 but results of such dilution testing for these specimens were not available. Nearly all of these dually reactive specimens had positive HIV-1 WB results by both criteria, but two were found to be HIV2 infections and one had negative HIV-1 NAT results. Because NAT uses primers specic for HIV-1 or HIV-2, obtaining a negative HIV-1 NAT result from a specimen that continues to be undifferentiated on the Multispot after dilution would require additional testing using HIV-2 specic NAT or WB to rule out HIV-2 infection.11 Although HIV-2 infection is rare in the United States, cases continue to be identied.3,4,17 Most assays approved in the US to screen for HIV infection contain specic antigens that can detect antibodies to both HIV-1 and HIV-2.18 When such tests are used, current recommendations stipulate that additional testing for HIV-2 be conducted for specimens that had a negative or an indeterminate HIV-1 WB.10 The alternative WB interpretive criteria that required two or more env bands reduced but did not eliminate the number of HIV-2 antibody-positive specimens that were incorrectly classied as HIV-1. However, in the NYS specimens, the number of additional indeterminate specimens from HIV-1 infected persons that would undergo testing for HIV-2 exceeded the number of true HIV-2 infections, and the diagnosis of HIV-1 in these persons would have been delayed. Most laboratories in the United States continue to use the HIV-1 WB to conrm a reactive IA or rapid HIV test, even though WB is less sensitive during early infection and yields negative or indeterminate results in some persons that are infected with HIV.1921 In this study, the Multispot test appeared to be slightly more sensitive than either WB criteria for detecting HIV-1 infections. It correctly identied HIV-1 infection in at least 15% of the specimens that were HIV-1 WB indeterminate and nearly 1% of those that were negative by WB. A recently proposed HIV testing algorithm recommends performing an HIV-1/HIV-2 differentiation assay such as Multispot on all specimens that are classied as reactive by an HIV screening test instead of WB.13 If the differentiation assay is negative for HIV antibodies, a NAT should be performed. In the proposed algorithm, specimens with reactive results on a screening HIV-1/HIV-2 assays that are also reactive for HIV-1 on the differentiation assay or on the HIV-1 NAT would be considered HIV-1-positive. Likewise, a sample that is reactive for HIV-2 on the differentiation assay and on a HIV-2 WB or HIV-2 DNA NAT would be considered HIV-2 positive. Among the NYS specimens, 17 that were classied as HIV-1 infected on the basis of IA-RR and Multispot results were negative or indeterminate by WB but were positive by HIV-1 NAT or follow-up testing. Two of these that had negative HIV-1 WB results would have been mistakenly reported as HIV-negative based on the current testing algorithm. Additionally, two specimens that were negative by Multispot but were positive by both WB criteria would have undergone NAT testing under the proposed algorithm, and thus, the infection would likely not have gone undetected. Multispot has been shown to be more sensitive during early infection than WB and performs with high specicity, 19,22 but at least one study noted it to have slightly lower than expected specicity.15 In this study, 34 (1.6%) of the IA-RR specimens that were HIV-1 positive by Multispot but classied as indeterminate or negative by WB would be classied as HIV-1 positive by the proposed algorithm. Because no NAT or follow-up test results were available for these specimens, their true status could not be established. Among the NYS specimens, HIV-2 infection was conrmed using a validated RUO HIV-2 WB and/or a laboratory-developed RUO

HIV-2 DNA NAT test. HIV-2 specic antibodies were detected using a validated RUO HIV-2 WB at CDC. There is no commercially available, FDA-approved supplemental test for HIV-2 diagnosis and, as a result, comparative performance evaluations of any HIV-1/HIV-2 differentiation test for identication of HIV-2 are challenging. Previous studies suggest that the detection of HIV-2 RNA is difcult due to low plasma viral loads.2326 In contrast, the levels of detectable proviral DNA in HIV-2 infected individuals were similar to proviral DNA found in HIV-1 infected individuals, and thus represents an alternative target for denitive identication of HIV-2 infection.24 Because the natural history and the treatment regimens for HIV-1 and HIV-2 infections differ substantially,9 accurate and commercially available tests for the diagnosis and the clinical monitoring of HIV-2 infected individuals are needed. There are several limitations associated with this analysis. The recorded results for the NYS specimens were used and no additional testing was done using fresh specimens. The lack of NAT or follow-up test results did not allow the true infection status to be established for the 36 specimens for which Multispot results disagreed with the WB results. Without the results of the dilution protocol, it was not feasible to resolve possible antibody crossreactivity that may have produced the undifferentiated Multispot results. Specimens in the HIV-2 panel had been selected on the basis of their HIV-2 WB and HIV-2 only Multispot results. Therefore, the performance of Multispot for the differentiation of HIV-1/HIV-2 infection was assessed on only a small number of specimens. Revising the WB interpretive criteria for a positive HIV-1 WB to require at least two env bands and one additional band reduced the number of HIV-2 specimens that were misclassied as HIV-1. However, all of the HIV-2 specimens that were reclassied as indeterminate HIV-1 WB using revised criteria would require additional testing to distinguish HIV-1 from HIV-2 infection. Unlike an HIV1 WB, an HIV-1/HIV-2 differentiation assay could directly identify HIV-2 antibody-positive specimens. Using HIV-1/HIV-2 differentiation assay followed by an HIV specic NAT may be a preferred algorithm for supplemental HIV testing in the United States. Funding None. Competing interests No nancial disclosures were reported by the authors of this paper. Ethical approval All specimens used in this study were unlinked from personal identiers; the study was determined to be research not involving identiable human subjects by the CDC. Disclaimer The ndings and conclusions in this report are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention or the U.S. Department of Health and Human Services. The use of trade names and commercial sources is for identication only and does not imply endorsement by the U.S. Department of Health and Human Services. Acknowledgements The authors would like to acknowledge the technical staff of the Wadsworth Centers Diagnostic HIV Laboratory for their

M. Nasrullah et al. / Journal of Clinical Virology 52S (2011) S23S27

S27

contribution in the processing and diagnostic testing of specimens for HIV, Julia Zhu and Huiseng Wang for their assistance with data cleaning and analysis for this project. References
1. Haute Autorite De Sante. Public health guidelines: HIV infection screening in France, laboratory tests and algorithms; 2008. Available from: http://www.hassante.fr/portail/upload/docs/application/pdf/2010-02/hiv infection screening in france - laboratory tests and algorithms-conclusions.pdf [cited 07.09.10]. 2. Constantine N, Saville RD, Dax EM. Retroviral testing and quality assurance. Ann Arbor, MI: Malloy Printers; 2005. p. 322323. 3. Torian LV, Eavey JJ, Punsalang AP, Pirillo RE, Forgione LA, Kent SA, et al. HIV type 2 in New York City, 20002008. Clin Infect Dis 2010;51(11):133442. 4. Kumar L, Selik R. Epidemiology of HIV-2 infection in the U.S., 19962006; 2007. Available from: http://www.hivtestingconference.org/hivtesting2007/ powerpoint/1 Kumar.pps [cited 07.09.10]. 5. Sullivan MT, Guido EA, Metler RP, Schable CA, Williams AE, Stramer SL. Identication and characterization of an HIV-2 antibody-positive blood donor in the United States. Transfusion (Paris) 1998;38(2):18993. 6. Irwin KL, Olivo N, Schable C, Weber JT, Janssen R, Ernst J. Absence of human immunodeciency virus type 2 infection among patients in a hospital serving a New York community at high risk for infection. Centers for Disease Control and Prevention/Bronx-Lebanon Hospital Center HIV Serosurvey Team. Transfusion (Paris) 1996;36(8):7313. 7. OBrien TR, George JR, Holmberg SD. Human immunodeciency virus type 2 infection in the United States. Epidemiology, diagnosis, and public health implications. JAMA 1992;267(20):27759. 8. De Cock KM, Porter A, Kouadio J, Maran M, Lafontaine MF, Gershy-Damet GM, et al. Cross-reactivity on western blots in HIV-1 and HIV-2 infections. AIDS 1991;5(7):85963. 9. Ntemgwa ML, dAquin Toni T, Brenner BG, Camacho RJ, Wainberg MA. Antiretroviral drug resistance in human immunodeciency virus type 2. Antimicrob Agents Chemother 2009;53(9):36119. 10. OBrien TR, George JR, Epstein JS, Holmberg SD, Schochetman G. Testing for antibodies to human immunodeciency virus type 2 in the United States. MMWR Recomm Rep 1992;41(RR-12):19. 11. Association of Public Health Laboratories and the Centers for Disease Control & Prevention. HIV testing algorithms a status report; 2010. Available from: http:// www.aphl.org/aphlprograms/infectious/hiv/Documents/StatusReportFINAL. pdf [cited 27.08.10]. 12. Bio-Rad Laboratories. Product insert, Human Immunodeciency Virus Type 1, genetic systems HIV-1 Western blot. Redmond, WA, USA: Bio-Rad; 2004.

13. Pandori MW, Branson BM. 2010 HIV diagnostics conference. Expert Rev Anti Infect Ther 2010;8(6):6313. 14. Bio-Rad Laboratories. Product insert, Multispot HIV-1/HIV-2 Rapid Test. Redmond, WA, USA: Bio-Rad; 2004. 15. Delaney KP, Branson BM, Uniyal A, Phillips S, Candal D, Owen SM, et al. Evaluation of the performance characteristics of 6 rapid HIV antibody tests. Clin Infect Dis 2011;52(2):25763. 16. Torian L, Forgione L, Punsalang A, Oleszko W. Comparison of Multispot enzyme immunoassay with Western blot for conrmatory serodiagnosis of HIV. In: 18th conference on retrovirus and opportunistic infections. 2011. Available from: http://www.retroconference.org/2011/Abstracts/41637.htm [cited 24.03.11]. 17. Myers R. Identifying HIV-2 infections using a differential HIV-1 (gp41)/HIV-2 (gp36) serological assays (select HIV or Multispot) by testing HIV EIA reactive specimens unconrmed as HIV-1 antibody positive by HIV-1 Western blot; 2007. Available from: http://www.hivtestingconference.org/hivtesting2007/powerpoint/ 2 Myers.pps [cited 20.10.10]. 18. U.S. Food and Drug Administration. Complete list of donor screening assays for infectious agents and HIV diagnostic assays. Rockville, MD: U.S. Department of Health and Human Services; 2010. Available from: http://www.fda.gov/ BiologicsBloodVaccines/BloodBloodProducts/ApprovedProducts/ LicensedProductsBLAs/BloodDonorScreening/InfectiousDisease/ ucm080466.htm [cited 12.11.10]. 19. Owen SM, Yang C, Spira T, Ou CY, Pau CP, Parekh BS, et al. Alternative algorithms for human immunodeciency virus infection diagnosis using tests that are licensed in the United States. J Clin Microbiol 2008;46(5):158895. 20. Zaaijer HL, v Exel-Oehlers P, Kraaijeveld T, Altena E, Lelie PN. Early detection of antibodies to HIV-1 by third-generation assays. Lancet 1992;340(8822):7702. 21. 2010 HIV Diagnostics Conference. 2007 HIV diagnostics conference summary; 2010. Available from: http://www.hivtestingconference.org/confsum.html# Toc192069911 [cited 09.09.10]. 22. OConell RJ, Peel SA. Multispot HIV-1/HIV-2 Rapid Test: advantages over other rapid HIV tests. Expert Rev Mol Diagn 2007;7(5):499505. 23. MacNeil A, Sarr AD, Sankale JL, Meloni ST, Mboup S, Kanki P. Direct evidence of lower viral replication rates in vivo in human immunodeciency virus type 2 (HIV-2) infection than in HIV-1 infection. J Virol 2007;81(10):532530. 24. Popper SJ, Sarr AD, Gueye-Ndiaye A, Mboup S, Essex ME, Kanki PJ. Low plasma human immunodeciency virus type 2 viral load is independent of proviral load: low virus production in vivo. J Virol 2000;74(3):15547. 25. Popper SJ, Sarr AD, Travers KU, Gueye-Ndiaye A, Mboup S, Essex ME, et al. Lower human immunodeciency virus (HIV) type 2 viral load reects the difference in pathogenicity of HIV-1 and HIV-2. J Infect Dis 1999;180(4):111621. 26. Andersson S, Norrgren H, da Silva Z, Biague A, Bamba S, Kwok S, et al. Plasma viral load in HIV-1 and HIV-2 singly and dually infected individuals in Guinea-Bissau, West Africa: signicantly lower plasma virus set point in HIV-2 infection than in HIV-1 infection. Arch Intern Med 2000;160(21): 328693.