- Email: [email protected]
group M subtyped samples
Journal of Virological Methods 69 (1997) 29 – 37
Reactivity of a new HIV-1 group O third generation A-HIV-1/-2 assay with an unusual HIV-1 seroconversion panel and HIV-1 group O/group M subtyped samples J. van Binsbergen a,*, W. Keur a, M. v.d. Graaf a, A. Siebelink a, A. Jacobs a, D. de Rijk a, J. Toonen a, L. Zekeng b, E. Afane Ze b, L.G. Gu¨rtler b b
a Organon Teknika, Boseind 15, 5281 RM Boxtel, The Netherlands Max 6on Pettenkofer Institut, Uni6ersita¨t Mu¨nchen, D-80336 Mu¨nchen, Germany
Received 14 July 1997; accepted 24 July 1997
Abstract It was shown previously that about 97% of the anti-HIV-1 group O strain-positive samples were detected by crossreaction with native HIV-1 gp160 (Van Binsbergen et al., Evaluation of a new third generation anti-HIV-1/antiHIV-2 assay with increased sensitivity for HIV-1 group O, J. Virol. Methods 60 (1996) 131 – 137). Fourteen out of 17 new anti-HIV-1 group O positive samples, selected with the Enzygnost HIV-1/2 plus assay, were already reactive when tested with HIV-1 gp160. When tested by the Vironostika HIV Uni-Form II plus O microELISA all 17 samples were reactive, demonstrating the necessity to implement an HIV-1 group O-specific antigen in the assay. On the other hand, it was surprisingly found that 40 out of 43 (93%) of anti-HIV-1 group M-positive samples, belonging to strain A, B, C, D, E or F, were detected by crossreaction with the HIV-1 group O (strain ANT70) synthetic peptide incorporated in the Vironostika HIV Uni-Form II plus O. Only HIV-1 subtype D-positive samples did not react with this peptide, presumably because of the presence of a histidine residue in the immunodominant region of HIV-1 subtype D gp41. Both crossreactions make the Vironostika HIV Uni-Form II plus O microELISA also sensitive for anti-HIV-1-positive samples originating from different geographical regions and resulting from different HIV-1 subtype infections. With an unusual seroconversion panel in which p24 Ag was present persistently, many anti-HIV-1/-2 assays produce alternating positive/negative results in anti-HIV antibody-positive bleeds. It was shown that the use of viral p24 and gp160 in a direct sandwich, allowing detection of anti-HIV IgG and IgM, explains the identification of all anti-HIV-positive bleeds by the Vironostika HIV Uni-Form II plus O. The high sensitivity of the plus O assay was confirmed with clinical samples of a so-called anti-HIV-1 low titer panel. The specificity of the Vironostika HIV Uni-Form II plus O determined in five blood transfusion centers, based on 135 070 tests, was 99.97%. © 1997 Elsevier Science B.V. Abbre6iations: HIV, human immunodeficiency virus; PBS, phosphate-buffered saline; S/COV, ratio sample/cut off value; HRP, horse radish peroxidase; ELISA, enzyme-linked immunosorbent assay. * Corresponding author. Tel.: +31 411 654465; fax: + 31 411 654527. 0166-0934/97/$17.00 © 1997 Elsevier Science B.V. All rights reserved. PII S 0 1 6 6 - 0 9 3 4 ( 9 7 ) 0 0 1 3 5 - 3
J. 6an Binsbergen et al. / Journal of Virological Methods 69 (1997) 29–37
30
Keywords: HIV-1 group O; HIV; AIDS; Diagnostics; ELISA; IgG; IgM
1. Introduction A new third generation anti-HIV-1/-2 screening assay was described recently in which the sensitivity for a-HIV-1 group O-positive samples was improved, the Vironostika HIV Uni-Form II plus O (Van Binsbergen et al., 1996). The enhanced sensitivity for HIV-1 group O was achieved by adding a specific HIV-1 group O peptide derived from the HIV-1 ANT70 strain (Vanden Haesevelde et al., 1994). The performance of the assay has been studied further with the most interesting clinical samples available. New anti-HIV-O-positive samples were collected in Cameroon. This new panel was tested in order. The increase in sensitivity observed when testing HIV seroconversion panels by third generation anti-HIV screening assays as compared to second generation assays, is largely based on the change of assay format, namely from an indirect to a direct assay and by the type of conjugate applied (HIV envelope and p24 HRP conjugates). This direct binding format also allows detection of anti-HIV IgM (Gallarda et al., 1992). Recently, an unusual seroconversion panel was described (BBI panel AG) showing a decreasing level of anti-HIV antibody in time, by some anti-HIV assays even below the detection limit. During the same period the p24 Ag level remained detectable. It was desirable to evaluate the third generation Uni-Form II plus O assay using this panel. The Vironostika HIV Uni-Form II plus O assay was also evaluated on a well-characterized antiHIV-1 group M-positive panel, containing 43 sera from groups A to F originating from different geographical regions in Asia and Africa, to study the possible crossreactivity between the HIV-1 strain subtype B antigen gp160 and other HIV-1 strain infections. 2. Materials and methods
2.1. HIV antigens All HIV specific antigens employed in the assays
were described previously by Van Binsbergen et al. (1996).
2.2. Assays The Vironostika HIV Uni-Form II and the Vironostika HIV Uni-Form II plus O (Organon Teknika, Boxtel, The Netherlands), the Behring Enzygnost HIV-1/2 plus (Behring, Marburg, Germany) and the Wellcozyme HIV-1 Recombinant (Murex, Dartford, UK) were used according to the manufacturer’s instruction. The Vironostika HIV Uni-Form II plus O is a direct sandwich anti-HIV assay in which the same HIV antigens (gp160, p24, HIV-1 group O peptide and HIV-2 peptide) are used as plate coat as well as HRP conjugate. The conjugate is present as a lyophilised sphere in the well.
2.3. Serum panels A new panel of 17 Cameroonian anti-HIV-1 group O-positive samples was used, collected from blood donor plasma obtained at Yaounde and Douala. These were identified with the Enzygnost HIV-1/2 plus ELISA (Behring) and the competitive ELISA (Wellcozyme HIV-1 recombinant, Murex) and confirmed by immunoblot based on HIV-1 group M (MVP899) and HIV-1 group O isolate MVP5180 (Gu¨rtler et al., 1995). HIV-1 seroconversion panels and the antiHIV-1 low titer performance panel PRB 104 was purchased from Boston Biomedica Inc. (BBI; West Bridgewater, MD, USA). All data presented on other assays were supplied by BBI. HIV-1 group M subtyped samples were obtained from BBI. This panel contains samples of subtype A, B, C, D, E and F, obtained from African and Asian countries (see Table 2). Subtyping was done with an indirect ELISA using the V3 loop peptides from the Centers for Disease Control (CDC; HIV Branch, Atlanta, GA, USA) (Schable et al., 1994).
J. 6an Binsbergen et al. / Journal of Virological Methods 69 (1997) 29–37
31
Table 1 Reactivity of anti-HIV-1 group O panel, presented as S/COV Sample MVP−
HIV UF-IIa (1/10 dil.)
HIV plus O b (1/10 dil.)
HIV plus O b (undil.)
Murex Comp. EIA
Behring HIV-1/2 plus
748 1639 3127 3675 6245 6250 6405 6599 8314 8525 8732 9433 9435 9437 9557 10474 13127
0.4 1.4 3.2 0.5 5.8 0.6 1.8 2.7 2.4 2.8 0.6 0.5 1.3 2.0 1.6 3.3 0.9
0.5 neg 2.4 3.7 1.1 4.3 0.8 neg 2.6 3.7 6.8 3.3 2.0 0.7 neg 2.2 3.3 2.6 3.5 2.6
1.0 n.t. n.t. n.t. n.t. 2.4 n.t. n.t. n.t. n.t. n.t. 2.1 n.t. n.t. n.t. n.t. n.t.
0.6 3.0 1.8 1.9 3.4 0.6 2.8 1.3 2.0 0.9 1.6 0.7 1.6 2.6 1.6 2.9 2.3
5.2 5.2 4.2 5.2 5.2 2.1 5.2 5.2 5.2 5.2 5.2 5.2 5.2 5.2 5.2 5.2 5.2
neg
neg neg
neg neg
neg
neg
neg
neg neg
a
HIV Uni-Form II. HIV Uni-Form II plus O. dil., diluted; undil., undiluted; neg, negative; n.t., not tested.
b
3. Results and discussion
3.1. HIV-1 group O sensiti6ity A new panel consisting of 17 anti-HIV-1 group O-positive samples, collected from Cameroonian blood donors, was evaluated by several a-HIV screenings assays: Behring HIV-1/2 plus; Murex Competitive EIA; Vironostika HIV Uni-Form II and HIV Uni-Form II assay in which the HIV-O specific peptide had been added; Vironostika HIV Uni-Form II plus O (Van Binsbergen et al., 1996). As expected, the Behring assay detected all HIV-1 group O samples because these were selected with this assay. Because of the limited amount of antiHIV-O sample available, these were first tested by both HIV Uni-Form II assays in an 1/10 dilution using a 5-ml sample. Table 1 shows that 11/17 samples reacted positively in the HIV Uni-Form II without HIV-1 ANT70 peptide, by crossreaction with gp160. In the plus O assay, 14/17 were detected when tested at a 1/10 dilution. The samples which were not reactive (MVP96-748, -6250
and -9433) were retested undiluted by the plus O assay according to the manufacturer’s instruction (50 ml) and these three were then also positive (Table 1). By the competitive ELISA from Murex the same samples were nonreactive when negative in a 1/10 dilution by the HIV Uni-Form II plus O. These three samples can only be detected by implementation of an HIV-1 group O sequence, e.g. HIV-1 ANT 70. The results presented here confirm the data described by Loussert-Ajaka et al. (1994) and Van Binsbergen et al. (1996) on Central West African anti-HIV-1 group O-positive samples and demonstrate the necessity to adapt anti-HIV assays with a group O-specific antigen.
3.2. HIV-1 group M subtype samples It is evident that because of the global spread of HIV the detection of the different HIV-1 group M subtypes is very important. To evaluate the sensitivity of the HIV Uni-Form II plus O assay on this issue a panel consisting of HIV-1 subtype A, B, C, D, E and F samples, collected from blood
J. 6an Binsbergen et al. / Journal of Virological Methods 69 (1997) 29–37
32
Table 2 Reactivity of anti-HIV-1 group M subtyped samples by HIV Uni-Form II plus O, and with single gp160 – HRP and HIV-O – HRP conjugates, expressed as S/COV ratio Sample BBI code Country of origin HIV-1 subtype
GM3-0505-0003 GM3-0505-0008 GM3-0505-0011 GM3-0911-0002 GM3-0911-0006 GM3-0911-0007 GM3-0911-0008 GM4-0308-0001 GM4-0308-0003 GM4-0905-0057 GM5-0111-0004 GM5-1807-0001 GM5-1807-0004 GM5-1807-0005 GM5-1807-0007 GM5-2206-0001 GM5-2206-0003 GM5-2206-0004 GM5-2206-0005 GM5-2206-0007 GN4-1409-0030 GN4-1409-0033 GN4-1409-0037 GN4-1409-0038 GN4-1409-0041 GN4-1409-0042 GN4-1409-0043 GN4-1409-0044 GN4-1409-0046 GN4-1409-0047 GN4-1409-0065 GN4-1409-0080 GN4-1409-0088 BR4-1310-0001 BR4-1310-0002 BR4-1310-0003 BR4-1310-0004 BR4-1310-0005 BR4-1310-0006 GL4-0707-0004 IH5-2811-0002 IH5-2811-0010 EW2-0307-0019
Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana India India India India India India India India India India India India India Thailand Thailand Thailand Thailand Thailand Thailand Thailand Uganda Uganda S.Africa
A,F A,C,E,F A,C,F C C C C C E C F C B A A C C C C C C A C C C C C C C F C C C A,C,E C,E C,E A,C,E,F A,C,E C,E C,E D D E
HIV Uni-Form II plus O plus O conjugate
Single gp160 – HRP
Single HIV-O – HRP
12.4 14.4 \14.6 11.9 \14.6 \14.6 \14.6 \14.6 \14.6 12.6 \14.6 13.7 11.5 13.1 \14.6 14.5 \14.6 13 14.3 \14.6 11.1 13.6 \14.6 \14.6 14.4 5.4 14.4 14.5 14 5.5 7.2 13.1 \14.6 11.6 9.3 10 \14.6 13.4 12.7 12 11.2 11.3 11.8
12.5 \12.6 \12.6 12.3 \12.6 \12.6 \12.6 \12.6 5.6 12.3 9.0 \12.6 \12.6 11.5 \12.6 \12.6 \12.6 \12.6 \12.6 \12.6 9.2 11.4 12.5 12.3 11.5 4.1 11.1 11.8 \12.6 4.9 6.8 \12.6 \12.6 10.8 9.4 10.4 \12.6 12.1 12.1 \12.6 11.8 10.7 4.9
2.5 6.2 7.8 1.4 1.8 4.7 2.9 6.2 3.7 1.7 4.8 3.5 1.3 1.2 9.4 2.7 11.8 1.2 5.6 12.9 2.5 2.6 7.9 10.1 5.2 3.0 10.1 7.1 3.1 0.5 neg 1.0 5.8 1.2 4.3 1.3 1.1 4.1 4.7 2.6 1.1 0.7 neg 0.5 neg 1.4
donors from African and Asian countries, was evaluated (Table 2). Subtyping of these samples was done with the V3 loop peptides of the respec-
tive HIV-1 strains. Some of these samples were reactive with two or more HIV-1 subtype peptides.
J. 6an Binsbergen et al. / Journal of Virological Methods 69 (1997) 29–37
33
Fig. 1. Sequence comparison between the gp41 immunodominant regions in gp160 of HIV-1 subtypes, based on the consensus sequences (Myers et al., 1995). The group O sequence is derived from HIV-1 ANT70. The prominent differences in charge between HIV-1 and HIV-1 group O strains and the difference between subtype D and the other HIV-1 strains are indicated (bold). The one-letter code for the amino acids is used.
Table 2 shows that all samples are strongly reactive by the HIV Uni-Form II plus O assay in which HIV-1 subtype B antigens (gp160 and p24) are applied. The reactivity was also studied with single HRP conjugates of gp160 and the HIV-O peptide, using the identical concentration of these conjugates as usually present in the plus O assay. Because all samples react with gp160 – HRP with nearly the same reactivity as with the plus O conjugate which contains both gp160 – HRP and HIV-O–HRP, it is concluded that epitopes of gp160 are very important for the sensitivity of the plus O assay. The results obtained with the single HIV-O–peptide conjugate are surprising. As can be seen in Table 2, 40/43 (93%) of the HIV-1 group M samples reacted positively. This was confirmed recently by Eberle et al. (1997) who found a comparable percentage when (non-subtyped) anti-HIV-1 group M-positive samples were tested in an indirect ELISA based on the gp41 immunodominant region of HIV-1 group O strain MVP-5180. It can be seen from Table 2 that, in contrast to the other HIV-1 subtypes, both subtype D samples, which originated from Uganda, were not reactive with the HIV-O peptide in the HIV Uni-Form II plus O. A comparison between the HIV-O peptide (HIV-1 ANT70 strain) and the homologue sequence of subtype D in the immunodominant region is shown (Fig. 1). The only difference between HIV-1 subtype D and the
other HIV-1 strains is the presence of a histidine residue present at position 607 in the cystinebridged region which is known to be a key HIV-1 immunodominant epitope (Gnann et al., 1987). Most likely, anti-gp41 antibodies in subtype D samples, directed against this subtype D epitope, do not crossreact with the HIV-1 group O amino acid sequence, possibly because this peptide does not contain the unique positive charge of His607. Although the sequence of the HIV-O peptide differs significantly from the other HIV-1 subtype consensus sequences, HIV-O-specific amino acids, such as Arg582, Thr590, Gln593, Asn594, Lys604, Tyr610, Lys614 and Arg617, thus do not hamper major crossreactivity. The HIV-O peptide selected for the assay contains apparently enough homologue motifs for crossreaction. It is concluded that the addition of the HIV-O peptide improves the anti-HIV-1 sensitivity of the Vironostika HIV Uni-form II plus O ELISA by crossreaction of antibodies directed against the immunodominant region of most HIV-1 strains. It is not clear why, looking at the consensus sequences presented in Fig. 1, the subtype F sample from India was not reactive with the HIV-O peptide. Perhaps the unique envelope sequence of this sample differs in this region from that of the F consensus. Sequence analysis should answer this question.
J. 6an Binsbergen et al. / Journal of Virological Methods 69 (1997) 29–37
34
Table 3 Reactivity of the Abbott 3rd Gen Plus and the HIV Uni-Form II plus O with new seroconversion panels Panel
BBI BBI BBI BBI BBI BBI BBI BBI BBI BBI BBI BBI BBI BBI
AF AG AH AI AJ AK AL AM AS AT AU AV AW AX
Abbott 3rd Gen Plusa HIV Uni-Form II plus O Sample no.
S/COV
Sample no.
S/COV
6 5 2 2 7 6 6 3 6 5 4 — 2 —
14.3 10.6 16.0 2.3 4.5 1.4 1.1 13.6 8.8 13.0 1.5 neg 6.4 neg
6 5 2 2 7 6 6 3 6 5 5 — 2 —
3.3 3.6 4.2 5.9 5.7 2.1 3.2 2.1 4.0 2.6 1.5 neg 2.3 neg
The first reactive sample of the panel is indicated with its corresponding S/COV. a Data obtained from BBI.
3.3. HIV serocon6ersion panels The sensitivity of the Vironostika HIV UniForm II plus O assay in clinical samples was studied further using seroconversion panels (Table 3). The results of the plus O assay are presented in parallel with the Abbott 3rd Gen Plus ELISA. In the majority of the samples the plus O assay
performs like the reference assay. Panel BBI AV and AX are not reactive by both assays because these panels are HIV antibody negative and contain only HIV antigen. A rather unusual seroconversion panel is BBI panel AG (PRB932). When HIV p24 Ag data of this panel are compared with those of other seroconversion panels, it is remarkable that the period of Ag positivity period is very long: in panel AG this period lasts for 167 days during which Ag levels are fluctuating, while in other seroconversion panels the period of p24 Ag positivity is limited to only 1 or 2 weeks. This is the first panel described with such a large deviation in the immune response (Garrett et al., 1997). The reason for this phenomenon could be virological but, because the virus is typed as subtype B, an immunological reason is more likely. The patient is known to have been Ag positive and asymptomatic during these 9 months. No acute ARC/ AIDS was developed during this period (Garrett et al., 1997). The CD4 count, however, declined from 250 cells/mm3 about the time of the first HIV diagnosis to 70 after 9 months. As a result of this prolonged Ag-positive period, the anti-HIV response by various screening ELISAs is fluctuating around the cut-off with occasional non-reactive results (for examples see Table 5). During this period HIV RNA was always present as determined by RT-PCR (Garrett et al., 1997).
Table 4 Reactivity of seroconversion panel AG PRB932 (BBI) in the presence and absence of 20% anti-human IgG presented as absorbance at 450 nm and as S/COV Sample
AG1 AG2 AG3 AG4 AG5 AG6 AG7 AG8 AG9
Undiluted plus O
1/6 diluted plus O, 20% anti-IgG
1/6 diluted plus O, 20% PBS
Undiluted gp160 – HRP
A450
S/COV
A450
S/COV
A450
S/COV
A450
S/COV
107 106 109 141 792 505 833 1521 1864
0.5 0.5 0.5 0.6 3.6 2.3 3.8 7.0 8.6
111 114 104 117 284 205 207 315 530
0.6 0.7 0.6 0.7 1.6 1.2 1.2 1.8 3.1
110 112 107 121 472 284 545 916 1436
0.5 0.5 0.5 0.6 2.2 1.3 2.5 4.2 6.6
67 63 130 151 424 290 486 827 1034
0.4 0.4 0.8 0.9 2.5 1.7 2.9 5.0 6.2
J. 6an Binsbergen et al. / Journal of Virological Methods 69 (1997) 29–37
35
Table 5 Reactivity of seroconversion panel AG PRB932 (BBI), presented as S/COVa Sample
Wellcozyme HIV 1/2
Abbott 3rd Gen Plus
Gen. System HIV 1/2
Coulter HIV Ag
BBI HIV IgM
AG1 AG2 AG3 AG4 AG5 AG6 AG7 AG8 AG9
0.4 0.3 0.3 0.5 3.2 1.3 1.2 6.6 7.6
0.2 0.1 0.2 0.8 10.6 1.7 0.9 3.1 5.0
0.1 0.1 0.2 0.2 2.6 3.4 1.3 0.9 0.8
0.1 0.0 0.0 18.6 17.8 5.4 9.2 4.8 2.9
0.3 0.3 0.3 1.0 11.5 5.1 0.6 0.5 0.4
a
Results obtained from BBI.
To determine the class of the antibody response detected by the Vironostika HIV UniForm II plus O assay the panel was tested in the presence and absence of anti-human IgG (Table 4) which was added to the assay medium. The absorbance values of sample AG5 and AG6 were decreased to about 30% in the presence of anti-human IgG, while anti-IgG decreases these values of sample AG7-9 to about 65%. Anti-HIV IgM was thus most likely present predominantly in samples AG 5 and 6, while in the latter three samples anti-HIV IgG determined the main immune response. This finding corresponds with the results of an HIV IgM assay (BBI, Table 5). Schmidt et al. (1992), Gallarda et al. (1992) and Hampl et al. (1995) reported that in seroconversion panels anti-HIV IgM is present as the first immune response indicator, about 7 days before anti-HIV-specific IgG in seroconversion panels. In panel AG the length of the anti-HIV IgM-positive response, however, was at least 23 days. Comparison of HIV Ag p24 data with the IgG response showed, as expected, a negative correlation between these two parameters (Table 5). It is concluded from Tables 4 and 5 that, despite the fluctuating anti-HIV immune responses in relation to the presence of p24 Ag, the Vironostika HIV Uni-Form II plus O was sensitive enough to detect anti-HIV IgG and IgM in anti-HIV-positive samples. The assay showed the usual, increasing level of anti-HIV response in
time. This is not only caused by the format of the assay which facilitates also anti-HIV IgM detection (Hampl et al., 1995). The results presented in Table 4 obtained with a single gp160– HRP conjugate showed that anti-HIV IgG and IgM detected in panel PRB932 were predominantly directed against gp160. This response was not affected by the changing levels of p24 Ag and it may explain why other assays sometimes test negative in this panel (Table 5).
3.4. Reaction with anti-HIV-1 low-titer performance panel (PRB 104) The sensitivity of the HIV Uni-Form II plus O was studied on a so-called anti-HIV-1-positive low-titer performance panel (PRB 104). The results are presented in Table 6, in parallel with Abbott HIV 1/2 and Wellcozyme HIV 1/2. Except for the anti-HIV Ig-negative samples 10 and 15, all samples of PRB104 were reactive in the plus O and Wellcozyme assay with comparable S/COVs. The Abbott HIV 1/2 tested negative on the so-called Western Blot p24—only samples 7, 8 and 14, which were true HIV-infected subjects as determined by the high HIV Ag levels (Table 6). The early seroconversion sample 15 which is HIV Ag positive, was not detected in any of the three antibody assays. A sample obtained from the same donor 4 days later (sample 14) tested positive by the Wellcozyme and by the plus O assay.
J. 6an Binsbergen et al. / Journal of Virological Methods 69 (1997) 29–37
36
Table 6 Reactivity of anti-HIV-1 low-titer performance panel (PRB 104), BBI Panel member
HIV UF-II plus O
Abbotta HIV 1/2
Wellcozymea HIV 1/2
Duponta HIV Ag
Organon Teknikaa Western Blot
Result
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
9.2 6.3 12.7 5.8 8.1 8.5 4.4 4.3 3.2 0.5 5.6 4.5 2.5 1.1 0.6
3.6 3.9 4.7 3.3 3.7 4.4 0.5 0.7 1.0 0.3 2.6 3.2 1.6 0.7 0.2
9.6 7.8 10.0 4.4 7.4 10.1 2.3 4.4 2.8 0.3 4.4 4.9 1.4 1.3 0.3
14.7 1.1 0.1 0.5 0.1 0.8 7.8 \35 0.9 0.0 0.2 0.9 0.9 12.6 2.3
24,41,120,160 24,120,160 18,24,41,65,120,160 24,160 24,f51,f65,120,160 24,65,120,160 24 24 24,41,120,160 no bands 24,120,160 24,120,160 24,f160 f24 no bands
Pos Pos Pos Pos Pos Pos Ind Ind Pos Neg Pos Pos Pos Ind Neg
PRB104-14 is a sample from the same donor as PRB104-15, obtained 4 days later; f, faint. Test results obtained from BBI.
a
Table 7 Specificity data of seven batches of HIV Uni-Form II plus O, evaluated at four blood banks in The Netherlands Lot
No. tested
IRa
%
61654 13 638 14 99.90 61658 5264 10 99.81 56008 40 201 71 99.82 53603 4123 7 99.83 56200 43 710 75 99.83 56196 26 236 45 99.83 61662 1898 1 99.95 Total 135 070 223 99.83 Samples tested of HIV Uni-Form II plus O lots in different blood banks Blood bank 1 20 481 22 99.89 Blood bank 2 35 430 70 99.80 Blood bank 3 36 312 39 99.89 Blood bank 4 42 847 92 99.79 Total 135 070 223 99.83 a
RRa
%
4 1 9 0 12 9 0 35
99.97 99.98 99.98 100 99.97 99.97 100 99.97
6 14 7 8 35
99.97 99.96 99.98 99.98 99.97
IR, number of initial reactives; RR, number of repeat reactives.
3.5. Specificity The specificity of the Vironostika HIV UniForm II plus O assay was monitored during 1 year by routine screening in bloodbanks. This was done by testing seven different production lots in four different blood banks in The Netherlands. The results are presented in Table 7. The initial
specificity between the lots was almost identical (99.81–99.95%), with an overall initial specificity of 99.83%. The initial specificity between the bloodbanks also did not show much variation (99.79–99.89%). The frequency of repeat reactive samples was also comparable between the assay lots (specificity of 99.97–100%) as well as between the transfusion
J. 6an Binsbergen et al. / Journal of Virological Methods 69 (1997) 29–37
centres (specificity of 99.96 – 99.98%). The number of repeat reactive samples was very low, that is 35 out of 135 070 samples, resulting in a final specificity of 99.97%. These 35 samples were indeterminate or false-positive reactions and could not be confirmed by Western blotting.
References Eberle, J., Loussert-Ajaka, I., Brust, S., Zekeng, L., Hauser, P.H., Kaptue, L., Knapp, S., Damond, F., Saragosti, S., Simon, F., Gu¨rtler, L.G., 1997. Diversity of the immunodominant epitope of gp41 of HIV-1 subtype O and its validity for antibody detection. J. Virol. Methods 67, 85– 91. Gallarda, J.L., Henrard, D.R., Liu, D., Harrington, S., Stramer, S.L., Valinsky, J.E., Wu, P., 1992. Early detection of antibody to Human Immunodeficiency Virus Type 1 by using an antigen conjugate immunoassay correlates with the presence of immunoglobulin M antibody. J. Clin. Microbiol. 30, 2379 – 2384. Garrett, P.E., Baird, I., Manak, M., Howell, R., Cosentino, M., Weissberger, H., Weiblen, B.J., Busch, M.P., Schumacher, R.T., 1997. Prolonged p24 antigenemia and weak immune response for nine months after HIV seroconversion. Association of State and Territorial Public Health Laboratory Directors, 12th Annual Conference on Human Retrovirus Testing, Houston, TX, February 25–28, 1997. Gnann, J.W., McCormick, J.B., Mitchell, S., Nelson, J.A., Oldstone, M.B.A., 1987. Synthetic peptide immunoassay distinguishes HIV type 1 and HIV type 2 infections. Science 237, 1346 – 1349. Gu¨rtler, L.G., Zekeng, L., Simon, F., Eberle, J., Tsague, J.M., Kaptue, L., Brust, S., Knapp, S., 1995. Reactivity of five
.
37
anti-HIV-1 subtype O specimens with six different antiHIV screening ELISAs and three immunoblots. J. Virol. Methods 51, 177 – 184. Hampl, H., Kapprell, H.P., Sawitzky, D., Wilske, W., Gu¨rtler, L., 1995. Detection of specific human immunodeficiency virus IgM antibodies. Med. Microbiol. Immunol. 184, 69 – 71. Loussert-Ajaka, I., Ly, T.D., Chaix, M.L., Ingrand, D., Saragosti, S., Courouce´, A.M., Brun-Ve´zinet, F., Simon, F., 1994. HIV-1/HIV-2 seronegativity in HIV-1 subtype O infected patients. Lancet 343, 1393 – 1394. Myers, G., Hahn, B.H., Mellors, J.W., Henderson, L.E., Korber, B., Jeang, K.-T., McCutchan, F.E., Pavlakis, G.N. (Eds.), 1995. Human Retroviruses and AIDS 1995. A compilation and analysis of nucleic acid and amino acid sequences, Los Alamos National Laboratory, Los Alamos, New Mexico, USA. Schable, C., Zekeng, L., Pau, C.-P., Hu, D., Kaptue, L., Gu¨rtler, L., Dondero, T., Tsague, J.-M., Schochetman, G., Jaffe, H., George, J.R., 1994. Sensitivity of United States HIV antibody tests for detection of HIV-1 group O infections. Lancet 344, 1333 – 1334. Schmidt, L., Bandick, F., Poschmann, A., 1992. Detection of HIV-antibodies and p24-Antigen in HIV Seroconversion sera using different enzyme immunoassays. Klin. Lab. 38, 511 – 515. Van Binsbergen, J., De Rijk, D., Peels, H., Dries, C., Scherders, J., Koolen, M., Zekeng, L., Gu¨rtler, L.G., 1996. Evaluation of a new third generation anti-HIV-1/anti-HIV2 assay with increased sensitivity for HIV-1 group O. J. Virol. Methods 60, 131 – 137. Vanden Haesevelde, M., Decourt, J.-L., De Leys, R., Vandenborght, B., Van der Groen, G., Van Heuverswyn, H., Saman, E., 1994. Genomic cloning and complete sequence analysis of a highly divergent African Human Immunodeficiency Virus isolate. J. Virol. 68, 1586 – 1596.
Reactivity of a new HIV-1 group O third generation A-HIV-1/-2 assay with an unusual HIV-1 seroconversion panel and HIV-1 group O/group M subtyped samples J. van Binsbergen a,*, W. Keur a, M. v.d. Graaf a, A. Siebelink a, A. Jacobs a, D. de Rijk a, J. Toonen a, L. Zekeng b, E. Afane Ze b, L.G. Gu¨rtler b b
a Organon Teknika, Boseind 15, 5281 RM Boxtel, The Netherlands Max 6on Pettenkofer Institut, Uni6ersita¨t Mu¨nchen, D-80336 Mu¨nchen, Germany
Received 14 July 1997; accepted 24 July 1997
Abstract It was shown previously that about 97% of the anti-HIV-1 group O strain-positive samples were detected by crossreaction with native HIV-1 gp160 (Van Binsbergen et al., Evaluation of a new third generation anti-HIV-1/antiHIV-2 assay with increased sensitivity for HIV-1 group O, J. Virol. Methods 60 (1996) 131 – 137). Fourteen out of 17 new anti-HIV-1 group O positive samples, selected with the Enzygnost HIV-1/2 plus assay, were already reactive when tested with HIV-1 gp160. When tested by the Vironostika HIV Uni-Form II plus O microELISA all 17 samples were reactive, demonstrating the necessity to implement an HIV-1 group O-specific antigen in the assay. On the other hand, it was surprisingly found that 40 out of 43 (93%) of anti-HIV-1 group M-positive samples, belonging to strain A, B, C, D, E or F, were detected by crossreaction with the HIV-1 group O (strain ANT70) synthetic peptide incorporated in the Vironostika HIV Uni-Form II plus O. Only HIV-1 subtype D-positive samples did not react with this peptide, presumably because of the presence of a histidine residue in the immunodominant region of HIV-1 subtype D gp41. Both crossreactions make the Vironostika HIV Uni-Form II plus O microELISA also sensitive for anti-HIV-1-positive samples originating from different geographical regions and resulting from different HIV-1 subtype infections. With an unusual seroconversion panel in which p24 Ag was present persistently, many anti-HIV-1/-2 assays produce alternating positive/negative results in anti-HIV antibody-positive bleeds. It was shown that the use of viral p24 and gp160 in a direct sandwich, allowing detection of anti-HIV IgG and IgM, explains the identification of all anti-HIV-positive bleeds by the Vironostika HIV Uni-Form II plus O. The high sensitivity of the plus O assay was confirmed with clinical samples of a so-called anti-HIV-1 low titer panel. The specificity of the Vironostika HIV Uni-Form II plus O determined in five blood transfusion centers, based on 135 070 tests, was 99.97%. © 1997 Elsevier Science B.V. Abbre6iations: HIV, human immunodeficiency virus; PBS, phosphate-buffered saline; S/COV, ratio sample/cut off value; HRP, horse radish peroxidase; ELISA, enzyme-linked immunosorbent assay. * Corresponding author. Tel.: +31 411 654465; fax: + 31 411 654527. 0166-0934/97/$17.00 © 1997 Elsevier Science B.V. All rights reserved. PII S 0 1 6 6 - 0 9 3 4 ( 9 7 ) 0 0 1 3 5 - 3
J. 6an Binsbergen et al. / Journal of Virological Methods 69 (1997) 29–37
30
Keywords: HIV-1 group O; HIV; AIDS; Diagnostics; ELISA; IgG; IgM
1. Introduction A new third generation anti-HIV-1/-2 screening assay was described recently in which the sensitivity for a-HIV-1 group O-positive samples was improved, the Vironostika HIV Uni-Form II plus O (Van Binsbergen et al., 1996). The enhanced sensitivity for HIV-1 group O was achieved by adding a specific HIV-1 group O peptide derived from the HIV-1 ANT70 strain (Vanden Haesevelde et al., 1994). The performance of the assay has been studied further with the most interesting clinical samples available. New anti-HIV-O-positive samples were collected in Cameroon. This new panel was tested in order. The increase in sensitivity observed when testing HIV seroconversion panels by third generation anti-HIV screening assays as compared to second generation assays, is largely based on the change of assay format, namely from an indirect to a direct assay and by the type of conjugate applied (HIV envelope and p24 HRP conjugates). This direct binding format also allows detection of anti-HIV IgM (Gallarda et al., 1992). Recently, an unusual seroconversion panel was described (BBI panel AG) showing a decreasing level of anti-HIV antibody in time, by some anti-HIV assays even below the detection limit. During the same period the p24 Ag level remained detectable. It was desirable to evaluate the third generation Uni-Form II plus O assay using this panel. The Vironostika HIV Uni-Form II plus O assay was also evaluated on a well-characterized antiHIV-1 group M-positive panel, containing 43 sera from groups A to F originating from different geographical regions in Asia and Africa, to study the possible crossreactivity between the HIV-1 strain subtype B antigen gp160 and other HIV-1 strain infections. 2. Materials and methods
2.1. HIV antigens All HIV specific antigens employed in the assays
were described previously by Van Binsbergen et al. (1996).
2.2. Assays The Vironostika HIV Uni-Form II and the Vironostika HIV Uni-Form II plus O (Organon Teknika, Boxtel, The Netherlands), the Behring Enzygnost HIV-1/2 plus (Behring, Marburg, Germany) and the Wellcozyme HIV-1 Recombinant (Murex, Dartford, UK) were used according to the manufacturer’s instruction. The Vironostika HIV Uni-Form II plus O is a direct sandwich anti-HIV assay in which the same HIV antigens (gp160, p24, HIV-1 group O peptide and HIV-2 peptide) are used as plate coat as well as HRP conjugate. The conjugate is present as a lyophilised sphere in the well.
2.3. Serum panels A new panel of 17 Cameroonian anti-HIV-1 group O-positive samples was used, collected from blood donor plasma obtained at Yaounde and Douala. These were identified with the Enzygnost HIV-1/2 plus ELISA (Behring) and the competitive ELISA (Wellcozyme HIV-1 recombinant, Murex) and confirmed by immunoblot based on HIV-1 group M (MVP899) and HIV-1 group O isolate MVP5180 (Gu¨rtler et al., 1995). HIV-1 seroconversion panels and the antiHIV-1 low titer performance panel PRB 104 was purchased from Boston Biomedica Inc. (BBI; West Bridgewater, MD, USA). All data presented on other assays were supplied by BBI. HIV-1 group M subtyped samples were obtained from BBI. This panel contains samples of subtype A, B, C, D, E and F, obtained from African and Asian countries (see Table 2). Subtyping was done with an indirect ELISA using the V3 loop peptides from the Centers for Disease Control (CDC; HIV Branch, Atlanta, GA, USA) (Schable et al., 1994).
J. 6an Binsbergen et al. / Journal of Virological Methods 69 (1997) 29–37
31
Table 1 Reactivity of anti-HIV-1 group O panel, presented as S/COV Sample MVP−
HIV UF-IIa (1/10 dil.)
HIV plus O b (1/10 dil.)
HIV plus O b (undil.)
Murex Comp. EIA
Behring HIV-1/2 plus
748 1639 3127 3675 6245 6250 6405 6599 8314 8525 8732 9433 9435 9437 9557 10474 13127
0.4 1.4 3.2 0.5 5.8 0.6 1.8 2.7 2.4 2.8 0.6 0.5 1.3 2.0 1.6 3.3 0.9
0.5 neg 2.4 3.7 1.1 4.3 0.8 neg 2.6 3.7 6.8 3.3 2.0 0.7 neg 2.2 3.3 2.6 3.5 2.6
1.0 n.t. n.t. n.t. n.t. 2.4 n.t. n.t. n.t. n.t. n.t. 2.1 n.t. n.t. n.t. n.t. n.t.
0.6 3.0 1.8 1.9 3.4 0.6 2.8 1.3 2.0 0.9 1.6 0.7 1.6 2.6 1.6 2.9 2.3
5.2 5.2 4.2 5.2 5.2 2.1 5.2 5.2 5.2 5.2 5.2 5.2 5.2 5.2 5.2 5.2 5.2
neg
neg neg
neg neg
neg
neg
neg
neg neg
a
HIV Uni-Form II. HIV Uni-Form II plus O. dil., diluted; undil., undiluted; neg, negative; n.t., not tested.
b
3. Results and discussion
3.1. HIV-1 group O sensiti6ity A new panel consisting of 17 anti-HIV-1 group O-positive samples, collected from Cameroonian blood donors, was evaluated by several a-HIV screenings assays: Behring HIV-1/2 plus; Murex Competitive EIA; Vironostika HIV Uni-Form II and HIV Uni-Form II assay in which the HIV-O specific peptide had been added; Vironostika HIV Uni-Form II plus O (Van Binsbergen et al., 1996). As expected, the Behring assay detected all HIV-1 group O samples because these were selected with this assay. Because of the limited amount of antiHIV-O sample available, these were first tested by both HIV Uni-Form II assays in an 1/10 dilution using a 5-ml sample. Table 1 shows that 11/17 samples reacted positively in the HIV Uni-Form II without HIV-1 ANT70 peptide, by crossreaction with gp160. In the plus O assay, 14/17 were detected when tested at a 1/10 dilution. The samples which were not reactive (MVP96-748, -6250
and -9433) were retested undiluted by the plus O assay according to the manufacturer’s instruction (50 ml) and these three were then also positive (Table 1). By the competitive ELISA from Murex the same samples were nonreactive when negative in a 1/10 dilution by the HIV Uni-Form II plus O. These three samples can only be detected by implementation of an HIV-1 group O sequence, e.g. HIV-1 ANT 70. The results presented here confirm the data described by Loussert-Ajaka et al. (1994) and Van Binsbergen et al. (1996) on Central West African anti-HIV-1 group O-positive samples and demonstrate the necessity to adapt anti-HIV assays with a group O-specific antigen.
3.2. HIV-1 group M subtype samples It is evident that because of the global spread of HIV the detection of the different HIV-1 group M subtypes is very important. To evaluate the sensitivity of the HIV Uni-Form II plus O assay on this issue a panel consisting of HIV-1 subtype A, B, C, D, E and F samples, collected from blood
J. 6an Binsbergen et al. / Journal of Virological Methods 69 (1997) 29–37
32
Table 2 Reactivity of anti-HIV-1 group M subtyped samples by HIV Uni-Form II plus O, and with single gp160 – HRP and HIV-O – HRP conjugates, expressed as S/COV ratio Sample BBI code Country of origin HIV-1 subtype
GM3-0505-0003 GM3-0505-0008 GM3-0505-0011 GM3-0911-0002 GM3-0911-0006 GM3-0911-0007 GM3-0911-0008 GM4-0308-0001 GM4-0308-0003 GM4-0905-0057 GM5-0111-0004 GM5-1807-0001 GM5-1807-0004 GM5-1807-0005 GM5-1807-0007 GM5-2206-0001 GM5-2206-0003 GM5-2206-0004 GM5-2206-0005 GM5-2206-0007 GN4-1409-0030 GN4-1409-0033 GN4-1409-0037 GN4-1409-0038 GN4-1409-0041 GN4-1409-0042 GN4-1409-0043 GN4-1409-0044 GN4-1409-0046 GN4-1409-0047 GN4-1409-0065 GN4-1409-0080 GN4-1409-0088 BR4-1310-0001 BR4-1310-0002 BR4-1310-0003 BR4-1310-0004 BR4-1310-0005 BR4-1310-0006 GL4-0707-0004 IH5-2811-0002 IH5-2811-0010 EW2-0307-0019
Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana Ghana India India India India India India India India India India India India India Thailand Thailand Thailand Thailand Thailand Thailand Thailand Uganda Uganda S.Africa
A,F A,C,E,F A,C,F C C C C C E C F C B A A C C C C C C A C C C C C C C F C C C A,C,E C,E C,E A,C,E,F A,C,E C,E C,E D D E
HIV Uni-Form II plus O plus O conjugate
Single gp160 – HRP
Single HIV-O – HRP
12.4 14.4 \14.6 11.9 \14.6 \14.6 \14.6 \14.6 \14.6 12.6 \14.6 13.7 11.5 13.1 \14.6 14.5 \14.6 13 14.3 \14.6 11.1 13.6 \14.6 \14.6 14.4 5.4 14.4 14.5 14 5.5 7.2 13.1 \14.6 11.6 9.3 10 \14.6 13.4 12.7 12 11.2 11.3 11.8
12.5 \12.6 \12.6 12.3 \12.6 \12.6 \12.6 \12.6 5.6 12.3 9.0 \12.6 \12.6 11.5 \12.6 \12.6 \12.6 \12.6 \12.6 \12.6 9.2 11.4 12.5 12.3 11.5 4.1 11.1 11.8 \12.6 4.9 6.8 \12.6 \12.6 10.8 9.4 10.4 \12.6 12.1 12.1 \12.6 11.8 10.7 4.9
2.5 6.2 7.8 1.4 1.8 4.7 2.9 6.2 3.7 1.7 4.8 3.5 1.3 1.2 9.4 2.7 11.8 1.2 5.6 12.9 2.5 2.6 7.9 10.1 5.2 3.0 10.1 7.1 3.1 0.5 neg 1.0 5.8 1.2 4.3 1.3 1.1 4.1 4.7 2.6 1.1 0.7 neg 0.5 neg 1.4
donors from African and Asian countries, was evaluated (Table 2). Subtyping of these samples was done with the V3 loop peptides of the respec-
tive HIV-1 strains. Some of these samples were reactive with two or more HIV-1 subtype peptides.
J. 6an Binsbergen et al. / Journal of Virological Methods 69 (1997) 29–37
33
Fig. 1. Sequence comparison between the gp41 immunodominant regions in gp160 of HIV-1 subtypes, based on the consensus sequences (Myers et al., 1995). The group O sequence is derived from HIV-1 ANT70. The prominent differences in charge between HIV-1 and HIV-1 group O strains and the difference between subtype D and the other HIV-1 strains are indicated (bold). The one-letter code for the amino acids is used.
Table 2 shows that all samples are strongly reactive by the HIV Uni-Form II plus O assay in which HIV-1 subtype B antigens (gp160 and p24) are applied. The reactivity was also studied with single HRP conjugates of gp160 and the HIV-O peptide, using the identical concentration of these conjugates as usually present in the plus O assay. Because all samples react with gp160 – HRP with nearly the same reactivity as with the plus O conjugate which contains both gp160 – HRP and HIV-O–HRP, it is concluded that epitopes of gp160 are very important for the sensitivity of the plus O assay. The results obtained with the single HIV-O–peptide conjugate are surprising. As can be seen in Table 2, 40/43 (93%) of the HIV-1 group M samples reacted positively. This was confirmed recently by Eberle et al. (1997) who found a comparable percentage when (non-subtyped) anti-HIV-1 group M-positive samples were tested in an indirect ELISA based on the gp41 immunodominant region of HIV-1 group O strain MVP-5180. It can be seen from Table 2 that, in contrast to the other HIV-1 subtypes, both subtype D samples, which originated from Uganda, were not reactive with the HIV-O peptide in the HIV Uni-Form II plus O. A comparison between the HIV-O peptide (HIV-1 ANT70 strain) and the homologue sequence of subtype D in the immunodominant region is shown (Fig. 1). The only difference between HIV-1 subtype D and the
other HIV-1 strains is the presence of a histidine residue present at position 607 in the cystinebridged region which is known to be a key HIV-1 immunodominant epitope (Gnann et al., 1987). Most likely, anti-gp41 antibodies in subtype D samples, directed against this subtype D epitope, do not crossreact with the HIV-1 group O amino acid sequence, possibly because this peptide does not contain the unique positive charge of His607. Although the sequence of the HIV-O peptide differs significantly from the other HIV-1 subtype consensus sequences, HIV-O-specific amino acids, such as Arg582, Thr590, Gln593, Asn594, Lys604, Tyr610, Lys614 and Arg617, thus do not hamper major crossreactivity. The HIV-O peptide selected for the assay contains apparently enough homologue motifs for crossreaction. It is concluded that the addition of the HIV-O peptide improves the anti-HIV-1 sensitivity of the Vironostika HIV Uni-form II plus O ELISA by crossreaction of antibodies directed against the immunodominant region of most HIV-1 strains. It is not clear why, looking at the consensus sequences presented in Fig. 1, the subtype F sample from India was not reactive with the HIV-O peptide. Perhaps the unique envelope sequence of this sample differs in this region from that of the F consensus. Sequence analysis should answer this question.
J. 6an Binsbergen et al. / Journal of Virological Methods 69 (1997) 29–37
34
Table 3 Reactivity of the Abbott 3rd Gen Plus and the HIV Uni-Form II plus O with new seroconversion panels Panel
BBI BBI BBI BBI BBI BBI BBI BBI BBI BBI BBI BBI BBI BBI
AF AG AH AI AJ AK AL AM AS AT AU AV AW AX
Abbott 3rd Gen Plusa HIV Uni-Form II plus O Sample no.
S/COV
Sample no.
S/COV
6 5 2 2 7 6 6 3 6 5 4 — 2 —
14.3 10.6 16.0 2.3 4.5 1.4 1.1 13.6 8.8 13.0 1.5 neg 6.4 neg
6 5 2 2 7 6 6 3 6 5 5 — 2 —
3.3 3.6 4.2 5.9 5.7 2.1 3.2 2.1 4.0 2.6 1.5 neg 2.3 neg
The first reactive sample of the panel is indicated with its corresponding S/COV. a Data obtained from BBI.
3.3. HIV serocon6ersion panels The sensitivity of the Vironostika HIV UniForm II plus O assay in clinical samples was studied further using seroconversion panels (Table 3). The results of the plus O assay are presented in parallel with the Abbott 3rd Gen Plus ELISA. In the majority of the samples the plus O assay
performs like the reference assay. Panel BBI AV and AX are not reactive by both assays because these panels are HIV antibody negative and contain only HIV antigen. A rather unusual seroconversion panel is BBI panel AG (PRB932). When HIV p24 Ag data of this panel are compared with those of other seroconversion panels, it is remarkable that the period of Ag positivity period is very long: in panel AG this period lasts for 167 days during which Ag levels are fluctuating, while in other seroconversion panels the period of p24 Ag positivity is limited to only 1 or 2 weeks. This is the first panel described with such a large deviation in the immune response (Garrett et al., 1997). The reason for this phenomenon could be virological but, because the virus is typed as subtype B, an immunological reason is more likely. The patient is known to have been Ag positive and asymptomatic during these 9 months. No acute ARC/ AIDS was developed during this period (Garrett et al., 1997). The CD4 count, however, declined from 250 cells/mm3 about the time of the first HIV diagnosis to 70 after 9 months. As a result of this prolonged Ag-positive period, the anti-HIV response by various screening ELISAs is fluctuating around the cut-off with occasional non-reactive results (for examples see Table 5). During this period HIV RNA was always present as determined by RT-PCR (Garrett et al., 1997).
Table 4 Reactivity of seroconversion panel AG PRB932 (BBI) in the presence and absence of 20% anti-human IgG presented as absorbance at 450 nm and as S/COV Sample
AG1 AG2 AG3 AG4 AG5 AG6 AG7 AG8 AG9
Undiluted plus O
1/6 diluted plus O, 20% anti-IgG
1/6 diluted plus O, 20% PBS
Undiluted gp160 – HRP
A450
S/COV
A450
S/COV
A450
S/COV
A450
S/COV
107 106 109 141 792 505 833 1521 1864
0.5 0.5 0.5 0.6 3.6 2.3 3.8 7.0 8.6
111 114 104 117 284 205 207 315 530
0.6 0.7 0.6 0.7 1.6 1.2 1.2 1.8 3.1
110 112 107 121 472 284 545 916 1436
0.5 0.5 0.5 0.6 2.2 1.3 2.5 4.2 6.6
67 63 130 151 424 290 486 827 1034
0.4 0.4 0.8 0.9 2.5 1.7 2.9 5.0 6.2
J. 6an Binsbergen et al. / Journal of Virological Methods 69 (1997) 29–37
35
Table 5 Reactivity of seroconversion panel AG PRB932 (BBI), presented as S/COVa Sample
Wellcozyme HIV 1/2
Abbott 3rd Gen Plus
Gen. System HIV 1/2
Coulter HIV Ag
BBI HIV IgM
AG1 AG2 AG3 AG4 AG5 AG6 AG7 AG8 AG9
0.4 0.3 0.3 0.5 3.2 1.3 1.2 6.6 7.6
0.2 0.1 0.2 0.8 10.6 1.7 0.9 3.1 5.0
0.1 0.1 0.2 0.2 2.6 3.4 1.3 0.9 0.8
0.1 0.0 0.0 18.6 17.8 5.4 9.2 4.8 2.9
0.3 0.3 0.3 1.0 11.5 5.1 0.6 0.5 0.4
a
Results obtained from BBI.
To determine the class of the antibody response detected by the Vironostika HIV UniForm II plus O assay the panel was tested in the presence and absence of anti-human IgG (Table 4) which was added to the assay medium. The absorbance values of sample AG5 and AG6 were decreased to about 30% in the presence of anti-human IgG, while anti-IgG decreases these values of sample AG7-9 to about 65%. Anti-HIV IgM was thus most likely present predominantly in samples AG 5 and 6, while in the latter three samples anti-HIV IgG determined the main immune response. This finding corresponds with the results of an HIV IgM assay (BBI, Table 5). Schmidt et al. (1992), Gallarda et al. (1992) and Hampl et al. (1995) reported that in seroconversion panels anti-HIV IgM is present as the first immune response indicator, about 7 days before anti-HIV-specific IgG in seroconversion panels. In panel AG the length of the anti-HIV IgM-positive response, however, was at least 23 days. Comparison of HIV Ag p24 data with the IgG response showed, as expected, a negative correlation between these two parameters (Table 5). It is concluded from Tables 4 and 5 that, despite the fluctuating anti-HIV immune responses in relation to the presence of p24 Ag, the Vironostika HIV Uni-Form II plus O was sensitive enough to detect anti-HIV IgG and IgM in anti-HIV-positive samples. The assay showed the usual, increasing level of anti-HIV response in
time. This is not only caused by the format of the assay which facilitates also anti-HIV IgM detection (Hampl et al., 1995). The results presented in Table 4 obtained with a single gp160– HRP conjugate showed that anti-HIV IgG and IgM detected in panel PRB932 were predominantly directed against gp160. This response was not affected by the changing levels of p24 Ag and it may explain why other assays sometimes test negative in this panel (Table 5).
3.4. Reaction with anti-HIV-1 low-titer performance panel (PRB 104) The sensitivity of the HIV Uni-Form II plus O was studied on a so-called anti-HIV-1-positive low-titer performance panel (PRB 104). The results are presented in Table 6, in parallel with Abbott HIV 1/2 and Wellcozyme HIV 1/2. Except for the anti-HIV Ig-negative samples 10 and 15, all samples of PRB104 were reactive in the plus O and Wellcozyme assay with comparable S/COVs. The Abbott HIV 1/2 tested negative on the so-called Western Blot p24—only samples 7, 8 and 14, which were true HIV-infected subjects as determined by the high HIV Ag levels (Table 6). The early seroconversion sample 15 which is HIV Ag positive, was not detected in any of the three antibody assays. A sample obtained from the same donor 4 days later (sample 14) tested positive by the Wellcozyme and by the plus O assay.
J. 6an Binsbergen et al. / Journal of Virological Methods 69 (1997) 29–37
36
Table 6 Reactivity of anti-HIV-1 low-titer performance panel (PRB 104), BBI Panel member
HIV UF-II plus O
Abbotta HIV 1/2
Wellcozymea HIV 1/2
Duponta HIV Ag
Organon Teknikaa Western Blot
Result
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
9.2 6.3 12.7 5.8 8.1 8.5 4.4 4.3 3.2 0.5 5.6 4.5 2.5 1.1 0.6
3.6 3.9 4.7 3.3 3.7 4.4 0.5 0.7 1.0 0.3 2.6 3.2 1.6 0.7 0.2
9.6 7.8 10.0 4.4 7.4 10.1 2.3 4.4 2.8 0.3 4.4 4.9 1.4 1.3 0.3
14.7 1.1 0.1 0.5 0.1 0.8 7.8 \35 0.9 0.0 0.2 0.9 0.9 12.6 2.3
24,41,120,160 24,120,160 18,24,41,65,120,160 24,160 24,f51,f65,120,160 24,65,120,160 24 24 24,41,120,160 no bands 24,120,160 24,120,160 24,f160 f24 no bands
Pos Pos Pos Pos Pos Pos Ind Ind Pos Neg Pos Pos Pos Ind Neg
PRB104-14 is a sample from the same donor as PRB104-15, obtained 4 days later; f, faint. Test results obtained from BBI.
a
Table 7 Specificity data of seven batches of HIV Uni-Form II plus O, evaluated at four blood banks in The Netherlands Lot
No. tested
IRa
%
61654 13 638 14 99.90 61658 5264 10 99.81 56008 40 201 71 99.82 53603 4123 7 99.83 56200 43 710 75 99.83 56196 26 236 45 99.83 61662 1898 1 99.95 Total 135 070 223 99.83 Samples tested of HIV Uni-Form II plus O lots in different blood banks Blood bank 1 20 481 22 99.89 Blood bank 2 35 430 70 99.80 Blood bank 3 36 312 39 99.89 Blood bank 4 42 847 92 99.79 Total 135 070 223 99.83 a
RRa
%
4 1 9 0 12 9 0 35
99.97 99.98 99.98 100 99.97 99.97 100 99.97
6 14 7 8 35
99.97 99.96 99.98 99.98 99.97
IR, number of initial reactives; RR, number of repeat reactives.
3.5. Specificity The specificity of the Vironostika HIV UniForm II plus O assay was monitored during 1 year by routine screening in bloodbanks. This was done by testing seven different production lots in four different blood banks in The Netherlands. The results are presented in Table 7. The initial
specificity between the lots was almost identical (99.81–99.95%), with an overall initial specificity of 99.83%. The initial specificity between the bloodbanks also did not show much variation (99.79–99.89%). The frequency of repeat reactive samples was also comparable between the assay lots (specificity of 99.97–100%) as well as between the transfusion
J. 6an Binsbergen et al. / Journal of Virological Methods 69 (1997) 29–37
centres (specificity of 99.96 – 99.98%). The number of repeat reactive samples was very low, that is 35 out of 135 070 samples, resulting in a final specificity of 99.97%. These 35 samples were indeterminate or false-positive reactions and could not be confirmed by Western blotting.
References Eberle, J., Loussert-Ajaka, I., Brust, S., Zekeng, L., Hauser, P.H., Kaptue, L., Knapp, S., Damond, F., Saragosti, S., Simon, F., Gu¨rtler, L.G., 1997. Diversity of the immunodominant epitope of gp41 of HIV-1 subtype O and its validity for antibody detection. J. Virol. Methods 67, 85– 91. Gallarda, J.L., Henrard, D.R., Liu, D., Harrington, S., Stramer, S.L., Valinsky, J.E., Wu, P., 1992. Early detection of antibody to Human Immunodeficiency Virus Type 1 by using an antigen conjugate immunoassay correlates with the presence of immunoglobulin M antibody. J. Clin. Microbiol. 30, 2379 – 2384. Garrett, P.E., Baird, I., Manak, M., Howell, R., Cosentino, M., Weissberger, H., Weiblen, B.J., Busch, M.P., Schumacher, R.T., 1997. Prolonged p24 antigenemia and weak immune response for nine months after HIV seroconversion. Association of State and Territorial Public Health Laboratory Directors, 12th Annual Conference on Human Retrovirus Testing, Houston, TX, February 25–28, 1997. Gnann, J.W., McCormick, J.B., Mitchell, S., Nelson, J.A., Oldstone, M.B.A., 1987. Synthetic peptide immunoassay distinguishes HIV type 1 and HIV type 2 infections. Science 237, 1346 – 1349. Gu¨rtler, L.G., Zekeng, L., Simon, F., Eberle, J., Tsague, J.M., Kaptue, L., Brust, S., Knapp, S., 1995. Reactivity of five
.
37
anti-HIV-1 subtype O specimens with six different antiHIV screening ELISAs and three immunoblots. J. Virol. Methods 51, 177 – 184. Hampl, H., Kapprell, H.P., Sawitzky, D., Wilske, W., Gu¨rtler, L., 1995. Detection of specific human immunodeficiency virus IgM antibodies. Med. Microbiol. Immunol. 184, 69 – 71. Loussert-Ajaka, I., Ly, T.D., Chaix, M.L., Ingrand, D., Saragosti, S., Courouce´, A.M., Brun-Ve´zinet, F., Simon, F., 1994. HIV-1/HIV-2 seronegativity in HIV-1 subtype O infected patients. Lancet 343, 1393 – 1394. Myers, G., Hahn, B.H., Mellors, J.W., Henderson, L.E., Korber, B., Jeang, K.-T., McCutchan, F.E., Pavlakis, G.N. (Eds.), 1995. Human Retroviruses and AIDS 1995. A compilation and analysis of nucleic acid and amino acid sequences, Los Alamos National Laboratory, Los Alamos, New Mexico, USA. Schable, C., Zekeng, L., Pau, C.-P., Hu, D., Kaptue, L., Gu¨rtler, L., Dondero, T., Tsague, J.-M., Schochetman, G., Jaffe, H., George, J.R., 1994. Sensitivity of United States HIV antibody tests for detection of HIV-1 group O infections. Lancet 344, 1333 – 1334. Schmidt, L., Bandick, F., Poschmann, A., 1992. Detection of HIV-antibodies and p24-Antigen in HIV Seroconversion sera using different enzyme immunoassays. Klin. Lab. 38, 511 – 515. Van Binsbergen, J., De Rijk, D., Peels, H., Dries, C., Scherders, J., Koolen, M., Zekeng, L., Gu¨rtler, L.G., 1996. Evaluation of a new third generation anti-HIV-1/anti-HIV2 assay with increased sensitivity for HIV-1 group O. J. Virol. Methods 60, 131 – 137. Vanden Haesevelde, M., Decourt, J.-L., De Leys, R., Vandenborght, B., Van der Groen, G., Van Heuverswyn, H., Saman, E., 1994. Genomic cloning and complete sequence analysis of a highly divergent African Human Immunodeficiency Virus isolate. J. Virol. 68, 1586 – 1596.