Bridging the Gap: Conference Record [Abstract book, International Conference on AIDS (12th: 1998: Geneva, Switzerland)]

794 Abstracts 42106-42110

12th World AIDS Conference

infection. Sera from 700 individuals were grouped into 7 pools of 100 of which
4 were positive. These 4 positive pools were subdivided into 8 pools of 50 of
which 7 were positive. 7 positive pools were subdivided into 35 pools of 10, of
which 10 were positive. Based on the 10 RNA-positive pools, the point estimate
of HIV incidence was 19.6% per year (95% CI: 7.3%-31.8%). Of the 700 samples
analyzed for p24 antigen, 8 were positive resulting in a point incidence estimate
of 18.5%/year (8.0-36.5%). In contrast, the incident rate based on the traditional
cohort method of follow-up was lower at 9.4%/year (4.8-16.4%) due to lower
follow-up rates of some individuals. Testing of samples in the final pools of 10
allowed for diagnosis of acute primary HIV infection at the individual level.
Conclusions: The multi-stage pooling method for detection of HIV RNA was
more sensitive than the p24 antigen method, detecting two additional cases of
acute primary HIV infection, and was tenfold more cost effective based on only
50 RNA assays compared to 700 p24 antigen assays. Pooling samples for RNA
detection was highly effective in estimating current incident rates which may be
useful for developing countries and in high-risk populations where it may be
difficult to follow cohorts to identify seroconverters.
166* /42106 RNA and DNA PCR for early diagnosis of infants
born to HIV-infected mothers, Thailand
Nancy L. Young1, N. Shaffer1, T. Chaowanachan2, N. Wanparapa3,
N. Waranawat4, K. Chokephaibulkit3, T. Chotpitayasunondh4,
R. Chuachoowong2, P. Wasinrapee2, N. Kaewpunt2, O. Suk Sripanich2,
T.D. Mastro. 1 CDC and HIV/AIDS Collaboration PO. Box 139 Nonthaburi
11000; 2HIV/AIDS Collaboration Nonthaburi; 3Mahidol University Bangkok;
4Children's Hospital Bangkok, Thailand
Background: A comparison of RNA and DNA PCR testing on infants born to HIVinfected mothers may offer new insights into optimal early diagnosis strategies,
timing of vertical transmission, and early disease progression.
Methods: Promptly processed infant plasma specimens were collected at birth
(within 2 days of birth), 2 months and 6 months in an ongoing clinical trial of late,
oral maternal zidovudine treatment for HIV-infected mothers to reduce perinatal
HIV transmission. These samples were tested prospectively by quantitative DNA
PCR (Roche AMPLICOR HIV-1 TEST) and later batch tested by quantitative RNA
PCR (Roche AMPLICOR HIV-1 MONITOR TEST, version 1.5). The RNA assay
had a lower quantitative detection limit of 400 copies/mL. Results were compared
with known final infection outcome of the infants. For infected infants, RNA results
at different time points were compared by paired t test.
Results: Most infections (93%) were HIV-1 subtype E. The DNA test sensitivity
was 32% at birth (13/41), and 100% at 2 months (41/41) and 6 months (25/25). By
RNA testing, 42% were positive at birth (17/39) and 100% were positive at both
2 and 6 months. All samples that tested DNA positive tested RNA positive. For
uninfected infants, there were no false positive tests, including 287 DNA and 50
RNA PCR tests of birth samples. Among infected infants, the mean log10 viral load
(copies/mL) was 3.8 at birth (5.4 for those RNA-positive), 6.6 at 2 months, and
6.3 at 6 months; 25% of 2 and 6 month samples had >5 million RNA copies/mL.
RNA levels were higher at 2 and 6 months than at birth (P < 0.001), and 2-month
levels were higher than 6-month levels (P = 0.02).
Conclusion: In testing to date, the Amplicor DNA and RNA PCR assays detected 32% and 42% of infected infants at birth and 100% at 2 months of age.
Both assays were 100% specific. Infected infants had extremely high viral loads
in the first 6 months, with highest levels at 2 months. RNA PCR may offer slightly
increased sensitivity over DNA PCR for diagnosis at birth and yields valuable
information on early viral dynamics.
S42107 Detection of HIV-1 seroconvertors by PCR in
antibody-negative pooled sera
Pierre Alain Morandi1, G.A. Schockmel1, S. Yerly1, P. Burgisser2, P. Erb3,
L. Matter4, L. Perrin1. 1Laboratory of Virology, University Hospital 1211, Geneva
14; 2Div. of Immunology and Allergy Chuv, Lausanne; 3lnst. for Medical
Microbiology Basel; 4 Inst. for Medical Microbiology, Bern, Switzerland
Background: Patients with early acute HIV-1 infection do not have HIV-1 specific
antibodies but have high levels of viremia. Systematic screening for HIV-1 RNA
could help to identify these patients.
Methods: Screening for HIV-1 RNA was performed using a pooling assay,
based on a modified commercial assay (Amplicor HIV-1 Monitor test). The main
modification consists of an increased input of viral RNA, obtained either by high
speed centrifugation or by precipitation by polyethylene glycol. The analytical
sensitivity of these assays is about 10 HIV-1 RNA copies per ml of sera, which
corresponds to a detection limit of about 500 copies per ml for individual sera
included in the pools.
Results: A total of 234 pools (average pool size: 45 sera) were prepared from
10692 consecutive HIV-1/2 antibody negative sera sent to five microbiological
laboratories for anti-HIV antibody testing. Five pools were identified as positive
and each contained one antibody negative and HIV-1 RNA positive serum, corresponding to an average of one infection per 2138 sera. The five HIV-1 RNA

positive specimens all originated from individuals with symptomatic primary HIV-1
infection at the time of sample collection. When individually tested, these antibody
negative sera contained 1 to 35 x 106 copies of HIV-1 RNA per ml.
Conclusions: Testing pools of HIV-1 antibody negative sera for the presence
of HIV-1 RNA is a sensitive, reliable as well as cost- and time-saving technique.
It allowed the detection of five HIV-1 seroconvertors out of 10692 consecutive

HIV-1 antibody negative sera sent to laboratories of microbiology for anti-HIV
antibody testing.
S42108 The new Cobas Core Anti-HIV-1/2 EIA DAGS II: A highly
sensitive assay based entirely on correctly folded HIV
antigens
Herbert Andres1, C. Boeck1, H. Brodeck1, S. Freiburghaus1, C. Hoffmann1,
B. Scheibling1, K. Schnell1, E.R. Schoenburner2, D.J. Sizmann1, K. Volz1,
B.T. Weinert1, M. Zauke1. 1F Hoffmann - La Roche Ltd. Building 205/3a4, Ch -
4070 Basel, Spain; 2F Hoffmann - La Roche Ltd. Alameda;, USA
Background: Early detection of HIV-infection requires sensitive recognition of
anti-HIV-IgM antibodies which are of low affinity and mainly recognize conformational epitopes. These specific IgM antibodies do not differ greatly in their affinity
compared to natural, multireactive IgM antibodies, posing an inherent problem
for the development of a sensitive and a specific anti-HIV assay.
Methods: The ectodomains of gp41 and gp36 were expressed in E. coli as
inclusion bodies and purified under denaturing conditions. The proteins were
refolded in a physiological buffer and analysed by CD-spectroscopy. For the new
Cobas Core Anti-HIV-1/2 double antigen sandwich immunoassay these folded
ectodomains were coated onto 6 mm polystyrene beads for capture of anti-HIV
antibodies and an antigen-conjugate with horseradish peroxidase was used as
detector. This assay was complemented with a native, correctly folded p24 protein as capture and detector and was evaluated with HIV-positive and negative
sera on the Cobas Core immunoanalyzer manufactured by Roche Diagnostics,
Switzerland.
Results: The CD-spectra of the ectodomains in physiological buffer revealed
an ellipticity of -28 000 [H]MRw(deg*cm2*dmol-1), indicating a high degree of ahelical secondary structure. This is in agreement with previous reports in the
literature that parts of the gp41 ectodomain show a high a-helical content. Likewise, coating of a solid phase using denatured envelope ectodomains leads to
false positive test results which can be abolished when coating the respective
folded antigens. Serological analysis using HIV-1 seroconversions, HIV-1 Subtype O variants and HIV-2 positive samples revealed an excellent sensitivity
compared to other commercial anti-HIV immunoassays, the specificity evaluated
in European blood donors was >99.8%.
Conclusions: Spectroscopic and serological data indicate that the ectodomains
of gp41/gp36 were sucessfully refolded in vitro. Due to their correct conformation
these antigens allow optimal bindung of low affinity HIV-IgM antibodies while
reducing the crossreactivity of multireactive IgM antibodies, resulting in a highly
sensitive and specific anti-HIV assay. To our knowledge this is the first anti-HIV
DAGS assay which is based entirely on correctly folded antigens or antigen
domains.
S42109 Screening with combined HIV ELISA (antibody plus p24
antigen) increases the detection rate of primary HIV
infection
Francois Simon1, S. Yerly2, C. Apetrei1, L. Perrin2. 1Laboratory of Virology,
Bichat Hospital, 46 Henri Huchard 75018, Paris, France; 2Laboratory of Virology,
Geneva, Switzerland
Background: ELISA for anti-HIV antibody might miss the identification of samples
from seroconverting individuals. Combined detection of anti-HIV antibody (Ab)
and p24 antigen (p24 Ag) using a single assay might improve the early detection
of primary HIV infection.
Methods: 263 HIV-positive samples (Western blot positive) and 128 successive
samples from 34 HIV seroconverting individuals (20 pre-seroconversion samples,
Ab negative, p24 Ag positive) were tested using 3 assays: Genscreen, Pasteur
(Ab), Abbott (Ab) and Duo, BioMerieux (Ab and p24 Ag).
Results: The 263 confirmed HIV-positive samples including 3 HIV-1 group 0
and 2 HIV-2 samples were found positive by the 3 assays. The 128 successive
samples from 34 seroconverting individuals were found positive using the Duo
assay, including the 20 pre-seroconverting samples (Ab negative, p24 Ag positive). The mean reduction of the window period using the Duo assay was 4 days
as compared to the 2 third generation antibody assays used for comparison.
Conclusions: All confirmed HIV-positive samples and all samples of seroconverting individuals including 20 Ab negative, p24 Ag positive samples were found
positive by the Duo assay. Screening with this combined HIV assay (co-detection
of anti-HIV antibody and p24 antigen) decreases the window period as compared
to third generation Ab assays and thus increases the detection rate of primary
HIV infection.
42110 IgG antibodies binding to CD4* T-cells are a sensitive
parameter to discriminate between HIVI-positive and
HIV1-negative children
Bernd Buchholz1, M. Muller2, T. Nebe2, K. Ramasubbu1, M. Beichert1,
P. Schafer1, K.H. Niessen1. 1Universitdts Kinder Klinik Mannheim, Haus 2

Grenadier Str. 68167 Mannheim; 2lnst. F. Klin. Chemie Am Klinikum Mannheim,
68167 Mannheim, Germany
Objectives: One hypothesis of autoimmunpathogenesis of HIV1-infection is the
antibody mediated depletion of CD4+T-cells by phagocytosis of macrophages.
To evaluate the first step of this mechanism immunglobulin G (IgG) coating of