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

12th World AIDS Conference

Abstracts 21111-21116 261

genesis. The characteristic manifestations of HIV-1 encephalopathy (perivascular
multinucleated giant cells, microgliosis and reactive astrocytosis) were shown in
both patients by retrospective histopathological analysis, confirming the presence
of HIV-1 in the studied brain regions.
Conclusion: This study is unique and is the first to demonstrate that HIV-1
variants from blood and different brain regions vary in their replicative capacity in
a number of cell types, confirming the in vivo existence of neurotropic variants
and their independent evolution in the brain. These studies may have important
implications in the therapeutic treatment of HIV disease and ADC in HIV-1 infected
individuals.
S21111 HIV infected CD4+ syncytia are commonly found in the
peripheral blood of HIV infected individuals
Jack Stapleton, J. McCoy, D. Wessels, D. Klinzman, K. Watkins, D.C. Shutt,
D.R. Soil. Univ of Iowa, 200 Hawkins Drive SW 54-GH Iowa City, IA, USA
Background: It has been assumed that HIV-induced syncytia are rarely present
in the peripheral blood of patients with HIV infection. Syncytia have been found in
lymphoid tissue, and this finding correlates with in vitro SI phenotype testing. Due
to the heterogeneity of T cell diameter, 3 to 5 cell syncytia are indistinguishable
from single T cells by standard light microscopy. Using advanced, high resolution
microscopy and immunofluorescent staining, we demonstrate for the first time the
presence of syncytia in the peripheral blood of HIV infected patients which stain
for CD4 and p24 antigen.
Methods: Nuclear staining methods were used to identify syncytia in PBMC's.
20 HIV infected patients (16 with detectable HIV RNA) and 5 HIV negative
controls were studied. Patient viral load (Roche Amplicor Method), CD4 counts,
and SI phenotype (in MT-2 cells) were determined by standard methods. Freshly
prepared PBMC's were stained for CD4 and p24 antigen, and nuclei were stained
with DAPI or PI. Cells were visualized by light, fluorescent, UV, and confocal
microscopy, and cell diameters and volumes were determined by measurement
of videorecorded images using DIAS technology (Solltech, Inc.).
Results: Mutinucleated (2 to 5 nuclei), CD4+ syncytia were identified among
the PBMC's of 9 of 16 patients, including 2 of 4 patients with SI phenotype, and
7 of 12 patients with NSI phenotype. No syncytia were seen among the 4 HIV+
patients with nondetectable HIV RNA, or the 5 HIV antibody negative controls.
HIV p24 antigen was detected in several of the syncytia. Absence of syncytia
appeared to be correlated with low CD4 counts, as patients in whom syncytia
were not visualized all had fewer than 200 CD4 cells.
Conclusions: In vivo, CD4+ T cell syncytia are commonly found in the peripheral blood of individuals with HIV infection, independent of standard (in vitro) SI
phenotypic testing results. Thus, current measurement of SI phenotype does not
reflect what occurs in vivo. Patients who have NSI phenotype virus (determined
in vitro) have T cell syncytia in peripheral blood. Contrary to current opinion, our
studies suggest that HIV induced syncytia are present in most patients with HIV
infection, and thus may play a major role in HIV pathogenesis.
269*/ 21112 Dynamic changes in naive and activated T cells in
blood and lymph nodes in patients receiving highly
active antiretroviral therapy (HAART)
Clive Gray', J. Lawrence2, J. Schapiro2, M. Vierra2, T.C. Merigan2. 1Center For
AIDS Research, Stanford University Medical Center, Stanford, CA; 2Stanford
University, Palo Alto, CA, USA
Objectives: To explore whether immunopathogenesis caused by persistent HIV-1
replication can be reversed in patients receiving quadruple drug HAART.
Design: a prospective, open label study consisting of 13 drug-naive patients
receiving 2 protease inhibitors (new formulation saquinavir and nelfinavir) and
2 reverse transcriptase inhibitors (3TC and D4T) were followed over 24 weeks,
where lymph node (LN) biopsies were obtained at baseline and at 24 weeks.
Methods: sequential changes in the frequency of naive and activated T cell
subsets in peripheral blood and LN were investigated by flow cytometry. Coexpression of CD45RA and CD62L was used to characterize naive subsets
and expression of CD69 and HLA-DR used to assess cell activation. Cell subset
changes were correlated with plasma and LN viral load as well as with architectural
changes as observed by histological assessment.
Results: 6 out of 13 individuals reaching 24 weeks of therapy showed a proportional increase in peripheral blood naive CD4+CD45RA+CD62L' cells from
a median of 39% (range: 5-80) to 62% (range: 54-88, P < 0.05) and in naive
CD8+CD45RA+CD62L+ cells from 15.5% (range: 3-44) to 28% (range: 26-65, P
- 0.05). There were consistently higher proportions of these cells in all LNs at
baseline, with a median of 60.5% (range: 48-75) increasing at 24 weeks of HAART
to 71% (50-88). Acutely activated CD4+CD69+ cells in the periphery declined
from a median of 6.5% (range: 2-77) to 3% (range: 1-7) at 24 weeks, whilst the
proportion of these cells remained elevated in the LN at a median of 33.5% (range:
28--49). The decline of these cells in the peripheral blood significantly correlated
(R = 0.66, P - 0.02) with reduced plasma viral load; from a mean baseline of 4.4 ~
0.7 to 2.36: 0.4 log copies/ml at 24 weeks (4/6 patients undetectable). The higher

proportion of acutely activated CD4+CD69+ cells in the LN appeared to be related
to a 100-fold difference in HIV-1 copy numbers between plasma and LN at 24
weeks (mean baseline of 7.2 ~ 0.44 declining to 4.47 ~ 0.3 log copies/mg tissue).
Conclusion: A parallel expansion of naive CD4+ and CD8+ cells in blood
and lymph nodes during HAART demonstrates that improved immune integrity
depends upon effectively suppressing high levels of replicating virus in both blood
and lymphoid compartments. These data also suggest that increased peripheral

naive T cell populations is not a result of redistribution from LN to blood during
therapy.
21113 Induction of HIV-1 replication by influenza vaccination (IV)
in patients on potent antiretroviral therapy
Huldrych F. GOnthard', J.K. Wong', C. Spina', C. Ignacio1, S. Kwok2
C. Christopherson2, D. Havlir3, D. Richman4. 'of California San Diego, Dept. of
Path, 9500 Gilman Dr, La Jolla, LA, 92093-0679; 2Roche Molecular Systems,
Alameda CA; 3Treatment Center, Univ of Calif. San Diego, San Diego CA;
4Univ. of California San Diego/Veterans Affairs Health Care System S.D., USA
Objectives: Influenza vaccine (IV) upregulates HIV replication in patients (pts)
with readily detectable plasma HIV RNA. The present study examined whether
IV could stimulate transcription and viral replication in pts with undetectable (. 50
copies/ml [c/ml]) or with low (50-400 c/ml) HIV plasma RNA.
Methods: 21 pts (20 pts on potent antiretroviral therapy) with a mean CD4
count of 404 cells/mm3 receiving IV were prospectively studied. HIV plasma RNA,
PBMC-associated RNA, proviral DNA, multiply-spliced RNA and CD4/CD8 T-lymphocyte phenotypes were measured at baseline, week (wk) 1, 2, 4, and 6/7/or 8.
Humoral immune responses (IR) were tested by influenza virus hemagglutination
inhibition. 5 healthy control subjects were included in parallel studies.
Results: At baseline, 11 pts [group 1] had <50 RNA c/ml [mean CD4 count:
389 cells/mm3], 5 pts [group II] had 50-400 c/ml [mean CD4 count: 461] and
3 pts [group III] had >400 c/ml [mean CD4 count: 380]. Significant elevations
of plasma RNA levels (>3 fold) after IV were found in 3 out of 11 pts in group
I. (n = 2, wk 1: RNA value 89 resp. 79 c/ml, n = 1, wk 2: RNA 79 c/ml). All 3
pts returned to undetectable levels within one wk. All 5 pts from group II had a
RNA boost with subsequent return to near baseline RNA values. The 3 pts from
group III experienced a boost in plasma RNA. PBMC RNA, proviral DNA and
multiply-spliced RNA correlated with plasma RNA in pts with poor but not in those
with strong viral suppression. PBMC RNA was clearly detected in 7 of 8 pts with
undetectable plasma viremia. Humoral and cellular immune responses to IV are
being analyzed for correlations with virologic responses.
Conclusions: 1. Transient HIV plasma viremia could be induced by IV in 3 of
11 patients receiving potent antiretroviral therapy with undetectable plasma RNA
values (<50 c/ml). Whether this represents in vivo stimulation of latently infected
cells or upregulation of low level replication remains to be determined. 2. The
RNA increase seen in all pts from group II confirms ongoing viral replication in pts
with 50-400 plasma RNA copies. 3. PBMC RNA (unspliced) is present in a high
proportion of patients despite undetectable plasma viremia. Whether this reflects
ongoing viral replication or a "latent" state requires further investigation.
21115 1 Plasma L-selectin level correlates with virus load in
pediatric HIV infection
Francis Lee', Athena Kourtis', C. Ibegbul, D.M. Thea2, A. Nahmias',
S. Nesheim'. 'Dept of Pediatrics, Emory University, 69 Butler St, Atlanta, GA,
30303; 2Harvard Institute for International Dev, Cambridge, MA, USA
Background: L-selectin is a cell surface glycoprotein expressed on lymphocytes
that regulates their migration to lymph nodes. It is shed after cell activation and can
be detected in the plasma as soluble L-selectin (s-LS). As a quantitative marker
of immune activation, we investigated the relationship of plasma s-LS to viral
replication in HIV+ infants infected by intrauterine or intra-partum transmission.
Methods:A solid-phase ELISA (R&D Systems, MN) was used to measure
plasma s-LS levels in 100 infants with perinatally transmitted HIV infection and
in 79 age-matched HIV exposed uninfected controls. The same plasma samples
from the infected children were assayed for HIV RNA viral copies using the NASBA
method (Organon, NC). Results from HIV DNA PCR and RNA (NASBA-QL) tests
on specimens obtained in the first week of life was used to assess timing of transmission. CHI-square and Pearson's correlation tests were employed for statistical
analyses.
Results: HIV infected infants <6 months of age had significantly elevated
levels of plasma s-LS (median 3843 ng/ml) when compared to that of uninfected
age-matched contols (2818 ng/ml, p < 0.01). Both were higher than normal adult
levels (1133 ng/ml). There was a significant correlation between s-LS levels and
the number of viral RNA copies (r = 0.60) in the plasma of the infected children.
Infants with higher levels of s-LS had lower CD4+ and higher CD8+DR cell counts.
Among HIV infected infants, s-LS levels in the first week of life were higher in
those with intrauterine, as compared to those with intrapartum infection (2644 vs.
806 ng/ml, P < 0.001).
Conclusions: Plasma s-SL level, reflecting state of immune activation, was
found to correlate with viral load. This may be due to HIV-induced immune activation. Conversely, cell activation may promote HIV replication, explaining the
higher viral loads generally found in younger infants, compared to infected adults.
s-LS elevation in the first week of life appears to be an indication of in utero HIV
infection.

21116 Symptomatic HIV-1 seroconversion correlates with
clinical, immunologic, and virologic markers
William Blattner. 725 W Lombard Street, Baltimore MD 21201, USA
Object: To test the hypothesis that immunologic and virologic markers differ
among patients who experienced symptomatic versus asymptomatic acute HIV-1
seroconversion. and impact the natural history of HIV-1 infection.