Abstract Book Vol. 2 [International Conference on AIDS (14th: 2002: Barcelona, Spain)]

XIV International AIDS Conference

Abstracts ThPeB7225-ThPeB7228 393

Results: 86 subjects were evaluated. The median age at admission was 14
months. There were 1,216 observations, median of 9 visits/patient. Lost to follow
up, defined as no visit occurring in a 6 months interval, was 7.0%. Progression
to Category C and death were observed in 19 and 10 patients, respectively. WAZ
< - 2 at admission was significantly correlated with risk for disease progression
or death in an univariate analysis (log rank p<0.0001) and it remained significant
when adjusted for age at admission and LT CD4+ counts (RR: 9.50, 95%Cl: 2.49
- 36.26).
Conclusion: In this cohort of Brazilian vertically infected children malnutrition
was a strong prognostic marker for disease progression or death. These findings
emphasize the need for an early and aggressive nutritional intervention among
HIV-infected children.
Presenting author: Jorge Pinto, Av. Alfredo Balena 190/40 andar, Belo Horizonte, MG, 30130-100, Brazil, Tel.: + (55-31) 3248 9822, Fax: + (55-31) 3273
0422, E-mail: [email protected]
ThPeB7225 Detection of ex-vivo activated HIV-specific CD8+ T
cells in HIV-infected children using the Elispot
assay: correlation with disease stage and
antiretroviral treatment
F. Buseyne', B. Corre1, F. Porrot 1, N. Bellall, M. Burgard2, C. Rouzioux2,
S. Blanche2, Y. Rivibre2. Institut Pasteur, Laboratoire dImmunopathologie
virale, B.timent SIDA et Retrovirus, Institut Pasteur, 28, rue du Dr Roux, 75015
Paris, France; 2 Hopital Necker, Paris, France
Background: HIV-specific CD8+ T cells play a major role in the containment of
HIV replication and disease progression. New techniques with high sensitivity are
now available for the study of virus -specific CD8+ T cells response. We present
data obtained with the IFN-y Elispot to study anti-HIV CD8+ T cell response of
freshly isolated PBMC from infected children.
Methods: Freshly isolated PBMC from 59 HIV-infected children were assessed
for their IFN-y production in response to HIV-proteins expressed by recombinant
Vaccinia viruses.
Results: The three structural proteins Env, Gag and Pol were the more frequently
recognized (69, 85, 81% of responders; median: 337, 489, 527 SFC/106 PBMC,
for Env, Gag, and Pol respectively). Responses of low intensity were detected
against Nef and Vif (40 and 25% of responders; median: 96 and 88 SFC/106
PBMC, respectively), but not against Tat and Rev. Frequency of CD8+ T cells
specific for Env-Gag-Pol were positively correlated with age for both children receiving HAART or not. In children under HAART, the intensity of the Env-Gag-Pol
specific CD8+ T cell response was higher in children with incomplete viral suppression
Conclusion: Evaluation of ex vivo activated HIV-specific CD8+ T cells can be
routinely performed in infected children. This antiviral response increases with
age and appears to be stimulated by viral replication.
Presenting author: Florence Buseyne, Laboratoire d'lmmunopathologie virale,
Bdtiment SIDA et Retrovirus, Institut Pasteur, 28, rue du Dr Roux, 75015 Paris,
France, Tel.: [phone], Fax: [phone], E-mail: [email protected]
ThPeB7226 Cerebral metabolite abnormalities in HIV-infected
children
M.A. Keller', T.N. Venkatraman1, M.A. Thomas2, A. Deveikis3, C. LoPresti2,
J. Hayes', N. Berman1, I. Walot', D. Osborn', S. Padilla', J. Johnston-Jones',
T. Ernst4, L. Chang4. 1Harbor-UCLA Medical Center, Harbor-UCLA Medical
Center, Bldg. N-25, Box 468, 1000 W Carson Street, Torrance, United States;
2University of California, Los Angeles, Los Angeles, CA, United States; 3Miller
Children's Hospital, Long Beach, CA, United States; 4Brookhaven National
Laboratory, Upton, NY, United States
Cerebral metabolite concentrations measured by Proton Magnetic Resonance
Spectroscopy (1 H MRS) have been shown to be abnormal in HIV-infected children. It is unclear how these findings relate to clinical severity or brain function.
Method: Localized 1H MRS and neurodevelopmental assessment were performed in 20 HIV-infected children and 13 control children, ages 6-16 years; measurements were correlated with age, CD4, and log viral load. MRI/MRS studies were conducted on a GE 1.5T MRI/MRS scanner using a water suppressed
PRESS sequence; voxel locations included midfrontal cortex (MFG), left and right
frontal white matter (LFW, RFW), right basal ganglia (RBG), and right hippocampal (RHIP) volumes. The LC-Model algorithm provided absolute concentrations
(mM) and metabolite/creatine (CR) ratios.
Results: Compared to controls, patients had decreased choline (CHO) in LFW
(-11.3%, P=0.0390). Patients with an HIV RNA viral load >10,000 had even
lower CHO (-14%, P=0.009) and CR (-13.4%, P=0.026) in the RBG compared
to controls. N-acetyl aspartate (NAA) correlated with age in RFW (r=0.599,
p=0.039), LFW (r=0.645, p=0.024) and RHIP (r=0.696, p=0.017) for controls,

not for HIV patients. The patients' RHIP total N-acetyl compounds showed inverse relationship with age (r= -0.590, p=0.034), but a positive correlation with
CD4 (r=0.569, p=0.043), and their inositol (INS) also correlated with age (LFW:
r=0.653, p=0.006: RFW: r=0.711, p <0.001; RFW INS/CR: r=0.66, p=0.002). Log
viral load correlated negatively with RBG INS (r= -0.520, p= 0.047) but positively
with MFG CHO (r= 0.494, p=0.037). HIV patients had poorer spatial memory
(CMS) relative to controls (p=0.031 1).

Conclusions: These data suggest that normal brain development may be affected in children infected with HIV, evidenced by lower CHO, CR and lack of
age-related increases in the neuronal marker NAA. Furthermore, patients with
higher viral load and lower CD4 might be more severely affected.
Presenting author: Margaret Keller, Harbor-UCLA Medical Center, Bldg. N-25,
Box 468, 1000 W. Carson Street, Torrance, United States, Tel.: [phone],
Fax: [phone], E-mail: keller@ humc.edu
ThPeB7227 Evaluation of an ultrasensitive p24 antigen assay
(UPTA) for use in the diagnosis of pediatric HIV-1
infection
D.G. Withum', S.A. Fiscus2, R.A. Respess', A. Cachafeiro2, D.R. Newman',
I. Cabruja3, B.M. Branson', O.E. Varnier4, T.J. Dondero1. 'US Centers for
Disease Control and Prevention, US Centers for Disease Control and
Prevention, 1600 Clifton Road, Mail-stop E-46, Atlanta, Georgia, 30333, United
States; 2 University of North Carolina at Chapel Hill, Chapel Hill, North Carolina,
United States; 3 Perkin Elmer Life Sciences, Milano, Italy; 4 University of Genova
School of Medicine, Genova, Italy
Background: Detection of HIV-1 by PCR is useful for the diagnosis of pediatric
infection. Commercially available PCR methods are based on elaborate amplification approaches, are expensive, and require skilled technicians with specialized
equipment and facilities, making their use prohibitive for many resource-poor areas. An inexpensive and less complex approach for the diagnosis of pediatric
HIV-1 infection would be valuable. An inexpensive ultrasensitive p24 assay, combining heat denaturation and ELISA technology, has been previously described,
however transfer of this approach has been limited. Methods: A laboratory protocol, test kit and software for an ultrasensitive p24 assay (UPTA) were developed.
To estimate the sensitivity and specificity of UPTA for use in pediatric
HIV-1 diagnosis, 97 specimens (60 HIV-1 subtype-B positive [by PCR or culture]
and 37 uninfected) from infants born to infected mothers were tested by UPTA.
Qualitative UPTA results were compared to known infection status determined by
PCR testing and clinical follow-up.
Results: UPTA correctly classified 53 of 60 HIV-1 infected specimens (sensitivity
= 88.3%) and all 37 uninfected specimens (specificity = 100%). The probability
the specimen was truly infected if the UPTA test was positive (predictive value
positive) was 100%. Probability the specimen was truly HIV-1 negative if UPTA
was negative (predictive value negative) was 84%. The association of UPTA results and known infection status of the specimens was statistically significant (Chi
Square = 72.05 [p < 0.0001]). 5 of 7 false negative UPTA tests occurred on specimens from very young infants (<1 month) also negative by PCR.
Conclusions: UPTA may be useful as a lower cost, lower technology approach
to assist the diagnoses in HIV-1 in infants. Research is needed to corroborate
these findings, determine at what age UPTA becomes reliably detectable, and to
evaluate the use of UPTA on infant specimens from non-B subtype areas.
Presenting author: David Withum, US Centers for Disease Control and Prevention, 1600 Clifton Road, Mail-stop E-46, Atlanta, Georgia, 30333, United States,
Tel.: [phone], Fax: [phone], E-mail: [email protected]
ThPeB7228 Identification of novel HIV-1 from perinatally
infected infants that can mediate CD4-independent
infection of the CD8+T lymphocytes: implications
for AIDS pathogenesis
B. Zerhouni, K. Saha, J. Zhang. Pediatrics/Molecular Medicine, The Ohio State
Univ, Children's Research Institute, Columbus, OH, United States
Background: HIV-1 primarily infects CD4+T cells using CD4 as a receptor. In
contrast, CD8+ lymphocytes play an important protective role against HIV-1. However, these cells eventually fail to protect against the development of AIDS. The
exact reason of their failure remains unclear. We have recently reported isolation
of HIV-1 mutants from an AIDS patient that could infect CD8+ cells independent of
CD4. Whether CD8-tropic variants are rare or commonly present in other patients
remains unknown.
Objective: To examine whether CD8-tropic HIV-1 may exist in the children infected perinatally.
Methods: We tested the viral quasispecies from HIV-1-infected subjects for the
presence of CD8-tropic variants (AIDS Research & reference Program, NIH).
CD8-tropism was tested in primary CD8+cells as well as in established CD8+Tcell lines. CD4-independent infection of the CD8+cells was further confirmed by
infection of CD8+ cells in the presence of anti-CD4 antibodies. The envelope sequences of the CD8-tropic isolates were sequenced and analyzed.
Results: CD8-tropic variants were isolated from two infants infected with HIV-1
of clade B. Like the previously identified CD8-tropic variants, these viruses also
maintained the CD4-tropism. Infection of the CD8+ cells could not be blocked
with anti-CD4 antibodies. Remarkably, CD8-tropic isolates from one infected infant induced striking syncitia in the CD8+cells and were acutely cytopathic for the

CD8+cells. Although no specific motif for CD8-tropism was detected in the envelopes, several possible "hot spots" were noticed in the gp41 regions of the CD8 -tropic viruses. CD8-tropic isolates from one patient could enter CD4-negative
quail cells using only CXCR4.
Conclusion: We provide evidence that CD4-independent CD8-tropic mutants
may exist in HIV-1 infected infants. Transmission of HIV-1 that can directly infect
CD8+ cells may have implications in AIDS pathogenesis in the pediatric patients.