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

148 Abstracts 13310-13314

12th World AIDS Conference

Conclusion: The magnitude of the adjusted risk of mortality for women with
albumin less than 3.5 gm/dl was similar to that for patients with CD4 < 200 or
viral load >500k. Baseline serum albumin is a major independent predictor of
survival in HIV+ women that is associated with survival even within the 'normal'
range (ie. >3.5 gm/dl).
S13310 HHV-8 seropositivity and risk of developing Kaposi's
sarcoma (KS) and other AIDS-related diseases among
individuals with known dates of HIV seroconversion
Giovanni Rezza1, M. Dorrucci1, P. Pezzotti1, M. Andreoni2, B. Ensoli3. 1IStituto
Superiore di Sanita'-COA V.LE Regina Elena 299, Roma; 2Universita' di Tor
Vergata, Roma; 3Lab. Virologia-lstituto Superiore di Sanita', Roma, Italy
Objectives: To estimate progression rates to KS and other AIDS-related diseases
in HIV-positive individuals coinfected/non-coinfected with HHV-8. To identify other
determinants of progression to KS in coinfected individuals.
Design: longitudinal study.
Methods: We studied individuals belonging to different exposure categories to
HIV (i.e., homosexual relationships, injecting drug use, and heterosexual contact)
with a negative HIV test followed by a positive test within two years. HHV-8 antibody testing was performed by IFA on the available serum sample stored closest
to the first HIV positive test. Rates of progression to KS and other AIDS-related
diseaseswere estimated through time-to-event statistical methods; CD4 decline
was estimated through linear regression analysis.
Results: 21 cases of KS and 95 non-KS AIDS cases occurred among the 366
participants. 140 (38.2%) participants were found to have anti-lytic HHV-8 antibodies, which were detected in 61% of the 171 homosexual men, in 15% of the
133 IDUs, and in 22% of the 56 heterosexual contacts. The estimated progression
rate to KS among HHV-8 coinfected persons was nearly 25% within 10 years of
HIV seroconversion. Of those who developed KS, only one individual was HHV-8
negative at time of testing. The risk of KS increased along with increasing HHV-8
antibody titers and with decreasing CD4 count among coinfected individuals.
The presence of HHV-8 antibodies was not significantly associated with a faster
progression to other AIDS-defining illnesses [adjusted relative hazard: 1.14 (95%
CI: 0.72-1.80)], and the rate of CD4 cell decline of the individuals with antibodies
to HHV-8 was similar to that of HHV-8 antibody negative persons.
Conclusion: Approximately one-fourth of HHV-8 coinfected individuals develop
KS within ten years of HIV seroconversion. Higher antibody titers appear to predict
faster progression to KS. There is no evidence that co-infection with HHV-8 is a
cofactor for the CD4 decline or progression to other AIDS-defining illnesses.
13311 i Chemokine receptor genotypes and HIV disease
progression: A preliminary meta-analysis
John P.A. loannidis1, D.G. Contopoulos-loannidis2. 1HIVRB - NIAID Solar 2 C31
Bethesda MD 20892; 2 George Washington Univ., Washington DC, USA
Objective: To evaluate whether chemokine receptor genotypes including heterozygosity for the CCR5 32-bp deletion (CCR5 A32) and the presence of the
CCR 641 mutation affect the rate of disease progression to AIDS (1993 CDC
definition) among HIV-infected patients.
Methods: Preliminary meta-analysis of 8 cohorts with n = 2,741 patients with
known CCR5 genotype (45% seroconverters [SC]) and 6 cohorts with n = 2,089
patients with known CCR2 genotype (43% SC). Due to measurement error, it
was anticipated that the observed strength of associations should be attenuated among seroprevalent (SP) patients, therefore SP patients were analyzed
separately from SC.
Results: Heterozygosity for CCR5 A32 conferred a highly significant protective
effect (odds ratio 0.69 [95% CI, 0.61-0.79] by fixed effects, 0.68 [95%, 0.58-0.80]
by random effects) with borderline heterogeneity (0.1 < p < 0.2) between cohorts.
As expected, the benefit was more clear among SC (odds 0.65 [0.53-0.80]) than
among SP patients (odds 0.73 [0.61-0.86]). CCR2 641 did not confer overall a
statistically significant protective effect (odds ratio 0.93 [0.81-1.06]; no heterogeneity [r2 = 0]), although the more accurate SC cohorts showed a more clear
protection (odds ratio 0.80 [0.65-0.99]; SP cohorts odds ratio 1.03 [0.86-1.23]).
The size-weighted difference in early viral load between CCR5 wild-type homozygotes and heterozygotes was a 2.5 fold lower median viral load in the latter group
which translates to an odds ratio of 0.7 for disease progression.
Conclusions: CCR5 A32 heterozygosity offers a protective benefit for disease
progression which is at least partially mediated through an effect on viral load.
The evidence for CCR2 641 is ambiguous, ranging from a sizable effect to no
effect at all. To avoid a posteriori selective reporting and subgroup analyses which
may yield misleading results, investigators in the field are invited to contribute
data to this ongoing international collaborative meta-analysis.
S13312 Evidence from the Australian Ashkenazi Jewish
population suggests an Eastern European Ashkenazi
origin of CCR5-A32A32

Marc Buhler', A. Proos2, V. Howell2, B. Bennetts1, L. Burnett2, G. Stewart1.
1Dept. of Clinical Immunology, Westmead Hospital NSW; 2Dept. of Molecular
Genetics, ICPMR, Westmead Hospital, Westmead NSW, Australia
Background: Homozygotes for the 32bp deletion mutation in the chemokine
receptor CCR5, the co-receptor for HIV virion entry into cells, are highly protected
from HIV infection whilst disease progression to AIDS is slower in heterozygotes.

Distribution of CCR5-A32 indicates a European Caucasian origin for the mutation
with a geographic dine within Europe and a high frequency in Ashkenazi Jews.
Methods: We report here on the CCR5-A32 genotype in 1379 Australians, 808
of whom are of Ashkenazi Jewish background, 129 "mixed" or Sephardic origin
Jews, and 442 non-Jewish people. Typing for CCR5-A32 was done by standard
PCR methods.
Results: Non-Jewish samples (n = 442) contained 67 heterozygotes and 3 homozygotes for CCR5-A32 (allele freq. = 0.08), compared with an allele frequency
of 0.14 (219 heterozygotes and 21 homozygotes) amongst the 937 individuals of
Jewish background (p = 0.0001). Analysis of birthplace of parents and grandparents showed a marked increase in CCR5-A32 in Eastern European Ashkenazim
compared with those of Western European or South African origin and with
Sephardic Jews. Several microsatellite markers in the 3p21.3 region are being
used to assess a common ancestral origin for individuals with the CCR5-A32
mutation using the 24 homozygotes and 286 heterozygotes identified in this study.
Conclusion: These data support the hypothesis that the A32 mutation arose
in Eastern European Ashkenazi Jews, an origin which may explain, at least in
part, the global distribution of the mutation.
S13313 HIV disease progression following newly acquired HIV
infection in Australia, 1991-1997
Ann McDonald1, J. Cui1, J.M. Kaldor2. 1NCHECR 376 Victoria Street
Darlinghurst Sydney NSW 2010; 2National HIV Surveillance Co Sydney, NSW,
Australia
Objective: Establish a new cohort of people in Australia with HIV infection and
describe the pattern of HIV disease progression from the known date of HIV
infection.
Methods: Cases of newly acquired HIV infection were defined as cases of
diagnosed infection with a negative test or diagnosis of an HIV seroconversion
illness within one year of diagnosis. Cases of newly acquired infection were linked
to the corresponding AIDS diagnosis by matching on name code and date of
birth. AIDS cases following newly acquired infection were analysed with respect to
evidence of newly acquired infection, age and CD4+ cell count at HIV diagnosis.
Median duration to AIDS was estimated using the Kaplan-Meier method.
Results: In 1991-97, 1,144 cases of newly acquired HIV infection (93% male)
were diagnosed in Australia. By 31 December 1997, 70 cases (6.1%) (94% male)
were diagnosed with AIDS. AIDS was diagnosed within 2.5 years and 4.8 years
of HIV acquisition in 5% and 10% of cases, respectively. No difference in the rate
of progression to AIDS was found between cases diagnosed with HIV prior to
or after 1993. Cases 30 years or older at HIV acquisition were more likely than
those aged less than 30 years to progress to AIDS (p = 0.05). No difference in
the rate of progression to AIDS was observed between cases with or without a
diagnosis of HIV seroconversion illness (p = 0.166). CD4+ cell count <500 at HIV
diagnosis was associated with progression to AIDS (p = 0.001).
Conclusion: National surveillance data suggests that, in Australia, HIV disease
progression is related to age and CD4+ count at HIV acquisition and that the rate
of disease progression has not changed over the past 7 years.
13314 Comparison of viral load levels between cohorts infected
with HIV-1 subtypes B and C
Noya Galai1, E. Shahar2, O. Naor3, A. Barzilai4, S. Pollack2, A. Kalinkovich5,
Z. Bentwich5. 1Department of Epidemiology, Ben-Gurion Univ. Beer-Sheva POB
653; 2Ramban Med Center, Haifa; 3Hadassah Med Center, Jerusalem; 4Shiba
Medical Center, Tel-Hashomer; 5Ruth Ben-Ari Instit. of Clinical Immun., Rehovot,
Israel
Objective: To assess potential differences in viral load levels between a cohort of
Ethiopian immigrants (ETH) in Israel, infected with African HIV-1 subtype C and
a cohort of non-Ethiopian Israelis (NETH) infected with HIV-1 subtype B.
Design: Cross sectional study.
Method: Measurements of viral load (VL) and CD4 counts were obtained for
149 ETH and 121 NETH followed in four centers in Israel. All participants had
no prior treatment with protease inhibitors (triple combination therapy). Plasma
VL was determined by NASBA on blood samples kept at-70 C. Simple statistical
comparisons (t-tests and X2 tests) were performed on groups stratified by CD4
counts: (a) < = 100, (b) 101-200, (c) 201-350, (d) > = 351.
Results: For individuals with detectable VL levels the mean log0o (VL) levels,
by CD4 category (a) - (d), for ETH Vs NETH were: (5.2 Vs 5.3), (4.7 Vs 4.7),
(4.5 Vs 4.7) and (4.5 Vs 4.6) with no statistically significant differences. Overall,
29% of all observations in each cohort had undetectable VL. The two cohorts
had similar rates of previous treatment with antiretrovirals (AZT, ddl, ddC): 48%
among the ETH and 41% among the NETH; Hence, this was not a confounding
factor.
Conclusions: The results indicate that within given CD4 categories, viral load
levels are very similar in Ethiopian immigrants infected with HIV-1 Subtype C
and non-Ethiopian Israelis infected with HIV-1 subtype B. These findings support
previous suggestions that the natural history of infection with these two subtypes

is similar, once the individuals are in the same environment. The findings have
direct implication for the planning of treatment strategies for these individuals.