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

12th World AIDS Conference

Abstracts 13324-13328 151

13324 Survival of healthy HIV-infected skin-test anergic and
non-anergic adults from a voluntary counseling and
testing clinic in Kampala, Uganda
Peter Nsubuga, A. Okwera, J.L. Johnson, C. Whalen, R. Mugerwa, J. Ellner.
Makerere University PO Box 663 Kampala, Uganda; Case Western Reserve
University Cleveland OH, USA
Objective: To compare the two-year survival of healthy Ugandans with skin test
anergy and non-anergy to tuberculin and candida at baseline.
Design: Prospective cohort study.
Methods: Between 1993 and 1995, 321 (mean age 30.6 ~ 6.5 yrs., 31% male)
anergic and 461 (30.4 ~ 6.6 yrs., 33% male) non-anergic to tuberculin and candida skin tests were recruited from an anonymous voluntary counseling testing
centre in Kampala, Uganda. These subjects were enrolled in the non-intervention
arm of a three-year clinical trial to assess the efficacy of TB chemoprophylaxis.
They received regular medical care for concomitant illnesses throughout the study
period.
Results: At baseline 15.9% of the non-anergic compared with 23.9% of the anergic had a prior history of herpes zoster (p <.001); and 35.5% of the non-anergic
compared with 45.5% of the anergic had present or past genital ulcers (p <.001).
The mean CD4 cell count in the non-allergic was 529 ~ 380 cells/ti. and 359 ~
313 cells//pl in the anergic (p <.001). Both groups had similiar haemoglobin, total
white cell counts, weights and body mass indices. After two years of follow-up,
there had been 68 deaths in the non-anergic compared with 96 in the anergic for
a product limit estimate log rank x2 of 23.0 (p -t.001). In a proportional hazards
model controlling for baseline CD4 cell count, the relative hazard for anergic was
1.14 p <.093, -2 Log-likelihood 8565, p -d.001, from a univariate relative hazard
of 1.24 (p --.004), 2 Log-likelihood of 8717.
Conclusions: In HIV-infected persons, TB was independently associated with
an increased risk for death, especially when the CD4+ count was greater than
200 cells//iL. In populations with high prevalence of co-infection, preventive
therapy for TB in HIV-infected persons may have beneficial effect on individual
survival.
S13325 Risk factors for progression to primary and subsequent
AIDS diagnoses in the Tricontinental Seroconverter Study
Birgit Van Benthem', K.A. Page-Shafer2, M.T. Schechter3, J.M. Kaldor4,
R.A. Coutinho5, G.J.P. Van Griensven5. 'Niewu Achtergracht 100 1018WT
Amsterdam; 5Municipal Health Service, Amsterdam, The Netherlands; 2Center
for AIDS Prevention Studies, San Francisco, USA; 3Centre of Excellence in
HIV/AIDS, Vancouver, Canada; 4National Centre on HIV-lepidemiology and
clinical Research, Sydney, Australia
Background: The objective was to evaluate risk factors for progression from
seroconversion to specific AIDS diagnoses like Kaposi's sarcoma (KS), Pneumocystis carinii pneumonia (PCP) and other opportunistic infections (01) and to
investigate changes in progression to specific AIDS diagnoses over time.
Methods: The population comprised 426 homosexual men with known intervals of seroconversion from five prospective cohort studies registered in the
Tricontinental Seroconverter Study. Active follow-up and matches with registries
determined the occurrence of AIDS. The censoring date was ultimately 1-7-1996.
Primary and secondary AIDS diagnoses combined were endpoints for survival
analysis. We examined effects of site, age, year of seroconversion and symptomatic primary infection (SPI) on progression time from seroconversion to KS,
PCP and 01 in a Cox model. Calendar time was included into the model as a
time-dependent variable to investigate time trends in progression rates to KS,
PCP and 01 and to correct indirectly for treatment.
Results: Of 426 men, 15% developed KS, 18% PCP and 30% 01. Controlling
for calender time and year of seroconversion, older age at seroconversion was
associated with faster progression to KS (adjusted relative hazard (ARH) 1.59 per
10 years, 95% confidence interval (Cl): 1.07-2.39), but not to PCP and 01 (ARH
1.04, Cl: 0.71-1.52 and ARH 0.94, Cl: 0.70-1.26, respectively). SPI was associated with faster progression to 01 and KS in multivariate analyses (ARH 1.89, CI:
1.11-3.21 and ARH 2.51, Cl: 1.22-5.16, respectively), but not to PCP (ARH 1.32,
Cl: 0.59-2.94). Although not significant, the analyses indicated that progression
rates to PCP decreased, progression to KS increased and progression to 01
remained stable over the years.
Conclusions: Risk factors for progression to specific AIDS diagnosis differed
between diagnoses: older age at seroconversion caused faster progression to
KS but not to PCP and 01. SPI was associated with progression to KS and
01. Although analyses were censored before protease inhibitors were widely
introduced, the data suggest that progression rates to PCP decreased, to KS
increased and to 01 remained stable over time. Because risk factors differed
between AIDS diagnoses, individual AIDS diagnoses should be considered as
outcomes.
13326 CD4 cell count and viral load in HTLV-II and HIV
coinfection in injection drug users

David Vlahov', Hong Zhang', C. Flynn2, K.E. Nelson'. 'Johns Hopkins School
of Hygiene & Public Health, Baltimore, Maryland; 2Maryland State AIDS
Adminstration, Baltimore, Maryland, US
Objectives: To examine CD4+ cell counts differed by HTLV-II serostatus, after
accounting for HIV plasma viral load, serum immune activation markers, and age.
Design: Cross-sectional

Methods: Participants in this study were selected from the ALIVE Study, a
natural history study of HIV among injection drug users (IDUs) recruited in 1988.
Eligibility for this analysis include age >18, history of IDU, HIV seropositive,
and specimen sufficient to conduct plasma viral load. All persons with coinfection were selected and from those with single HIV antibody was determined by
commerical ELISA and Western Blot, HTLV serostatus by p21e ELISA (Cambridge Bioscience), followed by a HTLV 1/Il Western Blot. CD4 cell counts were
determined using flow cytometry, and plasma viral load was determined using
a second generation bDNA assay. Statistical analysis compared median assay
using Wilcoxon 2 Sample Test, and linear regression was used to examine the
effect of HTLV serostatus on CD4 cell count after accounting for HIV viral load
and beta 2 microglobulin.
Result: In linear regression with CD4 cell count treated as the continuous
outcome variable, HTLV II serostatus did not provide discriminatory power univariately or after controls for plasma viral load, /12 microglobulin, and age; the
only significant association with CD4 cell count was plasma viral load.
Conclusion: Unlike HTLV-I, HTLV-II does not appear to be associated with
differences in CD4+ cell count among HIV-1 infected drug users. Prospective
studies are warranted to confirm these findings.
S13327 Chemokine receptor CCR2b 641 polymorphism and its
relation to CD4 T-cell counts and disease progression in
a Danish cohort of HIV infected individuals
Jesper Eugen-Olsen', A.K.N. Iversen2, T.L. Benfield', U. Koppelhus1,
P. Garred3. 'Department of Infectious Diseases 144, Copenhagen University
Hospitals, Hvidovre; 2Dept of Infectious Diseases Rio Hospital, Copenhagen;
3Dept of Clinical Immunology Copenhagen, Denmark
Background: We have investigated the role in Danish individuals of the recently
described mutation in CCR2b named 641 in relationship to HIV resistance, CD4
T-cell counts and disease progression.
Methods: The 641 allele was detected by PCR based methods. DNA from
long term non-progressors was sequenced for possible mutations in the CXCR4
and CCR5 genes. Two-hundred and fifteen Danish individuals were analysed
for 641 allele frequency. Disease progression was followed in 105 HIV-1 positive
homosexual Danish men from their first known positive HIV-1 test and up to 11
years. In 87 individuals the CD4 T cell count was monitored closely.
Results: There was no significant difference in 641 allele frequency between
HIV-1 seropositives (0.08), high risk HIV-1 seronegatives (0.11) and blood donors
(0.06). No significant difference was observed in annual CD4 T cell decline, CD4
T cell counts at the time of AIDS, in AIDS-free survival as well as survival with
AIDS between 641 allele carriers and wildtypes. However, pooling the 641 allele
carriers with CCR5A32 allele carriers into one group indicated that 641 allele
may have a slight influence on disease progression. Furthermore, among 9 long
term non-progressors 6 carried CCR5A32 allele and two carried the 641 allele.
By contrast, none of 9 fast progressors carried variant alleles (P = 0.0004). Long
term non-progression could not be explained by CXCR4 polymorphism or other
polymorphisms in the CCR5 gene than the CCR5A32 allele.
Conclusion: We were not able to detect any significant independent effect of
the 641 allele on AIDS free and overall survival and annual CD4 T cell decline in
this cohort. However, our results do not contrast the view that the 641 allele may
have some impact on HIV disease progression.
S133281 HIV-1 plasma viral load in tuberculosis patients with
single HIV-1 and dual HIV-1/HIV-2 infections, Abidjan,
C6te d'lvoire
Stefan Wiktor1, M. Sassan-Morokko2, L. Abouya2, A. Ackah2, A. Akaki2,
R. Lobognon2, 0. Tossou2, S. Koblavi2, J. Nkengasong2, D. Coubibaly3,
I.M. Coubibaly3, A.E. Greenburg4, T.H. Roelhs4. 1'01 BP 1712 Abidjan 01;
2Project Retro-CI, Abidjan; 3National AIDS/STD/TB/HIV Control Program,
Abidjan, Cote D' Ivoire; 4Centers for Disease Control And Prevention, Atlanta,
USA
Objective: To compare HIV-1 plasma viral load in tuberculosis (TB) patients with
PCR-confirmed HIV-1/HIV-2 dual infections with those who are HIV-1 seropositive.
Methods: Among HIV-infected TB patients enrolled in an ongoing cotrimoxazole chemoprophylaxis trial, CD4-count-matched blood samples obtained at
onset of TB therapy from HIV-1 seroreactive (Sero-1) and HIV-1/2 dually reactive (Sero-D) patients were selected. For a subset of patients (10 Sero-1, 10
Sero-D), 12 months post-onset TB therapy specimens were also available. HIV
seroreactivity was determined by a combination of monospecific peptide enzyme
immunoassays. In Sero-D samples, HIV-infection status was determined by PCR
using HIV-1 protease and HIV-2 LTR and protease genes primers. Plasma viral
load was measured by the Amplicor Monitor Assay (Roche).
Results: The mean age (37 vs. 34 years) and percent of males (60 vs 62)
was not significantly different in the 41 Sero-D and 45 Sero-1 patients. Among
patients with Sero-D samples, HIV- 1 and HIV-2 PCR were both positive in 26
(58%) samples (Sero- D/PCR-D), HIV-1 alone in 14 (Sero-D/PCR-1) and HIV-2

alone in one sample (Sero-D/PCR-2). Mean loglo) HIV-1 viral load were similar
for Sero-D/PCR-D (4.70 log10 copies/ml; range: 2.69-5.8), Sero-1 (4.62 loglo
copies/ml; range: 2.35-6.43) and Sero-D/PCR-1 (4.49 loglo copies/ml; range:
2.51-5.63); (all p > 0.05). The mean change in viral load between onset of TB
therapy and 12 months post-onset of therapy was similar for the 7 Sero-D/PCR-D,
three Sero-D/PCR-1, and 10 Sero-1 patients with available samples (logio HIV
viral load 0.57, 0.67, and 0.34 copies/ml respectively).