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

XIV International AIDS Conference

Abstracts WePeC6227-WePeC6231 137

WePeC6227 Prognostic value of plasma HIV-1 RNA among
patients receiving HAART in the Multicenter AIDS
Cohort Study (MACS)
P.M. Tarwater1, J.W. Mellors2, M.E. Gore1, J.P. Phair3, J.B. Margolick 1,
R. Detels4, A. Munoz1. ' Johns Hopkins School of Public Health, Johns Hopkins
Univ., School of Public Health, 615 N. Wolfe St., E7137, Baltimore, MD, 21205,
United States; 2University of Pittsburgh Graduate School of Public Health,
Pittsburgh, PA, United States; 3Northwestern University Medical School,
Chicago, IL, United States; 4 UCLA School of Public Health, Los Angeles, CA,
United States
Background: The value of plasma HIV-1 RNA as a prognostic marker is reduced
after initiation of highly active antiretroviral therapy (HAART) because of the profound effect HAART has on plasma HIV-1 RNA levels. However, the predictive
value of plasma HIV-1 RNA in patients receiving HAART has not been carefully
assessed. The study objective was to determine the prognostic value of plasma
HIV-1 RNA levels for AIDS development stratified by CD4 cell count in patients
receiving HAART
Methods: The study population comprised 387 men enrolled in the MACS who
were AIDS-free and initiated HAART between 1995 and 2001. Follow-up until
AIDS diagnosis (n=36, 9%) or the last AIDS-free visit was included. To determine
the predictive value of combining HIV-1 RNA and CD4 count, we constructed
a regression tree using eight categorizations of each variable. Regression tree
methods using recursive partitioning at the eight pre-specified cut off values of
both HIV-1RNA and CD4 count were used.
Results: The median time of follow-up after initiation of HAART was 3.8 years.
For all men with detectable HIV-1 RNA (> 50 copies/ml) on HAART, the relative
hazard (RH) of AIDS was 2.1 for each 10-fold increase in HIV-1 RNA (p <0.001),
which was equivalent to the RH of AIDS (2.1) associated with a CD4 cell count
that was lower by 166 cells/uL. Among men on HAART with CD4 cell counts >
250 cells/uL, an HIV-1 RNA level > 500 copies/ml was associated with a RH of
AIDS of 4.7 (95% confidence interval: 1.5, 15), compared with an HIV-1 RNA <
500 copies/ml. HIV-1 RNA levels were not prognostic for AIDS among men with
CD4 cell counts < 250 cells/uL.
Conclusion: Plasma HIV-1 RNA levels on HAART are prognostic for AIDS development among men with CD4 cell counts >250 cells/uL. These data indicate
that plasma HIV-1 RNA levels on HAART should influence the decision when to
modify therapy to reduce the risk of AIDS.
Presenting author: Patrick M. Tarwater, Johns Hopkins Univ., School of Public
Health, 615 N. Wolfe St., E7137, Baltimore, MD, 21205, United States, Tel.: +1
410 502 9771, Fax: [phone], E-mail: [email protected]
WePeC6228 Retrospective analysis of the HIV-1 RNA test (viral
load) using the model performance evaluation
program data from 1998 to 2000
M.B. Kahn, M.K. Simon, G.D. Cross, W.O. Schalla, H.B. Lipman. Centers for
Disease Control and Prevention, Atlanta, United States
Background: To evaluate laboratories performing HIV-1 related testing, the Centers for Disease Control and Prevention established the Model Performance Evaluation Program (MPEP) which includes performance evaluation of labs that test
for HIV-1 viral RNA (viral load). All samples used in the MPEP HIV-1 RNA performance surveys are undiluted, unpooled plasma obtained from individual donors
who are either HIV-1 infected or uninfected. Some samples are duplicated within
panels and replicated between shipment panels to evaluate testing results reproducibility.
Methods: A retrospective analysis of 6 biannual shipments of MPEP HIV-1 viral
RNA data was performed. We compared results reported by labs using 3 test kit
types: Roche~ Amplicor HIV-1 MonitorTM (RT-PCR), Bayer HIV-1 RNA 3.0 Quantitative Assay (bDNA), and Organon Teknika NucliSens~ HIV-1 QT (NASBA). Variability within and between test kits was investigated.
Results: For each of the 9 positive samples, the median reported viral load varied significantly between test kits (all Kruskal-Wallis p-values < 0.0001). To control for an assumed constant CV, a square root transformation was performed.
The transformed data still showed significant differences between test kits (8 out
of 9 Bartlett's test p-values < 0.0002). The differences in precision were further
demonstrated by the variation in CVs, ranging from 22% to 124%.
Conclusions: Reported viral load testing results differed between the 3 test kits
used. Some labs using the same test kits showed considerable differences in reported test results. There was significant variability in the reported testing results
both within and between the test kits used in the surveys. We concluded that if the
HIV-1 viral RNA test is being used to monitor a patient's viral load as a prognostic
tool, i.e., advancement of disease, or to monitor treatment efficacy, then viral load
testing should be performed by the same lab using the same manufactured test
kit to yield comparable results.
Presenting author: Marianne Simon, 4770 Buford Hwy., N.E., MS-G23, Atlanta,
GA 30341-3717, United States, Tel.: [phone], Fax: [phone], E
mail: [email protected]

WePeC6229 Studies on acute primary HIV infection with HIV-1
subtype C infection in India
R.S. Paranjape1, S.M. Mehendale1, M.R. Thakar1, R.G. Brahme1,
B.M. Mahajan1, S.S. Kulkarni1, S.PR Tripathy1, R.C. Bollinger2. 1NationalAIDS
Research Institute, National AIDS Research Institute, G-73, M.I.D.C., Bhosari,
Pune 411026, India; 2Johns Hopkins Medical School, Baltimore, United States
Background: Immunological and virological events during acute primary infection may impact HIV disease progression. However, few studies describe whether
the characteristics of acute primary HIV infection with non-subtype B HIV-1 follow
the same pattern as subtype B.
Methods: Consenting HIV antibody negative/p24 antigen positive individuals, as
well as those with documented HIV antibody seroconversions, from a prospective
cohort study of STD clinic attendees were followed for up to two years. In addition to collection of clinical data, lymphocyte subpopulations were enumerated
by flow cytometry. HIV-1 plasma viral load (VL) was measured and the subtype
of the HIV-1 env region was determined by heteroduplex mobility assay(HMA). T
cell responses to mitogens, recall antigens and HIV antigens were measured in
lymphocyte proliferation assay(LPA).
Results: 19 p24 antigenimic subjects and 23 seroconverters were enrolled. Only
2(10.5%) of 19 p24 antigenimic subjects reported clinical symptoms consistent
with acute retroviral syndrome before seroconversion. HMA of the provirus from
PBMC of all 42 subjects was homologous with subtype C HIV-1. A transient
increase in CD8+ and DR+ T cells was seen during the acute infection period. NK cell counts declined after serconversion. The mean VL and CD4 set
points, measured 6-12 months following estimated time of infection was 3.81 logRNA copies/ml (CI 3.38-4.25) and 413 cells/ml(CI389-622) with average annual
change of -122 cells/cmm and +3.52 log10 plasma VL. LPA responses to HIV
antigens (gag, env) was seen in about 50%, to recall antigens in 90% and to
mitogen in all patients tested.
Conclusions: Immune activation was evident during acute phase of infection.
HIV-specific, recall antigen and mitogen responses and VL and CD4 cell changes
do not suggest a faster progression of disease at least in early infection.
Presenting author: Ramesh Paranjape, National AIDS Research Institute, G-73,
M.I.D.C., Bhosari, Pune 411026, India, Tel.: [phone], Fax: [phone],
E-mail: [email protected]
WePeC6230 Evaluation of different techniques for the
estimation of the levels of CD4 cells in Lagos
Nigeria
R. Audul, J. Onyewuchel, O. Funsho-Adebayol, C. Oparugol,
D.I. Onwujekwe2, C.C. Onubogu2, J.A. Adedoyin3, E.O. Idigbe4. 1Nigerian
Institute of Medical Research, Human Virology Laboratory, Nigerian Institute of
Medical Research, Human Virology Laboratory 6 Edmond Crescent PMB 2013
Yaba, Lagos, Nigeria; 2Nigerian Institute of Medical Research, Yaba,
Microbiology Division, Lagos, Nigeria; 3Nigerian Institute of Medical Research,
Yaba Public Health Division, Lagos, Nigeria; 4 Nigerian Institute of Medical
Research, Lagos, Nigeria
Background: This study was carried out to evaluate and establish a cost effective
method for CD4 cell estimation in clinical samples in Nigeria.
Method: CD4 cell counts of 115 blood samples were determined. The blood samples were obtained from apparently healthy subjects (blood donors) as well as
subjects with various infections (HIV, TB, Malaria). The CD4 cell level estimation
for each sample was determined using three comparative methods, manual (Dynal beads), (Capceilla) (FACScount). All the data were collated and analyzed to
estimate the sensitivity and specificity of each of the methods.
Results: The sensitivity levels for the Manual, ELISA and Automated techniques
were recorded as 57.5%, 57.9% and 88.1% respectively There was no significant
difference of the sensitivity levels of the Manual and ELISA. However the sensitively level of the FACScount was significantly higher than for the other two methods. There was no significant difference between the specificity levels of 80.8%,
71.4% and 79.1%, recorded respectively for the Manual, ELISA and Automated
methods
Conclusion: Though the automated method was more sensitive than the other
two methods, it is equally more expensive.
The manual method at $4 per test is cheaper than the cost of $10 per test for
ELISA. Since there was no significant difference between these two methods, the
manual method is being recommended as a cost-effective method for CD4 cell
estimation in Nigeria and other resource poor countries.
Presenting author: Rosemary Audu, Nigerian Institute of Medical Research, Human Virology Laboratory 6 Edmond Crescent PMB 2013 Yaba, Lagos, Nigeria,
Tel.: [phone], Fax: [phone], E-mail: [email protected]
WePeC6231 Effective antiretroviral therapy reduces
degradation of tryptophan in patients with HIV-1

infection
R. Zangerle, M. Sarcletti, G. Quirchmair, B. Widner, G. Neurauter, D. Fuchs.
University, HIV Unit, University Hospital, Anichstrasse 35, A6020 Innsbruck,
Austria
Background: Antiretroviral therapy (ART) has a significant impact on HIV-1 RNA
levels, the CD4 cell count and immune activation. We examined whether these