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

12th World AIDS Conference

Abstracts 33165-33169 625

Median RNA copies/mL at baseline were 41,300 (E) and 32,700 (F) (P = 0.99).
Mean log10 viral load for the 2 groups was comparable at 36 W and 1 month
postpartum (1 MPP), and, in Group E, was similar throughout late pregnancy
and postpartum. In Group F, compared with baseline, there was a mean RNA
reduction of 0.5 log at 37 W and 0.7 log at 38 W and delivery; RNA levels returned
to baseline at 1 MPP.

Group n
E 19
F 21

36 W
4.6
4.6

37 W 38 W DEL 1MPP

4.6 4.7 4.6
4.1 3.9 3.9

4.6
4.6

(P. 0.001, compared with 36 W: P 0.01 compared with same week in other group)
Conclusions: Assuming that Group F is receiving ZDV, these data suggest
that with short-course oral ZDV treatment late in pregnancy, maternal plasma viral
load drops significantly (0.5 log) within the first week of treatment, drops lower at
24 weeks of treatment (0.7 log), and then returns to baseline after cessation of
treatment, without evidence of a rebound to a higher level at I month. The relation
of this reduction in viral load to reduction in transmission risk will be determined
with further testing and unblinding at the conclusion of the trial, scheduled to
occur by March 1998.
I33165 Cord blood ZDV concentration after intermittent oral
administration during labor, Thailand
Nathan Shaffer1, Wimol Siriwasin2, J. Mei3, P. Chaisilwattana4, R.J. Simonds3,
N.L. Young4, N. Meredith3, P. Mock1, A. Roongpisuthipong4, S. Chearskul4,
R. Chuachoowong1, H. Hannon3, T.D. Mastrol. 1HIV/AIDS Collaboration, PO.
Box 139, Nonthaburi 11000; 2Rajavithi Hospital, Bangkok; 4Mahidol University,
Bangkok; Thailand; 3CDC, Atlanta, GA, USA
Background: The ACTG 076 zidovudine (ZDV) regimen to reduce perinatal HIV
transmission includes intravenous (IV) ZDV administration during labor, which
poses significant logistical and cost barriers in developing countries. Little is
known about now well Oral ZDV is absorbed and tolerated during labor. We
conducted a study to determine every 3 hours from the onset of labor to delivery
to HIV-infected pregnant women. Umbilical cord blood was assayed for ZDV
concentration using a radioimmunoassay with a lower limit of detection of 0.115
ng/mL (with sample dilution, 2.41 ng/mL). We evaluated the association of ZDV
concentration and treatment group, time since last dose, number of labor doses,
body weight, and other delivery factors.
Results: The first 145 women who delivered were included in this analysis. All
women received at least one dose during labor (median = 4). Only I woman had
problems taking the oral labor drug, due to nausea and vomiting. No ZDV was
detected in the cord blood of women who received placebo. All 72 women in the
treatment arm had detectable ZDV. For these women, the median time between
last dose and delivery was 86 minutes, and 94% of women had their last dose
within 3 hours of delivery. The median ZDV concentration was 330 ng/mL (range,
7-1391). ZDV concentration was correlated with number of labor doses received
(P = 0.01), but the range was wide. ZDV concentration possibly was associated
with the interval between last labor dose and delivery (P = 0.06), but not with use
of other drugs during labor, mode of delivery, body weight, or duration of prenatal
ZDV treatment.
Conclusions: Oral administration of ZDV during labor was tolerated well and
appears to be a practical component of a ZDV regimen, in place of IV ZDV administration, for reducing perinatal HIV transmission in developing countries. The
mean ZDV cord blood concentration was in the therapeutic range, and the range
of concentrations observed was similar to that reported following intermittent IV
administration. Associations between transmission and cord blood concentration
will be analyzed at the completion of the trial, in March 1998.
33166 | Multicenter evaluation of HIV-1 RNA levels in dried
plasma and whole blood spots
S. Cassol1, A. Comeau2, S. Fiscus3, G. Aldrovandi4, J. Sullivan2, J. Bremer5,
B. Jackson6. 10ttowa General Hospital, Ottowa Box 411 501 Smyth Rd, Canada;
2 University of Massachusetts, Boston, MA; 3University of North Carolina,
Chapel Hill, NC; 4 University of Alabama, Birmingham, AL; 5Rush Presbyterian
Medical Ctr, Chicago, IL; 6Johns Hopkins University, Baltimore, MD, USA
Objectives: To evaluate the sensitivity, specificity, and precision in detecting
and quantifying HIV-1 RNA in dried plasma and whole blood filter paper spots
using the Organon Teknika Nuclisens HIV-1 QT assay and a modified Roche
AMPLICOR HIV-1 MONITOR assay.
Methods: Schleicher and Schuell 903 filter paper was spotted in duplicate
with 50 uL of seronegative plasma or whole blood to which had been added
5-fold serial dilutions of HIV-1 ranging from 500,000 to 800 copies/mL to no virus.
Coded specimens were tested by 3 experienced and 5 nonexperienced labs (3
by Nuclisens and 5 by a modified Roche assay). Zero, 15000, and 150000 copy
plasma controls were run with each assay.
Results: Both the Roche and Nuclisens assays were able to detect at least
4000 copies/mL or greater in both the plasma and whole blood spots by all
labs. Approximately half of the 800 copy/mL plasma and whole blood samples
had detectable HIV-1 RNA by both methods. There was 1 weak false positive
among the 31 zero copy samples. The median CV between copy number/mL in
duplicate plasma samples or between duplicate whole blood samples in the labs
as 21.6% and 21.5%, respectively, for the Roche assay and 20.1% and 19.1%
respectively, for the Nuclisens assay for samples -4000 copies/mL. The ratio of

copy number/mL in whole blood vs plasma was not significantly different from 1.0
for either the Roche or Nuclisens assay for samples with -4000 copies/mL.
Conclusion: Our preliminary data indicate that detection and quantitation of
HIV-1 RNA using commercial assays in testing either spiked dried plasma or
whole blood filter paper spots are comparable in terms of sensitivity, specificity,
and precision. The ease of this technology should prove useful in the diagnosis
and monitoring of HIV-1 infected infants.
S331671 Antiretroviral efficacy and viral resistance where no CD4
or RNA monitoring is available in Brazil
Christopher D. Pilcher1, D.M. Johnson2, M.D. Perkins3, Dietze4, J. Schock1,
A. Zago4, J.J. Eron1. 1CB #7030 547 Burnett-Womack, UNC-Chapel Hill Chapel
Hill, NC; 2Lacoratory Corporation of America, Research Triangle, NC; 3Duke
University Durham, NC, USA; 4Universidade Federal Do Espirito Santo. Vitoria,
ES, Brazil
Background: The Brazilian Ministry of Health provides free combination antiretroviral therapy for HIV-infected patients with symptoms or CD4 counts - 500 but
CD4 counts and RNA measurement are not routinely available in many areas in
Brazil. The consequences of this strategy were investigated in a cross-sectional
study of patients attending HIV clinics in Vitoria, Brazil.
Methods: Consecutive HIV-1 infected individuals from 2 clinics, on stable therapy, no therapy or initiating therapy were enrolled. Attending physicians answered
questionnaires on plans for antiretroviral therapy. CD4 counts and NucliSens
HIV-1 RNA in plasma were measured. HIV-1 Protease and reverse transcriptase
RNA sequence was analyzed on stored samples using the Affymetrix GeneChip
System in order to examine patterns of genomic resistance.
Results: 83 patients (36 naive asymptomatic, 11 naive with symptoms, 37 on
treatment) were enrolled. All 11 naive patients with symptoms were initiated on
therapy. However 33/36 asymptomatic patients had a high risk of progression as
defined by RNA >20,000 or CD4 < 500. Clinicians would have treated 26/33
high-risk asymptomatic patients had their CD4 count alone been known. 18/37
treated patients (on stable 2- or 3-drug therapy, median duration 13 mos) had
-20,000 copies RNA suggesting 10 or 20 virologic treatment failure. None of
these cases had been identified clinically. Genetic sequence has thus far been
determined on 6 patients (4 naive, 2 treated). No mutations associated with
drug resistance were observed in naive patients. Each of the treated patients
harbored RT mutations associated with resistance to their nucleoside therapy.
In one patient treated with indinavir no significant protease resistance mutations
were seen.
Conclusions: Clinical symptoms allowed clinicians to identify only a minority
of naive patients at high risk of progression and no cases of virologic treatment
failure. Discrimination of high-risk naive patients was greatly improved by access
to a CD4 count alone. Preliminary results indicate that treated patients may harbor
significant genomic drug resistance - sequence data for all samples will be presented and discussed. Using antiretroviral therapy without therapeutic monitoring
may not be an effective strategy in the developing world.
S33169 Prevalence of antiviral medication use in a probability
sample of HIV seropositive MSMs in major US cities
Ron Stall, L. Pollack, T. Mills, J. Catania. CAPS, 74 New Montgomery St.. 6th
Floor San Francisco, CA, USA
Objectives: To estimate the prevalence and identify correlates of antiviral use
in a household-based probability sample of HIV seropositive men who have sex
with men (MSMs) in major US cities.
Design: Cross-sectional, household-based probability survey (n = 2,530 for
the combined San Francisco, New York, Los Angeles, and Chicago samples).
Confirmatory HIV testing with ORASURE" home test kits was performed with a
sub sample of MSMs; all of the men who reported themselves to be seropositive
tested positive.
Method: Through use of existing data sources on MSM residence patterns
(MSM AIDS caseload data, census data on male couples and marketing lists),
geographical areas within each of the four cities were stratified for density of
MSMs and surveyed by telephone using random digit dialing. Men reporting sex
with men in the last five years were questioned regarding a broad range of health
issues, including (as appropriate) behavioral responses to being HIV seropositive.
Results: 17.3% of MSM in this sample reported themselves to be HIV seropositive. Of these, 57% (95% Cl 51-63%) reported current use of any protease
inhibitor and 74% (95% Cl 69-79%) reported current use of any other HIV antivirals. Only 55% (95% Cl 49-61%) reported current use of a protease inhibitor
in combination with another antiviral(s). Older men, more highly educated men
and men with lower CD4 counts were significantly (p - 0.05) more likely to
currently combine use of a protease inhibitor with another antiretroviral(s). City
of residence, race, income and a set of attitudinal measures did not distinguish
between HIV seropositive current users and non-users of combination antiviral
therapies.
Conclusion: We found no significant effects for race, income or city of residence on use of combination antiviral therapies, suggesting that HIV antiviral

medications are generally accessible to seropositive MSMs. However, use of
protease inhibitors in combination with other HIV antivirals is far from universal
among seropositive MSM residents of large American urban centers.