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

806 Abstracts 42166-42169

12th World AIDS Conference

suppress or reappears later on in a few subjects, mainly in those previously
pre-treated or in those having bad drug compliance. In these circumstances the
virus evade drug pressure, and cross-resistance becomes an important issue for
subsequent therapies.
Patients: We selected 21 patients ongoing triple therapy (3TC + D4T + indinavir). All showed a decrease in the viral load <500 copies/ml by the 2nd
generation bDNA assay. We collected samples at baseline, 3rd, 6th and 9th
months. All of them had good compliance.
Methods: Quantitation of plasma HIV-RNA using an ultrasensitive NASBA assay (Nuclisens, Organon) and detection of the codon M184V mutation in plasma
by LiPA (Murex), were performed.
Results: despite evidence of ongoing virus replication using the ultrasensitive
assay, the codon 184 mutation rarely appeared and only after more than 6 months
of therapy.

Viral load 3rd month
patients WT
- 20 copies/ml 9 9
<20 copies/ml 3 3
Total 12 12

MT
0
0
0

6th month
patients WT
9 9
8 8
17 17

MT
0
0
0

9th month
patients WT
10 6
3 2
13 8

MT
2
1
3

Conclusions: The low frequency of emergence of M184V codon mutation in
these individuals and the presence of very low but still detectable plasma HIVRNA supports the notion of "residual viraemia". This concept need to be kept
in mind when a therapeutic threshold for switching drugs or consider treatment
failure is recommended, specially considering that the number of available drugs
is still limited.
42166 Evaluation of the magnitude of virologic suppression in
patients on aggressive antiretroviral therapy
Brian Conway1, Mark Whaley2, Signe Fransen2, Andrew Shillington2,
Julio Montaner2, S. Kwok3, John Sninsky3. 1BC Centre for Excellence in
HIV/AIDS 667-1081 Burrard St., Vancouver, British Columbia; 2Canadian HIV
Trials Network, Pacific Region Vancouver BC, Canada; 3Roche Molecular
Systems Alameda CA, USA
Background: With the use of aggressive antiretroviral therapy regimens, plasma
viral load often falls below the limit of quantitation of both conventional and more
sensitive assays. This has often been interpreted as "maximal" suppression, the
first step towards viral eradication. There is a need to evaluate this concept more
rigorously, in light of data suggesting that HIV replication in the blood is ongoing
in patients with "undetectable" viral loads for over two years.
Methods: This is an observational study in 8 patients receiving triple drug
therapy (with two nucleoside analogs and a protease inhibitor or nevirapine) and
in whom plasma viral load was <400 copies/ml for 16-22 months. At the last time
point, a supplementary blood sample was obtained and the Ultra Direct Amplicor
assay (limit of quantitation 20 copies/ml) was performed on 11 aliquots taken from
this sample, and results reported according to the absolute value generated from
each aliquot.
Results: Virus was easily quantitated in 7/8 cases, at values ranging from undetectable (optical density of the aliquot below that of the corresponding negative
control) to a calculated value of 338 copies/ml plasma. The true plasma viral load
was taken as the mean of all results generated from a given sample, and was
found to be 13 copies/ml for the study group, with no significant difference based
on the duration of virologic suppression (11 vs. 14 copies/ml for patients with viral
load <400 for <20 and >20 months).
Conclusions: The insightful use of more sensitive viral load measurement
technologies allow us to appreciate that suppression of viral replication may be
incomplete (even in the plasma) in patients on long-term aggressive antiretroviral
therapy. This information will be useful in evaluating the true efficacy of such
regimens within research protocols.
42167 The prevalence of HIV-1 non-B subtypes in a London
HIV/AIDS outpatients: The implications for clinical care
Clive Loveday', H. Devereux1, A. Burke1, L. Dann1, A. Phillips2, M. Johnson3.
'Dept Retrovirology, Royal Free Hospital, Med. Sch., Roland Hill St., London
NW3 2 PF; 2Dept Primary and Biostatistics RFH, London; 3Dept HIV/AIDS
Royal Free Hospital, London, UK
Objectives: Divergent quantitative plasma HIV-1 RNA load results using different commercial assays suggested the presence of HIV-1 non-B subtypes in our
routine clinic population. This study explores the prevalence of non-B subtypes in
our clinic, the variability of quantification of their viral load using 4 different assays
and the implications of the results for their for clinical care.
Methods: One hundred and fifty consecutive clinic patients selected with epidemiological evidence of acquisition of non-clade B virus were characterised
using peptide EIA and/or sequencing. Viral loads were measured using RT-PCR
(Roche:cut-off 400 c/ml), RT-PCR including non-B primers (RT-PCRnb: cut-off
400 c/ml), NASBA (Organon, cut-off 2000 c/ml) and bDNA (Chiron: cut-off 500
c/ml). An interim analysis (n = 102) used linear regression with 95% CI, Students-t
and paired Wilcoxon tests.
Results: In this cohort 91/102 samples were characterised using the peptide
EIA, of these 68 (75%) were non-B and 23 (25%) B subtypes. Total patients attended the clinic during this period (n = 710) provided a denominator to determine
a crude prevalence of 9.2% for our population.

Detection and quantification of non-B subtypes was significantly higher using
RT-PCRnb (93%:27520 c/ml: p < 0.01 and p < 0.001) compared to RT-PCR
(72%: 6300 c/ml), NASBA (63%:13200 c/ml) and bDNA (74%: 6900 c/ml). There
was no significant differences for B subtypes. Using theoretical threshold values
of >5000, >10000 and >20000 c/ml for starting therapy we demonstrated that
patients with non-B subtypes had statistically less chance of starting therapy
according to the viral load assay used in their clinical evaluation.
Conclusions: Our data demonstrated a relatively high prevalence of non-B
subtypes (9.2%) in a clinic population from one of the most expensive area of
London. The variability in responses using different viral load assays requires
new strategies for patient and clinical trial management. As a consequence of
these data we have established baseline screening for HIV subtype in our clinical
service.
42168 Performance a semi-automated, high sensitivity,
quantitative assay for HIV-1 RNA using branched DNA
technology
Stephen Barr, G. Gorrin, D. Hendricks, J.A. Kolberg, J. Robertson, L.P. Shen,
B. Irvine. Chiron Diagnostics Corporation, 4560 Horton Street, Emeryville, CA,
USA
A Quantiplex'" HIV-1 RNA 3.0 Assay (bDNA) with sensitivity of 50 copies/mL
has been developed on a semi-automated platform (Quantiplex'" System 340).
The sensitivity improvements relative to Quantiplex'" HIV-1 RNA 2.0 Assay were
achieved by a combination of developments which lowered the assay background
and increased the level of amplification. Background reduction was achieved, in
part, by incorporating non-natural nucleotides (i.e. iso-G and 5'-methyl Iso-C) into
some of the amplification sequences. This has the effect of reducing non-specific
hybridization of these sequences to the other assay components. In addition,
background reduction was achieved in the redesign of the "target probes" which
mediate binding of the first layer of amplification (pre-amplifier) to the captured
target. The sequences on the target probes have been modified such that binding
of the pre-amplifier requires hybridization with sequences from two adjacent target
probes. This design reduces the non-specific hybridization of pre-amplifier. A new
target probe set was designed which increased the number of potential binding
sites for the pre-amplifier: this configuration achieved increased amplification. The
ease-of-use and throughput of the assay is greatly facilitated by use of a semiautomated instrument. This instrument system provides a platform wherein the
multiple incubations or hybridizations, washes, final luminescence readout and
data reduction are all automated. The instrument can process two 96 microwell
plates (i.e. 168 patient results per single run). One mL of plasma is required for
each specimen.
The limit of detection was 50 copies/mL with a specificity of 95%. The assay
has a dynamic range of greater than 4 logs (50 to 800,000 copies/mL). The
maximum observed difference of quantification between HIV-1 subtypes A-F was
two-fold. Precision (CV% of quantification) between 100 and 560,000 copies/mL
was <36%. HIV-1 of different subtypes can be quantified with precision, accuracy,
and at very low concentrations with this assay.
42169 Antiviral effect of AZT versus D4T in combination with
3TC and indinavir in the context of a population-based
study
Robert S. Hogg, J.S.G. Montaner, C. Christopherson, S. Kwok, J. Sninsky,
M. Harris, B. Conway. 1BC Centre for Excellence in HIV/AIDS 613-1081 Burrard
St. Vancouver, BC, V6Z 1Y6, Canada
Objectives: To compare the antiviral effect of AZT versus d4T given in combination with 3TC and IDV among antiretroviral (ARV) therapy naive individuals
enrolled in the drug treatment program of the province of British Columbia (BC),
Canada.
Methods: In BC antiretroviral therapies are distributed free of charge according
to therapeutic guidelines. Eligible for analysis were subjects who started therapy
with AZT or d4T given in combination with 3TC and IDV from 07/96 to 09/97,
had a baseline plasma viral load (pVL) determination and at least 2 pVL results
during therapy. The primary outcome was 2 consecutive decreases in pVL to
<500 copies/mL. All analyses were conducted on an intent to treat basis. Statistical comparisons were carried out using distribution-free and multivariate logistic
methods.
Results: 191 antiretroviral naive subjects (67 and 124 on AZT and d4T based
combination therapy, respectively) were eligible for analysis. There was no statistical difference between the two groups at baseline with respect to age (p =
0.373), gender (p = 0.068), AIDS (p = 0.438), CD4 cell count (p = 0.364) and pVL
(p = 0.606). In addition there was no statistical difference between the two groups
with respect to the proportion having 2 pVL below 500 copies/mL (p = 0.504).
In a multivariate analysis after adjusting for age, gender, CD4 cell count, pVL at
baseline, those on d4T based triple drug therapy were just as likely to have 2
consecutive pVL below 500 copies/mL (OR = 0.95; 95% CI: 0.44-2.06) as those
on AZT based triple therapy.

Conclusion: Preliminary results of this observational population-based study
among ARV therapy naive subjects suggest that AZT or d4T given in combination
with 3TC and IDV had similar plasma viral load effects. Formal comparisons of
these two regimens within a large prospectively designed randomized clinical trial
will be needed to definitively ascertain whether there is a statistically and clinically
significant difference in their antiviral potency.