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

12th World AIDS Conference

Abstracts 32284-32287 577

or >30 months and by prior treatment with a combination of ZDV/3TC. Viral
load measurements were performed using Roche Amplicor HIV-1 RNA PCR at
baseline, 8, 16, 24, 32, 40 and 48 weeks. Genotypic analysis of the HIV RT
gene was performed at baseline to determine pre-existing mutations that will be
correlated to viral load responses at week 4 and week 16. In addition, samples
from a subset of patients will be examined genotypically at week 8 and week 16
to look for RT mutations emerging post treatment. Phenotypic analysis will also
be performed on duplicate samples at baseline, week 8 and week 16 on a subset
of patients.
Results: Preliminary baseline genotypic data has been obtained for 59 patients
in this blinded study. Of these patients, 57.6% had a baseline RT mutation at
M184V, 6.8% at L74V, 28.8% had baseline RT mutations at 210, 215 or 219 and
only 6.8% of the patients had no known RT mutations at baseline. A subset of
these patients (21) had week 8 plasma genotyped in addition to baseline samples.
19% of these patients developed a mutation at M184V at week 8 that was not
present at baseline. 14% of the patients developed ZDV mutations at week 8 that
were not present in baseline samples.
Conclusions: The majority of ART experienced patients entering CNAA3006
had mutations at baseline associated with NRTI resistance. Approximately 20%
of the patients developed an additional mutation at M184V during the course of
this trial. A complete set of unblinded data will be presented correlating viral load
response with genotypic and phenotypic resistance.
563*/32284 Evidence for independent development of RT
inhibitor (RTI) resistance patterns in cerebrospinal
fluid (CSF) compartment
Philip Cunningham1, D. Smith1, C. Satchell', D.A. Cooper2, B.J. Brew3. Centre
for Immunology, St. Vincent's Hospital, Victoria Street; 2NCHECR University Of
NSW; 3Neurology Department, St. Vincent's Hospital, Sydney, NSW, Australia
Objective: (i) to evaluate the utility of the LiPA HIV-1 RT assay for detection
of drug selected mutations or polymorphisms of the HIV-1 reverse transcripase
gone in CSF and serum.(ii) to compare frequencies and positions of mutations in
both CSF and plasma compartments.
Methodology: A total of 56 paired CSF and serum samples collected from 53
patients with HIV-1 infection were tested using a commercially available line probe
assay (Murex, UK). Briefly, viral RNA is amplified using nested RT-PCR of the
reverse transcriptase gene. Biotinylated DNA is hybridised with oligonucleotide
probes immobilised as lines on a membrane strip allowing the simultaneous
detection of wild-type and mutations at codons 41, 69, 70, 74, 184, 214 and 215
conferring resistance to one or more drugs, AZT, ddl, ddC, and 3TC.
Results: HIV-1 RT sequences could be detected in both CSF and plasma samples in 31 (56%) patients. 19 (35%) CSF samples failed to amplify HIV RT cDNA,
highlighting the assays of detection limits in CSF. Twenty one (68%) patients had
identical resistance patterns in both CSF and serum. Ten (32%) patients had
variants conferring either wild-type or drug selected mutations in CSF but not in
serum. Of these, 3 (50%) patients demonstrated amino acid mutations associated
with high level ZDV resistance in CSF alone. One patient failed to amplify HIV RT
in serum but detected two (2) mutations conferring high level ZDV resistance in
the CSF. Nine patients exhibited resistance in plasma but remained sensitive to
therapy in CSF. There was no evidence of ddl resistance in any CSF specimen.
Conclusion: These results clearly demonstrate differences in the positions and
frequencies of wild-type a drug selected mutations in specimens derived from
both compartments and argue for independent development of drug resistance
in CSF and highlight important implications in guiding antiretroviral treatment in
HIV infection.
S32285 Genotypic and phenotypic resistance to stavudine (d4T)
after long-term monotherapy
Africa Molguint, U. Dietrich2, A. Immelmann2, V. Soriano3. BMS-020 Spanish
Study Group; Service Infectious Diseases, ISC-III, Calle Sinesio Delgado 10,
28029 Madrid; 2Georg-Speyer-Haus, Frankfurt; 31nstituto Salud Carlos III,
Madrid, Spain
Introduction: Protocol BMS 020 was a double blind, prospective clinical trial in
which two different doses of d4T, 20 and 40 mg bid, were compared in HIV+
patients with previous exposure to ZDV for longer than 16 weeks.
Material and Methods: Genotypic and phenotypic resistance to both ZDV
and d4T were examined after at least two years of D4T monotherapy. Direct
sequence analysis of the RT gene (aa 30 to 240) isolated from plasma HIV-RNA
was performed. Since pre-treatment specimens were not available, phenotypic
studies were carried out using as reference the Ill-B isolate.
Results: None of 35 tested individuals harboured the codon 50 and/or 75
mutations previously described to be associated with d4T resistance. However,
more than 80% of them carried mutations associated with ZDV resistance, despite all these patients had stopped ZDV at least two years before. Significant
phenotypic resistance to d4T was observed only in 2 of 5 tested individuals,
although ICso values were increased only 6.6 and 9.2 fold, respectively. These

two patients had suffered a decline in their CD4 count, and one of them had high
levels of plasma viraemia. The sequence analysis of the RT gene in these five
patients revealed a K104N change in only one specimen that could be involved
in d4T resistance. In contrast, and despite of having stopped treatment with ZDV
more than 2 years before, phenotypic resistance to ZDV was observed in all five
subjects, with IC50 values above 75-fold in all of them. Moreover, all harboured
codon substitutions within the RT gene associated with ZDV resistance, and these

mutations remained in viral genomes examined after virus co-culture, suggesting
that they yielded some biologic advantage to mutants in respect to WT viruses,
even in the absence of drug pressure.
Conclusion: both genotypic and phenotypic resistance to d4T seems to be a
rare event in patients exposed to the drug, even after long periods of time and
in subjects with apparent treatment failure. The significance of mutations at other
positions within the RT gene, such as at codon 104, should be further explored.
322861 Virologic findings in patients with paradoxical responses
to highly active antiretroviral therapy (HAART)
Jeffrey Fessel1, J.F. Krowka2, M.S. Ascher2, H.W. Sheppard2, B. Cuevas2,
S. Kwok3, C. Christopherson3. 12200 O'Farrel Street San Francisco California;
2California DHS-VRDL Virus Laboratory Berkeley, CA; 3Roche Molecular
Systems Alameda, CA, USA
Background: 5% to 10% of patients treated with HAART for 10 months had
a paradoxical response, i.e., CD4+ lymphocyte levels rose by >100 cells/mm3
whereas plasma HIV RNA levels remained >105 copies/ml (5th Retrovirus Conference, Chicago, 1998, abstract 145). We hypothesized that these patients with
paradoxical responses would have lower levels of infectious virus in plasma and/or
peripheral blood mononuclear cells (PBMC) in comparison with other patients
who are HAART failures as shown by plasma HIV RNA > 105 copies/ml and
CD4+ lymphocytes <100/mm3 after 6+ months of HAART.
Methods: We compared the results of virologic studies in 11 of these patients
with paradoxical response with those seen in 14 patients who had failed HAART.
Infectious HIV in plasma and infectious units/106 PBMC (IUPM) were quantified
by limiting dilution culture. HIV RNA in plasma was quantified by the Amplicor
Monitor test, and proviral DNA in PBMC was quantified by a prototype assay.
Results: We were able to culture HIV from plasma of significantly fewer patients
with paradoxical responses (27%) compared with patients who were complete
HAART failures (64%; p <.05). The mean level of infectious HIV in PBMC was
significantly lower in the group of paradoxical responders (IUPM = 25) compared
with patients who were in the group of complete HAART failures (IUPM = 278;
p <.05). Levels of plasma HIV RNA and of proviral DNA were not significantly
different in the two groups.
Conclusions: In some patients on HAART, virus of low infectivity persists in
plasma; and perhaps also in tissue sites. This allows reduced destruction of
CD4+ lymphocytes and a rise in the blood level of CD4+ lymphocytes despite the
maintenance of high levels of plasma HIV RNA.
229*/32287 HIV genotypic predictors of antiviral response to
saquinavir (SQV)/ritonavir (RTV) therapy in
patients who have failed prior protease inhibitors
(PIs): A clinical cohort study
Andrew Zolopa1, R. Shafer, A. Warford, J. Montoya, D. Katzenstein, T. Merigan.
Stanford University, 1School of Medicine, Div. of Infectious Disease, Grant Bldg.,
Stanford, CA, USA
Objective: To determine the genotypic predictors of an antiviral response to
SQV/RTV combination therapy in patients who have failed at least one prior
protease-containing anti-HIV regime.
Methods: In a university-based clinic we identified 50 HIV infected patients
who had failed multiple RT inhibitors and at least one prior PI-containing regime
(13 indinavir (IDV}, 5 nelfinavir (NFV}, 15 SQV then IDV, 8 SQV then NFV, and 8
SQV, IDV & NFV). Failure was defined as >0.5 log raise in plasma HIV RNA from
a nadir value while on a PI regime or a detectable viral load after being below
level of detection (<500 copies). Virological response was measured between 4
and 18 weeks after initiation of SQV/RTV therapy. HIV genotype was evaluated
on specimens stored within 1 month prior to initiation of SQV/RTV using population-based sequencing (ABI Foster City, CA). Protease mutations evaluated in
this analysis were at codons 30, 46, 48, 54, 82, 84, & 90.
Results: The subjects were 91% men, 72% White with a median age of 40.
The table shows the virological response for 6 different patterns of mutations
evaluated. Linear regression models demonstrated a strong relationship between
number of mutations and antiviral response (p < 0.001).
Mutation Subjects Past Pi's (medians) Baseline Follow up RNA
Patterns (N) # duration CD4 RNA (Ig) @ 4 wk @ 12 wk %. 500
O (none) 10 1 32 wks 235 4.86 -2.02 2.19 70%
30 7 1 23 wks 485 3.99 -1.97 2.02 100%
82 or 84 3 2 44 wks 200 4.00 1.35 1.77 67%
90 5 2 65 wks 420 4.72 1.38 -1.00 20%
any 2 9 2 69 wks 140 5.04 0.91 -0.47 11%
any >3 16 2 54 wks 165 5.02 -0.81 -0.01 0%
Conclusions: In patients who have failed protease inhibitors, HIV genotype
provides information that helps determine the likelihood of an antiviral response to
SQV/RTV combination therapy. The predictive power of a genotype in this study

appears to be independent of a detailed clinical history. Multivariate analyses are
ongoing and will be presented.