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

580 Abstracts 32297-32301

12th World AIDS Conference

132297|

Application of a genotypic driven rule-based expert
artificial intelligence computer system in treatment
experienced HIV-infected patients. Immunologic and
virologic response

Paul J. Cimoch1, D.M. See12, M.J. Pazzani2, W.M. Reiter3, R.A. Lathrop2,
J.G. Tilles2. 1Center for Special Immunology-lrvine, 100 Pacifica, #100 Irvine,
CA 92618; 2University of California-rvine, Irvine CA; 3 Center for Special
Immunology, Fort Lauderdale FL, USA
Background: A rule-based expert system, "Customized Treatment Strategies for
HIV" (CTSHIV), has been developed connecting the literature describing specific
HIV drug resistant mutations with the patient's genotype. Drug treatment recommendations are made based upon: known drug-resistant mutations, ranking
and weighting based upon antiviral activities, overlapping toxicities, relative levels
of drug resistance and proportion of drug-resistant clones in the patients HIV
quasispecies. All FDA approved antiretroviral (ARV) drugs are ranked by their
estimated ability to avoid current and nearby drug resistant mutants.
Methods: Fourteen patients failing ARV therapy were studied. CD4 counts and
viral load (VL) (Amplicor~ PCR) were performed at day-21, day zero and every
three months. For each patient, the reverse transcriptase and protease portions
of the polgene were amplified by RT-PCR, 5 clones were produced, plasmid DNA
extracted, and complete sequencing performed by an ABI sequencer (Applied
Biosystems) with the data directly downloaded into the CTSHIV program. The
five most effective 2, 3 and 4 drug regimens were displayed for each patient. The
clinician chose from the options based upon order of ranking from the program,
history of drug intolerance and patient preference. Patients were seen monthly for
tolerance, compliance and routine lab assessments. Every 3 months CD4 counts
and VL were performed.
Results: At 6 months, of the 14 patients treated according to CTSHIV, 7 (50%)
had undetectable VL, 3 (21%) had >1 log drop in VL and 4 (29%) were unchanged.
In the 10 patients who responded and remain on protocol, the various ARV regimens were well tolerated. Of the 4 patients that had an unchanged VL, 2 experienced severe drug toxicity and 2 were non-compliant with the selected regimens.
Conclusions: Patients with persistently elevated VL despite ARV therapy are
a particular challenge to manage and historically have poor treatment outcomes.
The use of the CTSHIV system with mutational prediction artificial intelligence
fields successfully assisted in the ARV management of these patients. With the
increasing prevalence of mutations in treatment naive patients, the system may
also potentially be used to optimize combination ARV therapy at its onset.
32298 1 Genotypic and phenotypic resistance in HIV-infected
children under AZT and 3TC therapy
Richard Liwde1, F.M. Funk Markus', Lynen Nils2, Cinatl Jinarich2,
Zielen Stefan1, Kreuz Wolfhart1, Hofmann Dietrich. 1 Universitat Skinder Klinik,
60596 Frankfurt; 2Departement of Medical Virology, 60596 Frankfurt, Germany
Issue: Appearance of drug resistance leads to disease progression under antiretroviral therapy. In order to evaluate drug resistance, genotypic and phenotypic
assays were compared.
Project: 12 children stage B (median age: 8.5 years at end of study) were included. 7 patients received AZT monotherapy initially, 5 of these patients received
3TC in a second line combination. AZT was given as a first line combination to
5 patients; in 3 patients AZT was combined with 3TC (AZT - dosage: 3 x 120
mg/sqm, 3TC-dosage: 2 x 4 mg/kg bw). Genotypic analysis based on RT-gene
sequencing (position 41-219) in plasma samples. Phenotypic susceptibility to
AZT and 3TC were assesed using MTT-MT4 assay (AntivirogramTM) and results
are expressed as an amount of resistance relative to wild type.
Results: Phenotypic resistance to AZT ranged 9-1500 fold (median 78 fold,
duration of therapy 11-49 months) in patients with initial monotherapy and 0-1200
fold (median 3 fold, duration of therapy 6-36 m) in patients with first line combination. Phenotypic resistance to 3TC ranged 11-27 fold (median 16 fold, duration
of therapy 6-17 m) under second line combination and 24-27 fold (median 24
fold, duration of therapy 6-8 m) under first line combination. Between phenotypic
and genotypic testing good correlation was seen for AZT first and second line
combination and 3TC second line combination. Under 3TC first line combination
mutation 184 was found only in one patient.
Conclusion: Under AZT first and second line combination phenotypic resistance were aproximately 10 fold in patients <12 months of therapy and and
100 fold in patients >12 months of therapy. Phenotypic 3TC-resistance reached
maximum values 6-12 m after onset of first and second line therapy.
32299 3TC resistance is associated with better HIV suppression
than 3TC sensitivity and confers little NRTI cross
resistance
Richard Paul Harrigan. BC Centre for Excellence in HIV/AIDS 613-1081 Burrard
St. Vancouver, BC, V6Z 1 Y6, Canada
Background: An antiviral and clinical benefit of 3TC persists despite the rapid
emergence of high level resistance to 3TC in most subjects taking partially suppressive therapies. One might expect an even greater anti-HIV response in those
patients whose plasma HIV remained 3TC sensitive.
Objective: To assess the association of 3TC resistance with antiretroviral
response and with nucleoside analogue cross-resistance.

Methods: Phenotypic resistance assays (VIRCO, N.V.) were performed retrospectively on stored samples from 48 ART-naive subjects in a recently-completed

comparative clinical trial (AVANTI 1). High level resistance was defined as a
greater than 10-fold difference in the IC50 of the subject strain relative to a lab reference strain. Longitudinal assessments of HIV-1 RNA (Amplicor, 400 copies/mL
detection limit) were measured prospectively.
Results: Virologic failure (<0.5 log drop in viral load) was observed in 83%
(5/6) subjects with < 10-fold resistance (median log reductin at wks 28, 40, & 52 =
0.33) and 15% (6/41) subjects with >10-fold resistance (median log drop at wks
28, 40, & 52 = 1.31) (p = 0.004). High-level ZDV, d4T, ddC, and ddl resistance
emerged in only 3, 0, 0, and 1 subjects, respectively.
Conclusions: Most patients were highly resistant to 3TC after one year of
therapy but had only low level cross-resistance to other RTIs. Counter-intuitively,
a more durable antiviral response was observed in those subjects with high level
phenotypic resistance to 3TC than those with 3TC sensitive virus. This may be in
part due to better adherence to the treatment regimens used, though, the effects
of other viral factors cannot be excluded.
S32300 Combination of the multidrug resistance mutation
Q151M/L and the AZT resistance mutation T215Y/F in the
same HIV-1 reverse transcriptase is compatible with
enzyme activity
Jan Balzarini1, H. Pelemans1, E. De Clerq1, A. DOnkler2, J.-P. Kleim3. 1Rega
Institute for Medical Research, Minderbroesstraat 10. B-3000 Leuven, Belgium;
2Hoescht AG, D-65926 Frankfurt, Germany; 3GlaxoWellcome, Stevenage, UK
Introduction: Multiple dideoxynucleoside (ddN)-resistant HIV-1 strains have been
isolated from an increasing number of patients under ddN combination therapy.
The Q151M/L mutation is mostly associated with several other mutations (i.e.
V751, F77L and F116Y). The resulting strains are highly resistant to AZT, ddl,
ddC, d4T and partially resistant to 3TC. On the other hand, the T215Y/F mutation
in the HIV-1 RT readily arises in patients treated with AZT. To the best of our
knowledge, patients harbouring a virus population containing both the Q151M/L
and T215Y/F mutations have not been reported yet.
Aim: We wanted to investigate whether the concomitant presence of both types
of mutations in the same RT enzyme would be compatible or not compatible with
enzymatic activity and virus replication.
Methods: An RT-BspMI-cassette was constructed from the HIV-1 MN RT wildtype gene whose first 800 base pairs were replaced by the homologous part of
the HIV-1 (1llB) RT. This RT cassette was subcloned into the expression vector
pQE30 (Qiagen), containing a poly His-tag at the amino terminus. The relevant
mutations were introduced using the Stratagene's Quickchange site-directed mutagenesis kit. After RT expression, purification of the (mutant) RTs was done using
a Ni-NTA-spin column (Qiagen). Recombinant viruses were constructed using a
slight modification of the recombinant virus assay of Kellam et al. (Antimicrob.
Agents Chemother., 38: 23-30, 1994). They were obtained through homologous
recombination of RT with RT-deleted proviral DNA which was propagated in
pHIVARTBstEII.
Results & Discussion: Enzyme activities were determined for the mutant RTs
containing the paired mutations Q151M + T215Y, Q151M + T215F, Q151L +
T215Y and Q151L + T215F. The mutant enzymes showed different catalytic activities depending the nature of the paired mutations. The differences in enzyme
activity not only differed in function of the mutations that were present in the
enzyme, but also depended on the nature of the template/primer that was used
to measure enzymatic activity (i.e. polyA/dT; polyC/dG; polyl/dC; heteropolymeric
template/primer). When electroporating MT-4 cell cultures with the DNA constructs to obtain recombinant mutant viruses, the Q151M + T215Y/F mutant virus
strains could be readily detected in the cell cultures.
Conclusion: Our data indicate that the concomitant presence of the Q151M/L
mutation with the T215Y/F mutation in the HIV-1 RT is compatible with RT activity.
32301 1Prevalence of resistance-associated genotypic mutations
in plasma HIV of patients failing tritherapy combination
Jacques Durant1, Philippe Clevenbergh2, P.H. Halfon3, P. Delgudic4
V. Rahelinirina2, S. Sayada3, P. Dellamonica2. 1Archet Hospital - Infectious
Diseases Department Nice 06202 Cedex 03; 2University of Nice, Nice;
3Actgene, Evry; 4Bonnet Hospital, Freyus, USA
Objective of the study: To determine the prevalence of drug resistance genotypes from plasma virus in HIV patients (pts) treated-for at least 6-months with
nucleoside analogues (Nc) and 3 months with protease inhibitor (IP) failing therapy
with a viral load above 10,000 copies/ml.
Mutant codons Resistance to Number of pts
75 d4T 4/59 (7%)
215 or 41 +/- (67, 70, 219) AZT 43/59 (73%)
184 3tC 30/59 (51%)
151 +/- (62, 75, 77, 116) MDR 1/59 (2%)
65 or 74 or 69 or 184 ddX 32/59 (54%)
103 or 106 or or 108 or 181 or 188 nevirapinei 2/59 (3%)
No major RT mutation 9/59 (15%)

48 or 90 +/- (63, 71, 84) saquinavir 14/61 (23%)
30 +/- (36, 46, 71, 77, 88) nelfinavir" 0
46 or 82 and two of the following (10, 20, ritonavir or indinavir 24/61 (39%)
24, 32, 54, 63, 64, 71,73, 77, 84)
No major IP mutation 28/61 (46%)
multidrug resistance, not in use at the study time