Abstracts Vol. 1 [International Conference on AIDS (11th: 1996: Vancouver, Canada)]

Track A: Basic Science

Mo.A. I II - Mo.A. I116

Mo.A. IIII
GENOTYPE RESISTANCE AT CODON 215 AND SI-NSI PHENOTYPE LONG-TERM
EVOLUTION OF HIV- I. CORRELATION TO CLINICAL OUTCOME IN A COHORT
OF 21 AZT-TREATED PATIENTS.
C. Arvieux*, A. Ruffault**. C. Michelet*, I. Renard**, C. Jacquelinet'', R. Colimon *, F.
Cartier*. *Department of infectious diseases, **Department of Virology, Unive rsity of
Rennes, France
Objectives: to evaluate the association between (I) genotypic resistance to AZT (codon
21I5) of HIV- I, (2) SI-NSI phenotype and clinical evolution in a cohort of 21 patients treated
with AZIT
Patients and Methods: Patients were enrolled in a first prospective study to determine the
correlation between viral load (Plasma Viremia, PV) and clinical outcome [J Acqu Immune
Defic SyndrVol 9, NO, 1995]. Samples of HIV isolates and plasm,, were stored every 3
months (M) for 24 months at -80~C and retrospectively analyzed in this second study
Assessment of viral load was done as described in the previous reference. Determination of
SI or NSI phenotype was determined by cultivation of HIV isolates with MT2 cells. For
genotypic resistance -i.e. emergence of the Thr - >Tyr mutation at codon 215- we
developped a rapid detection method using nested Digoxigenin PCR and a colorimetric
DNA-hybridisation assay.
Results: Patients who had persistent negative PV and patients who had a positive PV at
entry and for whom a 24 month negativation was obtained with AZT were considered as
Responders. Patients who experienced a clinical event resulting in a change of rclassification
(CDC 1987) during the 24 months of the study were considered as Progressors and the
other as Non-Progressors.
SI/NSI phenotype: 15 out of 21 patients had an NSI phenotype at Day 0, and one
became SI during the study SI phenotype was more frequent in Non Responders (p-0.04)
and in Progressors (p=0.049).The major difference between SI and NSI phenotype was
seen for the rate of CD4 decline, which was much higher in SI patients (p-0.0008).
Mutation at codon 215: 17 patients had a Wild (W) genotype at entry. One was
Mutated (M, contaminated with an AZT treated patient) and 3 patients had a mixed genotype of W and M virus. At M6, 6 out of 18 evaluable patients had always a W Genotype. At
M9, only four patients had still a W genotype.There was no correl nivn between Clinical
evolution at M24 and evolution of genotype in this series.
Conclusion: The SI/NSI phenotype seems to be well correlated to biological and clinical
evolution, but not, in this small series, the evolution of Genotype at codon 215.
A. Ruffault, Departement deVirologie, H6pital Ponchaillou, 35033 Rennes, FRANCE
TEL: [phone] FAX: [phone]
Mo.A. I 12
REVERSE TRANSCRIPTASE (RT) GENE ANALYSIS OF HIV-I MUTANTS DUALLY
RESISTANT TO AZT AND DDI
Kondo Makko* SaitoT*, lto A**,Akagi K***, Nishioka K*"-**, Imai M*. 4Kanagawa Prefectural Public
Health LaboratoryJapan, *Yokohama City UniversityJapan, *** Kanagawa Childen's Medical Center;
Japan, ***Viral Hepatitis Research Foundation of Japan
Objective: To study the relation between drug susceptibility and anmino acid mutation in
reverse transcriptase (RT) coding region of HIV- I isolates.
Methods: We isolated HIV- I mutants dually resistant to AZT and ddl from pl;ients who
received ddl treatment longer than one year after switching from AZT treatment due to
appearance of AZT resistant isolatesThe RT gene of these HIV-I mutants were amplified
by PCR, cloned to M 13 vector and sequenced.The amino acid nutations in he RT associated with resistance to AZT and ddl were analyzed.
Results: Isolats from patients treated with 13 months after switching fromI 18 m onths AZT
treatment grew in the presence of 10 pM AZT or 10 pM ddl. Mutation in 21 5 amino acid
to Tyr in RT region which is characteristic for AZT resistant mutants was observed in all
direct isolates from these patients as well as in cultured strain in the presence of AZT or
ddl. Mutation in 74 amino acid was not observed. Mutation in 41 amino acid (MetLeu) was
observed in 3 out of 7 clones from direct isolates. Moreover this mutation was observed in
all clones derived from strain grown in the presence of AZT or ddl but riot observed
before ddl treatment. After 2 years ddl treatment, all isolates showed 41 amino acid mutation. An isolate (038-4) from patients treated by AZT only but grew in the presence of ddl
showed mutation in 215 and 4I amino acid of RT region.
Conclusion: Our observation described above suggests that additional mutation in 4I
amino acid to 2 15 amino acid mutation in RT region correlates to the appearance of
mutants dually resistant to AZT and ddl.
Makiko Kondo, 52-2 Nakao-cho Asahi-ku Yokohama,24I Japan Telephone:[phone]
Fax: [phone]
Mo.A. 113
ZIDOVUDINE RESISTANCE AND HIV-I LOAD IN MULTIPLE AUTOPSY TISSUES OF
AIDS PATIENTS
EmerynVC#, Aitkins MR, Strappe PR, Kaye S*, Loveday C*, McLsAughli JL,. Johnson MAR,
Tedder PS*, Griftiths PD#. *Department of Virology, University College Lonclon, UK. #
Department of Virology and Hissopathotogy, Royal Free Hospital, I nu~des, UK.
Objective.This study aims to investigate the relationship between nthe priantitative prevalence of ZDV resistance in peripheral blood and lymphoid organs to that present in multiple other organs of the same individual.
Methods Proviral HIV- I load was measured by quantitative-competitive PCP in multiple
organs of I I patients dying with AIDS. Nine of these patients tact been proscrihed zidovudine.The percentage of wildtype and mutant sequences at the positions 4 I, 67, 70. 2 15 and
219 of the reverse transcriptase was assessed using a point mutation assay.
Results HIV- I proviral load in 90 samples from multiple organs obtdmio~es for 12 patients
dying with AIDS was significantly associated with increasing resistance to ZDV (p --0.008).
Proviral biads wei-e significantly higher in lymph node and spleen tar ill other organs
analysed (p <0.01 ).The distribution of wildtype and mutant sepueincrs it cocfoiis 41, 67, 70,
21 5 and 2 19 was not uniform and could differ markedly between lymptsoreticular system
and other organs such as brain.

Conclusion These results demonstrate that treatment of HIV- I infection with zidovudine
does not exert uniform selective pressures in multiple organs.These findings have implications for the interpretation of resistance data and design of treatment strategies for HIV,
arguing in particular, that alterations in therapeutic regimens should consider the likelihood
of different resistance patterns being present in sites other than the peripheral circulation.
Dr VC Emnery, Department ofVirology Royal Free Hospital, Rowland Hill Street, London
NW3 2PF Tel: [phone] x3109 Fax: [phone]
Mo.A.I 114
DIFFERENTIAL PLAQUE ASSAY FOR STUDYING POPULATION DYNAMICS OF
WILD-TYPE AND DRUG-RESISTANT MUTANT HIV IN MIXED INFECTIONS.
Rayner; Marlene M, Jackson, David A. The DuPont Merck Pharmaceutical Company
Wilmington. DE, USA
Objective: We wished to develop a sensitive assay that would differentially detect wild-type
(WT) and drug-resistant HIV in a mixed infection for the purpose of studying HIV population dynamics in response to drug selective pressure.
Methods: As a model system we studied mixtures ofWT HIV and the mutant virus containing a single amino acid change from Ile to Val at position 84 in the HIV protease (184V).
HIV car ying tle 184V mutation is 10-fold less sensitive to the cyclic urea HIV protease
inhibitor DMP450 than VVt HIV.Viral titers were determined by plaque assay.The titer of
WT plus 184V was measured in the absence of inhibitocThe titer of 184V was selectively
determined by performing the plaque assay in the presence of an IC99 concentration of
DMP450.The titer ofWT HIV was then selectively determined by subtraction.
Results: In the presence of an IC99 concentration of DMP450, less than I% the number of
plaques of WT HIV are detected as in its absence, whereas the same number of 184V
plaques are detected under both conditions.-he differential plaque assay can thus be used
to measure the titer of 184V in the presence ofWT HIV in mixed infections. A 95% to 5%
(WT:184V) mixture grown for 7 days in the absence of drug increases in total titer 20-fold
and changes in ratio to 99.75% WVFT to 0.25% 184V. In contrast, the same mixture grown in
an IC99 concentration of DMP450 for 7 days remained constant in total titer but changed
in ratio of WF to 184V to 50:50 at 3 days and 2:98 at 7 days.
Conclusions: Our results indicate that the differential plaque assay can be used to monitor
changes in mixtures of WT and drug-resistant HIV. In the absence of drug, 184V is at a significant selective disadvantage to WT, as first shown by EI-Farrash et al. [J.Virology (1994)
68: 233-239], whereas in the presence of drug, 184V has a strong selective advantage over
Wt. Such laboratory studies of the behavior of mixtures of WT and drug-resistant HIV
under var ious selective conditions may prove useful in understanding and predicting the
population dynamics of HIV in drug-treated patients.
Marlene M. Rayner DuPont Merck Pharmaceutical Company Wilmington, DE USA
Mo.A.I 115
HIV- I ISOLATES FROM SUBJECTS ON PROLONGED STAVUDINE THERAPY REMAIN
SENSITIVE TO STAVUDINE
Deminie, C., Bechtold, C., Riccardi, K., Lin, P-F, Colonno, Richard. Bristol-Myers Squibb
Company Wallingford, Connecticut, USA
Objective: To determine if stavudine-resistant variants arise following prolonged therapy and
to identify a genetic marker for any resistance observed.
Methods: Nineteen matched pairs of pre- and post-treated samples were obtained flom
subjects on stavudine therapyTwo approaches were employed to obtain viruses for susceptibility testing: I) viruses were isolated following co-cultivation with PBMC using established
ACTG protocols or 2) the reverse transcriptase gene resident in clinical samples was
obtained by PCR and cloned into a recombinant HIV-I LAI clone and assayed. Drug sensitivity testing on viruses from method I were assayed using a PBMC-based p24 assay, while
viruses obtained from method 2 were tested in a MT-2 cell based p24 assay Specific assay
conditions have been described previously [JID 170:1157-1164(1994)].
Results: Selected subjects from Phase 11(006) and Phase III (019) clinical studies had a
median CD4 count of 267 and received stavudine therapy for 0.5 to 2.4 yrs (mean =1.6
yrs). Nearly all had received prior zidovudine therapy before beginning stavudine treatmeent
Post-treatment isolates had a mean EC50 of 0.15 pM (range - 0.01-0.39 pM), a slight
increase fiom the 0.13 pM mean EC50 of pre-treatment isolates. Genotypic analysis on the
reverse transcriptase gene of several pairs of isolates failed to identify any amino acid
changes that could convey stavudine resistance to recombinant HIV- I clones.
Conclusions: This data combined with our previous studies on I I additional pairs of isolates suggests that stavudine resistance is infrequent and modest in degree. A genetic marker for in vivo resistance has yet to be identified.
Richard J. Colonno, Bristol-Myers Squibb Co., 5 Research Parkway, Wallingford, CT 06492
USA Telephone: [phone] Fax: [phone]
Mo.A.I I16
HYDROXYUREA AS ANTIVIRAL DRUG AGAINST HIV- I RESISTANT MUTANTS
tori, France*, Malykh, A.**,Wainberg, M.***,Villani, R*, Maserati, P.*, Gallo, P. C. *,
Lisziewicz, J.. *Research Institute for Genetic and Human Therapy, Policlinico S. Matteo,
Pavia, Italy, and Gaithersburg MD, USA: **Laberatory of Tumor Cell Biology, NCI, Bethesda,
MD, USA: ***Jewish Geneial Hospital, Montreal, Canada.
Objeccive:To assess the sensitivity to nucleoside analogues (AZT, ddl) of nucleoside resistant sariarits of HIV- I in the presence of Hydroxyurea at concentrations achievable is vive.
Methods: Ii nitne. Peripheral blood mononuclear cells were infected with high multiplicity of
infertion with HIV- I clones or primary isolates resistant to either ddl or AZT. Cells were
subsequrently treated with the proper nucleoside analogue (i.e. ddl in she case of ddl resistasnt virust either alone or in combination with Hydroxyurea. In vive. 60 patients have been
einiolled in.n phase I/IIB, randomized, controlled clinical trial in Pavia, Italy. 20 of there have
been snrted with 200 mg/bid ddl, and 40 patients have been treated with 200 mg/bid ddl
plan 5011 nng/bnd Hydroxyurea. Phimary objective: to analyze drug toxicity secondary objectives: to analyze drug efficacy and pharmacokinetics of low doses Hydroxyurea.
Results: In vise. Hydroxyurea was well absorbed and eliminated linearly with a half-life of 2.5
haunt. Plasota concentrations ranged between I135 and 8.5 p M. In vitro. Hydroxyurea concntr~tsons used in our experiments ranged between 100 and I 0 pJM, similar to these
achirvarble in HIV- I positive patients. In the absence of Hydioxyurea, 50 p M AZT or- 100

I'O
O\
-)
0
cc
-ct
to
0
Q)
C
Q)
C3
0
0
cc0
cX
7)
c74