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

66 Abstracts 12269-12272

12th World AIDS Conference

who stopped primary or secondary prophylaxis while using 3 or more antiretrovirals (the regimen including a PI in 52/68), none developed PCP or cerebral
toxoplasmosis and only one patient died during 34.4 person years of follow-up
(PYFU) (median CD4 cell count at time of stopping 263 cells/mm3 (IQR 158-342).
In contrast, in the remaining 922 patients who stopped prophylaxis while using
<2 antiretrovirals, 25 developed PCP or cerebral toxoplasmosis and 620 patients
died during 335.5 PYFU (median CD4 cell count at time of stopping prophylaxis
35 cells/mm3 (IQR 10-177).
Conclusion: Almost 25% of patients had discontinued their prophylaxis; most
patients died shortly after stopping treatment, suggesting that their disease was
terminal at stopping. Among patients on HAART who have stopped, to date there
has been no PCP or cerebral toxoplasmosis; however, the total PYFU among
these patients is short. Updated figures including a further 6 months follow-up will
be available for presentation.
12269 Significant increases in serum cholesterol are seen
among HIV patients taking protease inhibitors
Jane Pollner1, N.E. Aronson1, S. McHugh1, R. Nielson2, C. Hawkes1. 1 Walter
Reed Army Medical Center, 6825 Georgia Ave, NW, Washington, DC
20307-5001; 2HMJF Rockville, MD, USA
Background: Increases in cholesterol among HIV patients taking both ritonavir
and indinavir have been reported. We looked at changes in cholesterol among
HIV patients on all approved protease inhibitors (PI).
Methods: Charts of 340 HIV patients followed at the Walter Reed Army Medical
Center were reviewed for age, race, gender, CD4 count, viral load, cholesterol
before and after starting (PI) therapy. Patients on no antiretroviral therapy and
those on other than PI therapy comprised the two control groups. For these
groups, cholesterol measurements within similar time frames as those on PI were
abstracted.
Results: 95 patients were in the group on no antiretroviral therapy (group 1),
were 90.5% male, 69.5% black, mean age 35 years. 87 patients were in the group
on therapy other than PI (group 2), were 86.2% male, 69.8% black, mean age 38.
158 patients were on a PI (group 3), were 88.6% male, 58.2% black, mean age
39. Median CD4 counts and viral loads were 477 (1-1319), 384 (8-1189), 314
(3-1411) and 9,965 (108-536,333), 3,545 (126-462,249), 1,237 (104-14,417) for
groups 1.2, and 3. Mean change in (d) cholesterol was.084 +/- 22.5 for group
1, 3.5 +/- 24 for group 2, and 28.9 +/- 39.7 for group 3. The comparison of
(d) cholesterol for group 3 with the 2 control groups was statistically significant,
p value <0.0005. (d) cholesterol for patients who had ever taken indinavir (n =
119) was 26.4, for ritonavir (RIT) (n = 36) was 40.9, for saquinavir (SAQ) (n =
31) was 30.8, for RIT/SAQ (n = 9) was 47.3, and for nelfinavir (n = 21) was 25.
The comparison of all of these PI groups with the control groups was statistically
significant with p values <0.001. 16% of patients had increases in cholesterol
>65 mg/dl, and 5.7% of patients had increases >100 mg/dl. All of these patients
were on PI.
Conclusions: Cholesterol levels can be expected to increase significantly in
HIV patients on PI therapy, most dramatically for patients on RIT and SAQ.
Clinical consequences of this increase in cholesterol are as of yet unknown.
12270 Anti-HIV-1 activity of and HIV-1 resistance profiles against
JE-2147 (KNI-764), a novel inhibitor of HIV-1 protease
Takamasa Ueno1, T. Mimoto1, M. Shintani1, H. Sato1, K. Yusa2, H. Mitsuya2,
H. Hayashi'. 1Pharm. & Biotech. Lab., Japan Energy Co. 3-17-35, Niizo-Minami,
Toda, Saitama, 335, Japan; 2Medicine Branch, National Cancer Inst. Bethesda,
MD, USA
Background: To define the anti-HIV-1 activity of and HIV-1 resistance profiles
against a novel allophenylnorstatine-containing HIV protease inhibitor JE-2147
(formerly called KNI-764).
Methods: Drug concentrations that inhibit HIV-1 replication by 50% (IC50)
were determined in antiviral assays using multiple viral isolates and cell types.
Endpoints employed were: (i) inhibition of HIV-1-induced cytopathic effects using
the tetrazolium reagent (XTT) or (ii) decrease in reverse transcriptase activity in
culture supernatants. Synergism of JE-2147 when combined with other protease
inhibitors was also determined using the computer program MacSynergy II. To induce JE-2147-resistant HIV-1 variants, CEM-SS cells were infected with HIV-1LAI
and were cultured in the presence of increasing concentrations of JE-2147.
Results: JE-2147 inhibited the replication of HIV-1, -2, and SIV with IC5o values
ranging 17 to 160 nM in vitro. JE-2147 was comparably active against HIV-1 in
the presence (ICso = 31 nM) and absence (IC5o = 26 nM) of 50% human serum in
vitro, indicating the low protein-binding property of the compound. JE-2147 also
showed substantial synergism in vitro (synergy volume >100) when combined
with saquinavir, indinavir, ritonavir, nelfinavir, or 141W94. When serially passaged
in the presence of JE-2147 in vitro, HIV-1 variants with decreased sensitivity to
JE-2147 (15-fold) emerged while the variants were still sensitive to currently available protease inhibitors (3-fold difference in ICso). In addition, JE-2147 remained
active (<2-fold difference in IC5o compared to wild type HIV-1) against various

HIV-1 variants resistant to saquinavir, ritonavir, and indinavir, indicating the resistance pattern for JE-2147 differs from other protease inhibitors examined. Amino
acid mutations found in the protease-encoding gene of the JE-2147-resistant
variants (LeulO-Phe, Ile47->Val, and Val82-- lie) were unique to the mutations
associated with other protease inhibitors.

Conclusions: A novel protease inhibitor JE-2147 has favorable profiles of antiHIV-1 activity and resistance in vitro. Further evaluation of antiretroviral activity of
JE-2147 is warranted.
12271 Determinants of failure of highlyactive antiretroviral
therapy (HAART)
Ferdinand Wit, R. Van Leeuwen, G.J. Weverling, S.A. Danner, P. Reiss, F. De
Wolf, J.M.A. Lange. Natec, Academic Medical Centre, F5-166 Meibergdreef 9
1105 AZ Amsterdam, Netherlands
Objectives: So far only limited information is available on the long term effect
of HAART. We compared the virological outcome of protease inhibitor containing
HAART in an outpatient clinic in Amsterdam.
Methods: Consecutive HIV-1 infected patients starting their first protease inhibitor containing combination therapy were studied. Patients who (1) never
exhibited an undetectable viral load or (2) after having reached undetectability,
exhibited a detectable viral load sometime during the follow up of 36 weeks were
considered to be virological treatment failures.
Results: 243 patients were studied, 76% were nucleoside RT inhibitor experienced. Protease inhibitors used were: indinavir (95), ritonavir (41), saquinavir
1200 mg tid (86), ritonavir+saquinavir (21). The median baseline viral load was
4.5 loglo (IQR 3.7-4.9). The median baseline CD4 cell count was 150 cells/mm3
(IQR 50-280). The proportion of patients with an undetectable viral load after 36
weeks was 77% (IDV 88%, RTV 90%, SQV 58%, RTV+SQV 81%). The median
CD4 cell count increased to 300 cells/mm3 (IQR 170-430) after 36 weeks. The
rate of virological treatment failure after 36 weeks was 33% (IDV 19%, RTV
24%, SQV 54%, RTV+SQV 29%). In a multivariate analysis the risk factors for
failure were baseline viral load (OR 1.5 per loglo increase, 95% CI 1.0-2.4),
baseline CD4 cell count (OR 0.6 per 100 CD4 cells increase, 0.4-0.8), use of
saquinavir (OR 4.6, 2.4-9.0), and the use of a protease inhibitor as the only new
drug (OR 2.9, 1.1-7.6). After 36 weeks only 53% of the patients still used their
original combination (IDV 58%, RTV 39%, SQV 49%, RTV+SQV 71%). Change
of therapy was caused by drug related toxicity in 56%.
Conclusion: Predictors for long-term failure of HAART were high baseline viral
load, low baseline CD4 cell count, and the use of a protease inhibitor as the
only new drug. The use of saquinavir (Invirase formulation, 1200 mg tid), was
associated with a higher risk of virological treatment failure. Extended follow up
will be available at the time of conference.
S437*/12272 1 Regional survival differences across Europe in HIV
positive patients reflect differences in antiretroviral
treatment. Data from the Eurosida study group
Antonio Chiesi1, Stefano Vella1, L.G. Dally1, A. Macroft2, A.N. Phillips2,
V. Miller3, S. Stazewski3, J.D. Kundgren4. 11stituto Superiore Di Sanita', Lab.
Virology, Viale Regina Elena 299, Rome 00161, Italy; 2Royal Free Hospital,
London, UK; 3 Frankfurt University, Frankfurt, Germany; 4Hvidovre Hospital,
Hvidovre, Denmark
Background: The expanded access to combination therapy and the introduction
of protease inhibitors (PI) in clinical practice seems to be producing a marked
drop in mortality rates of patients with HIV disease.
Objectives: To compare mortality rates of patients with HIV in 3 different regions
of Europe (Northern - NE, Central - CE and Southern - SE) from 1994 to 1997,
with respect to RTI and PI treatments, PCP prophylaxis and CD4 lymphocyte
count.
Design: Data from EuroSIDA, a prospective, European, multicentre cohort
study on patients with HIV were analysed. In a subgroup of subjects with
CD4 > 200/mm3 at enrollment, Kaplan-Meier curves were used to investigate
survival from the first time that the CD4+ cell counts fell below 200/mm3. A
'per-100-person-years' analysis was used to compare crude death rates in 3
CD4+ categories (<50, 50-99, 100-200/mm3) within regions. Cox proportional
hazards models were used to compare death rates between regions after adjusting for diagnosis of AIDS, treatment and CD4+ lymphocyte count (time-dependent
covariates), as well as sex and risk group.
Results: Of 4,491 patients recruited to the study, 20.5% were females, median
age 35.5 years and median CD4 lymphocyte count 209 cells/mm3 (90% range
10-558). SE had the highest proportion of intravenous drug users (43.2%) and
heterosexuals (24.2%), while NE had the highest proportion of homosexuals
(63.9%). SE also had the youngest population (median 33.5 years) and the highest proportion of females (25.6%). NE had the lowest median CD4 lymphocyte
count (190 cells/mm3) and the highest proportion of AIDS patients (34.3%), while
SE had the highest median CD4 lymphocyte count (221 cells/mm3) and the lowest
proportion of AIDS patients (29.4%). Overall, 31.7% of patients were AIDS at
enrolment. A total of 73.4% of patients had already been treated with reverse
transcriptease inhibitors (RTI) and only 0.8% had received PI treatment. A greater
proportion of patients in SE and CE had already taken RTI (77.6% and 75.9%,
respectively) compared to NE (66.8%, p < 0.001). CE had the highest proportion
of PI treated patients (1.5%). Follow-up: during a median follow-up of 19 months

(range 2-38 months) there were 1,262 deaths (28.1%). At any time before or
during follow-up, 86.8% of NE patients had received RTI compared to 91% of SE
and 92% of CE patients. Similarly, a significantly (p < 0.001) greater proportion
of CE patients (51%) had taken PI compared to 42.3% of NE and only 30.3% of
SE patients. Crude death rates, stratified by CD4 category, were significantly (p
< 0.05) lower in CE compared to SE and NE in patients with <50 cells/mm3 (31.6
vs 55.5 & 45.0, resp.) and with 51-100 cells/mm3 (6.1 vs 14.5 & 13.1, resp.). In