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

334 Abstracts 22351-22355

12th World AIDS Conference

were also observed between ritonavir and indinavir and between ritonavir and
nelfinavir.
The combination of ritonavir and saquinavir has antiviral synergy in vitro. This
finding may explain the profound antiviral effect observed in clinical trials using this
regimen. The use of in vitro virological combination data modeled onto observed
plasma levels for protease inhibitor combinations may provide a means for the
prediction of the efficacy of dual protease inhibitor regimens.
22351 Risk factors for virologic failure in a cohort of HIV
patients receiving highly active antiretroviral therapy
(HAART)
Zelalem Temesgen, Elie Berbari, J.H. Henely, L.L. Estes, A.J. Wright. Mayo
Clinic, 200 First Street, SW, Rochester, MN, USA
Background and Objective: While HAART has had a significant impact on the
morbidity and mortality of HIV infected persons, recent reports have indicated up
to 53% virologic failure rate in a clinical, non-research, setting. We performed
a retrospective analysis of patients seen in our clinic to determine the virologic
failure rate and identify possible risk factors.
Methods: All patients with HIV attending our clinic between March 1, 1996 and
December 31, 1997 were identified, and their medical records reviewed. Patients
were followed until the occurrence of HAART failure or the last visit date. Patients
were eligible for analysis if they had a minimum of 3 months follow-up on their first
HAART regimen. HAART was defined as a protease inhibitor containing regimen
of three or more antiretrovirals. HAART failure was defined as a reduction of
less than 1 log in plasma HIV RNA (viral load) at 3 months of therapy or a
viral load of >500 copies/ml at any measurement thereafter. Compliance was
defined as adherence by the patient to 80% or more of the prescribed regimen.
The cumulative probability of treatment failure was estimated by the technique
of Kaplan-Meier survival method and the comparisons between subgroups were
performed with the use of Cox's proportional hazards model.
Results: Of a total of 119 patients seen in our clinic during the study period,
54 met the eligibility criteria and underwent further analysis. The mean follow-up
was 322 days, range (75-574). The cumulative incidence of HAART failure was
6% (95% confidence interval [CI], 0%-12%) and 31%, (95% CI, 16%-46%) at
6 and 12 months respectively. The only identifiable risk factor for HAART failure
in this cohort was previous antiretroviral therapy (relative risk [RR], 3.9; 95%
CI, 1.3-12.2) P = 0.0179. The cumulative incidence of HAART failure at 1 year
among antiretroviral naive and experienced patients was 16% (95% CI, 0%-33%)
and 47%, (95% CI, 25%-69%) respectively. The following variables did not attain
statistical significance as variables associated with HAART failure: Prior history of
opportunistic infections (RR, 1.2; 95% CI, 0.4-3.3), compliance with antiretrovirals
(RR, 0.4; 95% CI, 0.1-1.2), viral load prior to starting HAART more than 50,000
copies/ml (RR, 0.5; 95% CI, 0.2-1.4) or 20,000 copies/ml (RR, 1.4; 95% Cl,
0.4-4.9), a CD4 cell count of more than 50 cells/mm3 (RR, 0.8; 95% Cl, 0.3-2.6)
or 100 cells/mm3 (RR, 1.4; 95% CI, 0.5-4) prior to starting HAART.
Conclusion: In this patient population, 31% of patients failed HAART after
one year of follow-up. Prior antiretroviral therapy was identified as a significant
predictor of virologic failure to subsequent HAART.
122352 Co-administration of indinavir and nelfinavir:
Pharmacokinetics, tolerability, anti-viral activity, and
preliminary viral resistance
Alfred Saah1, Sharon Riddler2, D.V. Havlir3, K. Squires4, R. Anderson5,
B. Kerr5, K. Yeh1, P. Deutsch1. 1Merck & Co, PO. Box 4, West Point, PA;
2 University of Pittsburgh, Pittsburgh, PA; 3 University of California, San Diego,
CA; 4University of Alabama, Birmingham, AL; 5Agouron Inc, La Jolla, CA, USA
Objective: 1) To determine safety, optimal dosing/pharmacokinetics (PK), and
antiviral activity of combination indinavir (IDV) and nelfinavir (NFV). 2) To evaluate
the resistance mutations for those patients (pts) with incomplete viral suppression.
Methods: Initial dosing with IDV 1000 mg &NFV 750 mg q12h given to HIV-1
infected, protease inhibitor naive pts with serum HIV RNA > 30,000 copies/ml,
CD4 > 100 cells/mm3. NFV dose was increased to 1000 mg q 12 h when steady
state PK evaluation demonstrated lack of IDV effect on NFV levels.
Results: IDV 1000 mg & NFV 750 mg q12h resulted in steady state IDV levels
similar to q8h dosing but low NFV trough (0.7 mg/L compared to 1.5 mg/L for
750 mg TID). NFV 1000 mg q12h with IDV resulted in a 35% increase in NFV
trough over the NFV 750 mg q12h dose. IDV PK at this dose level is pending. An
additional dose increase of NFV to 1250 mg q12h is planned with PK evaluation.
After a median of 24 weeks, 11/18 (61%) who received protocol therapy (11/21
overall) had HIV RNA < 400 copies/ml, of whom 6 were negative by ultradirect
(<50 copies/ml). CD4 cell counts increased a median of 133 cells/mm3. Treatment
was discontinued in 5 pts (1 rash, 2 virologic failure, 2 noncompliance). Other
adverse events (_ grade 2) included: diarrhea (6), bloating/cramps (4), nausea
(4), kidney stone (2). Genotyping on viral isolates from 2 pts demonstrated an
isolated protease substitution at L90M. Genotyping of additional isolates is in
progress.
Conclusions: Concomitant administration of IDV & NFV results in enhanced

IDV levels. The optimal twice-daily NFV dose in this regimen will be defined and
repeat PK of both drugs will be performed. Combination IDV/NFV was generally
well tolerated. In this study, 61% of pts receiving IDV/NFV had HIV RNA below
400 copies/ml at week 24.

22353 Long-term effectiveness of HAART in a large clinical
group and the role of viral resistance in treatment failure
William M. Reitner1, P.J. Cimoch2, D.S. Berger3, D.M. See2, D.E. Vorce4,
P. Darmanin4, B. Baker3. 12020 NE 48h Court, Fort Laderdale, FL.; 2Center for
Special Immulogy, Irvine; 3Center for Special Immulogy, Chicago; 4 Center for
Special Immulogy, Fort Lauderdale, USA
Objective:To determine the long-term effectiveness of Highly Active Antiretroviral
Therapy (HAART) in a general clinical population and to consider the role of viral
resistance in treatment failure.
Methods: In patients receiving HAART at three different treatment sites, longitudinal HIV-1 plasma load was determined by Chiron's Quantiplex'" HIV-1 RNA
2.0 assay (bDNA; undetectable at <500 copies/ml). HAART was administered in
accordance with the principals later outlined in the consensus Guidelines developed by the Panel on Clinical Practices for Treatment of HIV Infection, November
1997.
Results: A total of 420 patients were followed for an average duration of 13.1
months (SD = 6.2). Each had at least three data points with group averages
of 5.8 data points obtained every 11.8 weeks. HIV-1 RNA remained detectable
despite treatment efforts in 140 patients (33%) followed for an average duration of
12.7 months (SD = 6.3). HIV-1 RNA became undetectable in 280 patients (67%).
However, in 141 of these patients (50%) HIV-1 RNA again became detectable
during average follow-up of 15.1 months (SD = 5.7). In only 139 of these patients
(50%) did HIV-1 RNA remain persistently undetectable during average follow-up
of 11.4 months (SD = 5.7). In a sub-group of 29 patients with insufficient or
non-sustained therapeutic responses, partial or full HIV-1 genotype determination
(Specialty Laboratories, Santa Monica, CA, or Univ of Cal, Irvine, CA) indicated
that all 29 had mutations consistent with resistance to at least one of the agents
in their therapeutic regimens. In a parallel group of 6 treatment naive patients full
plasma HIV-1 genotype determination (Univ of Cal, Irvine, CA), revealed that all
6 had pre-treatment mutations consistent with resistance to various antiretroviral
agents often included in HAART regimens.
Conclusions: Despite HAART, a significant number of patients did not have
a full therapeutic response and continued to produce measurable plasma
HIV-1RNA. In half the patients whose plasma HIV-1RNA became undetectable,
the response was not sustained. Determination of pre-treatment viral resistance
probability patterns with appropriately chosen drug combinations may be necessary to achieve better long-term viral suppression.
S22354 Combination therapy with protease-inhibitor during six to
twenty one months in HIV infected patients followed by a
single physician
M. Kirstetter1, A. Daguenel-Nguyen2, G. Kreplak3, J.B. Guiard-Schmid2
J. Frottier2. 112 Rue Theophile Roussel, 75012 Paris; 2Hopital St. Antoine,
Paris; 3Laboratoire Chemin Vert, France
Objectives: Retrospective study to assess the immuno-virological efficacy of PI
correlated to compliance in patients with advanced disease, heavily pretreated
for the major part. This study concerns only patients treated in private practice.
Methods and Design: Data of each patient were filed according to specific
criteria. 87 patients (Mean age of 41 years, 79 men, 9 women) were evaluated
at day 0, M1, M3, M6, M9, M12, M15, M18, M21 on clinical evolution and events,
CD4 cell count, viral load (bDNA Chiron), tolerance and compliance. 50.6% of
patients were at CDC stage C3, 30.5% at stage B and 12.9% at stage A. Mean
follow-up was 9 years [1-12]. Mean duration of prior antiretroviral therapy (nucleoside analogs) was 56 months [16-96]. 60.3% of patients received two NA,
39.7% three NA. Only 14 pts were naive. When PI was started, change of 2 NA
was possible in 24.1% of patients (Group 1) change of 1 NA in 41.4% (Group 2),
and change of NA was impossible in 18.4% (Group 3). PI was IDV in 78.2% of
cases, and RTV in 21.8%. The most common NA combination were D4T-3W and
AZT-3TC. Compliance was assessed during follow-up according to interviews
personnal history and tolerance.
Results: At baseline, mean CD4 cell count was 136/mm3 and mean viral
load 5.9 log for all groups (Group 1: CD4 164/mm3, VL 5.0 log; Group 2: CD4
75.5/mm3, VL 4.6 log, Group 3: CD4 82.5/mm3, VL 4.8 log, and naives patients:
CD4 170.5/mm3, VL 5.1 log). At M12, for all groups, mean CD4 increase was
106/mm3, and mean VL decrease was 1.5 log. Results were better for naive
patients and group 1 patients. At M18, 59% of all patients have VL under 500
copies/ml. Clinical score globaly improved and results were directly correlated to
compliance.
Conclusion: Combination therapy with PI is associated with clinical and immunovirological improvement even in heavily pretreated patients. This study
performed in private practice shows that compliance is determinant for a better
efficacy of treatment. In addition, it is paramount to establish with the patient a
relationship based on trust and communication.
| 22355 Patterns of antiretroviral therapy (ART) use from
1988-1998 in a cohort of 120 HIV+ patients

MichaelS. Saag1, Stephanie Call1l, A. Westfall1, S.V. Pham2, A. Chatwood1,
G. Cloud1, J. Raper1. 1 University of Alabama at Birmingham 908 20th Street,
South Birmingham AL; 2Glaxxo Wellcome, Res. Triangle Pk. NC, USA
Background: With the increased availability of new antiretroviral agents and the
use of combination therapy, the number of potential combinations has increased