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

12th World AIDS Conference

Abstracts 22346-22350 333

22346 A two years analysis of the viral load of participants of
the Swiss HIV cohort study with and without
antiretroviral therapy
Peter ERB', M. Battegay2, M. Egger3, M. Richenbach4, L. Jeannerod4. Swiss
HIV Cohort Study; 1Institute for Medical Microbiology, Petersplatz 10 CH-4003
Basel; 2University Hospital Outpatient Dep.; 4Chuv Coordination and Data
Center SHCS, Switzerland; 3University of Bristol Dep. Soc. Medicine, Bristol, UK
Objectives: To examine the effect of antiretroviral treatments on viral load levels
in the Swiss HIV Cohort Study.
Methods: Retrospective study of HIV-infected adults in 7 centers in Switzerland
with clinical and laboratory follow-up every 6 months. Roche HIV Amplicor Monitor
was used for viral load measurements. 3687 participants with 10963 follow-up
visits between June 95 and June 97 were included. 1678 received antiretroviral
therapy at the time of first viral load measurements. Out of 2009 participants
without treatment, 755 started with antiretroviral therapy and were followed up.
Monotherapy with 1 RT-inhibitor (RT-mono, n = 57), combination therapy with at
least 2 RT-inihibitors (RT-combo, n = 460) and combination therapy with RT and
PI-inhibitors (RT/PI-combo, n = 238) were compared.
Results: Antiretroviral treatment led to reductions of viral laod, with pronounced
differences between groups: -2.45 log maximal reduction for RT/PI-combo, -1.46
log for RT-combo and -0.64 log reduction for RT-mono. Viral load decreased by
more than 0.5 log in 82% of the patients with RT/PI-combo, in 76% of RT-combo
and in 37% of RT-mono. Only 3% of patients were below detection limit (400
copies/ml) before starting therapy. This proportion increased to 16% for RT-mono,
30% for RT-combo and 45% for RT/PI-combo. The percentage of patients with
more than 100'000 copies/ml dropped from 32% before therapy to 8% and 10%
with RT/PI-combo and RT-combo, respectively and remained unchanged for
RT-mono. A decrease followed by increase of viral load was found in 7% with
RT/PI-combi and 11% with RT-combo. No reduction or an increase of viral load
was observed in 11% with RT/PI-combo, 13% with RT-combo, but in 63% with
RT-mono.
Conclusions: In a large cohort outside the setting of randomised trials
RT/PI-combo lead to a significant reduction of viral load in a high proportion
of participants. Only few patients in this group experienced a viral rebound due
to virus resistance or non adherence or showed no response. Highly active
antiretroviral therapy with PIs was effective in a community setting in Switzerland.
22347 Response to therapy with protease inhibitors in clinical
practice
Richard Moore1, J.C. Keruly2, J. Gallant2, R.E. Chaisson2. 11830 E Monument
St Room 8059 Baltimore Maryland 21205; 2Johns Hopkins University Baltimore
MD, USA
Objective: Combination antiretroviral therapy (COMBO) has been highly efficacious in clinical trials in maintaining an undetectable viral load (UNDET) and
slowing clinical disease progression. The utilization and effectiveness of this
therapy in clinical practice is less well established.
Methods: We analyzed response to therapy, viral load and CD4 in patients
receiving longitudinal care a large urban academic HIV specialty practice in the
US.
Results: COMBO+Protease Inhibitor (PI) was used in 422 patients in 1996-97.
Of these, dual PI was used in 18%. Concomitant dual nucleosides were used in
90% and non-nucleoside/nucleosides in 10%. Median CD4 = 133 (range: 1-518)
and median HIV-1 RNA = 58,000 (range: 800- >750,000) at first PI use; 79%
had received prior antiretroviral therapy. Patients who were illicit drug users (IDU)
were less likely to be prescribed protease inhibitors (PI) than non-IDUs, as were
people who had compromised living or economic situations (p <.05). Only 55%
of patients who started a PI had UNDET at 6 months, and 48% at 12 months.
Factors associated with UNDET included evidence of good adherence, good
tolerance, CD4 > 200 cells/mm3 at start, HIV-1 RNA < 50,000 copies at start,
and those who had all new antiretroviral drugs at first use of PI. Age, sex, race,
and specific past antiretroviral therapy were not associated. IDUs were less likely
to have UNDET, but the difference was not significant.
Conclusions: COMBO+PI is associated with lower rates of UNDET in clinical
practice than in clinical trials. Several factors associated with both prescribing
and response to therapy were identified.
|223481 OZCOMBO 1: A randomised, open-label, comparative,
phase IV trial to evaluate the efficacy and safety of triple
drug combination antiretroviral regimens for the
treatment of HIV infection
Jeffrey Hudson1, Andrew Carr2, D.A. Cooper1, J. Chuah1, M. Law1. 1NCHECR
Level 2 376 Victoria Street Darlinghurst New South Wales; 2St. Vincent's
Hospital Sydney, NSW, Australia
Objectives: To generate comparative, randomised data describing antiviral efficacy, safety and durability of three indinavir containing triple-drug regimens in

antiretroviral naive patients.
Design: Open label, randomised, multicentre phase IV study.
Methods: Adult HIV+, antiretroviral naive patients with HIV RNA > 30,000
copies/mi were randomised to AZT/3TC/indinavir, ddl/d4T indinavir or d4T/3TC
indinavir, and assessed monthly for 52 weeks for adverse events, HIV RNA and
CD4 counts. Plasma was collected and is being stored for subsequent viral resistance testing and for batched viral load analysis (Roche HIV monitor assay by

the Standard and Ultra Sensitive methods; range of quantification 50 to 750,000
copies/ml plasma).
Results: As of December, 1997, 109 patients were enrolled with a median
follow up 24 (range 4-57) weeks. Most (82%) were male with a median age of
38 (range 22-69) years, median weight 72 (range 43-110) kg. The median CD4+
cell count at baseline was 298 (range 4-804) and the median HIV RNA was
86,100 copies (log 4.9) with range (1800->750,000). 83% had no prior AIDS
defining condition. As of January 1998, 21 patients have withdrawn. No patients
were withdrawn from the study due to failure of randomised therapy (as defined
by viral load). Of these 21 withdrawals, 11 were for serious adverse events (SAE)
related to study drugs, 1 SAE not study drug related, 9 due to patient requests
related to factors such as mild symptoms/adverse events or personal reasons
such as pill burden.
Conclusion: At least 21/109 (19%) patients cannot tolerate combination therapy of double nucleoside/protease inhibitor after a median of 24 weeks. This
study will produce valuable data comparing the three double nucleoside/protease
inhibitor regimens for treatment initiation and provide additional information on
strategy evaluation concerning combination therapies.
S22349 Incidence and predictors of virologic failure of
antiretroviral triple drug therapy in a community based
cohort
Stefan Zimmerli1 2, D. Paris3, B. Ledergerber3. 1Friedbuehlstrasse 51 Ch-3010
Bern; 2lnst. For Medical Microbiol. Bern University Bern; 3Zurich University
Hospital Zurich, Switzerland
Objective: To determine virologic response to antiretroviral triple drug therapy
including a protease inhibitor and two nucleoside analogues, in an unselected
cohort of HIV-infected individuals.
Methods: Retrospective study of all patients seen at the HIV outpatient Clinic
of Zurich University Hospital who started triple drug therapy between 10/95 and
3/97, had a viral load (VL) assay >500 copies/ml (PCR; Roche Amplicor) and
CD4 cell count within 6 weeks before the start of therapy, and at least 2 VL assays
during a follow-up of no less than 6 months (n = 274; 21% female; 26% IDU).
Results: Failure of treatment to reduce viral load to <500 copies/ml within
6 months (initial failure) occurred in 51/274 (19%) of patients [5/74 (7%) of nucleoside analogue (NA) naive and 46/200 (33%) of NA-experienced patients].
Logistic regression analysis revealed the following independent risk factors for
initial failure: High viral load (OR = 2.41, per log VL, p = 0.0009); addition of a
protease inhibitor to established NA treatment (OR = 5.18, compared to naive
patients, p = 0.008); treatment with saquinavir (OR = 3.55, compared to indinavir,
p = 0.03) and duration of prior NA treatment (OR = 1.33, per year, p = 0.04).
After initially decreasing to <500 copies/ml, VL rebounded above this threshold
in 71/223 (32%) of patients [14/69 (20%) of nucleoside analogue (NA) naive and
57/154 (37%) of NA-experienced patients]. In proportional hazard analysis none
of the potential risk factors (including high VL) was significantly associated with
VL rebound. However, in 40/71 (56%) of patients with VL rebound, adherence
problems (25/71) or treatment interruptions (15/71) were mentioned in the charts.
Conclusion: At 45%, the overall virologic failure rate of antiretroviral combination therapy in an unselected community based cohort is substantially higher
than was reported for clinical trials. High baseline viral load, addition of new drugs
to an existing regimen, use of saquinavir, and long prior NA treatment are risk
factors for initial virologic failure. Improving adherence may help to make initial
viral success more durable.
S22350 Ritonavir-containing dual protease inhibitor regimens
may have synergistic antiviral effects in patients - Based
on in vitro model
Ann Hsu, R. Granneman, A. Molla, S. Vasavanonda, A. Japour, D. Kempf,
E. Sun. Abbott Laboratories, 100 Abbott Park Road, D4PK Bldg. 13A, Abbott
Park, IL, USA
The combination of ritonavir (R) and saquinavir (S) with or without reverse transcriptase inhibitor (RTI) regimens has beenshown to be highly active in numerous
clinical studies. In study M96-462, over 90% of patients treated for 48 weeks
have nondetectable viral loads with the majority of these patients receiving R +
S without RTIs. We report our investigation of the roles of individual drugs in
the net antiviral effect observed in dual protease inhibitor therapies. The in vitro
anti-HIV3B activity in MT4 cells was determined in a medium containing 50%
human serum and 10% fetal calf serum for six concentration (cone) levels of
R-alone and S-alone, and for 36 R-S conc combinations. For each of R and S,
the cone range was selected based on the estimated protein-binding corrected
in vitro EC50 (0.125 to 4 /ng/mL for R and 0.0625 to 2 uig/mL for S). A total
of 12 replicates were performed for each cone or cone combinations. The in
vitro EC50 values were determined using curve fitting assuming an Emax model
with a shape factor, g, for R and S, respectively. With these fitted protease
inhibitor-alone EC50 values, the extent of inhibition of viral activity was modeled
for the 36 combination concs assuming an additive relationship. The observed

activity of the two inhibitors in combination was significantly higher than predicted,
indicating in vitro synergism. The synergistic effect was most pronounced when
both drugs were at cones near the EC50. Since the R/S conc ratio in patients
is approximately 4:1 in the 400R/400S combination, the ECso for this conc ratio
was also determined using the Emax model. Based on this EC50, the calculated
inhibition of viral relication ranged from 92.2 to 99.7% over a 12-h dosing interval, compared to 70.1 to 98.4% assuming additivity. Similar in vitro synergy