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

12th World AIDS Conference

Abstracts 12291-12295 71

12291 Protease inhibitors among injecting drug users in
Amsterdam: Cumulative incidence, determinants and
impact
Janke Schinkel, R.A. Coutinho, E.J.C. Van Ameijden. Municipal Health Service
Nieuwe Achtergracht 100, 1018 WT Amsterdam, The Netherlands
Objective: To investigate in our cohort of injecting drug users (IDUs) 1) the
proportion of subjects that started combination treatment with protease inhibitors
(PIs) 2) determinants of initiating treatment and 3) the impact of treatment on
CD4 counts, viral load and drug use.
Methods: 1) We estimated the cumulative incidence of initiating combination
treatment between 1 July 1996 (when PIs generally became available in the
Netherlands) and 30 June 1997 applying the Kaplan-Meier method. Persons
entered the risk set at 1 July 1996 if they had a CD4 count < 500/p~ (n = 124)
or if their CD4 count dropped below -500/pIl during follow-up (n = 12). 2) Potential determinants of initiating treatment were identified in a case-control study.
Cases were IDUs being treated with PIs (n = 29), controls were IDUs eligible for
treatment, but who were not being treated with PIs (n = 74). 3) CD4 counts and
drug use before and after initiating treatment were compared in a subgroup with
sufficient follow-up (n = 16, mean follow-up 23 weeks, range 15-31 weeks) using
the Wilxocon paired test. Viral load after initiating treatment was available for 23
of the 25 cases who continued treatment.
Results: The cumulative incidence of getting treatment among IDUs was 29%
(95% Cl 20%-40%) in the year following 1 July 1996 (n = 29). Being treated was
independently associated with having received antiretroviral treatment in the past,
stable housing and a CD4+ count -200//1. After a mean interval of 23 weeks
a significant increase in CD4 counts was observed (from 191//p to 233 //, p =
0.01), the viral load was below detection level in 70% (<1000 copies/ml). Mean
frequency of using drugs decreased from 32 to 15 times per months (p = 0.03).
Conclusion: Among IDUs eligible for treatment, the proportion of IDUs being
treated with PIs is small (29%) one year after the date that protease inhibitors
generally became available in the Netherlands. Moreover, 56% of IDUs who are
not being treated have a CD4 count -200//p. However, after a mean follow-up of
23 weeks, treatment was effective in 70% (using viral load below detection level
as criterium), mean CD4-count increased significantly and a significant reduction
in drug use was observed.
12292 Effectiveness of a systematic educational approach to
alter suboptimal dosing of indinavir
Sarah Lewis, Andrea Ho-Kean, L.K. Gajewski, G.J. Vanscoy. Stadtlanders
Managed Pharmacy Services, 600 Penn Center Boulevard Pittsburgh PA 15235,
USA
Issue: Determination of the long-term effectiveness of a prescriber-targeted educational process directed at optimizing indinavir dosing in a large US patient
base.
Project: HIV/AIDS patients on indinavir therapy must be maintained on an
optimal dose of indinavir to delay emergence of resistant viral strains. Aside
from 800 mg q8h, other doses including 1200 mg bid and 600 mg q8h may be
warranted by patient compliance issues and concomitant disease/drug issues,
respectively. The need for a consistent, effective and efficient mechanism for
pharmacist management of indinavir suboptimal dosing issues was identified. A
proactive prescriber-targeted educational process was developed to address this
need. All prescriptions received by the US's largest specialty HIV/AIDS pharmacy
for indinavir doses that were outside of suggested guidelines were identified by
pharmacists and forwarded to a clinical pharmacy specialist for further assessment. Information addressing indinavir dosing and viral resistance was sent to
the physician in reference to each sub-optimal case. A physician action form
was enclosed to collect patient-specific clinical and laboratory data, as well as to
provide justification or modification of the current indinavir dose. Prescribers were
asked to complete and return the form within 24 hours of receipt.
Results: This process was evaluated after one year and was successful in optimizing indinavir dosining in 92% (351/381) of patients initially prescribed ndinavir
outside of suggested dosing guidelines. A subsequent six month reassessment
reveals that over this time period, the frequency of suboptimal indinavir prescribing
has decreased dramatically.
A prescriber-targeted educational process maintains effectiveness in optimizing
indinavir therapy. The decrease in the frequency of suboptimal indinavir prescribing over the past six months may at least partly be attributed to the long-term
effectiveness of a prescriber-targeted educational process.
S12293 Persistent CD8+ T lymphocyte activation despite
prolonged protease inhibitor therapy suggests residual
viral activity
Gilbert R. Kauffman1 2, J. Zaunders2, A.D. Kelleher2, C. Duncombe3, A. Carr4,

D.A. Cooper5. 16 Ilumba Place Bangor NSW 2234; 2Centre for Immunology St.
Vincent's Hosp. Sydney; 3Holdworth House General Practice, Sydney; 4HIV
Medicine unit St. Vincent's Hosp, Sydney; 5Natl. Centre in HIV Epidemiology,
Sydney, Australia
Objectives: To analyse the activation of circulating CD8+ T lymphocytes in HIV
positive patients on a highly active anti-retroviral therapy (HAART) with low to
undetectable viral load.
Methods: 23 male HIV patients (age 40 ~ 7) were included, who were treated
with SQV/RTV and two nucleosides for at least 48 weeks. 4-colour flow cytometry

was used to measure CD 38, HLA-DR, CD45RA, CD28, LFA-1 and CD62L on circulating CD3+CD8+ T lymphocytes cross-sectionally at 48 weeks. Patients could
be grouped into complete responders (CR, viral load -400 copies/ml, n = 16) and
incomplete responders (IR, viral load >400, n = 7). They were compared with (i)
a HIV negative control group (HIV, n = 35), (ii) a treatment-naive asymptomatic
HIV+ group (HIV', n = 18) and (iii) a HIV+ group on nucleoside therapy >24
weeks (HIVnuc, n = 25).
Results: Median CD4+ and CD8+ T lymphocyte count in the whole cohort was
441 (interquartile range: 259-476) and 1141 (913-1653) cells/pl, respectively.
CD38+HLA-DR+ cells as a percentage of CD3+CD8+ lymphocytes were 16%
(12-24) compared with 2.5% (1.6-3.4) for the HIV-, 42% (36-47) for the HIV~
and 39% (26-48) for the HIV"uc group. CD28-CD8+ T lymphocytes were higher
in the cohort than in the HIV group (59% vs 26%), but lower than the HIV~ and
HIVnuc groups (73%, 69% respectively). However, there was no difference of
these activation markers between the CR and IR groups within the cohort. The
number of naive CD45RA+CD62L+ CD8+ T lymphocytes was 187 (107-287)
cells/l in the cohort, 138 (108-261) in HIV and 82 (64-127) in HIVnuC. There
was a higher percentage of naive CD8+ T lymphocytes in the CR vs the IR group
(23 vs 12, P < 0.05).
Conclusion: Despite an undetectable viral load during an observation period
of 48 weeks the increased activation of CD8+ lymphocytes suggests the continuous presence and activity of HIV. We could not detect a difference in activation
between patients with undetectable viral load and very low viral load, but the
trend towards fewer naive CD8+ T lymphocytes in patients with detectable viral
load may indicate a less favourable prognosis in this group.
S12294 Virological response to protease inhibitors in an HIV
clinic cohort
Schlomo Staszewski1, Andrew Phillips2, V. Miller1, C.A. Sabin2, A. Carlebach1,
R. Brodt1, A. Berger', A. Hill3. Zentrum Der Inneren Medizin Theodor-Stern-Kai
7 60590 Frankfurt, Germany; 2Royal Free Hospital School of Medicine, London;
3Glaxo Wellcome Research and Development, GreenFord, USA
Objectives:- To assess factors associated with virological response to protease
inhibitors in a large clinic population.
Design: Prospective study of patients starting protease inhibitor therapy.
Methods: 663 patients from a large outpatient clinic were followed a median
of 10 months from the time of starting a protease inhibitor until 1 September
1997. Viral load and CD4 count measurements were made on average every 4.2
weeks. All new AIDS diseases were recorded.
Results: Overall there was a 71% (95% Cl 67%-75%) probability of the patients achieving an undetectable viral load (<500 copies/mL) by 24 weeks after
starting the protease inhibitor. In a multiple Cox regression model, those with
lower initial viral load (relative hazard 0.63; p = 0.0001), those starting more other
new drugs at same time as the protease inhibitor (RH 1.99 for two vs. none; p =
0.0001) and those using indinavir or nelfinavir were more likely to achieve such
levels. In those 413 patients achieving undetectable viral load within 24 weeks,
there was an estimated 50% (95% Cl 43%-57%) probability of re-appearance
of detectable viral load by 52 weeks from the first undetectable value. Again,
those who had started more other new drugs at the same time as the protease
inhibitor (RH 0.53; p = 0.003) tended to experience a lower probability of viral load
re-arrearance, as did those with higher initial CD4 count (RH 0.75 per 100/mm 3
higher; p = 0.0001). Those who took Saquinavir achieved less durable virological
responses than those who took other protease inhibitors. There was no evidence
that drug-naive patients experienced significantly poorer Virological effects than
drug-experienced patients so long as they started the same number of new
drugs.
Conclusion: Starting at least two other new antiretroviral drugs simultaneously
with protease inhibitor therapy offers the best chance of achieving sustained
undetectable viral load.
12295] Ritonavir (RTV)-saquinavir (SQV) in protease
inhibitor-naive patients after 72 weeks
John Mellors1, A.J. Japour2, J. Leonard2, E. Sun2, Y. Xu2, M. Salgo3. M97-462
Study Groups; ' Univ. Pittsburgh A450 Crabtree Hall, 130 De Sotto Pittsburgh Pa
15261; 2Abbott Laboratories, Abbott Park, IL; 3Hoffmann Laroche, Nutley, NJ,
USA
Objectives: Summarize the safety and efficacy of RTV-SQV regimens through 72
wks in protease inhibitor naive HIV-positive pts with CD4 counts of 100-500/ L.
Design: Prospective, open-label randomized study with four RTV-SQV dose
regimens. RTIs were added for pts whose viral load did not reach or maintain
<200 copies/mL after 12 wks. After wk 48 all pts were allowed to dose reduce to
400-400 mg BID and add up to two additional RTIs (nucleoside or non-nucleoside)
for any reason.
Results: The baseline plasma HIV RNA was 4.63 logio copies/mL and CD4
count was 273///L. 73% (103/141) remain on study at Wk 72 of follow-up. Wk 72
data is available from 93% (96/103). Overall, 90% (86/96) have plasma HIV RNA

<200 copies/mL More than half had been dose reduced to RTV-SQV 400-400
mg BID by 24 wks. Median CD4 cell increase was 188//uL at Wk 72. Thirty-three
pts had RTIs added to their regimen after a median of 186 days; three of these
pts discontinued the study. Twenty-seven pts added RTI for HIV RNA >200
copies/mL. In 85% (23/27) HIV RNA decreased to <200 copies/mL at wk 72
(preliminary data). Updated safety data will be presented. There have been no
reported CDC-defined AIDS events and no deaths reported through Wk 72.