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

12th World AIDS Conference

Abstracts 12301-12304 73

durable-93/105 (89%) continue to respond after 48 weeks of treatment, while
the 39 PCR but not Ultra responders had 5/39 (13%) continue to respond; the
remaining 34 patients relapsed prior to 38 weeks of response. Time to response
(baseline to response) had no significant influence on the duration of response
at 48 weeks, r = -0.178 (p = 0.26). Time to response may be an indicator of
whether a PCR responder will be a Ultra responder. At 750 mg NFV, patients who
also qualified as Ultra responders had earlier response to PCR than the PCR
responder-only patients (medians of 6 vs. 12 weeks).
Conclusions: The findings above can be utilized to formulate treatment strategy customized for individual patient to maximize the duration of viral load
suppression utilizing the patient's viral load status with respect to category of
response and time to response.
2301 Fortovase" (saquinavir; SQV) soft gel capsule (SGC) in
combination with AZT and 3TC in antiretroviral-naive
HIV-1 infected patients
Charles Farthing', M. Sension2, R. Pilson3, P. Siemon-Hryczyk3. 1AIDS Health
Care Foundation 1300 N Vermont Ave Los Angeles CA 90027; 2North Broward
Hospital District, FT Lauderdale FL; 3Roche Laboratories Inc., Nutley NJ, USA
Background: Saquinavir is the most potent protease inhibitor in vitro. Based on
recent clinical studies, 1,200 mg TID FORTOVASE'" (saquinavir) soft gel capsule
(SGC) delivers plasma drug concentrations 10 times those of the approved hard
gel formulation dose. The objective of this study is to evaluate the efficacy of
FORTOVASE'" 1,200 mg TID in combination with two nucleoside analog reverse transcriptase inhibitors, AZT and 3TC, in antiretroviral-naive HIV-1 infected
patients.
Methods: This open-label, non-comparative study is evaluating the efficacy of
FORTOVASE'" (SQV) SGC 1200 mg TID+AZT 300 mg BID+3TC 150 mg BID in
42 antiretroviral-naive HIV-1 patients, as measured by HIV-1 RNA PCR, CD4 and
CD8 cell counts, over 48 weeks. Patients with HIV-1 RNA > 10,000 copies/mL
and CD4 count,100 cells/mm3 were enrolled. Patients with chronic hepatitis B
and C were excluded.
Results: At baseline, mean viral load was 4.8 (4.0-6.1) logio copies/mL, and
mean CD4 cell count was 419 (118-951) cells/mm3. At 32 weeks (n = 20), the
mean viral load reduction was 3.35 loglo, and mean CD4 count had increased
by 209 cells/mm3. Four patients had serious adverse events unrelated to study
drug. Most adverse events were mild, the commonest being nausea, headache,
dyspepsia, diarrhea, fatigue and loose stool. One patient had grade III AST/ALT
levels at week 4 that resolved upon treatment discontinuation; 1 patient had grade
IV AST/ALT levels at week 12 associated with Hepatitis A; 1 patient had grade
III AST/grade IV ALT levels at week 20 associated with Hepatitis A; 1 patient
had grade Ill neutropenia at week 16; and 1 patient had a grade III triglycerides
at week 2 associated with interruption of cholesterol-reducing medication. Two
patients were withdrawn due to adverse side effects; 2 patients did not cooperate;
5 patients refused treatment; 6 patients were lost to follow-up; 1 patient had an
insufficient therapeutic response.
Conclusions: FORTOVASE'" (SQV) SGC combined with AZT and 3TC is a
highly potent antiretroviral therapy. The regimen is well tolerated with few adverse events. 32 week preliminary data show a mean viral load decrease of 3.35
logio copies/mL, with 90% of patients below the limit of Amplicor detection (400
copies/mL), and 70% below the UltraDirect limit (20 copies/mL). The latest data
will be presented.
12302 Lymphocyte subsets and functional changes in HIV-1
infected patient during HAART
Ivano Mezzaroma, E. Pinter, G. Cunsolo, L. Bernardi, V. Tiracchia, R. Paganelli,
F. Aiuti. Chair of Allergy and Clinical Immunology, Viale Universita' 37, 1-00185
Rome, Italy
Objectives: To investigate the mechanisms of immune reconstitution during
HAART in naive and experienced HIV-1 infected subjects with severe (CD4+ _
50/cu.mm) or mild (CD4+ > 100 < 500/cu.mm) immunosuppression.
Design: Prospective, open study.
Methods: Thirty-three patients (out of 76) were stratified in two groups on
the basis of CD4+ cells count and followed up to 24 months. The following
phenotypical and functional parameters were studied: HIV-RNA plasma viremia,
lymphocyte subsets (CD3, CD4, CD8, CD45RA, CD45RO, CD28 and CD38) and
in vitro proliferative responses to PHA, anti-CD3, HIV-1 rgpl60, tetanus toxoid
(TT) and C. Albicans mannoprotein (MP). The study also included v-/f repertoire
analysis and mortality rate.
Results: An increase of total lymphocytes was consistently observed in all
subjects soon after therapy and during the follow up, accompanied by a significant reduction of viral load (p = 0.02). Both CD4+ and CD8+ T cells increased
in percentage and in absolute numbers (p < 0.001). In particular, within the
CD4+ subset CD45RA+ (naive) and CD45RO+ (memory) both increased as absolute number from baseline (p < 0.001). This increase occurred later after 1-2
years of follow up, in contrast to previous data. A similar increase was observed

for CD8+CD45RO+CD38+ and for CD8+CD28+ lymphocyte subsets. Results of
functional studies have shown a restoration of lymphoproliferative responses to
PHA and anti-CD3, whereas a week response to specific antigens (rgpl60, TT
and MP) was sporadically observed.
Conclusions: HAART can lead to an increase of both CD4+ and CD8+ lymphocyte subsets and partly restore their function in the most immunocompromised
HIV+ patients. This occurs in conjunction with a dramatic decrease of viral load.

These results will be compared in less severely immunocompromised patients
(CD4+ > 100 < 500/cu.mm) both in peripheral blood and at bone marrow level.
These data on other 20 cases will be reported in extensive.
12303 Combined nelfinavir/saquinavir protease inhibitor
treatment in a BID regimen as salvage therapy in
advanced HIV disease
Juergen Lohmeyer, J. Selhorst, G. Friese, J. Teichmann, T. Discher. Departmen
of Internal Medicine of JLU; Kliniks Trasse 36D-3 5385 Giessen, Germany
Objectives: To evaluate on the basis of intention to treat the immunological and
anti-retroviral efficacy, pharmacokinetics, safety and tolerance of a four drug BID
regimen combining nelfinavir/saquinavir double protease inhibitor (PI) treatment
with two nucleoside reverse transcriptase inhibitors (NRTI) as salvage therapy in
heavily pretreated patients with advanced HIV-disease.
Design: 25 HIV-infected patients with CD4 cells <300/p/L who had previously
been treated with 2 NRTI/PI combination therapy and had failed or were intolerant
to such treatment were enrolled in an exploratory, single arm, open-label study
with nelfinavir 1250 mg BID + saquinavir 1000 mg BID in combination with 2
NRTI's.
Methods: Patients were were assessed monthly for CD4 cell count and plasma
HIV-RNA (NASBA; detection limit 40 copies/mL). In addition, plasma concentrations of protease inhibitors were quantified in heparinized blood samples obtained
before intake and at 1, 2, 3, 4, 8, 12 hr intervals using validated high-performance
HPLC.
Results: Nineteen of 25 patients have completed 24 weeks of treatment.
Baseline median and mean HIV RNA levels were 4.3 and 4.5 loglo copies/mL
respectively (range 3.9-6.6). CD4 cell counts at baseline ranged from 1 to 295
cells/pL with median and mean values of 80 and 118 CD4 cells//iL respectively.
Mean decrease in HIV RNA levels was 1.8 logio at 8 weeks and 1.6 loglo at 24
weeks. Mean increase in CD4 cells was 118 (week 8) and 140 cells/piL (week
24). The proportion of patients with plasma RNA levels below 500 and 40 copies,
respectively, were 60% and 47% at week 8 and 45% and 35% at week 24. The
treatment regimen has been generally well tolerated. No treatment related grade
3 or 4 adverse effects or clinical significant laboratory abnormalities were seen.
Two of 25 patients withdrew due to non-tolerated diarrhea.
Conclusion: Nelfinavir-saquinavir combination treatment in a four drug BID
schedule with 2 NRTI's is effective as rescue therapy in late HIV-infection in terms
of prompt and sustained HIV plasma RNA level reduction and CD4 cell count
increase.
S12304 Virologic response-plasma drug concentration
relationship in Phase III study of nelfinavir mesylate
(Viracept")
Bradley Kerr, Yazdi Pithavala, M. Zhang, M. Knowles, R. Daniels, R. Anderson.
Agouron Pharmaceuticals, Inc., 10350 N. Torrey Pines Rd., La Jolla, CA, USA
Purpose: To characterize the virologic response-plasma drug concentration relationship for nelfinavir (NFV).
Methods: Pharmacodynamic analysis was performed of results from a Phase
III study in which treatment-naive patients received AZT + 3TC plus NFV 500
mg TID (n = 97), NFV 750 mg TID (n = 99), or Placebo (n = 101). Predose and
2 hr postdose plasma samples were collected at Week 2 and Week 8; average
drug concentrations for the two visits were used in the analysis. The probability
of undetectable viral load on Week 24 was determined by logistic regression.
Predictor variables evaluated were: predose and 2 hr postdose plasma concentrations of NFV, M8 (major NFV metabolite in human plasma, active in vitro),
and AZT; baseline viral load (range 10,978-750,000); age (21-63 yr); gender
(90% male); race (78% Caucasian, 12% African-American, 2% Asian, 8% other).
A stepwise selection procedure was used to identify regression models having
significant variables.
Results: Viral load response variables using ultrasensitive assay data (LLOQ,
lower limit of quantitation = 50 HIV-1 viral RNA copies/mL) or standard PCR
(LLOQ = 400 copies/mL) provided more significant relationships compared to the
bDNA assay (LLOQ = 1200 copies/mL). Log baseline viral load was the best
single predictor of viral response and was a significant variable in all models.
Correlation between Week 2 and Week 8 NFV concentrations was better for timed
2 hr samples (r = 0.54) than for untimed predose samples (r = 0.18). Among
plasma drug concentration variables, the log of the sum of NFV and M8 2 hr
postdose concentrations (p = 0.0017) or the log of NFV 2 hr concentration without
M8 (p = 0.0021) were the best predictors of response (viral load <50 copies/mL
on Week 24) in the presence of log baseline viral load (p =- 0.0001). Viral response
was independent of patient age, gender, race, or AZT plasma concentration at
predose or 2 hr postdose.
Conclusions: In treatment-naive patients receiving NFV + AZT/3TC, the probability of achieving undetectable viral load by ultrasensitive assay was dependent
upon both baseline viral load and plasma NFV concentration. Timed 2 hr postdose concentrations were more highly predictive of virologic response than were

untimed predose trough concentrations. Interpatient differences in plasma drug
concentration appear to be of greater clinical significance for patients with higher
baseline viral load.