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

12th World AIDS Conference

Abstracts 12254-12258 63

12254 Response to ritonavir in severely immunocompromised,
antiretroviral experienced children with perinatal HIV-1
infection
Nancy Hutton, M.L. Joyner, A.M. Butz, P.D. Fosarelli. CMSC 144 Johns Hopkins
Hospital, 600 N Wolfe St, Baltimore, MD 21287-3144, USA
Objectives: To determine the virologic and immunologic response to the addition
of a protease inhibitor in antiretroviral experienced, severely immunocomprised
children with perinatal HIV-1 infection.
Design: Retrospective review of clinical care.
Methods: Children ages 1-12 y with perinatal HIV-1 infection and CDC class
3 immunodeficiency despite combination therapy with nucleoside reverse transcriptase inhibitors (NRTIs) were offered ritonavir (350 mg/m2/dose PO q12h)
in combination with 1 or 2 NRTIs. Quantitative HIV-1 RNA by RT-PCR and
CD4 counts were obtained at baseline and periodically as needed for clinical
care. Mean log copies of HIV-1 RNA/ml plasma and median CD4 cells/mm3 are
reported in the table below by age group at each time point on therapy.
Results: 21 children (1-5 yr = 7; 6-12 yr = 14) initiated 2 or 3 drug combination
therapy with ritonavir between 8/1/96 and 5/11/97. Previous therapy included ZDV
(21), 3TC (16), DDI (17), DDC (1), and NVP (3).

Baseline
(n =21)
CD4 RNA
1-5 yrs 26 5.42
6-12 yrs 18 5.12

1 month
(n = 17)
CD4 RNA
146 4.09
79 3.36

6 month
(n = 20)
CD4 RNA
641 4.66
247 3.80

9 month
(n = 17)
CD4 RNA
711 4.58
443 4.05

12 month
(n = 9)
CD4 RNA
539 4.11
463 4.67

Conclusion: Children with extremely low CD4 counts and viral loads > 100,000
copies/ml despite NRTI combination therapy respond to the addition of a protease
inhibitor to their regimen. Greatest decrease in viral load was seen 1 month into
therapy. However, most children had dramatic, continued increases in CD4 count
despite viral loads between 10,000 and 100,000 copies/ml.
128*/12255 Antiretroviral activity and safety of abacavir (1592,
ABC) with 3TC/ZDV in therapy-experienced
children
Russell van Dyke1, X. Saez-Llorens2, R.P. Nelson3, P. Emmanuel4, A. Wiznia5,
C.A. Gilbert6, A. Keller6. 1 Tulane University 1430 Tulane Ave Box SL 37 New
Orleans; 3U South Florida All Childrens Hospital St. Petersburg FL; 4 Tampa FL;
5Bronx Lebanon Hospital Bronx NY; 6Glaxo Wellcome Inc. RTP 27709 NC, USA;
2Hospital Del Nino, Panama City, Panama
Background: Fewer antiretroviral agents are available for children than adults,
and limited data regarding safety and efficacy in children are available at the
time of regulatory approval for some new agents. CNAA3006 is a phase III trial
comparing the safety and efficacy of 3TC/ZDV versus 3TC/ZDV plus abacavir in
HIV-infected children ages 90 days to 13 years.
Methods: Treatment-experienced children were randomized to receive either
3TC/ZDV or 3TC/ZDV/ABC (stratified according to prior treatment with 3TC/ZDV).
At 16 weeks and every 8 weeks thereafter, viral load (VL) was determined in real
time (Roche Amplicor). Subjects with VL < 10,000 c/mL continued on blinded
study medication, while those above this level (confirmed with a second sample)
were considered to have reached a virologic "switch" criteria and were offered
the option to continue blinded study treatment, receive combination treatment including open label ABC, or go off study. Additional assessments included clinical
endpoints, recording of adverse events, and CD4+ cell counts.
Results: Two hundred five children were randomized to study treatment. Seventeen (17) children have left the study: 9 because of compliance or parental
choice, 6 because of adverse events (2 suspected abacavir hypersensitivity reactions), and 2 died from non-medication-related events. As of 1/1/98, 135 children
have reached 16 weeks of treatment. Forty-eight (48) have reached the confirmed
virologic "switch" criterion; 22 switched to open label ABC, 22 elected to remain
on blinded treatment, and 4 withdrew from the study. Additionally, 16 subjects
have VL results pending.
Conclusions: Abacavir was well tolerated by HIV-infected children in this study.
The drop out and switch frequencies are within the estimates assumed for the
design of this study. During the 16 weeks, only 2 subjects discontinued because of
suspected abacavir hypersensitivity. All subjects will reach 16 weeks of treatment
by Feb. 1, 1998. Data analysis could demonstrate that abacavir is a significant
addition to the group of antiretroviral agents available for the treatment of children
with HIV infection.
12256 Viral load and CD4-cell count under a triple-therapy with
nelfinavir and two RT-inhibitors in previously untreated
HIV-infected children
Markus B. Funk1, Richard Linde1, Uwe Wintergerst1, Gundula Notheis1,
Monika Kaiml', Jindrich Cinatl2, Wolfhart Kreuz3. 1 Universitats-Kinderklinik, St.
31-1 Theodor Stern, Kai 7, 60596, Frankfurt; 2Departement of Medical Virology
Universitaet, Frankfurt 3 University-Children's Hospital, Frankfurt, Germany
Issue: Benefit and side-effects of a triple therapy with protease-inhibitor and
two nucleosi-de-analogues were investigated in pediatric HIV-infected patients
without any antiretroviral pretreatment.
Project: 14 children (median age at study onset: 68 months) in stage A
or B (CDC 94) were included. Informed consent was obtained by their par
ents. The dosage of nelfinavir was 3 x 20-30 mg/kg bw. 7 patients received
a RT-lnhibitor combination of zidovudine (AZT-dosage: 3 x 120 mg/sqm) and
lamivudine (3TC-dosage: 2 x 4 mg/kg), 7 patients received a combination of didanosine (ddl-dosage: 2 x 90 mg/sqm) and stavudine (d4T-dosage: 2 x 1 m/kg).
Viral load (Amplicor-test; Roche) and CD4-cell count were measured every 4-8
weeks.
Results: After 4, 12 and 24 weeks of treatment median reduction of viral load
(VL) was >2 log. 9 patients had a VL - 500 copies/ml (reduction >2.0 log) and 5
patients a reduction of 0.9-2.0 log (median: 1.9 log) after 12 weeks. CD4 cells//il
raised from median value of 650 (range: 40-2250) to 925 after 12 weeks and
fell to baseline level after 24 weeks. Main problems were difficulties in application, especially of nelfinavir in small children, hyperactivity in patients under both
regimes and rash in children under d4T combination.
Conclusion: Beside application problems both regimes were well tolerated. A
clear reduction of viral load could be demonstrated in most of the children over a
period of 6 months. CD4 cell raise was temporary.
12257 Mutation pattern of the RT and protease genes in
HIV-infected children receiving combination therapy
Catherine Tamalet1, Novara Yahi1, I. Thuret2, G. Michel2, C. Perrimond2.
1 Hpital de la Timone, Laboratoire Virologie, 264 Rue St Pierre, Marseille;
2Hematologie Pediatrique La Timone, Marseille, France
Objective: To analyse genotypic changes in relation with antiretroviral treatment
in two groups of children with undetectable (<200 HIV copies/mi) or detectable
(>200 HIV copies/mi) plasma viral load.
Design: 29 patients were studied (10 under bitherapy without protease inhibitor,
17 under tritherapy with a protease inhibitor, and 2 under qudritherapy with two
protease inhibitors).
Methods: DNA extracted from mononuclear cells was amplified by nested
PCR. A 1,200 nucleotide amplicon was generated of which the entire protease
and codons 1-240 of RT were sequenced on a Applied Biosystems 377 DNA
sequencer.
Results: From the 29 patients of this study, only 4 had undetectable plasma
viral load. One of these, under quadritherapy, displayed 7 mutations in the RT
coding region including M41L, M184V, T215F, and K219Q, and 6 mutations in the
protease coding region including M361, 154V, and V82F, but not at positions 48
and 90 associated with resistance to saquinavir. In 86% of cases (25/29 patients),
plasma viral load was >200 HIV copies/ml. This group included all 10 patients
under bitherapy, 14 under tritherapy, and 1 under quadritherapy. All displayed
a polymutated pattern of the RT gene including M46L (46%), M1841/V (30%),
T215F/Y (69%), and K219E/Q (38%). Mutations associated with resistance to
ritonavir, which was given to 13 of the 14 patients under tritherapy in this group
with high viral load, were found in 58% of cases. The most common mutations of
the protease gene were L63P (96%) and M361 (35%), followed by V771 (27%),
V82A/F (23%), 154LV (15%) and M461 (4%).
Conclusion: These data demonstrate a polymutated pattern in both RT and
protease genes probably due to step-by-step selection of resistant variants during
sequential treatment of the patients with various RT and/or protease inhibitors.
The most striking result concerns the mutation of codon 184 in the RT gene which
was recognized in only 10 of the 25 patients receiving 3TC for long periods of
time up to recent weeks before blood collection for the current analysis.
12258ý UK experience of nelfinavir in paediatric HIV-1 infection
Zuber Mitchla1, M. Sharland1, N. Delmas2, D. Gibb3. 1Pharmacy Dept St
Georges Hospital, Blackshaw Road, London SW170QT; 2Roche Herts; 3Institute
of Child Health, London, UK
Background: The use of protease inhibitors in adults has been widely demonstrated to provide superior clinical efficacy when used as part of the antiretroviral
combination therapy. However, to date very little data is available on the clinical
effect of using a protease inhibitor in paediatrics.
Objective/Design: To audit the clinical experience of nelfinavir in the UK paediatric named patient programme. This was carried out retrospectively by an
observational analysis of CD4 count (CD4), count and viral load (VL) measurements in HIV infected children who had been enrolled into the programme.
Results: Preliminary results to December 1997 were available for 30 patients
who had completed at least 3 months treatment in the programme. Median
age 7 years (5 months to 16 years), 43.3% male and 56.7% female. 83.3%
had taken prior antiretroviral therapies but only 3 patients were protease experienced. Median CD4 and VL before nelfinavir treatment were 91 cells/mm3, and
111,000 copies/ml respectively. Following treatment median CD4 increased to 349
cells/mm3 (p = 0.00021, Wilcoxon matched-pairs) (95% Confidence Interval (CI)
140 to 725) and VL reduced to 1360 copies/ml (p = 0.0021, Wilcoxon matchedpairs) (95% CI 57,100 to 201,275). Diarrhea reported in 4 patients was the main
side-effect requiring cessation of therapy in 1 patient. A further 3 patients also
stopped treatment, 2 due to refusal to take medication and 1 following a drug
induced rash.

Conclusion: Use of nelfinavir as part of a triple combination of antiretrovirals
in children provides a marked increase in CD4 and reduction in VL in patients
enrolled in this programme. Nelfinavir is generally well tolerated in the short
term. However, evidence from this programme suggests that although nelfinavir
is easier to take than other protease inhibitors, there is still difficulty in maintaining
compliance with the current formulation, particularly for young children.