703 
Immunological Success Is Predicted by Enfuvirtide but Not Interleukin-2 in Immunocompromised Patients, Final Results of the ANRS 123 ETOILE Randomized Trial
Jean-Paul Viard*1, C Fagard2, C Rouzioux1, E Pereira2, M Bentata3,4,5, N Colin de Verdiere4, G Pahlavan5, L Weiss6, Y Levy7, G Chene2, and The ANRS 123 Study Group
1Ctr Hosp Univ Necker, Paris, France; 2INSERM U593, Bordeaux, France; 3Ctr Hosp Univ Avicenne, Bobigny, France; 4Ctr Hosp Univ St Louis, Paris, France; 5Ctr Hosp Univ Bichat, Paris, France; 6Ctr Hosp Univ Georges Pompidou, Paris, France; and 7Ctr Hosp Univ Henri-Mondor, Creteil, France
Background:
The effect of Interleukin-2 (IL-2) in patients with low CD4 counts and multiple
treatment failure is unknown. The ANRS 123 ETOILE trial was designed to
evaluate the efficacy of adding IL-2 to an optimized background treatment in
patients with no remaining treatment options.
Methods:
Patients with CD4 ≤200/mm3, viral load >10,000 copies/ml
and a genotypic score based on available genotypic results showing ≤2
active drugs, were randomized to receive either 8 subcutaneous IL-2 cycles (4.5
MUI twice daily, 5 days, from Week 2 to W42) with optimized background
treatment, or optimized background treatment alone. Background regimen was
optimized as much as possible according to genotypic score, and enfuvirtide
(T20) was added in T20-naive patients. Final evaluation was performed at week 52
on the proportions of patients with a CD4 count ≥200/mm3
(primary outcome) and with an increase ≥50 CD4 cells/mm3 from week
0 to week 52, on median HIV RNA, incidence of HIV-related events and tolerance.
Results: From
June 2004 to December 2005, 28 patients were randomized to each arm: 61% were
stage C, mean age was 46 years, median baseline CD4 count, and viral load were
63/mm3 and 4.5 log10 copies/mL, and 43% of patients had a
genotypic score = 0. Background treatment could be optimized in 23 patients
(T20 as only active drug: n = 9, 1 or 2 other active drugs according to
genotypic score: n = 7, both: n = 7), while treatment could not
be optimized in 33 (regimen unchanged, T20 in non-naive patients, non-active
drugs). At week 52, the proportion of patients with ≥200 CD4 was 14% in
the IL-2 arm vs 18% in the control arm (p = 1.00), the proportion of
patients with a CD4 increase ≥50 was 25% in the IL-2 arm vs 32% in the
control arm (p = 0.56) and median HIV RNA was 4.5 vs 4.6 log10
copies/mL (p = 0.63). In multivariate analysis, optimization with T20
was the only factor associated with immunological success defined either by CD4
≥200 (50% with T20 vs 2.5% without T20, p = 0.004) or a CD4
increase ≥50 (75% with T20 vs 5% without T20, p <0.001). The
incidence of AIDS events did not differ between arms (n = 15 in the
control arm, n = 11 in the IL-2 arm, including 2 lymphomas considered as
possibly related to IL-2, occurring in profoundly immunodeficient patients).
IL-2 was well tolerated.
Conclusions:
IL-2 failed to increase CD4 cells in severely immunocompromised patients with
multiple therapeutic failures. Use of T20, even on a poorly optimized
background, was highly associated with success, underlining the importance of
using a new drug family in salvage therapy.
|
