58
A 16-week Treatment Interruption Does Not Improve the Virologic Response to Multidrug Salvage Therapy in Treatment-experienced Patients: 48-week Results from ACTG A5086
C Benson*1, G Downey2, D V Havlir3, F Vaida2, M Lederman4, R Gulick5, M Glesby5, S Patel6, M Wantman2, C Bixby6, C Pettinelli7, A Rinehart9, S Snyder8, J Mellors6, and the ACTG A5086 Study Team
1Univ. of Colorado Hlth. Sci. Ctr., Denver, USA; 2Harvard Sch. of Publ. Hlth., Boston, MA, USA; 3San Francisco Gen. Hosp., CA, USA; 4Case Western Reserve Univ., Cleveland, OH, USA; 5Weill Med. Coll. of Cornell Univ., New York, NY, USA; 6Univ. of Pittsburgh, PA, USA; 7NIAID, NIH, DHHS, Bethesda, MD, USA; 8Social and Sci. Systems, Inc., Silver Spring, MD, USA; and 9VircoLab Inc., Tibotec-Virco, Durham, NC, USA
Background: Data are conflicting on the clinical and
virologic response to antiretroviral therapy (ART)
following structured treatment interruption (STI)
in patients with multidrug resistant HIV-1.
Methods: A5086 was a phase 3, open-label, prospective,
randomized, multicenter trial evaluating the efficacy of STI in heavily pretreated patients failing ART. Patients were randomized to a 16-week STI followed by salvage ART
(STI) or immediate salvage ART (No STI);
treatment was selected at entry using ART
history and genotype plus virtual or actual phenotype. The primary comparison
was the proportion of patients in each arm with HIV-1 RNA <400 copies/mL 48
weeks after randomization (Fisher’s exact test). Phylogenetic analyses were
performed by the neighbor joining method (PHYLIP).
Results: A total of 41 patients were randomized
(enrollment was halted due to slow accrual and data from other studies); 39
completed 48 weeks (1 death at week 48; 1 off-study at week 12). Median entry
CD4 count was 226 cells/μL, median HIV-1 RNA was
38,000 copies/mL mean number of drugs in salvage ART
was 4.4, and similar between arms. Of 39 evaluable patients (21 on STI and 18 on No STI)
at week 48, 4 (19%) and 6 (33%), respectively, had HIV-1 RNA <400 copies/mL
(p = 0.46), and 3 (14%) and 5 (28%),
respectively, had HIV-1 RNA <50 copies/mL (p = 0.43). Median ∆CD4 was +10 for the STI arm versus +17.5 for the No STI arm; median ∆HIV-1
RNA was -0.65 log for the STI arm
versus -1.15 log for the No STI
arm. One patient in the STI arm
developed a new or recurrent AIDS-defining event (lymphoma) at week 66 versus 0
patients in the No STI arm.
Genotypes at end of STI showed
reversion of baseline mutations in 5/18, partial reversion in 7/18, and
little/no reversion in 6/18 patients (2/21 had no baseline resistance and 1/21
had no genotype). In patients with confirmed virologic failure despite
reversion of mutations (n = 4),
phylogenetic analyses showed that the virus population at failure clustered
with the population at entry and not that at the end of STI.
Reappearance of virus with 3 drug class resistance occurred without inclusion
of all 3 classes in the salvage regimen.
Conclusions: A 16-week STI
prior to starting best available salvage ART
did not improve virologic response through week 48. STI
led to partial or no reversion of mutations in most patients; in those with
complete reversion, genetic analyses suggest that virologic failure resulted
from reselection of variants present before STI
that encode 3 class resistance mutation on the same genome.
Keywords: treatment interruption; virologic response; salvage therapy
