581 
Randomized Pilot Study of Immediate Enfuvirtide-based Therapy vs a Treatment Interruption followed by Enfuvirtide-based Therapy in Highly Treatment-experienced Patients
George Beatty*1, J Lu2, P Hunt1, W Huang1, J Martin1, D Kuritzkes2, and S Deeks1
1Univ of California, San Francisco, USA and 2Partners AIDS Res Ctr, Harvard Univ, Boston, MA, USA
Background: Enfuvirtide (ENF) has limited long-term
effectiveness when the drug is the only fully active drug in a treatment
regimen. The level of drug-resistant HIV-1 in plasma decreases dramatically
after interruption of therapy (STI).
We hypothesized that the suppression of drug-resistant virus with a regimen
containing a single new therapeutic class would be more likely after an STI.
Methods: We randomized 30 three-class experienced
ENF-naïve individuals to either immediate therapy with ENF/optimized background
or a 16-week STI followed by ENF/optimized
background. Eligible subjects had a viral load of > 500 copies/mL and
genotypic or historical evidence of resistance to ³ 2 protease inhibitors (PI), ³ 2 nucleoside reverse transcriptase inhibitors (NRTI),
and ³ 1 non-NRTI (NNRTI). A
segment of the gp41-coding region encompassing the HR-1 and HR-2 domains was
amplified at baseline and weeks 16 and 24, and multiple independent clones
sequenced to assess resistance mutations. The study was powered based on
observing a 30% success rate in the control arm.
Results: The median CD4 and viral load at study entry
were 47 cells/mm and 4.72 log10 copies RNA/mL, respectively. As a
consequence of the STI, CD4
decreased by a median of 27 cells/mm (IQR –59 to –7) and HIV RNA increased by
0.40 (IQR +0.13 to +0.57) log copies/mL. At 24 weeks of therapy, 8 out of 15
(53%) subjects in the immediate therapy group vs 5 of 14 (36%) of patients in
the STI group had a viral load of <
75 copies/mL (p = NS). In
multivariate analysis, only baseline phenotypic susceptibility score was
predictive of treatment response at week 24 (p = 0.03). Baseline susceptibility to ENF, despite a 10-fold
variation, did not predict outcome. Analysis of HR-1 sequences showed rapid
emergence of ENF-resistance mutations within 2 to 4 weeks of treatment in most
patients with virologic failure.
Conclusions: Interrupting therapy prior to initiating
salvage therapy with ENF did not result in an improved virologic response at 24
weeks. The collective predictive activity of an ENF-containing regimen, but not
ENF baseline susceptibility, was important in predicting treatment response.
Keywords: enfuvirtide; structured treatment interruption; salvae therapy
