622b
CD4-guided Scheduled Treatment Interruptions: Low Incidence of Resistance Mutations in the Staccato Trial
Jintanat Ananworanich*1, B Hirschel2, H Furrer3, S Ubolyam1, A Gayet-Ageron2, A Gayet-Ageron, S Yerly2, T Jupimai1, A Hill4, L Perrin2, K Ruxrungtham1,5, and the Staccato Study Group
1HIV Netherlands Australia Thailand Res Collaboration, Bangkok; 2Geneva Univ Hosp, Switzerland; 3Univ Hosp, Berne, Switzerland; 4Univ of Liverpool, UK; and 5Chulalongkorn Univ, Bangkok, Thailand
Background: The potential
risk of increased selection of drug resistance during structured treatment
interruptions (STI) must be assessed and balanced with benefits of decreased
costs and toxicities.
Methods: We randomized 430 patients 1:2 to either receive continuous therapy (n = 146), or therapy that ceased as long
as the CD4 count exceeded 350 (n = 84).
In Thailand,
352 patients (79.1%) received 1600 mg of saquinavir (SQV) with 100 mg of
ritonavir (RTV) once daily together with 2 nucleoside reverse transcriptase
inhibitors (NRTI): didanosine/stavudine
(ddI/d4T) from 2002 until March 2003, and tenofovir/lamivudine (TDF/3TC) after
that. In Swiss (n = 80) and
Australian (n = 6) patients, the type
of HAART used was decided by the treating physician (53% non-NRTI [NNRTI], 29% protease
inhibitor [PI], 18% 3-NRTI). Direct double-stranded
sequencing was performed on an automatic sequencer (Applied Biosystems). Genotypic
resistance was defined as specified by the consensus mutation figures of IAS-USA,
May/June 2002 version.
Sequencing of the reverse transcriptase and protease
genes was attempted in 153 patients where resistance was most likely, i.e.
those with ³2
STI-retreatment cycles (n = 108 of 284),
or those where viral load was >500 despite treatment (n = 45). In patients with viral rebound during STI, the sample
analysed was always the first of those with a viral load >500 during the
last STI.
Results: The median nadir of CD4 counts,
before the trial, was 265, range 2 to 1143. The median time on HAART before
randomization was 21.6 months. The trial accrued 484 patient-years of
observation in the STI group and 262 in the continuous therapy group. Sequence
data were available from 125 patients. We observed no resistance mutations in
the RT and P genes in 115; 4 patients had resistance to 3TC (all with the 184V
mutation), 1 to 3TC and NNRTI (184V plus 103N), 1 to NNRTI only (Y181C), 1
possibly to zidovudine (ZDV) (M41L, and D67N without a history of exposure),
and 3 to indinavir (IDV) (M46L/I plus M36I). The relative risk of resistance,
compared to treatment with RTV-boosted SQV, was 0.95 (95%CI 0.11 to 7.91) for
NNRTI (p = 0.72), and 14.96 (95%CI
3.02 to 74.08) for 3 NRTI (p = 0.01).
Conclusions: Few resistance mutations occurred during STI.
In contrast to other studies, where resistance was more frequently observed,
most patients in Staccato were treated with a boosted PI, and HAART duration
before STI was relatively short.
|
