594 
Phidisa II: A Randomized 2x2 Factorial Trial Comparing Initial Therapy of Efavirenz with Lopinavir/Ritonavir and Zidovudine + Didanosine with Stavudine + Lamivudine in Treatment-naïve HIV-infected Persons with < 200 CD4+ Cells/mm3 or a Prior AIDS Diagnosis
Andrew Ratsela*1, M Polis2, and the Phidisa II Study Group
1South African Military Hlth Svc, South African Natl Defense Forces, Pretoria and 2NIAID, NIH, Bethesda, MD, US
Background: When the South African Phidisa II study began
in 2004 few NRTI were available, and the relative efficacy of
lopinavir/ritonavir (LPV/r) and efavirenz (EFV) was not known.
Methods: Phidisa II enrolled ART-naive HIV-infected members
of the South African National Defense Forces or family members at 6 sites. Participants
>13 years with <200 CD4+ cells/mm3 or prior AIDS
diagnosis were randomized to EFV or LPV/r; also to zidovudine + didanosine
(ZDV+ddI) or stavudine + lamivudine (d4T+3TC). HIV RNA and CD4+ cell
count were centrally determined monthly for 3 months, then every 3 months until
March 2008 when the study closed. The primary endpoint was AIDS or death. Cox
models were used to estimate hazard ratios (HR). All analyses are by intention
to treat.
Results: We randomized 1771 persons and followed them for
24.7 months (median). At baseline, median HIV RNA level and CD4+
count were 144,000 copies/mL and 106 cells/mm3. In the EFV group, 163
persons reached primary endpoints, as did 157 in the LPV/r group (HR = 1.04;
95%CI 0.84 to 1.30). Virologic suppression was faster with EFV than LPV/r (68%
vs 58% <400 copies/mL at 3 months; p <0.001); the difference was
negligible by 12 months (66% vs 65%) and through study closure. Persons assigned
LPV/r had greater CD4+ cell responses; CD4+ count was 272
for EFV and 317 for LPV at 36 months (p = 0.004). In the ZDV+ddI group, 170
participants reached primary endpoints, as did 150 in the d4T+3TC group (HR = 1.15;
95%CI 0.93 to 1.44). HIV RNA was significantly lower over follow-up (by 0.25
log; p <0.001) for d4T+3TC compared with ZDV+ddI. A correspondingly
greater CD4+ count response was also seen; CD4+ count was
309 for d4T+3TC and 279 for ZDV+ddI at 36 months (p = 0.04). Effects of
the 2 treatment factors were additive. Fastest HIV RNA suppression was with EFV+d4T+3TC
(73% <400 copies/mL at 3 months); greatest CD4+ response with LPV+d4T+3TC
(326 cells/mm3 at 36 months). Those assigned EFV+ZDV+ddI had lowest 36-month
CD4+ count (251 cells/mm3) and highest rate for the primary
endpoints (10.6/100 person-years vs 7.6 to 8.8 for the other groups).
Conclusions: EFV led to more rapid HIV RNA suppression than
LPV but CD4+ counts were increased more with LPV than EFV. AIDS/deaths
rates were similar for EFV and LPV. A trend towards better clinical outcomes
and significant virologic and immunologic improvement favored d4T+3TC over ZDV+ddI.
The nucleoside combination of ZDV with ddI is not recommended in this setting.
|
