Does use of antiretroviral therapy regimens with high central nervous system penetration improve survival in HIV-infected adults?
H McManus ; PCK Li ; E Wright ; BJ Brew ; M Law ; J Zhou ; K Petoumenos ; B Mulhall ; JY Choi ; S Kiertiburanakul ; M Bloch ; D Nolan
HIV Medicine, Vol.12(10) : 610~619, 2011
OBJECTIVES: The aim of the study was to determine whether combination antiretroviral therapy (cART) with high central nervous system penetration-effectiveness (CPE) rank (neurocART) is associated with increased survival benefit compared with non-neurocART.
METHODS: Prospective data were examined for HIV-positive patients in the Asia Pacific HIV Observational Database who had commenced cART. CPE rank was calculated using the 2010 rankings process. NeurocART status was assigned to regimens with a CPE rank of 8 or more. Survival was analysed using Cox proportional hazards models with covariates updated at changes in cART regimen and with deaths up to 90 days after regimen cessation attributed to that regimen. Sensitivity analyses were conducted to examine the robustness of analysis assumptions.
RESULTS: Among 5882 patients, 308 deaths occurred. The hazard ratio (HR) for neurocART use was 0.89 (P=0.35) when data were stratified by cohort and adjusted for age, mode of HIV exposure, hepatitis B virus coinfection, AIDS-defining illness, CD4 count (cells/μL) and regimen count. Sensitivity analyses showed similar nonsignificant results. We also examined a composite endpoint of AIDS-defining illness or death (HR=0.93; P=0.61), baseline regimen as neurocART (HR=0.95; P=0.69), CPE category (P=0.71) and prior neurocART duration (P=0.16). No association between CD4 cell count and neurocART use was observed (P=0.52).
CONCLUSIONS: Our findings do not show a significant overall survival benefit associated with neurocART compared with cART. The potential benefit associated with neurocART in terms of prevention of neurocognitive impairment did not translate into an improvement in overall survival in this population. These findings were limited by the low incidence of associated mortality. Further studies and more extensive data are needed to address these limitations.