Publication Abstract Display
Type: Poster
Title: Low-level HIV RNA declines over time in csf but not in plasma.
Authors: Letendre S, McClernon D, Deutsch R, Cherner M, Cookson, Crescini M, Grant I, Ellis R
Date: 03-05-2013
Abstract:Background: The central nervous system (CNS) is a protected compartment that presents challenges for HIV eradication. We previously showed that 41% of 300 subjects taking suppressive antiretroviral therapy (ART) had between 2 and 50 HIV c/mL in CSF and that this was associated with worse neurocognitive (NC) functioning. The primary objective of the current analysis is to validate these findings in an independent cohort. Methods: 98 subjects were taking suppressive ART (HIV RNA ≤ 50 c/mL), did not have severe NC comorbidities, and had up to 3 assessments (mean 2.1 visits, median duration between visits 188 days) that included veni- and lumbar-puncture. HIV RNA between 2 and 50 c/mL (i.e., detectable) was measured in CSF and plasma using a validated qualitative assay that incorporates molecular beacons for detection. NC functioning was assessed using standardized comprehensive testing and summarized by the global deficit score (GDS) method. Statistical analyses were performed using parametric and non-parametric tests as well as mixed effects modeling. Results: Most subjects were middle-aged (mean 47 years) men (89%) who had been previously diagnosed with AIDS (72%), and were HCV seronegative (68%). Median CD4+ T-cell count was 436/mm3, median duration of the current ART regimen was 9 months, and mean CPE value was 8.4. Global NC impairment was present in 37%. The proportion of specimens with detectable HIV RNA declined from the first to the third visit in CSF (21% to 16% to 5%, p = 0.03) but not in plasma (48% to 41% to 39%, p = 0.48). Multivariable mixed effects modeling identified that detectable HIV RNA in CSF was associated with shorter durations of ART and detectable HIV RNA in plasma but not CPE. Multivariable mixed effects modeling also identified that worse GDS values were associated with detectable HIV RNA in plasma and more NC comorbidities. Conclusions: Low-level HIV RNA in CSF declines over time with longer durations of ART. In this independent, longitudinal cohort of subjects, the presence of low-level HIV RNA in plasma, but not in CSF, was associated with worse NC functioning. This finding does not match the prior analysis but does continue to support a link between the presence of low-level HIV RNA and worse NC functioning.

return to publications listing