| Authors: Munoz-Moreno J, Letendre S, McClernon D, Ellis RJ, LeBlanc S, Rosario D, Clifford D, Collier A, Gelman B, Marra C, McArthur J, McCutchan A, Morgello S, Simpson D, Franklin D, Heaton RK, Grant I, and the CHARTER Group |
| Abstract:Background: Cognitive impairment can occur or
persist during antiretroviral therapy (ART). Explanations
include comorbidities, neurotoxic ART,
persistent neuroinflammation, or persistent HIV
replication in central nervous system (CNS). This
analysis assessed whether low levels of HIV in
cerebrospinal fluid (CSF) were associated with
inter-individual differences in ART regimens and
neuropsychological (NP) performance.
Methods: 329 participants (pts) were selected
from the CHARTER cohort because they were taking
ART, and had HIV RNA levels below 50 c/mL with
an ultrasensitive assay (Roche Amplicor) in CSF and
blood. Paired CSF and blood plasma (PL) specimens
that were obtained within 1 hour of each other were
assayed with a more sensitive assay, a modified
version of the NucliSens EasyQ (bioMerieux) assay
capable of qualitatively detecting HIV at 2 c/mL. To
determine the stability of low-level HIV in CSF, a
follow-up specimen from 61 pts was assayed (median
duration between visits, 7.0 months). Penetration
of ART into the CNS was estimated by CNS
Penetration-Effectiveness (CPE) ranks. NP performance
was summarized by the Global Deficit Score
(GDS), a validated method which integrates relevant
information about 7 NP performance domains.
Results: Pts were mostly non-white (55%), middle-
aged (mean 45 yr) men (76%) with AIDS (75%)
who were HCV seronegative (68%). Median duration
of the current ART regimen was 15 months.
Median CD4 count was 467/mL. By the more sensitive
assay, 136 (41%) had detectable HIV in CSF and
216 (66%) had detectable HIV in plasma. Detectable
HIV in CSF was associated with worse CPE scores
(mean 1.49 vs. 1.70, d0.29, p0.009) and detectable
HIV in PL (71% pts with HIV detected in CSF
vs. 62% undetected, p0.077). Other demographic
or disease characteristics were not found in association.
Of this group of pts, 39 (28%) had detectable
HIV in CSF but not in PL, had worse global deficit
scores (GDS) (0.63 vs. 0.37, p0.012), and were
particularly likely to have at least moderate global
impairment (GDS greater than 0.93, 28% vs. 8%,
p0.005). Multivariate analyses identified that
worse global deficit scores were associated with
detectable HIV in CSF but not in PL (B0.13, p
0.007), HCV seropositivity, shorter durations of
ART, and ethnicity other than white (model R2
0.29, pB0.0001). In the subgroup with a follow-up
CSF specimen assayed, CSF values became undetectable
in 18 (30%) pts although remained detectable
in the other 43 (70%). Transition from
detectable to undetectable was associated with
higher CD4 counts at the second visit (46% of
undetectable pts had CD4500 vs. 12% in detectable,
p0.09), and HCV seropositivity (57% vs.
24%, p0.04), but not CPE scores or improved
global NP performance.
Conclusions: Despite having achieved virologic
suppression in CSF and plasma by the Roche
ultrasensitive assay, we found that ART-treated
individuals frequently (41%) had low, detectable
levels of HIV in CSF, and this was associated with
less penetrant ART. At least a quarter of these
individuals (28%) had persistent HIV in CSF, but
not in PL, and this was linked to the presence of
worse neurocognitive functioning. People living
with HIV may have cognitive impairment as a result
of ART that is incompletely effective in the CNS |