CROI 2019: Neurologic Complications of HIV Disease - IAS-USA
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
CROI 2019: Neurologic Complications Volume 27 Issue 1 April 2019
Invited Review
CROI 2019: Neurologic Complications of HIV Disease
Beau M. Ances, MD, PhD, MSc; Scott L. Letendre, MD
Investigators reported many new neuroHIV research findings at the 2019 HIV-Associated Neurocognitive
Conference on Retroviruses and Opportunistic Infections (CROI). These find- Disorders
ings included confirmation that HIV-associated neurocognitive disorder
(HAND) remains common with an increasingly recognized role for comor- HIV-associated neurocognitive disorder
bidities (eg, obesity) and neurodegenerative conditions (eg, Alzheimer’s (HAND) remains common and contin-
disease), especially as persons living with HIV (PLWH) advance into their ues to persist despite ART. Within a large
seventh decade of life and beyond. HAND is increasingly recognized as a cohort of ART-naive PLWH who resided
heterogeneous disorder that differs between individuals (eg, by sex) in the in Uganda, the presence of HAND at ini-
trajectory of specific neurocognitive abilities (eg, executive functioning). A tial evaluation was associated with 68%
more recent focus at this year’s conference was toxicity of combination anti- increased odds of death at 2 years and
retroviral therapy: neurocognitive performance and neuroimaging data from a 98% increased odds of death within
several studies were presented but did not consistently support that integrase 5 years (Abstract 425). These results in-
strand transfer inhibitors are associated with worse neurologic outcomes. dicate that HAND diagnosis carries sub-
Neuroimaging studies found that white matter changes reflect a combina- stantial morbidity and mortality risks. In
tion of the effects of HIV and comorbidities (including cerebrovascular small the WIHS (Women’s Interagency HIV
vessel disease) and best correlate with blood markers of inflammation. The Study), greater immune activation be-
pathogenesis of HIV in the central nervous system (CNS) was the focus of a fore the initiation of ART was assoc-
plenary lecture and numerous presentations on HIV compartmentalization iated with higher rates of neurocogni-
in the CNS and cerebrospinal fluid viral escape. Novel findings were also pre- tive impairment on subsequent follow-
sented on associations between HIV-associated neurologic complications and up (Abstract 407). In a cohort of indi-
glycomics, neuron-derived exosomes, and DNA methylation in monocytes. viduals with acute and early HIV in-
This summary will review findings from CROI and identify new research and fection from Peru, Robertson and col-
clinical opportunities leagues showed that early initiation of
ART improved cognition (Abstract 445).
Keywords: CROI, 2019, HIV, neurology, HAND, comorbidities, central nervous PLWH who were recently infected (CROI 2019: Neurologic Complications Volume 27 Issue 1 April 2019
infected PLWH can fluctuate over time. nitive decline (both normal and impaired Perez and colleagues longitudinally
De Francesco and colleagues (Abstract at baseline) that was attributable to investigated the relationship between
420) evaluated changes in cognition several risk factors including reduced obesity, frailty, and cognition over time
over 2 years in virologically well con- neurocognitive reserve (less educa- (Abstract 129). Similar to De Frances-
trolled PLWH (n=173) compared with tion, more unemployment, and de- co and colleagues (Abstract 420), 78%
HIV-seronegative individuals (n=77). At of PLWH had no cognitive deficits and
baseline evaluation, 20% of the PLWH Higher baseline Athero- continued to have normal cognition
and 3% of the HIV-seronegative individ- over a 3-year interval, and 10% had cog-
sclerotic Cerebrovasuclar nitive impairment at both time points,
uals had cognitive impairment using a
multivariate normative comparison Disease or Framingham 6% had an improvement in cognition
(MNC) score. At 2-year follow-up, 13% Heart Study risk score at the second time point, and 6% de-
of PLWH and 6% of the HIV-uninfected veloped impairment over the 3-year
individuals had cognitive impairment was associated with interval. Obesity and older age, but not
based on the MNC. Although none of worsening in cognition frailty, were the greatest risk factors for
the cognitively impaired HIV- unin- developing cognitive impairment over
over 4 years 3 years. Finally, Chow and colleagues
fected participants changed over the
2 years of follow-up, 46% of the PLWH pression) and metabolic factors (obe- investigated the association between
improved (changed from cognitively sity, diabetes, and substance use). In a presence of cardiovascular disease
impaired to not cognitively impaired). cohort of PLWH and HIV-uninfected (CVD) as assessed by the Atheroscle-
For those individuals who were cog- persons followed up at the US National rotic Cerebrovascular Disease [ASCVD]
nitively normal at baseline, 2% of the Institute of Health and the US Depart- and Framingham Heart Study CVD Risk
PLWH and 4% of the HIV-uninfected ment of Defense (n=397), risk factors Score (FRS) and risk of developing neu-
participants developed cognitive im- for worse cognitive impairment in- rocognitive impairment in the HAILO
pairment. Among PLWH, 10% had a cluded currently smoking, history of al- cohort (Abstract 128). In unadjusted and
reliable decline in cognition, 79% re- cohol abuse, and unemployment (Ab- adjusted models, higher baseline ASCVD
mained stable, and 11% had improved stract 414). Overall, these results sug- risk score or FRS was associated with
cognition. Among HIV-uninfected in- gest that cognitive impairment seen in worsening in cognition over 4 years. Al-
dividuals, 7% had a reliable decline in PLWH may reflect changes in the brain though the negative impact of cerebral
cognition, 92% remained stable, and 1% due to HIV early in the disease and ad- small vessel disease (CSVD) on cogni-
improved. These results suggest that ditional risk factors later in the disease tion was seen for both men and women,
most PLWH who are virologically well process. effects were significantly greater in
controlled remain cognitively stable women. Overall, these results point to
over 2 years. In contrast to other neuro- focusing on modifiable risk factors in-
Comorbidities and HAND
degenerative disorders, in which there cluding anemia, diabetes, and meta-
are progressive declines, approximately Several comorbidities appear to in- bolic factors as intervenable targets for
half of all PLWH who have cognitive crease the risk of cognitive impair- potentially stabilizing neurocognitive
impairment at a given time point may ment in virologically well-controlled function in PLWH, especially women.
improve over time. HAND is character- PLWH. Within cohorts of PLWH at the Preventive interventions geared toward
ized by fluctuations in cognition over University of California San Diego, ane- these comorbidities in PLWH may be
time rather than a gradual progressive mia was associated with worse overall important for patient care.
decline seen in other neurodegenera- neurocognitive performance cross-sec-
tive diseases. tionally and longitudinally (Abstract
Neurodegenerative Diseases
PLWH who have HAND may be con- 426). Changes in cognition were ob-
and HAND
siderably heterogeneous regarding the served in several domains including
domains that contribute to neurocog- speed of information processing, motor A symposium presentation by Valcour
nitive impairment. Fitzgerald and col- functioning, and working memory. The focused on the potential increasing
leagues identified distinct clusters of authors postulate that chronic inflam- prevalence of aging-related neurode-
age-related changes in declarative mem- mation affects iron metabolism that generative diseases in PLWH (Abstract
ory in PLWH and HIV-uninfected indi- leads to anemia. Within the HAILO (HIV 159). Questions remain if clinicians
viduals (n=1752) followed up in the Infection, Aging, and Immune Function can successfully distinguish Alzheim-
WIHS (Abstract 408). Using a Bayesian Long-term Observational) study, gait er's disease (AD) from HAND and if an
Dirichlet process mixture model, 4 sub- speed and cognition were assessed in accelerated phenotype exists within
groups were identified: normal slow de PLWH (n=929). Increased levels of he- older (>60 years old) PLWH. Inflam-
cline, normal accelerated decline, im- moglobin A1C, cognitive impairment, mation persists in virologically sup-
paired accelerated decline, and impair- and African American race were assoc- pressed PLWH with impairment, in
ed but stable cognition. Approximately iated with declines in gait speed (Ab- the periphery (eg, plasma measures)
55% of the women had accelerated cog- stract 703). Within this HAILO group, and centrally (eg, positron emission
27IAS–USA Topics in Antiviral Medicine
tomography [PET] measures). However, longitudinal studies of neuroimaging significant changes in cognitive per-
the contribution of inflammation due and CSF biomarkers in older cohorts formance or distress were observed
to HIV is contentious, as a recent PET of PLWH are needed. after switching. Two other studies re-
imaging study did not observe elevated ported on participants who were ini-
neuroinflammation in virologically sup- tiating or currently taking InSTIs with-
CNS Effects of Integrase Strand
pressed PLWH compared with matched out an observed therapy switch. A
Transfer Inhibitors
HIV-uninfected persons (Abstract 460). study from Barcelona assessed partic-
Controversy also remains regarding Integrase strand transfer inhibitors (In- ipants who initiated InSTI-containing
whether HIV and aging accentuate or STIs) are potent components of initial ART during early HIV infection (n=12)
accelerate changes in brain integrity. ART regimens and their use is growing similar to the SEARCH study, but in
Some studies have demonstrated a worldwide. With the report of more contrast, compared them with those
greater rate of brain atrophy in PLWH frequent neuropsychiatric adverse who initiated InSTI-containing ART
than in HIV- uninfected persons and events (NP-AEs) in a clinical popula- during chronic HIV infection (n=15)
others have demonstrated that HIV and tion,3 questions have arisen about the and to persons without HIV (Abstract
aging independently cause changes. CNS safety of InSTIs. 439). In addition to a 12-test neuro-
Several studies at CROI 2019 re- cognitive test battery, participants
ported the CNS effects of initiating were assessed with structural neuro-
Several studies at CROI or switching to InSTI-containing ART. imaging and an assessment of daily
2019 reported the CNS Mora-Peris and colleagues, for exam- functioning. Cognitive performance
effects of initiating or ple, evaluated 8 PLWH who remained improved in all 3 groups over time
on a raltegravir-containing regimen and did not differ among the groups.
switching to InSTI- and 12 PLWH who switched to a dolute- Functional assessments identified that
containing antiretroviral gravir-containing regimen (Abstract the early HIV infection group had evi-
443). Neurocognitive performance, dence of greater stress levels 4 weeks
therapy neuroimaging, and CSF measures were after initiating ART (but was similar to
assessed at baseline and after 120 other groups at 48 weeks) and that the
Differences may reflect the presence of days. Although the group sizes were chronic HIV infection group trended to-
CSVD or the sample cohorts studied. small, no statistically significant dif- ward having worse depressive symp-
Several markers could potentially ferences were observed between the toms at 48 weeks. Neuroimaging ident-
distinguish AD from HAND. PET im- groups in any of the assessments, sup- ified that the chronic HIV infection
aging of amyloid and tau, pathologic porting the conclusion that switching to group also had evidence of a reduction
hallmarks of AD, were not abnormal in dolutegravir-containing ART is safe for in medial orbitofrontal gray matter vol-
small cohorts of PLWH compared with the CNS. This conclusion was also sup- ume at weeks 4 and 48 that did not ap-
HIV-uninfected individuals.1,2 Although ported by data from the Thai SEARCH pear to be present in the other groups.
these studies were performed in younger (South East Asia Research Collabo- Since both HIV infection groups were
PLWH, cognitively impaired individuals ration in HIV) study (Abstract 440). taking InSTI-containing ART, however,
were included. Furthermore, if PLWH Participants (n=254) diagnosed with this is more likely to be due to later
do have accelerated aging, PLWH who acute HIV infection had taken at least initiation of ART than to InSTI-specific
are 50 years of age or older should be at 24 weeks of ART (median, 144 weeks) declines.
increased risk for developing AD. How- and were subsequently switched to a In a cross-sectional analysis, O’Hal-
ever, the few studies that have been dolutegravir-containing regimen. They loran and colleagues compared neuro-
performed in this age range have not were evaluated before and after the cognitive performance and neuroim-
shown an increase in the prevalence of switch using multimodal assessments aging measures in participants who
AD. The presence of AD in older PLWH (neurocognitive testing, the Patient were taking InSTI-containing (n=99)
may reflect aging and genetic risk fac- Health Questionnaire-9 [PHQ-9], ma- or non-InSTI-containing (n=103) ART
tors and may not be specifically due to jor depression screening, and an as- (Abstract 442). The specific InSTI drugs
HIV. CSF amyloid and tau also serve as sessment of distress). After switching, used by participants were raltegra-
useful biomarkers for distinguishing participants reported more somatic vir (40.4%), dolutegravir (30.3%), and
AD from HAND. Conflicting results have symptoms on the PHQ-9 and more elvitegravir (29.3%). InSTI users had
been observed with some studies dem- symptoms of depression, although worse global neurocognitive perfor-
onstrating mild alterations in CSF amy- they only trended toward having more mance (specifically in the combined
loid but not CSF tau in PLWH. Trunfio symptoms on the cognitive/affective learning/memory domain) than non-
and colleagues evaluated PLWH (n= subscale of the PHQ-9 and were not InSTI users and this did not appear to
181) who were 45 years of age or older more likely to have evidence of moder- differ by InSTI drug (ie, the effect size
and virologically suppressed and only ate-to-severe depression. Importantly, for dolutegravir was similar to ralte-
1 individual (CROI 2019: Neurologic Complications Volume 27 Issue 1 April 2019
with decreases in volumes throughout disoproxil fumarate (TDF)/FTC or abaca- lished studies have reported that PLWH
the brain. This evidence of InSTI neu- vir/lamivudine (Abstract 436). Twenty typically take more concomitant pre-
rotoxicity is generally consistent with PLWH were evaluated with cognitive scribed drugs than the general pop-
the prior report from deBoer et al,1 but testing and CSF assessments at 3 and ulation, including drug classes with
stands in contrast to other reports at 12 months after switching. No statis- known neurocognitive adverse events
tically significant changes were seen (eg, Rubin et al).7 Consistent with this,
Anxiolytics, anti- in cognitive performance or CSF bio- Ma and colleagues reported in cross-
psychotics, opioids, and markers (neopterin, neurofilament light sectional analyses that polypharmacy,
[NFL], β2-microglobulin, IgG index) in or using 5 or more concomitant drugs,
antimicrobials were the this small study, supporting the CNS was associated with worse cognitive
classes of concomitant safety of switching to TAF/FTC. In a performance (Abstract 437), particu-
drugs that were most much larger analysis, Li and colleagues larly in learning, memory, and verbal
aimed to determine if cognitive perfor- fluency. Anxiolytics, antipsychotics,
commonly associated mance of men enrolled in the MACS opioids, and antimicrobials were the
with worse cognitive (Multicenter AIDS Cohort Study) im- classes of concomitant drugs that were
proved after discontinuing efavirenz most commonly associated with worse
performance (Abstract 441). This analysis of nearly cognitive performance. Statistically ad-
CROI, and could be confounded by the 2000 PLWH failed to show differences justing for the underlying conditions
cross-sectional design of the study. in the cognitive trajectory over time be- for which these drugs were prescribed
Concerns about the CNS safety of tween men who either discontinued did not substantially weaken the asso-
InSTI drugs were also supported by or continued efavirenz, supporting the ciations, but careful longitudinal analy-
in vitro and animal experiments (Ab- long-term CNS safety of efavirenz. The ses are required to clearly delineate
stract 435). Oligodendrocytes are not longitudinal design and large sample whether the observed adverse impact
easily infected by HIV, but interest in size of this analysis were strengths, but on the CNS is due to the underlying
these understudied myelin-producing an important limitation was that only condition, the drug, or both.
glial cells is growing since white matter 44 (2.2%) men remained on efavirenz
abnormalities are common in PLWH,4,5 throughout the period of observation. Structural neuro-
even in those taking suppressive ART, DeFranceso and colleagues also fo- imaging measures may
and have been linked to worse cogni- cused on the CNS safety of nucleoside/
tive performance.6 In carefully planned nucleotide reverse transcriptase inhib- detect changes not seen
experiments, Jordan-Sciutto and col- itors (nRTIs) by comparing concentra- with cognitive
leagues administered elvitegravir or tions in blood of 4 nRTIs (abacavir,
raltegravir to primary rat oligodendro- TDF, FTC, and lamivudine) with cog- performance testing
cytes and monocyte-derived macro- nitive performance in more than 600
Neuroimaging in NeuroHIV
phages (MDMs, either uninfected or participants of the POPPY (Pharma-
infected with HIV) (Abstract 435). They cokinetic and Clinical Observations in Neuroimaging is currently not included
observed that HIV-infected MDMs in- People Over Fifty) study (Abstract 419). Neuroimaging is currently not included
hibited oligodendrocyte differentiation Population pharmacokinetic modeling in the evaluation for HAND, but several
but, somewhat unexpectedly, that elvi- estimated maximum and trough drug studies demonstrated the potential
tegravir (but not raltegravir) did as well. concentrations as well as the area- relevance of this technique in PLWH.
In animal experiments, investigators under-the-time-concentration curves. As previously noted, CSVD may lead
induced demyelination in mice with Higher concentrations of TDF and FTC to vascular cognitive impairment. A
cuprizone and found that elvitegravir were associated with worse cognitive combination of both CSVD and HIV
inhibited remyelination. Experiments performance in unadjusted analyses may lead to the substantial cognitive
that are more mechanistic in design but these associations weakened above changes that are observed despite ART.
are being performed but these findings statistical significance after adjustment However, it can be difficult to differ-
suggest a possible biologic basis for for potential confounding factors such entiate the contributions of HIV from
InSTI-associated neurotoxicity. as age, sex, efavirenz use, and recrea- CSVD. Within the MACS, Wu and col-
Other scientists reported on inves- tional drug use. In contrast, higher aba- leagues longitudinally evaluated HIV
tigations regarding the potential neu- cavir concentration was associated with uninfected controls (n=46) and PLWH
rotoxicity of other antiretroviral drugs better cognitive performance and this (n=76) (Abstract 456). Annualized rates
and concomitant drugs at CROI 2019. association remained statistically signi- of change in white matter hyperinten-
For example, instead of focusing on ficant even after adjustment. Although sities (WMH), a proxy of CSVD, were
switching to an InSTI-containing regi- antiretroviral drugs may have neuro- similar between HIV–uninfected con-
men, one study focused on the switch toxicity, they are not alone: other drug trols and PLWH. PLWH who had diabe-
to tenofovir alafenamide (TAF)/emtri classes such as anticholinergics can tes or hypertension had a greater an-
citabine (FTC) from either tenofovir also adversely affect the CNS. Pub- nual increase in WMH volume. Sanford
29IAS–USA Topics in Antiviral Medicine
and colleagues also longitudinally ev- Effects of HIV on with either approach. Concerns about
aluated changes in WMH (using CSVD) Neuropathogenesis the durability of the initial response of
in virologically suppressed PLWH (n= CSF viral escape to ART optimization
119) compared with HIV-uninfected Many published studies have iden- were raised by Ferretti and colleagues,
persons (n=55) (Abstract 453). They tified HIV characteristics that may in- who identified that 5 of 21 (23.8%)
also examined if an interaction oc- fluence its neurovirulence, including PLWH with CSF viral escape who had
curred between CSVD and HIV for neu- HIV subtype,8,9 envelope sequence,10,11 previously responded to ART optimiza-
roimaging and cognitive measures. macrophage tropism,12 and CD4 and tion had a recurrent episode of es-
chemokine receptor type 5 (CCR5) af- cape (Abstract 447). Recurrence only
5 of 21 (23.8%) people finity.13 These and other issues related occurred, however, if the optimized
with HIV infection with to how HIV interacts and adapts to the regimen was simplified (n=4) or was
brain were summarized in a plenary not taken (n=1). Although CSF viral
CSF viral escape who had lecture by Swanstrom (Abstract 121). escape remains uncommon and these
previously responded to His presentation highlighted the im- data are sparse, patients and clinicians
portance of distinguishing HIV that should be educated to continue the
ART optimization had a uses R5 to enter T cells from HIV that optimized regimen and efforts should
recurrent episode uses CCR5 for entry specifically into be made to support adherence. In ad-
of escape macrophages (R5-macrophage tropic), dition to use of protease inhibitors, the
which express approximately 25-times risk of CSF viral escape has been linked
WMH burden was similar for PLWH less CD4 than T-cells. He also reviewed to low nadir or current CD4+ cell count
and HIV-uninfected individuals. Older important data supporting that approx- in chronic HIV infection. To date, no
age and the presence of hypertension imately 25% of PLWH have evidence of one has reported on the incidence of
were associated with a greater risk of an compartmentalized HIV in CSF even at CSF viral escape in early HIV infection,
increased WMH burden. These results the time of early infection.14 The pres- a shortcoming that was addressed by
suggest that both HIV and CSVD may ence of compartmentalized HIV in the Handoko and colleagues by analyzing
independently contribute to brain at- CNS may be associated with viral es- data from the Thai SEARCH 010 Study
rophy. Modifiable risk factors (eg, hy- cape from ART in the CSF15,16 (Abstract (Abstract 450). Among PLWH who ini-
pertension and diabetes) should be 449) and has implications for eradica- tiated ART during early HIV infection
aggressively treated in PLWH. Struc- tion of HIV from the CNS. (Fiebig I-V) (n=89), only 1 (1.1%) met cri-
tural neuroimaging measurements (in- Several abstracts presented new teria for CSF viral escape at 24 weeks.
cluding magnetic resonance spectros- data on CSF viral escape. A CSF Viral
copy and diffusion tensor imaging) Escape Consortium was organized by Deep sequencing was
were also obtained from several co- the National Institute of Mental Health used to identify that
horts of virologically suppressed PLWH. and proposed an approach to classify
Using principal components analysis different forms of CSF viral escape.17 64% of people with
of neuroimaging data, Cysique and Kincer and colleagues identified 14 HIV infection (n=50)
colleagues defined a composite neu- PLWH who had one form, symptomatic
rochemical marker (CNM) or “signature CSF viral escape, and they commonly
had evidence of HIV
of HIV disease,” which strongly cor- had T-cell tropic and drug-resistant compartmentalization in
related with CSF NFL concentrations HIV in CSF (Abstract 446). Similar to the cerebrospinal fluid
but not neurocognitive impairment the seminal report from Canestri et
(Abstract 454). Ruiz-Saez and col- al in 2010,15 nearly all participants re- Of 46 PLWH evaluated after 96 weeks
leagues demonstrated that perinatally sponded to optimization of their ART of ART, none had CSF viral escape.
infected adults living with HIV have regimen. Dravid and colleagues, who These data add to prior evidence that
substantial reductions in frontal brain previously published evidence link- initiating ART early in disease protects
volumes compared with matched HIV- ing CSF viral escape to use of protease the CNS. Smith and colleagues identi-
uninfected individuals (Abstract 458). inhibitors,18 reported follow-up data on fied that participants who had CSF viral
Overall, these results suggest that struc- CSF viral escape (n=41) after one of escape were approximately twice as
tural neuroimaging measures may de- 2 interventions, ART optimization or likely to have the HIV-encoded protein,
tect changes not seen with cognitive per- intensification (Abstract 451). Inten- Tat, detected in CSF (Abstract 417). The
formance testing. Observed changes sification may be a more clinically putative neurotoxicity of extracellular
may reflect neurodegeneration and implementable strategy since it does Tat remains controversial but this anal-
inflammation that occurred soon after not require genotypic resistance testing ysis found that participants who had a
seroconversion and before the initia- of HIV from CSF, which is not feasible Tat concentration that exceeded 1000
tion of ART. Longitudinal neuroimaging in many clinical settings. Approxi- pg/mL were nearly 4-fold more likely
studies of acutely infected PLWH who mately 80% of participants had sup- to have cognitive impairment than
were administered ART are needed. pressed CSF HIV RNA (≤20 copies/mL) those who had lower concentrations.
30CROI 2019: Neurologic Complications Volume 27 Issue 1 April 2019
The presence of Tat in CSF was also blood-brain permeability and better perimental autoimmune encephalitis,
associated with lower CSF amyloid-β executive functioning. Although these a disease model of autoimmune en-
1-42 concentrations, suggesting that it findings generally support that better cephalitis that has some features simi-
may be associated with AD-type neuro- suppression of HIV RNA, even at very lar to HIV encephalitis.21 Another re-
pathology. low levels, may lead to better outcomes, port from the SEARCH 010 study team
In addition to these informative pre- the findings of this study may be con- built on published research about DNA
sentations, other scientists presented founded by group differences in base- methylation signatures in monocytes
new findings relevant to how HIV inter- line HIV RNA concentrations and nadir in HAND (particularly those associated
acts with the CNS. In Uganda, where CD4+ T-cell counts. (Abstract 126). with the nervous system and the im-
non-B HIV subtypes (predominantly mune response to HIV) 22 to identify
subtypes A and D) may affect the CNS that similar signatures are present in
Effects of Host Mechanisms
differently than subtype B that is com- early HIV infection (median, 17.5 days
on Neuropathogenesis
mon in North America, Joseph and col-
leagues used deep sequencing to iden- Even though HIV can adapt to the CNS HIV was associated with
tify that 64% of PLWH (n=50) had evi- environment, which may increase its persistent alterations
dence of HIV compartmentalization in neurovirulence, the host environment
the CSF (Abstract 449). The frequency also plays a critical role, particularly in plasma and IgG
of compartmentalization did not dif- among a population that is more likely glycomes, including
fer by HIV subtype. Few details were than the general population to be ad-
provided about the cognitive assess- versely affected by comorbid condi-
decreases in anti-
ment, but the investigators noted that tions, such as obesity, cardiovascular inflammatory highly
CSF compartmentalization was asso- disease, and drug toxicity, as discussed sialylated glycans.
ciated with worse verbal fluency in un- above. Observations from cohort stud-
treated PLWH, although this difference ies and clinical trials are crucially im- after infection) (Abstract 409). Nearly
was no longer significant after ART ini- portant elements of translational re- a year after initiating ART, most DNA
tiation. de Oliveira and colleagues se- search, but the development of clini- methylation changes were minimally
quenced HIV envelope DNA by high- cally useful biomarkers and beneficial restored except for interferon-related
throughput single genome amplifica- interventions ultimately hinges on a genes (eg, IFI27, IRF7, and MX1), sug-
tion from brain tissue collected at au- clear, mechanistic understanding of gesting that DNA methylation of these
topsy from 12 donors enrolled in the pathogenesis. New findings were re- genes in blood-derived monocytes ident-
North American National NeuroAIDS ported at CROI 2019 that advance our ifies HAND risk very soon after infec-
Tissue Consortium, identifying that a understanding of the mechanisms by tion and might be a future, clinically ac-
third of the participants had evidence which the host environment increases cessible biomarker.
of compartmentalization compared the risk of CNS disease in PLWH. Nearly all research in the neuroHIV
with HIV DNA from lymph node or Four presentations focused on the field is challenged by the heterogene-
spleen (Abstract 452). immune system, a key contributor to ity of the HAND phenotype: numerous
Measuring CSF HIV RNA down to the HIV pathogenesis in the CNS. One conditions contribute to HAND risk and
single-copy level may have value,19 but novel report focused on CD30, a CD4+ these differ from individual to individ-
single-copy assays are not clinically T-cell surface protein that is enriched ual.23 Chief among these differences
available. The Cobas-TaqMan HIV-1 in infected cells. Concentrations of may be sex: women and men appear
Assay v2.0 is commonly used in the soluble CD30 in the CSF may indicate to differ substantially in the conditions
clinic and has a lower limit of quan- the extent of ongoing migration of that predispose to impaired cognition
tification (LLQ) of 20 copies/mL. If HIV transcriptionally active T-cells during and mental health disorders.24 Rubin
RNA is suppressed below the LLQ, suppressive ART, although CD30 may and colleagues extended their work
the report for this assay will indicate also be solubilized from the surface of in this area by using novel methods
whether the HIV RNA concentration an as-yet unidentified cell type within (Dynamic matrix factorization; Cluster
is at or below 20 copies/mL or below the CNS. Peluso and colleagues mea- Identification using Frobenius residual;
the limit of detection, which may be sured soluble CD30 in CSF from 130 Ingenuity Pathway Analysis) to analyze
lower than 10 copies/mL. Motta and PLWH and identified that concentra- data from a 42-plex biomarker array
colleagues previously identified that tions in CSF, but not blood, remained el- measured at several time points in par-
having HIV RNA in CSF below the limit evated during suppressive ART. Higher ticipants in the WIHS (Abstract 407).
of detection is associated with lower CSF concentrations of soluble CD30 They found that biomarker profiles, in-
CSF neopterin concentrations than correlated with higher concentrations cluding biomarkers classified as being
having HIV RNA suppressed below 20 of NFL, an axonal protein that has been associated with the antiviral immune
copies/mL.20 At CROI, Farhadian and strongly linked to risk for HAND (Ab- response, oxidative stress, and vascular
colleagues extended these findings to stract 125). Of note, the CD30/CD30 dysfunction within 2 years of initiating
link lower HIV RNA in CSF to lower ligand axis has been implicated in ex- ART distinguished women living with
31IAS–USA Topics in Antiviral Medicine
HIV from women not living with HIV sialylated glycans, correlated with bet- to disclose. Dr Letendre has had travel paid
as well as predicted cognitive trajec- ter neurocognitive performance. Data for by ViiV Healthcare.
tory over 12 years. Among women liv- on the CSF glycome were also pre-
ing with HIV, biomarkers classified sented and were similar to the findings
as “Myeloid, T Cell, and Endothelial from blood. Although high-dimension, Additional References Cited
Cell Communication” or “Microglial discovery-driven methods such as gly- in Text
Chemokine-Mediated T Cell Recruit- comics have limitations, the reported 1. Ances BM, Benzinger TL, Christensen JJ,
ment to Brain” seemed to be broadly results are promising and strongly sup- et al. 11C-PiB imaging of human immuno-
deficiency virus-associated neurocognitive
deleterious (as estimated by their as- port the value of additional research. disorder. Arch Neurol. 2012;69(1):72-77.
sociation with performance in cogni- In addition to the glycomic exo- 2. Cooley SA, Strain JF, Beaumont H, et al.
tive domains) and those classified as some work, Pulliam and colleagues Tau positron emission tomography bind-
ing is not elevated in HIV-infected individu-
“Immune Activation and Vascular Dys- presented impactful data on neuron- als. J Infect Dis. 2018;[in press]
function” or “Leukocyte Recruitment derived exosomes (NDEs) in blood (Ab- 3. de Boer MG, van den Berk GE, van HN, et
to Brain” appeared be more beneficial stract 411). A non-exosomal neuronal al. Intolerance of dolutegravir-containing
combination antiretroviral therapy regi-
over time. This distinction between biomarker, NFL, has been measured mens in real-life clinical practice. AIDS.
neuropathogenic and neuroprotective in blood and may have clinical util- 2016;30(18):2831-2834.
mechanisms highlights an important ity, 26 but its measurement in blood 4. Sanford R, Strain J, Dadar M, et al. HIV
infection and cerebral small vessel dis-
issue in the field. To date, neuroHIV currently requires a specialized instru- ease are independently associated with
research has focused more on dele- ment (Quanterix Simoa) and its concen- brain atrophy and cognitive impairment.
AIDS. 2019;
terious mechanisms associated with trations in blood can be very low during 5. Su T, Caan MW, Wit FW, et al. White mat-
the neurologic complications of HIV suppressive ART. In this analysis, the ter structure alterations in HIV-1-infected
than on mechanisms associated with investigators identified sex-based dif- men with sustained suppression of vi-
raemia on treatment. AIDS. 2016;30(2):
resilience. ferences in NDEs. In women, cognitive 311-322.
In this regard, Giron and colleagues impairment was not associated with 6. Alakkas A, Ellis RJ, Watson CW, et al. White
presented very novel glycomics data, an NFL concentrations but was associated matter damage, neuroinflammation, and
neuronal integrity in HAND. J Neurovirol.
area that has not yet been addressed with concentrations of 7 NDE proteins 2019;25(1):32-41.
in the neuroHIV field (Abstract 124). (eg, microtubule associated protein tau 7. Rubin LH, Radtke KK, Eum S, et al. Cogni-
HIV causes a persistent state of hypo- and neuronal cell adhesion molecule), tive burden of common non-antiretrovi-
ral medications in HIV-infected women.
sialylation that interferes with binding with a consistent pattern being that the JAIDS. 2018;79(1):83-91.
of sialic acid to sialic acid binding pro- proteins were higher in women who 8. Ranga U, Shankarappa R, Siddappa NB,
tein and that does not appear to reverse had asymptomatic neurocognitive im- et al. Tat protein of human immunodefi-
ciency virus type 1 subtype C strains is
with ART. Sialyic acid binding proteins pairment (ANI) and lower in women a defective chemokine. J Virol. 2004;78:
are expressed on monocytes, macro- with symptomatic mild neurocognitive 2586-2590.
phages, and other cells and the binding disorder (MND). In men, the expected 9. Sacktor N, Nakasujja N, Skolasky RL, et
al. HIV subtype D is associated with de-
of sialic acid to them may contribute association between higher NFL con- mentia, compared with subtype A, in
to the persistent inflammation that oc- centrations and cognitive impairment immunosuppressed individuals at risk of
cognitive impairment in Kampala, Ugan-
curs in PLWH.25 In this initial cross- was present, but impairment was also da. Clin Infect Dis. 2009;49(5):780-786.
sectional analysis (n=108), HIV was associated with 12 NDE proteins (eg, 10. Dunfee RL, Thomas ER, Gorry PR, et al.
associated with persistent alterations mesencephalic astrocyte-derived neu- The HIV Env variant N283 enhances
macrophage tropism and is associated
in plasma and IgG glycomes, including rotrophic factor and “a disintegrin and with brain infection and dementia. Proc
decreases in anti-inflammatory highly metalloproteinase” [ADAM] metallo- Natl Acad Sci U S A. 2006;103(41):15160-
15165.
sialylated glycans, compared with con- protease 23) that differed from those 11. Strain MC, Letendre S, Pillai SK, et
trols. The investigators found that 7 of women and were higher in both ANI al. Genetic composition of human im-
glycan structures (eg, A2G3S3, LacNac and MND than in unimpaired PLWH. munodeficiency virus type 1 in cerebro-
spinal fluid and blood without treatment
Glycans) differed between participants Although exosome methods remain a and during failing antiretroviral therapy.
who had cognitive impairment and specialized method, the prospect of J Virol. 2005;79(3):1772-1788.
those who did not. In general, main- identifying biomarkers of CNS neuro- 12. Arrildt KT, LaBranche CC, Joseph SB, et
al. Phenotypic correlates of HIV-1 macro-
tenance of higher levels of sialylation nal injury using blood is promising and phage tropism. J Virol. 2015;89(22):11294-
in blood plasma was protective: higher requires additional research. 11311.
levels of sialylated oligosaccharides 13. Gorry PR, Taylor J, Holm GH, et al. In-
creased CCR5 affinity and reduced CCR5/
correlated with better cognitive perfor- All cited abstracts appear in the CD4 dependence of a neurovirulent prima-
mance (or conversely, higher levels of CROI 2019 Abstracts eBook, ry human immunodeficiency virus type
1 isolate. J Virol. 2002;76(12):6277-6292.
hypo-sialylated oligosaccharides were available online at 14. Sturdevant CB, Joseph SB, Schnell G, Price
associated with worse performance). www.CROIconference.org RW, Swanstrom R, Spudich S. Compart-
Investigators also assessed the exo- mentalized replication of R5 T cell-tropic
HIV-1 in the central nervous system early
somal glycome and found that higher Financial affiliations in the past 12 months: in the course of infection. PLoS Pathog.
levels of several glycans, including α2-3 Dr Ances has no relevant financial affiliations 2015;11(3):e1004720.
32CROI 2019: Neurologic Complications Volume 27 Issue 1 April 2019
15. Canestri A, Lescure FX, Jaureguiberry 19. Anderson AM, Munoz-Moreno JA, McCler- consensus report of the Mind Exchange
S, et al. Discordance between cerebral non DR, et al. Prevalence and correlates program. Clin Infect Dis. 2013;56(7):
spinal fluid and plasma HIV replication of persistent HIV-1 RNA in cerebrospinal 1004-1017.
in patients with neurological symptoms fluid during antiretroviral therapy. J Infect 24. Maki PM, Rubin LH, Springer G, et al. Dif-
who are receiving suppressive antiretro- Dis. 2017;215(1):105-113. ferences in cognitive function between
viral therapy. Clin Infect Dis. 2010;50(5): 20. Motta I, Allice T, Romito A, et al. Cerebro- women and men with HIV. J Acquir Im-
773-778. spinal fluid viral load and neopterin in mune Defic Syndr. 2018;79(1):101-107.
16. Mukerji SS, Misra V, Lorenz D, et al. Tem- HIV-positive patients with undetectable 25. Lubbers J, Rodriguez E, van KY. Modu-
poral patterns and drug resistance in CSF viraemia. Antivir Ther. 2017;22(6):539-543. lation of immune tolerance via Siglec-
viral escape among ART-experienced 21. Shinoda K, Sun X, Oyamada A, et al. Sialic acid interactions. Front Immunol.
HIV-1 infected adults. JAIDS. 2017;75(2): CD30 ligand is a new therapeutic target 2018;9:2807.
246-255. for central nervous system autoimmu- 26. Gisslen M, Price RW, Andreasson U, et al.
17. Joseph J, Cinque P, Colosi D, et al. High- nity. J Autoimmun. 2015;57:14-23. Plasma concentration of the neurofila-
lights of the global HIV-1 CSF escape con- 22. Corley MJ, Dye C, D'Antoni ML, et al. ment light protein (NFL) is a biomarker
sortium meeting, 9 June 2016, Bethesda, Comparative DNA methylation profiling of CNS injury in HIV infection: a cross-
MD, USA. J Virus Erad. 2016;2(4):243-250. reveals an immunoepigenetic signature sectional study. EBioMedicine. 2016;3:
18. Dravid AN, Natrajan K, Kulkarni MM, et of HIV-related cognitive impairment. Sci 135-140.
al. Discordant CSF/plasma HIV-1 RNA in Rep. 2016;6:33310.
individuals on virologically suppressive 23. Mind Exchange Working Group. Assess-
antiretroviral therapy in Western India. ment, diagnosis, and treatment of HIV- Top Antivir Med. 2019;27(1):26-33.
Medicine (Baltimore). 2018;97(8):e9969. associated neurocognitive disorder: a ©2019, IAS–USA. All rights reserved
33You can also read
