Identifying persons with HIV (PWH) at increased risk for Alzheimer’s disease (AD) is complicated because memory deficits are common in HIV-associated neurocognitive disorders (HAND) and a defining feature of amnestic mild cognitive impairment (aMCI; a precursor to AD). Recognition memory deficits may be useful in differentiating these etiologies. Therefore, neuroimaging correlates of different memory deficits (i.e., recall, recognition) and their longitudinal trajectories in PWH were examined.

We examined 92 PWH from the CHARTER Program, ages 45–68, without severe comorbid conditions, who received baseline structural MRI and baseline and longitudinal neuropsychological testing. Linear and logistic regression examined neuroanatomical correlates (i.e., cortical thickness and volumes of regions associated with HAND and/or AD) of memory performance at baseline and multilevel modeling examined neuroanatomical correlates of memory decline (average follow-up = 6.5 years).

At baseline, thinner pars opercularis cortex was associated with impaired recognition (p = 0.012; p = 0.060 after correcting for multiple comparisons). Worse delayed recall was associated with thinner pars opercularis (p = 0.001) and thinner rostral middle frontal cortex (p = 0.006) cross sectionally even after correcting for multiple comparisons. Delayed recall and recognition were not associated with medial temporal lobe (MTL), basal ganglia, or other prefrontal structures. Recognition impairment was variable over time, and there was little decline in delayed recall. Baseline MTL and prefrontal structures were not associated with delayed recall.

Episodic memory was associated with prefrontal structures, and MTL and prefrontal structures did not predict memory decline. There was relative stability in memory over time. Findings suggest that episodic memory is more related to frontal structures, rather than encroaching AD pathology, in middle-aged PWH. Additional research should clarify if recognition is useful clinically to differentiate aMCI and HAND.

Methamphetamine and cannabis are two widely used substances with possibly opposing effects on aspects of central nervous system functioning. Use of these substances is prevalent among people with HIV (PWH), though their combined effects on HIV-associated neurocognitive impairment (NCI) are unknown. Adverse effects of methamphetamine use on cognition are well documented. Cannabis may disturb cognition acutely, though its longer-term effects in PWH are not well understood. Our prior analysis of people without HIV (PWoH) found that cotemporaneous cannabis use was associated with better neurocognitive outcomes among methamphetamine users. The aim of this study was to assess how lifetime cannabis and methamphetamine use disorder relate to neurocognitive outcomes in PWH.

HIV-positive participants (n=472) were on average 45.6±11.5 years of age, male (86.4%), White (60.6%), and educated 13.9±2.5 years. Most participants were on ART (81.9%) and virally suppressed (70%). Participants were stratified by lifetime methamphetamine (M-/M+) and cannabis (C-/C+) DSM-IV abuse/dependence disorder into four groups: M-C- (n=187), M-C+ (n=68), M+C-, (n=82) and M+C+ (n=135) and completed a comprehensive neurobehavioral assessment. Demographically corrected T-scores and deficit scores were used for analyses. Group differences in global and domain NC performances (i.e., T-scores) were examined using multiple linear regression, holding constant covariates that were associated with study groups and/or cognition. Specifically, M+ participants displayed higher rates of Hepatitis C infection (p=.004), higher current depressive symptom scores (p<.001), and higher rates of detectable plasma HIV RNA (p=.014). Multiple logistic regression was used to test for group differences in probability of neurocognitive impairment (i.e., deficit scores>0.5), including the same covariates. Pooling data with a sample of HIV-negative participants (n=423), we used generalized linear mixed effect models to examine how neurocognitive performance and impairment profiles varied by methamphetamine and/or cannabis use group, HIV disease characteristics, and their interactions.

Compared to M+C+, M+C- performed worse on measures of executive functions (ß=-3.17), learning (ß=-3.95), memory (ß=-5.58), and working memory (ß=-4.05) and were more likely to be classified as impaired in the learning (OR=2.93), memory (OR=5.24), and working memory (OR=2.48) domains. M-C- performed better than M+C+ on measures of learning (ß=3.46) and memory (ß=5.19), but worse than M-C+ on measures of executive functions (ß=-3.90), learning (ß=-3.32), memory (ß=-3.38), and working memory (ß=-3.38). Generalized linear mixed effect models indicate that detectable plasma HIV RNA (ß=-1.85) and low nadir CD4 T-cell counts (nadir CD4<200; ß=-1.07) were associated with worse neurocognitive performance, and these effects did not differ in size or direction by substance use group.

In PWH, lifetime methamphetamine use disorder and both current and legacy markers of HIV disease severity are associated with worse neurocognitive outcomes. Cannabis use disorder does not appear to exacerbate methamphetamine-related deficits in PWH. Instead, results are consistent with findings from preclinical studies that cannabis use may protect against methamphetamine’s deleterious effects. Profile analysis models showed that participants with a history of cannabis use disorder display better overall neurocognitive performance than comparison (M-C-) participants. Mechanisms underlying a potential protective effect of cannabis may be elucidated by examining the temporal relationship between cannabis and methamphetamine consumption and neurocognitive performance.

Many people with HIV (PWH) are at risk for age-related neurodegenerative disorders such as Alzheimer’s disease (AD). Studies on the association between cognition, neuroimaging outcomes, and the Apolipoprotein E4 (APOE4) genotype, which is associated with greater risk of AD, have yielded mixed results in PWH; however, many of these studies have examined a wide age range of PWH and have not examined APOE by race interactions that are observed in HIV-negative older adults. Thus, we examined how APOE status relates to cognition and medial temporal lobe (MTL) structures (implicated in AD pathogenesis) in mid- to older-aged PWH. In exploratory analyses, we also examined race (African American (AA)/Black and non-Hispanic (NH) White) by APOE status interactions on cognition and MTL structures.

The analysis included 88 PWH between the ages of 45 and 68 (mean age=51±5.9 years; 86% male; 51% AA/Black, 38% NH-White, 9% Hispanic/Latinx, 2% other) from the CNS HIV Antiretroviral Therapy Effects Research multi-site study. Participants underwent APOE genotyping, neuropsychological testing, and structural MRI; APOE groups were defined as APOE4+ (at least one APOE4 allele) and APOE4- (no APOE4 alleles). Eighty-nine percent of participants were on antiretroviral therapy, 74% had undetectable plasma HIV RNA (<50 copies/ml), and 25% were APOE4+ (32% AA/Black/15% NH-White). Neuropsychological testing assessed seven domains, and demographically-corrected T-scores were calculated. FreeSurfer 7.1.1 was used to measure MTL structures (hippocampal volume, entorhinal cortex thickness, and parahippocampal thickness) and the effect of scanner was regressed out prior to analyses. Multivariable linear regressions tested the association between APOE status and cognitive and imaging outcomes. Models examining cognition covaried for comorbid conditions and HIV disease characteristics related to global cognition (i.e., AIDS status, lifetime methamphetamine use disorder). Models examining the MTL covaried for age, sex, and

relevant imaging covariates (i.e., intracranial volume or mean cortical thickness).

APOE4+ carriers had worse learning (ß=-0.27, p=.01) and delayed recall (ß=-0.25, p=.02) compared to the APOE4- group, but APOE status was not significantly associated with any other domain (ps>0.24). APOE4+ status was also associated with thinner entorhinal cortex (ß=-0.24, p=.02). APOE status was not significantly associated with hippocampal volume (ß=-0.08, p=0.32) or parahippocampal thickness (ß=-0.18, p=.08). Lastly, race interacted with APOE status such that the negative association between APOE4+ status and cognition was stronger in NH-White PWH as compared to AA/Black PWH in learning, delayed recall, and verbal fluency (ps<0.05). There were no APOE by race interactions for any MTL structures (ps>0.10).

Findings suggest that APOE4 carrier status is associated with worse episodic memory and thinner entorhinal cortex in mid- to older-aged PWH. While APOE4+ groups were small, we found that APOE4 carrier status had a larger association with cognition in NH-White PWH as compared to AA/Black PWH, consistent with studies demonstrating an attenuated effect of APOE4 in older AA/Black HIV-negative older adults. These findings further highlight the importance of recruiting diverse samples and suggest exploring other genetic markers (e.g., ABCA7) that may be more predictive of AD in some races to better understand AD risk in diverse groups of PWH.

Methamphetamine and cannabis are two widely used, and frequently co-used, substances with possibly opposing effects on the central nervous system. Evidence of neurocognitive deficits related to use is robust for methamphetamine and mixed for cannabis. Findings regarding their combined use are inconclusive. We aimed to compare neurocognitive performance in people with lifetime cannabis or methamphetamine use disorder diagnoses, or both, relative to people without substance use disorders.

423 (71.9% male, aged 44.6 ± 14.2 years) participants, stratified by presence or absence of lifetime methamphetamine (M−/M+) and/or cannabis (C−/C+) DSM-IV abuse/dependence, completed a comprehensive neuropsychological, substance use, and psychiatric assessment. Neurocognitive domain T-scores and impairment rates were examined using multiple linear and binomial regression, respectively, controlling for covariates that may impact cognition.

Globally, M+C+ performed worse than M−C− but better than M+C−. M+C+ outperformed M+C− on measures of verbal fluency, information processing speed, learning, memory, and working memory. M−C+ did not display lower performance than M−C− globally or on any domain measures, and M−C+ even performed better than M−C− on measures of learning, memory, and working memory.

Our findings are consistent with prior work showing that methamphetamine use confers risk for worse neurocognitive outcomes, and that cannabis use does not appear to exacerbate and may even reduce this risk. People with a history of cannabis use disorders performed similarly to our nonsubstance using comparison group and outperformed them in some domains. These findings warrant further investigation as to whether cannabis use may ameliorate methamphetamine neurotoxicity.

Emotional functioning is linked to HIV-associated neurocognitive impairment, yet research on this association among diverse people with HIV (PWH) is scant. We examined emotional health and its association with neurocognition in Hispanic and White PWH.

Participants included 107 Hispanic (41% primarily Spanish-speakers; 80% Mexican heritage/origin) and 216 White PWH (Overall age: M = 53.62, SD = 12.19; 86% male; 63% AIDS; 92% on antiretroviral therapy). Emotional health was assessed via the National Institute of Health Toolbox (NIHTB)-Emotion Battery, which yields T-scores for three factor-based summary scores (negative affect, social satisfaction, and psychological well-being) and 13 individual component scales. Neurocognition was measured via demographically adjusted fluid cognition T-scores from the NIHTB-cognition battery.

27%–39% of the sample had problematic socioemotional summary scores. Hispanic PWH showed less loneliness, better social satisfaction, higher meaning and purpose, and better psychological well-being than Whites (ps <.05). Within Hispanics, Spanish-speakers showed better meaning and purpose, higher psychological well-being summary score, less anger hostility, but greater fear affect than English speakers. Only in Whites, worse negative affect (fear affect, perceived stress, and sadness) was associated with worse neurocognition (p <.05); and in both groups, worse social satisfaction (emotional support, friendship, and perceived rejection) was linked with worse neurocognition (p <.05).

Adverse emotional health is common among PWH, with subgroups of Hispanics showing relative strengths in some domains. Aspects of emotional health differentially relate to neurocogntition among PWH and cross-culturally. Understanding these varying associations is an important step towards the development of culturally relevant interventions that promote neurocognitive health among Hispanic PWH.

Recent cannabis exposure has been associated with lower rates of neurocognitive impairment in people with HIV (PWH). Cannabis’s anti-inflammatory properties may underlie this relationship by reducing chronic neuroinflammation in PWH. This study examined relations between cannabis use and inflammatory biomarkers in cerebrospinal fluid (CSF) and plasma, and cognitive correlates of these biomarkers within a community-based sample of PWH.

263 individuals were categorized into four groups: HIV− non-cannabis users (n = 65), HIV+ non-cannabis users (n = 105), HIV+ moderate cannabis users (n = 62), and HIV+ daily cannabis users (n = 31). Differences in pro-inflammatory biomarkers (IL-6, MCP-1/CCL2, IP-10/CXCL10, sCD14, sTNFR-II, TNF-α) by study group were determined by Kruskal–Wallis tests. Multivariable linear regressions examined relationships between biomarkers and seven cognitive domains, adjusting for age, sex/gender, race, education, and current CD4 count.

HIV+ daily cannabis users showed lower MCP-1 and IP-10 levels in CSF compared to HIV+ non-cannabis users (p = .015; p = .039) and were similar to HIV− non-cannabis users. Plasma biomarkers showed no differences by cannabis use. Among PWH, lower CSF MCP-1 and lower CSF IP-10 were associated with better learning performance (all ps < .05).

Current daily cannabis use was associated with lower levels of pro-inflammatory chemokines implicated in HIV pathogenesis and these chemokines were linked to the cognitive domain of learning which is commonly impaired in PWH. Cannabinoid-related reductions of MCP-1 and IP-10, if confirmed, suggest a role for medicinal cannabis in the mitigation of persistent inflammation and cognitive impacts of HIV.

Frascati international research criteria for HIV-associated neurocognitive disorders (HAND) are controversial; some investigators have argued that Frascati criteria are too liberal, resulting in a high false positive rate. Meyer et al. recommended more conservative revisions to HAND criteria, including exploring other commonly used methodologies for neurocognitive impairment (NCI) in HIV including the global deficit score (GDS). This study compares NCI classifications by Frascati, Meyer, and GDS methods, in relation to neuroimaging markers of brain integrity in HIV.

Two hundred forty-one people living with HIV (PLWH) without current substance use disorder or severe (confounding) comorbid conditions underwent comprehensive neurocognitive testing and brain structural magnetic resonance imaging and magnetic resonance spectroscopy. Participants were classified using Frascati criteria versus Meyer criteria: concordant unimpaired [Frascati(Un)/Meyer(Un)], concordant impaired [Frascati(Imp)/Meyer(Imp)], or discordant [Frascati(Imp)/Meyer(Un)] which were impaired via Frascati criteria but unimpaired via Meyer criteria. To investigate the GDS versus Meyer criteria, the same groupings were utilized using GDS criteria instead of Frascati criteria.

When examining Frascati versus Meyer criteria, discordant Frascati(Imp)/Meyer(Un) individuals had less cortical gray matter, greater sulcal cerebrospinal fluid volume, and greater evidence of neuroinflammation (i.e., choline) than concordant Frascati(Un)/Meyer(Un) individuals. GDS versus Meyer comparisons indicated that discordant GDS(Imp)/Meyer(Un) individuals had less cortical gray matter and lower levels of energy metabolism (i.e., creatine) than concordant GDS(Un)/Meyer(Un) individuals. In both sets of analyses, the discordant group did not differ from the concordant impaired group on any neuroimaging measure.

The Meyer criteria failed to capture a substantial portion of PLWH with brain abnormalities. These findings support continued use of Frascati or GDS criteria to detect HIV-associated CNS dysfunction.