Plasma Biomarkers as Potential Indicators for HIV-associated Neurocognitive Disorders

Purpose: There are approximately 1.1 million people living with human immunodeficiency virus (HIV) in the United States (US). By race, African Americans and Hispanics are disproportionately affected by HIV. Despite improved life expectancy and a decline in HIV-associated dementia with the advent of antiretroviral therapy, HIV-associated neurocognitive disorders (HAND) remain a major comorbidity, affecting 15-50% of HIV+ individuals. HAND severity correlates with immune activation, suggesting that inflammatory mediators play a significant role in disease progression. This study aims to identify biomarkers which correlate to neurocognitive impairment and associate with gender and race. Methods: HIV seropositive African Americans, Caucasians and Hispanics (10 men and 10 women in each category) were recruited in compliance with National Institutes of Health guidelines. Subject visits included informed consent, drug screening, HIV relevant medical history review, socio-demographic survey, neurocognitive assessment and blood collection. Neurocognitive assessment was further categorized into functional domains, domain 1 (memory, psychomotor speed, reaction time, complex attention and cognitive flexibility) and domain 2 (processing speed and executive functioning). Blood samples were processed to collect plasma, PBMCs, extra-chromosomal DNA, and RNA. Biomarker mRNA levels in PBMCs were quantified and protein levels were measured in plasma and PBMC culture supernatants. Differences in protein and mRNA levels across gender, cognitive status, race and socio-demographic factors were analyzed. HIV 2-LTR circles were quantified and correlated to neurocognitive status. Results: Multivariate analysis identified monocyte chemoattractant protein 2 (MCP2) and tissue inhibitor of metalloproteinases 1 (TIMP-1) as significantly different across cognitive domain 1 and 2. Interleukin-10 was significantly different across cognitive domain 2. A significant difference by level of education, occupation and income was observed in cognitive domains 1 and 2. Univariate analysis was performed to identify specific domain tests, which correlated with biomarker levels. Conclusions: Our data suggest that plasma protein levels likely correlate with neurocognitive impairment. We also demonstrate a relationship between several socio-demographic factors and neurocognition. Further studies to characterize and validate these potential biomarkers as indicators of HAND are being conducted.