Vascular injury markers associated with cognitive impairment in people with HIV on suppressive antiretroviral therapy

Abstract

Objective: Human immunodeficiency virus (HIV)-associated neurocognitive disorders (HAND) remain prevalent despite viral suppression on antiretroviral therapy (ART). Vascular disease contributes to HAND, but peripheral markers that distinguish vascular cognitive impairment (VCI) from HIV-related etiologies remain unclear.

Design: Cross-sectional study of vascular injury, inflammation, and central nervous system (CNS) injury markers in relation to HAND.

Methods: Vascular injury (VCAM-1, ICAM-1, CRP), inflammation (IFN-γ, IL-1β, IL-6, IL-8, IL-15, IP-10, MCP-1, VEGF-A), and CNS injury (NFL, total Tau, GFAP, YKL-40) markers were measured in plasma and CSF from 248 individuals (143 HIV+ on suppressive ART and 105 HIV- controls).

Results: Median age was 53 years, median CD4 + cell count, and duration of HIV infection were 505 cells/μl and 16 years, respectively. Vascular injury, inflammation, and CNS injury markers were increased in HIV+ compared with HIV- individuals ( P < 0.05). HAND was associated with increased plasma VCAM-1, ICAM-1, and YKL-40 ( P < 0.01) and vascular disease ( P = 0.004). In contrast, inflammation markers had no significant association with HAND. Vascular injury markers were associated with lower neurocognitive T scores in age-adjusted models ( P < 0.01). Furthermore, plasma VCAM-1 correlated with NFL ( r = 0.29, P = 0.003). Biomarker clustering separated HAND into three clusters: two clusters with high prevalence of vascular disease, elevated VCAM-1 and NFL, and distinctive inflammation profiles (CRP/ICAM-1/YKL-40 or IL-6/IL-8/IL-15/MCP-1), and one cluster with no distinctive biomarker elevations.

Conclusions: Vascular injury markers are more closely related to HAND and CNS injury in PWH on suppressive ART than inflammation markers and may help to distinguish relative contributions of VCI to HAND.

Figure 1.. HIV infection and HAND are associated with increased plasma vascular injury markers.

(A) Association of plasma ICAM-1, VCAM-1, and CRP with HIV infection, HAND, and vascular disease. (B) Association of CSF ICAM-1, VCAM-1, and CRP with HIV infection, HAND, and vascular disease. Medians and IQRs are indicated as horizontal and vertical lines, respectively. Statistical significance was calculated using Mann–Whitney U test; significant differences (p<0.05) are indicated.

Figure 2.. Vascular injury markers are associated with lower neurocognitive T scores and correlate with CNS injury markers.

(A) Association of plasma (top) and CSF (bottom) NFL, total Tau, GFAP, and YKL-40 with HIV infection, HAND, and vascular disease. Medians and IQRs are indicated as horizontal and vertical lines, respectively. Statistical significance was calculated using Mann–Whitney U test; significant differences (p<0.05) are indicated. (B) Plasma NFL, YKL-40, ICAM-1, and VCAM-1 levels correlate negatively with global neurocognitive T scores, plasma VCAM-1 correlates positively with plasma NFL, and plasma VCAM-1, ICAM-1, and CRP correlate positively with plasma YKL-40 in PWH (middle panels). Inter-relationship of plasma and CSF CNS injury and glial activation markers with plasma vascular injury markers are shown in the bottom panels. Relationships between continuous variables were analyzed by Spearman’s rank correlation (significant correlations p<0.05).

Figure 3.. Association of plasma inflammation markers with HIV infection, HAND, and vascular disease.

(A) Association of plasma IL-1β, IFN-γ, IL-6, IL-8, IL-15, IP-10, MCP-1, and VEGF-A with HIV infection, HAND, and vascular disease. Medians and IQRs are indicated as horizontal and vertical lines, respectively. Statistical significance was calculated using Mann–Whitney U test; significant differences (p<0.05) are indicated. (B) and (C) K-means clustering of 15 plasma biomarkers combined with dimensionality reduction by principal component analysis (PCA) among PWH with HAND (n=71) identifies three clusters distinguished by inflammation marker profiles, prevalence of prior vascular disease, and progression to HAD within 2.5 years. Contributions of each biomarker to Dim1 and Dim2 are shown on the PCA plot (B) and Supplementary Digital Content 7. Statistical significance was calculated using the Kruskal-Wallis H test.