Abstract
We examined relationships between neurocognition and immune activation in Ugandan adolescents with perinatally acquired HIV (PHIV). Eighty-nine adolescents in Kampala, Uganda (32 virally suppressed [<400 copies/mL] PHIV and 57 sociodemographically matched HIV-negative controls), completed a tablet-based neurocognitive test battery. Control-derived z-scores for 12 individual tests and a global/overall z-score were calculated. We measured plasma (soluble CD14 and CD163), monocyte (proportions of monocyte subsets), and T-cell (expression of CD38 and HLA-DR on CD4 + and CD8 + ) activation and gut markers. Spearman rank correlations and median regressions examined associations between test performance and immune activation. The median [IQR] age was 15 [13-16] years, and 40% were girls. The median time on antiretroviral therapy was 10 years [7-11] for PHIV; 87% had viral load <50 copies/mL. Compared with controls, global z-scores were lower among PHIV ( P = 0.05) and significantly worse on tests of executive functioning and delayed recall ( P 's ≤ 0.05). Overall, monocyte activation significantly correlated with worse test performance on global z-score (r = 0.21, P = 0.04), attention, processing speed, and motor speed (r = 0.2-0.3, P ≤ 0.01). T-cell activation was significantly correlated with worse performance on tests of learning, executive functioning, and working memory (r = 0.2-0.4, P ≤ 0.04). In PHIV, after adjusting for age, sex, and antiretroviral therapy duration, activated CD4 T cells remained associated with worse memory (β-0.3, 95% CI: -0.55 to -0.07, P = 0.01). PHIV with virologic suppression on antiretroviral therapy shows evidence of worse neurocognitive test performance compared with controls. Monocyte and T-cell activation is correlated with worse neurocognition in Ugandan youth with and without HIV, which has not been previously investigated in this setting.