Quantifying brain-age gap in people with HIV on antiretroviral therapy using plasma neurofilament light chain

OBJECTIVE: To estimate the number of additional years above chronological age ("brain-age gap") at which observed plasma neurofilament light chain (NfL) levels in cognitively unimpaired people with HIV (PWH) correspond to published age-stratified standards from cognitively unimpaired individuals in the general population. DESIGN: Secondary analysis of the multicenter observational cohort study (ACTG HAILO) involving older, cognitively unimpaired PWH on antiretroviral therapy (ART). METHODS: Plasma NfL concentrations were quantified using the Quanterix® Simoa assay in 340 PWH aged ≥45 years. Values were compared with two age-stratified reference standards from cognitively unimpaired general-population controls. The brain-age gap was estimated using a generalized logistic model with repeated measures so that 5% of observed NfL concentrations in PWH would exceed the reference upper 95 th percentile for the imputed age. RESULTS: Median plasma NfL concentration was 10.1 pg/mL (IQR: 7.4-13.5). Using Simrén et al. standards, 24.1% (95% CI: 19.8-29.0) of observed values exceeded the age-specific upper 95 th percentile, and 10.0% (95% CI: 7.2-13.7) using Bornhorst et al. standards. Model-based results suggested brain age was 11.0 years (95% CI: 9.1-13.7) or 6.4 years (95% CI: 5.1-8.5) higher than chronological age, depending on the standard. The estimated brain-age gap was significantly greater in individuals with subsequent cognitive decline than in those without (Simrén: 13.0 vs. 8.4 years; Bornhorst: 7.6 vs. 5.0 years). CONCLUSION: Plasma NfL reflects a substantial brain-age gap in cognitively unimpaired PWH on ART, supporting evidence of advanced aging. The brain-age gap was higher in PWH with subsequent cognitive decline.