Elevation of brain ADC (apparent diffusion coefficient) in HIV-associated neurocognitive disorders and evolution after treatment: A pilot study

Abstract

BackgroundThirty to 50% of HIV-infected patients develop HIV-Associated Neurocognitive Disorders (HAND) despite virological control. The previously published Neuro+3 study showed their neurocognitive status can be improved by intensifying antiviral therapy. Our study is a part of the Neuro3+ study and aims to study apparent diffusion coefficient (ADC) as a biomarker for neurological improvement. Patients and methodsWe prospectively included 31 patients with HAND. They received therapy with better CNS Penetration Effectiveness (CPE) score with two-year follow-up. Cognitive status was assessed at day 0 (D0) and week 96 (W96) using Frascati 3-stage classification and Global Deficit Score (GDS). Brain MRI at D0 and W96 assessed morphological data (white matter hyperintensities, opportunistic infections, ischemic lesions, atrophy) and measured whole brain apparent diffusion coefficient (ADC). We compared their data with a control group of 20 healthy patients with similar ages and sex ratio. ResultsAfter ARV intensification, cognitive status was significantly improved: GDS (n = 1,4 vs 1,0 p = 0.01) and Frascati scale (2HAD/22MND/7ANI vs 1HAD/8MND/17ANI p = 0.001). Mean ADC was significantly higher in patients at inclusion than in controls (0.88 × 10−3 mm2/s ± 0.06 vs 0.81 × 10−3 mm2/s ± 0.04, p = 0.0001). ADC decreased after treatment (0.88 × 10−3 mm2/s ± 0.06 vs 0.85 × 10−3 mm2/s ± 0.06 (p = 0,04). In subgroup analysis, ADC significantly decreased in clinically improved patients (0.89 × 10−3 mm2/s ± 0.07 vs 0.85 × 10−3 mm2/s ± 0.07 (p = 0,03)) and did not significantly change in non-clinically improved patients (0.86 × 10−3 mm2/s ± 0.07 vs 0.84 × 10−3 mm2/s ± 0.07 (p = 0,31)). After treatment, there was no significant difference between patients and controls (0.85 × 10−3 mm2/s ± 0.06 vs 0.81 × 10−3 mm2/s ± 0.04, p = 0.17). ConclusionWhole-brain ADC is a good biomarker of HIV-associated neurocognitive disorders. It is significantly increased in patients with HAND compared with controls and significantly decreases after treatment. It is all the more important to have a quantitative biomarker as conventional imaging does not contribute to the diagnosis.