Redistribution of brain glucose metabolism in people with HIV after antiretroviral therapy initiation

Abstract

Objective: We evaluated brain glucose metabolism in people living with HIV (PWH) with [18F]-Fluoro-Deoxyglucose (FDG) PET/computed tomography (CT) before and after antiretroviral therapy (ART) initiation.

Design: We conducted a longitudinal study wherein ART-naive late-presenting untreated PWH with CD4+ cell counts less than 100 cells/μl were prospectively assessed for FDG uptake at baseline and at 4-8 weeks (n = 22) and 19-26 months (n = 11) following ART initiation.

Methods: Relative uptake in the subcortical regions (caudate, putamen and thalamus) and cortical regions (frontal, parietal, temporal and occipital cortices) were compared across time and correlated with biomarkers of disease activity and inflammation, in addition to being compared with a group of uninfected individuals (n = 10).

Results: Before treatment initiation, putaminal and caudate relative FDG uptake values in PWH were significantly higher than in uninfected controls. Relative putaminal and thalamic uptake significantly decreased shortly following ART initiation, while frontal cortex values significantly increased. FDG uptake changes correlated with changes in CD4+ cell counts and viral load, and, in the thalamus, with IL-6R and sCD14. Approximately 2 years following ART initiation, there was further decrease in subcortical relative uptake values, reaching levels below those of uninfected controls.

Conclusion: Our findings support pretreatment basal ganglia and thalamic neuroinflammatory changes in PWH, which decrease after treatment with eventual unmasking of long-term irreversible neuronal damage. Meanwhile, increased frontal cortex metabolism following ART initiation suggests reversible cortical dysfunction which improves with virologic control and increased CD4+ cell counts. Early initiation of treatment after HIV diagnosis and secondary control of inflammation are thus necessary to halt neurological damage in PWH.

Figure 1:

Longitudinal changes in relative SUV in the caudate, putamen, thalamus and frontal cortex between the baseline (BL), short-term follow-up (ST, n=22) (A) and long-term follow-up after ART initiation (LT, n=11) (B). For the thalamic dataset, one subject was excluded as an outlier.

Figure 2:

A. Differences in relative SUV in the putamen, caudate, thalamus and frontal cortex between ART-naïve PLWH at baseline (BL< n=22) and uninfected controls (n=10). B. Differences in relative SUV in the caudate, putamen, thalamus and frontal cortex between PLWH on continuous long-term ART for approximately two years (LT, n=11) and uninfected controls (n=10). For the thalamic dataset, one subject was excluded as an outlier.

Figure 3:

Repeated measures correlations between regional relative SUV values and CD4 counts/plasma viral loads in the short-term (ST, n=22) and long-term (LT, n=11) follow-up after ART initiation.

Figure 4:

Repeated measures correlations between relative SUV values and inflammatory markers in the short-term (ST, n=15) follow-up after ART initiation.

Figure 5:

Relative SUV parametric maps in one participant at (A) baseline, (B) short-term follow-up and (C) long-term follow-up. Axial, sagittal and coronal maps showing visually detectable decreasing putaminal (black arrows) and increasing cortical uptake (arrowheads) after initiation of ART.