FDG PET/computed tomography can detect region-specific neuronal changes following antiretroviral therapy in HIV-infected patients.

Abstract

Antiretroviral therapy (ART) has dramatically extended the life expectancy of people with HIV (PWH) by effective control of viral replication and immune recovery [1]. However, despite the significant success of ART, HIV-associated neurocognitive disorders (HAND) continue to persist, with dysfunction in activities observed in up to 15% of patients who are also at increased risk for worsening neurocognitive impairment [2,3]. Neuroimaging findings and altered volumetric changes observed in virally suppressed PWH, coupled with elevated cerebrospinal fluid levels of proinflammatory markers, provide ample support for persistent neuroinflammation in the setting of suppressive ART [4,5]. Variations in the clinical presentation of HAND sub-phenotypes in PWH [6,7] as well as the postmortem evidence of region-specific synaptic and neuronal dysfunction in the ART era [8] highlight the presence of distinct pathological mechanisms responsible for HAND. In this issue of AIDS, Wang et al.[9] report their findings of a neuroimaging study that evaluated changes in brain glucose metabolism following treatment initiation in a cohort of 22 ART-naïve PWH with low CD4+ counts. The authors utilized 18F-fluorodeoxyglucose (FDG) PET/computed tomography (PET/CT) to explore the temporal relationship between ART initiation and short-term and long-term changes in FDG uptake as an indicator of neuroinflammation and neuronal function in specific cortical and subcortical regions. The authors also explored the correlations of the observed changes with viral and immune activation parameters. The study does not simply propose FDG-PET/CT as a modality to monitor disease progression as this advanced imaging technique relies on substantial infrastructure and expertise. Rather, Wang et al.[9] provide a unique insight into the central nervous system (CNS) metabolic response to ART initiation, revealing the differential susceptibility of specific brain regions to HIV infection and permanence of neuronal injury following ART in these regions. The authors observed predominant hypermetabolism in basal ganglia and thalamus of ART-naive PWH compared with uninfected controls; however, this change was reversed within a short period of about 2 months after treatment initiation. More intriguingly, hypometabolism was detected in these subcortical regions after nearly 2 years of ART, despite effective viral control and CD4+ recovery. Furthermore, the parallel assessment of different cortical regions allowed the authors to determine that there was a short-term increase in FDG uptake following ART initiation primarily in the frontal cortex. In contrast to the observation in the subcortical regions wherein changes in FDG uptake were positively correlated with soluble IL-6 (sIL-6) and soluble CD14 (sCD14), the FDG update in the frontal cortex was accompanied with decreased viral load and sCD14 levels. These longitudinal and region-specific changes in the FDG uptake and their correlation with sIL-6 and sCD14 illustrate a more nuanced pathological process in the setting of effective viral control than that initially considered in the early days of the HIV pandemic. The early subcortical hypermetabolism followed by decreased hypometabolism in the long-term cases agrees with neuroimaging evidence of volume loss and pathological evidence of persistent damage in these regions. The short-term increase in FDG uptake in the frontal cortex likely reflects recovery from neuronal injury following ART initiation. Despite the small cohort size, the clear correlation of viral load and CD4+ count with persistent neuronal damage in subcortical regions confirms the clinical and pathological presentation of neurocognitive impairment in PWH in the ART era and provides several clinical implications. The study findings raise important questions about the neuropathogenesis of neurocognitive impairment. What are the factors involved in permanent versus reversible neuronal injury observed in specific regions in the setting of ART? Evidently, timing of treatment initiation is important to prevent persistent injury as reflected by the presence of very early CNS damage within days after HIV infection [5]. Additionally, as suggested by the authors, the persistence of neuropathology and neuroimaging findings in striatal areas, including caudate nucleus and putamen, indicates a role for dopamine in this regional specificity. Among other host factors that might contribute to these findings, regional differences in the expression of the antioxidant/anti-inflammatory enzyme heme oxygenase-1 (HO-1) have also been proposed to explain why recovery might be possible in certain brain regions but not others [10]. It would, therefore, be interesting to explore the possible correlation between region-specific changes observed in FDG uptake and changes in oxidative stress and dopamine uptake using neuroimaging studies in future longitudinal studies. One other consideration in the interpretation of the study findings is the potential confounding effect of ART. Although some studies suggest insufficient CNS access of ART is associated with worse neurocognitive outcomes, other studies, including a recent meta-analysis of diffusion tensor imaging (DTI) of PWH, reveal significant changes in white matter microstructures and a correlation between white matter loss and duration of ART exposure [11,12]. The study by Wang et al.[9] was not designed to address the role of ART in persistent neuronal injury. However, given that ART compliance is critical to effectively control HIV replication, the potential unintentional deleterious effects of ART in the CNS should be considered in future studies. The article by Wang et al.[9] is timely, and the authors should be congratulated for their findings providing a unique clinical perspective. The findings of Wang et al.[9] drive home the necessity of early ART initiation for improved neurocognitive outcomes, while providing further evidence that neuroinflammation should be resolved with approaches complementary to ART. Acknowledgements Conflicts of interest There are no conflicts of interest.