Metabolic brain networks associated with cognitive function in Parkinson's disease

Abstract

The motor manifestations of Parkinson's disease (PD) have been linked to an abnormal spatial covariance pattern involving basal ganglia thalamocortical pathways. By contrast, little is known about the functional networks that underlie cognitive dysfunction in this disorder. To identify such patterns, we studied 15 non-demented PD patients using FDG PET and a voxel-based network modeling approach. We detected a significant covariance pattern that correlated (p < 0.01) with performance on tests of memory and executive functioning. This PD-related cognitive pattern (PDCP) was characterized by metabolic reductions in frontal and parietal association areas and relative increases in the cerebellar vermis and dentate nuclei. To validate this pattern, we analyzed data from 32 subsequent PD patients of similar age, disease duration and severity. Prospective measurements of PDCP activity predicted memory performance (p < 0.005), visuospatial function (p < 0.01), and perceptual motor speed (p < 0.005) in this validation sample. PDCP scores additionally exhibited an excellent degree of test–retest reliability (intraclass correlation coefficient, ICC = 0.89) in patients undergoing repeat FDG PET at an 8-week interval. Unlike the PD-related motor pattern, PDCP expression was not significantly altered by antiparkinsonian treatment with either intravenous levodopa or deep brain stimulation (DBS). These findings substantiate the PDCP as a reproducible imaging marker of cognitive function in PD. Because PDCP expression is not altered by routine antiparkinsonian treatment, this measure of network activity may prove useful in clinical trials targeting the progression of non-motor manifestations of this disorder.

Introduction

Parkinson’s disease (PD) is a progressive degenerative neurological disorder characterized clinically by its motor manifestations. However, cognitive deficits and behavioral abnormalities are also well documented as part of the illness, with a prevalence of dementia ranging from 24% to 31% (Aasland et al., 2005). The dementia of PD typically includes difficulties in executive and visuospatial functions, as well as memory and language deficits (Bosboom et al., 2004). By contrast, more circumscribed executive deficits, as well as secondary disturbances of memory and visuospatial function, have been described in non-demented PD patients (Green et al., 2002; cf., Zgaljardic et al., 2003, Bosboom et al., 2004 for reviews). Such subtle behavioral/cognitive deficits may have a major impact on the quality of life of PD patients (Schrag et al., 2000).

The best recognized pathological change in PD is the loss of dopaminergic cells of the substantia nigra pars compacta, directly altering the activity of motor cortico-striatopallido-thalamocortical (CSPTC) pathways (Wichmann and DeLong, 2003). The depletion of striatal dopamine has also been suggested as a cause for the cognitive deficits of PD (Green et al., 2002). The direct dopaminergic connections between the ventral tegmental area and the prefrontal cortex may also influence changes in cognition in this disease (e.g., Mattay et al., 2002; cf., Cools, 2006 for review). Nonetheless, the modulation of dopamine systems cannot fully explain the cognitive deficits in PD, and other neurotransmitter systems may be influential (Emre, 2003, Pillon et al., 2003, Bosboom et al., 2004). Moreover, recent postmortem studies in PD suggest a correlation between dementia and cortical Lewy body formation as well as coincident Alzheimer's disease-associated neurofibrillary pathology (Braak et al., 2003, Braak et al., 2005, Kovari et al., 2003).

Positron emission tomography (PET) has been useful in the investigation of cognitive functioning in PD. Functional imaging studies have generally focused upon activation experiments to detect abnormal patterns of neural activity in PD patients during task performance (e.g., Dagher et al., 1999, Nakamura et al., 2001, Owen, 2004). Relevant information has also been obtained through imaging studies conducted in the rest state. For instance, PET imaging has been used to examine the relationship between neuropsychological test performance and the integrity of nigrostriatal dopaminergic projections (e.g., Marié et al., 1999, Brück et al., 2004, Brück et al., 2005, Cheesman et al., 2005). Recently, PET has also been utilized to contrast dopaminergic and cholinergic projection systems in PD patients with and without dementia (Hilker et al., 2005).

PET imaging in the rest state can also provide inferences regarding the status of neural pathways in PD patients. In this regard, ¹⁸F-fluorodeoxyglucose (FDG) PET is particularly relevant as a measure of local synaptic activity and the biochemical maintenance processes that dominate the rest state (cf., Jueptner and Weiller, 1995, Eidelberg et al., 1997). Indeed, the effects of pathology on these local cellular functions appear to be greater than other factors that influence the variation in regional brain function that is observed in the resting condition (Ma et al., in press). Moreover, neuropathological processes, even if highly localized, can alter functional connectivity across the entire brain in a disease-specific manner (Eidelberg, 1998, Eckert and Eidelberg, 2005).

Spatial covariance analysis has proven useful as a means of identifying the network abnormalities that are associated with neurological disease (Alexander and Moeller, 1994). Using this network-modeling approach, we have found that PD patients express an abnormal metabolic pattern characterized by increased pallido-thalamic and pontine activity associated with relative reductions in cortical motor regions (Eidelberg et al., 1994, Eidelberg et al., 1997; cf., Carbon et al., 2003a). To date, this PD-related pattern (PDRP) has been detected in eight independent patient populations (Moeller et al., 1999, Feigin et al., 2002, Lozza et al., 2004, Asanuma et al., 2005, Eckert et al., in press) and its expression has been found to be highly reproducible in individual subjects (Ma et al., in press). In addition to accurately discriminating between PD patients and controls (Asanuma et al., 2005, Ma et al., in press), PDRP expression has been found to correlate consistently with Unified Parkinson's Disease Rating Scale (UPDRS) motor scores (Eidelberg et al., 1995, Lozza et al., 2004, Asanuma et al., 2006) and with clinical responses to therapy (e.g., Trošt et al., 2006, Asanuma et al., 2006; cf., Carbon et al., 2003a, Eckert and Eidelberg, 2005).

Similar network approaches can be used to identify potential imaging biomarkers of cognitive functioning in PD (see Carbon and Marié, 2003 for review). Using a covariance mapping approach in the analysis of FDG PET and neuropsychological data, Mentis et al. (2002) identified discrete patterns of resting metabolic activity associated with cognitive and affective functions in non-demented PD patients. In a subsequent FDG PET study, Lozza et al. (2004) used the scaled subprofile model (SSM) and principal components analysis (PCA) (Alexander and Moeller, 1994; cf., Habeck et al., 2005) to characterize a specific metabolic pattern associated with executive functioning in less severely affected patients. Interestingly, their data showed that the patterns associated with motor and cognitive functioning were orthogonal, i.e., statistically independent. This finding supported the notion that the neural pathways mediating these aspects of PD are functionally segregated, at least in the rest state (Wichmann and DeLong, 2003).

In the current study, we utilized a voxel-based adaptation of this approach (Carbon et al., 2003b; cf., Scarmeas et al., 2004) to identify specific spatial covariance patterns relating to cognitive function in non-demented PD patients. To validate these networks as potential metabolic markers of cognitive decline in PD, we determined whether pattern expression predicted performance in a prospective patient cohort. We further explored the potential of these patterns as biomarkers in clinical trials of interventions to improve cognitive function in PD patients. To this end, we evaluated the reproducibility of pattern expression in individual subjects, as well as the effects of routine antiparkinsonian therapies on measurements of network activity.