Director: David Eidelberg, M.D.
Title: Functional Brain Networks: A Novel Approach to Address Clinical Challenges in PD
The Udall Center at The Feinstein Institute for Medical Research focuses on a unique patient-oriented approach using validated
functional brain networks to develop novel approaches/solutions to basic problems of diagnosis and management confronted each
day by Parkinson’s disease (PD) patients, clinicians and caretakers. Findings from Center projects will shed light on the
relationship between functional brain organization in the resting state (PET, rest state fMRI) and neural activation during
cognitive processing (event-related fMRI), and provide unique information on how both interact to shape performance in PD
patients and healthy people. Project 1 addresses the serious clinical problem of levodopa-induced dyskinesias, which ultimately
affect nearly all PD patients. Project 2 examines the network basis for individual differences in the cognitive response to
dopaminergic treatment with a view to predicting which patients will develop untoward cognitive side effects under different
treatment conditions. Project 3 aims to establish the feasibility of a new network-based algorithm for providing earlier and
more accurate differential diagnosis than is currently possible.
Six cores support the research activities of three Projects: the Administrative, Clinical, Cognitive and Behavioral, Statistics
and Data Management, Imaging, and Training and Education cores. The activities of each core are independent. That said, these
activities are highly coordinated across the Clinical, Cognitive and Behavioral, and Imaging Cores in that the data output
of these cores are entered into a database created in the Statistics and Data Management core.
The specialized Udall Center mechanism allows Feinstein investigators to collaborate with scientific leaders in important
translational research areas. For instance, the involvement of Dr. Angela Cenci-Nilsson at Lund University (Sweden) offers
a true translational collaboration in the study of the role of the microvasculature in the pathogenesis of levodopa-induced
dyskinesias (LID). Professor Cenci-Nilsson’s unique expertise provides the basis for a unique collaboration spanning the gap
between experimental animal studies and human investigation in Parkinson’s disease. Similarly, investigators in Dr. Eidelberg’s
laboratory have developed a productive working relationship with Dr. Mark Gluck and his team at Rutgers University (Newark,
NJ) in the area of cognitive processing in Parkinson’s disease. This collaboration has developed further during the course
of the fMRI/behavioral studies of probabilistic category learning and neuro-economic decision making. Moreover, studies conducted
at the sleep disorders centers at North Shore – LIJ Health System, Stanford University, and the University of Pennsylvania
offer insights into the network abnormalities that appear in individuals at risk for PD prior to the onset of motor symptoms.
Recent Significant Advances
Project 1: Microvascular Changes in PD: Relationship to Levodopa-Induced Dyskinesia
- An interim analysis of the Feinstein Udall Center data confirms that levodopa treatment is associated with significant flow-metabolism
dissociation in Parkinson’s disease patients at both regional and network levels. Moreover, the data thus far support the
hypothesis that levodopa-induced dyskinesias (LID), a common side effect of treatment, is associated with an exaggerated hemodynamic
(rCBF) response to medication. Lastly, off-medication assessments of rCBF responses to 5% hypercapnia suggest that local capillary
reserve is increased in areas of flow-metabolism dissociation in LID patients. We are currently determining if these changes
are associated with changes in blood-brain-barrier (BBB) permeability during levodopa administration. These observations are
consistent with a new mechanism to explain the occurrence of LID and raise the possibility of a new treatment strategy.
- As part of our collaboration with Dr. Cenci’s laboratory in Lund, Sweden, we have begun to apply a microPET method to measure
blood flow and glucose metabolism in rats utilizing the same multi-tracer approach as performed in the human PET studies.
The data collected thus far supports the overarching hypothesis of abnormal flow-metabolism dissociation with levodopa. In
the next 6-12 months, we will determine whether an analogous exaggeration of these responses occurs in LID animals.
- Relation to the NINDS PD2014 Research Recommendations: Our translational project (Project 1) is directed at the innovative use of functional imaging tools to study levodopa-induced dyskinesias (LID), the most limiting
physical side-effect of dopaminergic treatment in today’s PD patients. We are taking a body of observations regarding microvascular/hemodynamic
changes in LID, rigorously delineating the phenomena in human patients and bringing the results for closer scrutiny in an
existing, validated rodent model. The results will likely lead to the identification of new therapeutic targets for this major
complication of PD treatment. Moreover, the flow of information from clinic to lab should inform future studies of vasculogenesis
in the brain, a topic that is only recently beginning to garner attention in the neurodegeneration field.
Project 2: Metabolic Networks and the Cognitive Treatment Response in Parkinson's Disease
- The data collected thus far are consistent with the hypothesis that the effect of levodopa on cognitive performance in individual
Parkinson’s disease patients is determined by baseline (off-state) metabolic network activity levels, and presynaptic caudate
dopamine tone. As more data are acquired, we are examining the influence of additional factors, such as the deposition of
β-amyloid in network regions and genotypic factors.
- Relation to the NINDS PD2014 Research Recommendations: The studies conducted under Project 2 of the Udall Center at Feinstein bring a new set of intellectual tools to PD research:
innovative behavioral assessments from cognitive neuroscience, as well as validated, widely available psychometric instruments.
With these we are probing the neural basis for the cognitive changes associated with dopaminergic treatment. Levodopa can
affect cognition in various ways, with a great deal of individual variability. Some effects are quite problematic, and it
would benefit patients tremendously if we could predict their response to dopaminergic therapies.
Project 3: Early Differential Diagnosis of Parkinsonism with Metabolic Imaging and Pattern Analysis
- Preliminary analysis of network expression data from Parkinson’s disease (PD) and atypical parkinsonian syndrome (APS) subjects
with established clinical diagnoses support the utility of the automated logistical classification algorithm. Moreover, blinded
categorization based on individual scan data from two independent cohorts (India: 134 subjects; South Korea: 128 subjects)
were highly accurate (> 90%) in both populations. These findings are overall encouraging for the ongoing early differential
diagnosis Udall study.
- An abnormal metabolic network was found to be specifically associated with corticobasal ganglionic degeneration (CBD). The
expression of this network is useful in the differential diagnosis of CBD vs. progressive supranuclear palsy (PSP), PD and
multiple systems atrophy (MSA). The significance of these findings will be prospectively tested on an individual basis in
the CBD cohort recruited in the ongoing study.
- Parkinson’s disease-related metabolic pattern (PDRP) expression was consistently elevated above normal levels in separate
REM behavior disorder (RBD) cohorts. Subject scores of this network can be used as an indicator of subsequent phenoconversion
to PD/diffuse Lewy Body disease (DLB).
- Relation to the PD2014 research recommendations: Project 3 studies test our current algorithm prospectively in several distinct patient populations, including a group at
high risk for developing PD. If successful, we could at last have a reliable diagnostic biomarker for PD for research and
clinical applications. Validation of such a novel disease biomarker cannot take place without extensive patient-based research,
in addition to the functional imaging tools and unusual depth of expertise in bioinformatics that we have at the Feinstein
Animal Models: Rat model of L-dopa-induced dyskinesia (LID) at Lund University.
Other Resources: Clinical ratings (off-state UPDRS), a standardized neuropsychological testing battery, off-state FDG PET data, off-state,
routine anatomical MRI, DNA banking, and brain donation/banking.
Plans for the Coming Year
Project 1: (1) We plan to continue recruitment for Aims 1 and 2 of the project. We will update the interim analyses on the newly collected
data, particularly for local vasoreactivity in drug-naïve patients; (2) In the coming year, we will start a new series of
microPET studies to measure blood flow and glucose metabolism in rats utilizing the same multi-tracer approach as already
performed in the human PET studies.
Project 2: (1) We will continue to aim for our recruitment goals in this project as well as perform interim analyses on the data.
Project 3: (1) We will continue to enroll subjects for Aims 1 and 3. (2) Network analysis will be performed on REM behavior disorder
(RBD) and healthy control subjects to identify potential metabolic covariance patterns relating specifically to prodromal
disease states. As more RBD subjects return for follow-up imaging and sufficient longitudinal scan data become available,
we will search for specific "preclinical" progression-related metabolic topographies. Such patterns may be particularly valuable
as biomarkers in disease modification trials targeting individuals with early/preclinical disease.
Relation to the NINDS PD2014 Research Recommendations: The data collected to date along with our plans for the current year will shed light on the relationship between functional
brain organization in the resting state (PET, rest state fMRI) and neural activation during cognitive processing (event-related
fMRI) and provide unique information on how both interact to shape performance in PD patients and healthy people. These studies
will also systematically probe the role of the neurovascular unit in PD, particularly the unique alterations of structure
and function that appear to occur with the development of treatment complications. This is an unchartered research area, with
potential mechanistic relevance to other neurodegenerative disorders. In the same vein, the validation of an accurate and
objective image-based diagnostic algorithm for the parkinsonian disorders would likely stimulate the development of analogous
methods for other classes of brain disease.
Select Recent Publications
- Ko JH, Spetsieris P, Ma Y, Dhawan V, Eidelberg D. Quantifying significance of topographic similarities of disease-related brain metabolic patterns. PLoS One, 2014; 9(1):e88119 [PMC 3909315]
- Holtbernd F, Gagnon JF, Postuma RB, Ma Y, Tang CC, Feigin A, Dhawan V, Vendette M, Soucy JP, Eidelberg D, Montplaisir J. Abnormal metabolic network activity in REM sleep behavior disorder. Neurology, 2014: 82(7):620-7 [PMC3963420]
- Peng S, Ma Y, Spetsieris PG, Mattis P, Feigin A, Dhawan V, Eidelberg D. Characterization of disease-related covariance topographies with SSM/PCA toolbox: effects of spatial normalization and PET
scanners. Hum Brain Map, 2014; 35(5): 1801-14 [PMC 3778048]
- Holtbernd F, Eidelberg D. The utility of neuroimaging in the differential diagnosis of parkinsonian syndromes. Semin Neurol, 2014; 34(2): 202-9 [PMC: in process]
Recent Feinstein Udall Center efforts in outreach to the Parkinson’s Disease community include:
- Feinstein Udall Center members participated in the Michael J. Fox Foundation’s Clinical Trials Fair in New York in October,
2013, which provided information about ongoing studies and led to recruitment of new volunteers to Udall Center research projects.
- The Feinstein Institute was a sponsor of the local American Parkinson Disease Association (APDA) Optimism Walk and Paws for
Parkinson's event in Massapequa, NY, October 2013. Udall Center participants provided updates to people with PD and their
family members about the current Udall Center projects.
- Udall Center members met with the Director of National Programs for the Parkinson’s Disease Foundation (PDF) to discuss novel
recruitment strategies for ongoing research studies.
- As the result of interactions with PDF, a Feinstein Udall research participant has completed training in the PDF's Parkinson's
Advocates in Research (PAIR) program, and will now work directly with our Center to inform the local community about ongoing
PD research and other resources available through the Feinstein Udall Center.
Public Health Statement
The Udall Center at The Feinstein Institute for Medical Research employs rigorously validated Parkinson’s disease (PD)-related
networks to address vital issues that impact heavily on the care of today’s PD patients.
Because dopaminergic treatment is generally so effective for the motor symptoms of PD, at least early on, it is easy to dismiss
the very real problems that ultimately develop: levodopa-induced dyskinesias and cognitive and behavioral changes for some
patients. Understanding these phenomena should not only help us improve the lives of patients, but will provide unique insight
into the pathophysiology of PD and perhaps other neurodegenerative disorders. Likewise, the validation of an automated pattern-based
method for early diagnosis will help streamline trials of new therapies for PD as well as for atypical parkinsonian syndromes.