ARTICLE IN BRIEF
Investigators described network changes in a dozen people who carried the mutant Huntington's disease (HD) gene, a third of whom developed symptoms while in the study. The study measured the rate of disease progression in these HD carriers by first identifying the network, tracking changes over time, and then ultimately linking the network changes to the onset of symptoms.
People who carry the mutant huntingtin (HTT) gene for Huntington's disease (HD) have allowed scientists to peer into their brain years, even decades, before the onset of clinical symptoms. Now, a team of New York scientists has identified an abnormal brain network that worsens over the course of these pre-symptomatic years, suggesting its use as a potent biomarker of the earliest events of the neurodegenerative condition.
In a paper published in the Aug. 29 issue of the Journal of Clinical Investigation, investigators described network changes in a dozen people who carried the HTT gene, a third of whom developed symptoms while in the study. The study measured the rate of disease progression in these HD carriers by first identifying the network, tracking changes over time, and then ultimately linking the network changes to the onset of symptoms. The genetic information included the number of triplicate repeats in the HD gene.
“It's always hard to know what it means if a patient is worse or better,” said David Eidelberg, MD, director of the Center for Neurosciences at the Feinstein Institute for Medical Research in New York, who is the principal author of the study. “Even rating scales in symptomatic patients are lousy predictors of disease progression. Then how do you measure change in a person who has yet to have symptoms? We really needed a way to identify brain changes before symptoms begin.”
Clinical rating scales, such as the Unified Huntington's Disease Rating Scale (UHDRS) are not sensitive to disease progression and are not useful in pre-symptomatic HD, he said. Other investigators searched the periphery for biomarkers but came up short. Volumetric MRI of the caudate is meaningful, but it only shows regional changes.
“We identified a network that changes over time and then extracted the information and validated the network,” Dr. Eidelberg said. “It is a faithful recorder of disease progression, whether a person has clinical symptoms or not.”
The study was funded by the NIH National Institute of Neurological Disorders and Stroke and the National Institute of Biomedical Imaging and Bioengineering.
Dr. Eidelberg and colleagues recruited a dozen people between 25 and 62 years old who had genetic testing and knew that they carried the HD mutation. The CAG repeat length ranged from 39 to 45.
“HD gene carriers vary in the extent of their mutations but fundamentally they have the same profile,” he explained.
The investigators scanned the HD mutation carriers with [18F]-fluorodeoxyglucose PET to measure cerebral metabolic activity at baseline and again at 18 months, four years, and seven years. At each time point, the subjects were also scanned with [11C]-raclopride PET and structural MRI to measure concurrent declines in caudate/putamen D2 neuroreceptor binding and tissue volume.
They compared the measurements with the rate of metabolic network progression in another cohort of 21 pre-symptomatic HD gene carriers in Holland who had two scans over the course of two years.
In both groups, they identified a significant “covariance pattern characterized by progressive changes in striato-thalamic and cortical metabolic activity.”
The network activity increased linearly over the course of the seven years, and was consistent across all time points. The rate of progression was virtually the same in both samples, the study authors noted. Dr. Eidelberg added that the network progressed at about twice the rate of the single region measurements, suggesting that it may well be a more sensitive biomarker than MRI and regional PET D2 binding to assess HD progression and test potential disease-modifying drugs.
The pattern expression had a similar linear increase as the disease progressed (p<0.0001) at a nearly identical rate of 0.19 standard deviations (from normal) a year, said Dr. Eidelberg. On the seventh year they had nine people return for testing; and four of them had phenoconverted. When the scientists analyzed the network data from those who had officially been diagnosed with HD they found that these subjects had “substantial elevations in network activity, more than 3 standard deviations above the normal mean.”
The scientists believe that the metabolic network progression begins two decades before phenoconversion. The network comes on about three years before the volume-loss pattern and continues to change at a constant rate. The network activity is about 2 standard deviations from normal at about a decade before clinical signs. This is about the same time that the caudate D2 receptor binding is showing deficits but before any volume loss is picked up on the MRI. At the time of phenoconversion, the network activity is about 4 standard deviations above normal.
While structural MRI is also being tested as a potential biomarker of HD, as is functional MRI, Dr. Eidelberg said that PET shows areas of low metabolic activity that was not picked up on structural MRI.
He added that having a more sensitive biomarker to pick up changes could mean that scientists could enroll fewer subjects for clinical trials. “Randomizing 120 premanifest HD carriers may be sufficient for detecting a 30% reduction in the progression rate at a power of 0.90 in a 2-year disease modification trial,” the study authors wrote.
Dr. Eidelberg said that this study is a proof of principle that network scanning is sensitive to neurodegenerative disease, adding that ultimately researchers would be able to identify networks in Alzheimer's disease and other conditions.
“We can measure HD progression beautifully,” he added. “Once we have a potentially successful drug candidate there will be ways to rapidly determine if it works.” The New York researchers collaborated with colleagues at Columbia University, University of Iowa Carver College of Medicine, University Medical Center Groningen, Netherlands, and University of Toronto. They will be conducting further studies to see whether the metabolic network activity continues to increase in the years following the onset of symptoms. They showed continued changes in the first five years after phenoconversion but they are not sure what happens further down the road.
EXPERTS WEIGH IN
“The advantage of identifying brain networks in Huntington's is these changes start earlier than the volume loss that is measured with MRI,” said Rajiv Ratan, MD, PhD, executive director of Burke Medical Research Institute and professor of neurology and neuroscience and associate dean at Weill Cornell Medical College. “Being able to identify changes in brain networks involved early on in Huntington's is going to improve the power of clinical trials, and ultimately to allow people to be treated much earlier in the disease process.”
“While MRI is easier to access, the PET networks are more robust with much higher signal to noise so it is easier to detect a change,” he added. “They showed a significant change in the network during the five years prior to phenoconversion. If you didn't want to give all-comers a drug 20 years before symptoms you would want a biomarker that can show significant changes even during the two years before phenoconversion. Volume loss is like a tombstone. The network changes reflect one of the earliest events in the disease process before cells die. What we have been looking for is a biomarker that gives you the greatest sensitivity to change, particularly near the time of phenoconversion.”
“Any biomarker we can find in presymptomatic HD will be important,” said Elizabeth Aylward, PhD, associate director of the Center for Integrative Brain Research at Seattle Children's Research Institute. “The scientists did everything they could to validate their findings,” said Dr. Aylward, who studies structural MRI in HD. She is not certain, however, whether it would be “feasible to put HD patients through repeated PET scans. Very few sites have access to PET, and unlike MRI, patients are exposed to radiolabeled ligands.”
Christopher Ross, MD, PhD, a professor of psychiatry, neurology and neuroscience at Johns Hopkins University and director of the Division of Neurobiology, favors MRI for its availability and cost. Still, Dr. Ross said that more studies using FDG-PET could prove its benefit as a stronger biomarker.