Mild Cognitive Impairment in Prediagnosed Huntington Disease

September 10, 2010


Kevin Duff, PhD, Associate Professor in the Department of Neurology of the University of Utah, discusses his paper "Mild cognitive impairment in prediagnosed Huntington disease" that was recently published in Neurology (2010;75:500-507). He spoke with José Merino, MD, Science Editor of Please describe the PREDICT-HD study.

Duff: PREDICT-HD (Neurobiological Predictors of Huntington's Disease) is a longitudinal, observational study that seeks to identify the earliest detectable changes in motor, cognitive, and psychiatric functioning in persons at risk for Huntington disease (HD). A host of biomarkers, including brain imaging, are also being examined.

The study, which is funded by the National Institutes of Health and CHDI Foundation, Inc., hopes to inform clinical decision-making (e.g., how and when to diagnose HD) and research (e.g., most accurate measures to use in clinical trials of disease-modifying agents). Since the study began in 2001, we have used yearly visits to follow the progression of over 1,000 participants who have the HD gene mutation but do not yet have signs and symptoms to warrant a diagnosis of HD.

We are also following a smaller cohort of individuals who do not have the HD gene. Although the study is based out of the University of Iowa, it has 30 data collection sites throughout North America, Europe, and Australia. What are the cognitive domains affected in Huntington disease (HD)? Why do patients with HD have cognitive impairment?

Duff: HD affects numerous cognitive domains, including learning and memory, executive functioning, complex attention, and visuospatial perception and construction. However, processing speed might be the cognitive domain most affected by HD. Processing speed generally refers to the amount of time taken to perform cognitive acts (e.g., mental arithmetic, divided attention, matching symbols and shapes).

Processing speed may also be one of the cognitive domains affected earliest in the development of HD. The declines in processing speed and other cognitive abilities are likely due to a range of brain abnormalities observed early in the development of HD. PREDICT-HD has observed smaller striatal volumes, less cerebral white matter, and thinner cerebral sulci in those with the genetic mutation for HD compared to nonmutated peers.

Although we still have a long way to go in making the link between these brain changes and cognitive decline, the evidence appears to be rapidly accumulating. Please describe the methods and key findings of your study.

Duff: Using the nongene expanded controls in PREDICT-HD, we determined cutoff scores for four cognitive tasks (memory, processing speed, executive functioning, and visual perception) that fell 1.5 standard deviations below the mean. This 1.5 standard deviation cutoff has been used in other studies of mild cognitive impairment (MCI). These cutoffs were then applied to the large cohort of gene-expanded participants in PREDICT-HD to examine the prevalence of MCI in prodromal HD.

Our key finding was that nearly 40 percent of “presymptomatic” HD individuals met criteria for MCI. A second major finding was that individuals closer to HD diagnosis (based on current age and CAG repeat length) had higher rates of MCI. Our third major finding was that impairments in processing speed were most frequent in this sample. Do patients with pre-HD MCI have impairment of the same domains as patients with full-blown HD? What is the pattern of the impairment in patients with pre-HD MCI (amnestic vs. nonamnestic, single- vs. multiple-domain)?

Duff: As would be expected with a neurodegenerative condition like HD, the impairments in the prodromal period appear to be quite similar, albeit less severe, than those in the manifest period.

Domains other than memory (i.e., nonamnestic) appear to be more significantly affected. Processing speed, in particular, was the most commonly affected cognitive domain in our sample. For example, in our entire sample of prodromal HD, 18.8 percent of cases had impairments in processing speed, either alone or in combination with impairments in other cognitive domains.

Despite PREDICT-HD's extensive cognitive battery, some cognitive domains are not well-represented (e.g., attention, construction), and the results could have been different if other measures were used. Nonetheless, the current findings are consistent with our hypotheses based on the existing literature. Does the gene expansion length affect the incidence or severity of pre-HD cognitive impairment?

Duff: Not surprisingly, those individuals closer to manifesting HD (based on current age and CAG repeat length in our study) had higher rates of MCI. For example, in those individuals estimated to be over 15 years from a diagnosis of HD, about 27 percent percenthad some type of MCI. Even though this is a sizable number, twice as many individuals had MCI (54 percent) if they were less than nine years from diagnosis. Is there a specific pattern of MCI that may identify patients without a family history of HD who should be screened for the gene expansion?

Duff: Although this is a provocative question that could be examined in future studies, the data from PREDICT-HD did not allow us to address this point. All individuals enrolled in PREDICT-HD are "at risk" by family history. Some are found to have the expansion and others are found not to have the expansion. You found an association between motor symptoms and cognitive impairment in pre-HD patients. Why do you think this is?

Duff: Yes, although all participants in the current analyses did not show sufficient motor signs to be diagnosed with HD at the time of their enrollment into the study, some did display minor motor signs. And those who showed more motor signs did have higher rates of MCI (e.g., no motor signs = 33.8 percent MCI, motor signs that may be HD = 47.8 percent MCI).

Some of the cognitive tasks, especially the processing speed task, have motor demands (e.g., writing numbers to match symbols), and it might be that this combination of cognitive and motor functioning is what makes these tests so sensitive to the early changes in HD. What are the implications of your findings for clinical neurologists?

Duff: One important clinical implication of these findings seems to be that cognitive evaluations are indicated for patients with a family history of HD, especially those who have already been found to have the CAG expansion.

Unlike MCI trials in Alzheimer's disease, where the average age of patients is 65+ years old, the participants in PREDICT-HD are middle-age (e.g., mean age <45 years). They drive, work, raise families, and participate in community activities. Yet, nearly half of these individuals have significant cognitive impairments that might be interfering with their abilities to optimally function in their usual daily activities.

Although a formal neuropsychological evaluation might not be indicated in all cases, some regular screening of cognition, especially processing speed, seems warranted. Are there diagnostic criteria for MCI in this population?

Duff: Unfortunately, the diagnostic criteria for HD remain underdeveloped. Although some groups, such as the investigators in PREDICT-HD and the Huntington Study Group, have developed research criteria for a diagnosis of HD, these have yet to be formally moved to the clinic.

Additionally, these research criteria largely rely on motor signs, and they appear to underappreciate the cognitive and psychiatric manifestations of developing HD. As a field, medicine needs to use research information, like that found in PREDICT-HD, to better define HD and its prodromal period. Could appearance and progression of MCI be a surrogate outcome in early neuroprotection trials in carriers of the gene expansion?

Duff: One of the goals of PREDICT-HD is to better inform clinical trials in HD. As such, our findings suggest that individuals with prodromal HD and MCI be targeted for enrollment in clinical trials for compounds thought to target cognition in HD.

Additionally, progression from MCI to dementia might serve as a surrogate outcome in these neuroprotection trials. Although the clinical trials in manifest HD are numerous today, we expect that they will begin to move earlier and earlier in the development of HD (i.e., the prodromal period), and cognition will continue to be an important outcome measure.

Author Disclosure:

Dr. Huff has received funding from the following: National Institutes of Health (NIH), Janssen, National Alliance for Research on Schizophrenia and Depression, the American Cancer Society, the University of Iowa Cancer and Aging Program, and Eli Lilly and Company.

These funds have supported various research grants and clinical trials related to atomoxetine in Huntington's Disease and Alzheimer's disease.

Dr. Duff serves as Associate Editor for Archives of Clinical Neuropsychology and serves on the editorial board for the Clinical Neuropsychologist.

Dr. Merino performed a one-time consultation with staff from Bell, Falla and Associates.