24 May 2012. Alzheimer’s disease research has received a huge boost in recent years through the development of large pooled datasets. Alas, the same cannot be said for frontotemporal lobar degenerations (FTLD). FTLDs are as devastating as Alzheimer’s, eroding a person’s judgment, empathy, and social behavior as their frontal or temporal lobes wither. FTLDs are 200-fold less common than AD, but the estimated 25,000 Americans affected (Knopman and Roberts, 2011) cluster among people in their 50s and 60s. While recent progress into the genetics and pathology of these disorders has jumpstarted molecular biology and drug discovery in the field, much remains unknown (see ARF related news series; ARF related news series). Now a new initiative by the National Alzheimer’s Coordinating Center (NACC) seeks to fuel the momentum with a new resource enabling a large clinical data foundation.
NACC, which is funded by the National Institute on Aging (NIA) and based at the University of Washington School of Public Health, Seattle, promotes collaborative AD research. NACC curates large quantities of clinical and neuropathological data gathered since 1984 from the nationwide set of federally funded Alzheimer’s Disease Centers (ADCs). Since 2005 NACC has collected a longitudinal Uniform Data Set (UDS) from more than 25,000 patients seen at the centers for dementia and mild cognitive impairment (MCI). The UDS includes cognitively healthy elderly and patients’ annual follow-up visits. The center is now expanding this dataset by adding a module specific for FTLD. The module began its pilot phase of data collection in March 2012. It will enable the centers to gather standardized data on their FTLD patients, all of which will be combined into a single database and made freely available to researchers around the world.
The acronym FTLD encompasses many different molecular forms of brain disease that are distinct from Alzheimer’s. The clinical syndromes (that is, the symptoms a patient expresses) are also different from dementia of the Alzheimer type. The main ones are primary progressive aphasia (PPA), in which people have trouble naming objects and speaking, and behavioral variant frontotemporal dementia (bvFTD), in which personality and significant behavioral changes (e.g., social inappropriateness) occur very early in the disease. A third presentation of FTLD is early motor disorders such as progressive supranuclear palsy and corticobasal degeneration. These variants have been characterized according to their pathologic features, such as deposits of tau and TDP-43 protein.
The database project will help nail down the relationship between the various clinical syndromes and their underlying pathologies, Walter (Bud) Kukull told Alzforum. Kukull is at the University of Washington and directs NACC. Despite recent progress in understanding the physiological basis of FTLD and other dementias, “We are still unsure of what the clinical and pathological overlap is,” Kukull noted. With this database, “maybe we can get a better idea of the characteristics that are specific or span different groups.” In the future, NACC hopes to create clinical modules tailored to Dementia with Lewy Bodies and possibly other dementia subtypes, Kukull added.
The FTLD dataset will make it easier to do multicenter investigations, said David Knopman at the Mayo Clinic in Rochester, Minnesota, who led the working group that designed the new module. “It’s critical to be able to combine data across sites for this group of disorders, because they’re relatively uncommon, and strength lies in numbers,” he pointed out. Another goal is to increase awareness and interest in this group of disorders throughout the nation’s ADCs. Right now, only a handful of the 29 centers specialize in FTLDs. “Disseminating that expertise more broadly will have a huge benefit for patients and their families in terms of more timely diagnosis,” Knopman said. It will also create a larger pool of people who might be interested in participating in future clinical trials, he noted.
The new module is currently being rolled out in the ADCs, and should be fully implemented this month, Kukull said. As new patients come into the centers with clinically diagnosed or suspected FTLD, they will take FTLD-specific tests in addition to the standard UDS. The module will lengthen a person’s testing time by an hour, Knopman said. It includes four sections: a clinical and diagnostic portion, genetics, imaging, and neuropsychological and functional assessments. The clinical portion uses newly published diagnostic criteria for PPA and bvFTD, which help distinguish between these two (see Gorno-Tempini et al, 2011; Rascovsky et al., 2011). The genetics component adds FTLD-specific genes to those already included in the UDS. The imaging section collects data on scans the patients have previously had, or get during their current visit to help with diagnosis. These may include structural MRIs, FDG-PET and amyloid PET, or cerebral blood flow SPECT scans. At this point, however, the module does not mandate specific imaging tests.
The neuropsychological portion is the biggest part. Knopman noted that the neuropsychologists and neurologists who assembled this section faced a challenge in that the tests had to be brief, available to all centers (which meant choosing paper-and-pencil tests over computerized ones), and well-validated. Ultimately they came up with language tests not found in the standard UDS that expose deficits specific to FTLDs. "The committee honed the tests down to a time-efficient, information-rich battery," Knopman said. Sandra Weintraub at Northwestern University Feinberg School of Medicine, Chicago, Illinois, said the committee focused on tests that were not proprietary, so that they would be freely available to centers at no or low cost. Beyond the language tests, the neuropsychological portion includes a set of questionnaires administered to a family member or other informant, which assess the patient’s interpersonal relationships, empathy, and social judgment. The tests are not designed to be diagnostic; rather, they serve to collect longitudinal data, detecting small changes over time in people with FTLD.
The existing UDS has been a boon to AD research, said Michael Shelanski at Columbia University, New York City, who has contributed to the database. “It allows for very good exchange of data between sites, greatly improving the ability to do multicenter studies and to facilitate AD research,” he said. Jeff Cummings, director of the Cleveland Clinic Lou Ruvo Center for Brain Health in Las Vegas, Nevada, noted that while the NACC database is less well publicized and less widely used than the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database, it has several strengths. It is large, comprising more than 90,000 patients seen since 1984 compared to ADNI’s roughly 1,600. It includes a broad spectrum of data and has already been following people over many years. “NACC is providing a very valuable service,” Cummings said, adding that the FTLD module “is a thoughtfully put-together and very useful additional dataset. We certainly need to have well-characterized frontotemporal patients, as well as well-characterized AD patients.” Cummings added that NACC staff has been helpful and responsive when his group has requested data.
“One of the strongest aspects of the NACC dataset is the presence of pathology for many of the patients,” Cummings pointed out. That is true for more than 12,000 patients at this time. As an example of how NACC data can be used, a recent study led by Thomas Beach at Banner Sun Health Research Institute, Sun City, Arizona, compared clinical and neuropathological diagnoses on more than 900 dementia patients from the NACC database (see Beach et al., 2012). The authors found that diagnostic accuracy was low, with about 25 percent of cases showing a mismatch between clinical and neuropathological findings.
The neuropathology forms are currently in revision, said Kukull. This is being done to bring them into compliance with newly published NIA/Alzheimer’s Association criteria that adjust the way pathologists score beta amyloid accumulation in the brain (see ARF related news story; ARF related news story). Thomas Montine at the University of Washington, Seattle, who leads the NACC Neuropathology Committee in charge of the revision, noted that the new forms will also include guidelines specific for FTLD pathology. Montine told Alzforum that his group expects to submit their draft revision to the meeting of the ADC directors in Boston, Massachusetts, this October.
Furthermore, the standard UDS itself is undergoing revision this year. John Morris at Washington University, St Louis, Missouri leads the ADC clinical task force for this. Weintraub, who sits on the task force and leads the neuropsychology portion of this revision, noted that NACC wants to replace proprietary neuropsychological tests in the UDS with free or low-cost alternatives for ease of sharing and distribution. The ADCs nowadays see so many patients that commercial tests have become expensive to administer in an era of tight budgets, Weintraub said. All other portions of the UDS are freely available for download from the NACC website.
Still another reason exists to revise these forms, Weintraub said. The new diagnostic criteria for AD distinguish a preclinical stage of the disease, in which cognitive evaluations are normal but biomarkers are abnormal (see ARF related news story). There is a big push in the field to find new tests that are sensitive enough to pick up subtle cognitive changes that may occur during this phase, she said. Computerized tests could possibly accomplish this. They are frequently too difficult for people in later stages of the disease to use. One test being considered, among others, is the Cognitive Function Battery from the NIH toolbox. It is brief, computerized, inexpensive, and applicable for people ages 3 to 85. Weintraub said the NACC expects to introduce new cognitive tests into the UDS within the year.
NACC collects other types of data as well. Currently the center is gearing up for a study on the side effects of lumbar punctures (LP), modeled after the European study led by Kaj Blennow at the University of Gothenborg, Sweden, and Philip Scheltens at VU University Medical Center, Amsterdam, Holland (see ARF Webinar). The idea is to find out how common side effects such as headache actually are after LP, and to develop better procedures to minimize their occurrence. Kukull told Alzforum that the LP study is just getting off the ground in the U.S., but eventually all the data will be combined with the European numbers. The European study has been running for more than a year and has more than 20 centers enrolled, Blennow said. It will continue for another year or more, until enrollment has reached at least 5,000 participants. This large number is necessary because only about two percent of patients are expected to develop complications. Blennow said he hopes the study results will increase interest in the LP procedure among clinicians and patients. —Madolyn Bowman Rogers.