2 November 2010. Frontotemporal dementias pose a challenge for diagnosis because numerous overlapping clinical syndromes are associated with the disease. Scientists at the 7th International Conference on Frontotemporal Dementias, held 6-8 October 2010 in Indianapolis, Indiana, held out hope that this situation will soon change, thanks to advances in neuroimaging and greater knowledge of fluid biomarkers. “The biggest change since our last [FTD] meeting has been the proliferation of rational imaging approaches to frontotemporal dementia,” said Bruce Miller of the University of California in San Francisco. “I think we really are understanding these diseases from a circuit perspective.”
For example, William Seeley of UCSF showed that neurodegenerative syndromes map onto specific brain networks (see ARF related news story). In behavioral variant FTD, he said, connectivity falters primarily in the salience network (see Zhou et al., 2010). Consisting of regions of the prefrontal cortex, anterior cingulate, insula, and striatum, the salience network is activated by events with emotional or survival significance, and so is dubbed the “here and now” network. In AD, by contrast, the default mode network sustains the most damage. This network includes regions of the medial temporal lobe, posterior cingulate, and inferior parietal cortex, and is active during daydreaming and in between a person’s focus on a specific task, prompting Seeley to refer to this network as the “there and then” system. These two networks have opposing functions in the brain, and lesions in one system seem to increase connectivity in the other, Seeley said. Thus, people with AD tend to have overactive salience networks, whereas people with FTD have more connectivity in the default mode network than do healthy people. Because of these changes in brain circuits, AD patients are acutely emotionally sensitive, whereas bvFTD patients tend to be emotionally flat and more captivated by non-emotional stimuli in the outside world. The clinical potential of this finding, Seeley said, is that by comparing blood oxygen-level dependent (BOLD) magnetic resonance imaging signals from the two networks, research physicians can readily discriminate between behavioral variant FTD and AD.
On a larger scale, consider this: As scientists have grasped the importance of imaging for FTD diagnosis, they took a page from the AD book and established the FTLD Neuroimaging Initiative (see press release), announced study leader Howard Rosen of UCSF. Funded in late 2009 by the National Institute on Aging and the National Institute of Neurological Disorders and Stroke, the project is patterned after the Alzheimer’s Disease Neuroimaging Initiative. It will study 120 FTLD patients for 18 months using positron emission tomography (PET) and structural MRI, as well as diffusion tensor imaging, which measures axon integrity and connectivity. The study will also collect CSF, blood, and urine to look for molecular biomarkers in collaboration with the University of Pennsylvania in Philadelphia. UCSF is collaborating with the Mayo Clinic in Rochester, Minnesota, to enroll and test patients. The goal is to develop standardized, shared datasets, Rosen said. He noted that studies show that the use of imaging plus a fluid biomarker can provide a 100 times more sensitive diagnosis than can cognitive tests. This is the second large neuroimaging initiative ADNI has inspired for related neurodegenerative diseases, the other being the Parkinson’s Progression Markers Initiative sponsored by the Michael J. Fox Foundation for Parkinson’s Research (see ARF related news story).
Diagnosis Goes Modern
How will imaging advances affect diagnosis? Murray Grossman of the University of Pennsylvania in Philadelphia proposed that a two-stage diagnostic process for FTLDs would be most effective. First, the clinician could distinguish FTLD from AD using imaging techniques. For example, positron emission tomography with Pittsburgh compound B (PET-PIB) detects amyloid deposits, which are florid in AD but largely absent in FTD. Grossman said that the pattern of degeneration in FTLD involves frontal and anterior temporal lobes, whereas AD strikes the medial temporal and parietal lobes, but even so, the distinction between the two diseases is not always clear on the MRI of an individual patient. To increase diagnostic sensitivity, Grossman suggested combining structural MRI with diffusion tensor imaging. If the images are analyzed with support vector machines, a type of software that recognizes patterns, patients can be reliably diagnosed as having one disease or the other (see Grossman, 2010).
Rik Vandenberghe, of University Hospitals Leuven in Belgium, said that functional MRI imaging can distinguish between people with AD and people who have FTLD with primary progressive aphasia (PPA). In early-stage AD patients, connectivity in the language circuit is preserved, he said, in contrast to people with PPA, where those circuits disconnect.
Once FTLD has been diagnosed, Grossman said, biomarkers in blood and cerebrospinal fluid (CSF) generally are the most sensitive method for identifying subtypes. In this area, tau is fairly established and TDP-43 is a newcomer. Grossman showed that CSF levels of TDP-43 at present make a poor biomarker because patient and control levels overlap. However, several other proteins may prove useful, including adrenocorticotropic hormone, Agouti-related protein, Fas, Interleukin-17, and Eotaxin-3 (see Grossman, 2010). For tracking disease progression, current data suggest that imaging methods are more sensitive than fluid biomarkers, Grossman said, although some fluid markers show potential as well. Other scientists agreed with Grossman’s findings. For example, Keith Josephs of the Rochester Mayo Clinic found no consistent differences in the patterns of atrophy between FTLD-tau and FTLD-TDP patients, suggesting that these disorders must be separated instead by fluid biomarkers.
Primary Progressive Aphasia
Primary progressive aphasia presents another diagnostic challenge. Scientists now recognize three categories of PPA, which are distinguished by their most common underlying pathology. As described by Marsel Mesulam of Northwestern University in Chicago, Illinois, in an overview talk on the disease, the categories include agrammatic PPA (FTLD-tau), semantic PPA (FTLD-TDP), and logopenic PPA, found in Alzheimer’s disease (see Bonner et al., 2010). Clinicians do not yet have reliable tests for sorting out these subtypes in living patients.
In agreement with Vandenberghe’s findings, Mesulam suggested that impaired connectivity, rather than brain atrophy, may be the culprit in primary progressive aphasia. Mesulam found that atrophied language areas retain some activity and are still functional in some patients. Mesulam also discussed his findings that people with learning disabilities seem to be more vulnerable to developing PPA (see ARF related news story). He mentioned the IMPPACT website that seeks to register PPA patients from around the world for studies and clinical trials.
For his part, Brad Dickerson, of Brigham & Women’s Hospital and Massachusetts General Hospital in Boston, described the development of a Progressive Aphasia Severity Scale (PASS) that uses both performance-based language tests and measures of cortical thickness to more accurately diagnose and track PPA. Initial results show that the scale is a valid and reliable tool, Dickerson said, and should be useful in clinical trials (see Sapolsky et al., 2010).
Behavioral Variant FTD
Continuing the theme of better diagnosis, Katya Rascovsky of the University of Pennsylvania in Philadelphia presented a report on proposed new diagnostic criteria for behavioral variant FTD. Rascovsky coordinates the International bvFTD Criteria Consortium (FTDC). The new criteria allow clinicians to distinguish between possible, probable, and definite bvFTD. They are expected to be more flexible and sensitive than the previous Neary criteria (see Neary et al., 1998). Patients need meet only three of six criteria to receive a diagnosis of possible bvFTD, as opposed to five of five features to meet the Neary standard. “The Neary criteria have served us well for many years,” said consortium member Boeve, “but we increasingly appreciate that some patients who clearly have FTD didn’t formally meet the Neary criteria.” In one comparison, the proposed criteria were able to detect possible bvFTD with 85 percent sensitivity, as opposed to 52 percent with the Neary criteria, Rascovsky said. She added that the consortium is still in the process of evaluating just how reliable and specific the new criteria truly are. For news on animal models and molecular findings in FTD, see Part 3 of this series.—Madolyn Bowman Rogers.
- Indianapolis: Frontotemporal Dementia Research Comes of Age
- Indianapolis: Dissecting the Pathways Behind Frontotemporal Dementia
- Indianapolis: Clinical Trials a Ripple, Scientists Hope for a Wave
- Network Connections: Missing Links in Neurodegeneration?
- PPMI: Parkinson's Field’s Answer to ADNI
- Primary Progressive Aphasia—Learning Disabilities Point to Early Susceptibility
- Zhou J, Greicius MD, Gennatas ED, Growdon ME, Jang JY, Rabinovici GD, Kramer JH, Weiner M, Miller BL, Seeley WW. Divergent network connectivity changes in behavioural variant frontotemporal dementia and Alzheimer's disease. Brain. 2010 May;133(Pt 5):1352-67. PubMed.
- Grossman M. Biomarkers in frontotemporal lobar degeneration. Curr Opin Neurol. 2010 Dec;23(6):643-8. PubMed.
- Bonner MF, Ash S, Grossman M. The new classification of primary progressive aphasia into semantic, logopenic, or nonfluent/agrammatic variants. Curr Neurol Neurosci Rep. 2010 Nov;10(6):484-90. PubMed.
- Sapolsky D, Bakkour A, Negreira A, Nalipinski P, Weintraub S, Mesulam MM, Caplan D, Dickerson BC. Cortical neuroanatomic correlates of symptom severity in primary progressive aphasia. Neurology. 2010 Jul 27;75(4):358-66. PubMed.
- Neary D, Snowden JS, Gustafson L, Passant U, Stuss D, Black S, Freedman M, Kertesz A, Robert PH, Albert M, Boone K, Miller BL, Cummings J, Benson DF. Frontotemporal lobar degeneration: a consensus on clinical diagnostic criteria. Neurology. 1998 Dec;51(6):1546-54. PubMed.