En Español | Plain Text Version | Skip Menus

You are here:  Home > News and Events > Proceedings

Frontotemporal Dementia Workshop

Skip secondary menu

THE GENETICS WORKING GROUP SUMMARY AND SUGGESTIONS FOR
FUTURE RESOURCE DEVELOPMENT AND RESEARCH INITIATIVES
FRONTOTEMPORAL DEMENTIA WORKSHOP
January 18-19, 2007 

Genetics Working Group

Michael Hutton (Chair) PhD, Mayo Clinic College of Medicine, Jacksonville FL.
John Hardy PhD, NIH/NIA
Katrina Gwinn-Hardy MD, NIH/NINDS
Christine van Broeckhoven PhD, DSc, University of Antwerp-CDE, Antwerpen Belgium
Rosa Rademakers PhD,  Mayo Clinic College of Medicine, Jacksonville FL.
Diane Murphy PhD, NIH/NINDS 

1. Incidence and Prevalence of FTD
Only a limited number of studies have been performed to estimate the prevalence and incidence rates of FTD. The incidence of FTD in Rochester, Minnesota, was 2.2 for ages 40 to 49, 3.3 for ages 50 to 59, and 8.9 for ages 60 to 691.  The prevalence of FTD was relatively high (15 per 100,000) in a 45- to 64-year-old UK population2, however, the estimates from a study performed in The Netherlands were lower with a prevalence of 3.6 per 100,000 among the 50- to 59-year-old and 9.4 per 100,000 in the 60- to 69-year-old population3.

Since the current estimates have been limited by small study populations or patient selection bias in the study design, it is clear that this research needs to be updated. Moreover, recent advances in the diagnosis and definition of FTD should allow more accurate determination of disease prevalence and incidence.  The identification of TDP-43 immunoreactive inclusions as a major feature of the histopathology of FTLD-U also means that retrospective analyses of patient brain autopsy series can be performed to provide improved data on FTD misdiagnosis rates.

2. Identification of additional genetic loci for familial and idiopathic FTD
To date, mutations in four different genes (MAPT, PGRN, VCP and CHMP2B) have been shown to cause familial forms of FTD.  In all cases the pattern of inheritance is autosomal dominant although at this point there is no clear evidence of a direct mechanistic link between any of these proteins.  Mutations in VCP and PGRN give rise to FTLD-U neuropathology with TDP-43 immunoreactive neuronal cytoplasmic and intranuclear inclusions4, 5.  In contrast, pathogenic mutations in MAPT are inevitably associated with tau histopathology of varying types6, 7.

Despite this progress over the last 10 years, it is clear that additional unidentified genetic loci exist for familial FTD.  At least one major gene for FTD and MND appears to reside on chromosome 9p8, 9.  Considerable effort is currently being focused on this locus by multiple groups as an increasing number of families are reported with evidence of linkage to this region.  The identification of the relevant disease gene on chromosome 9p is a major priority for genetic research in FTD.  However, many families with FTD, including those with an autosomal dominant pattern of inheritance, do not show evidence of mutations in the previously reported FTD genes and also do not appear to be linked to chromosome 9p.  As a result it is anticipated that additional loci for familial FTD remain to be mapped.  Creation of a funded DNA bank for FTD families would clearly assist with progress in this area (see Specific recommendations).

Little attempt has thus far been made to systematically identify common genetic risk factors within the genome for idiopathic FTD.  The only exception has been the identification of the MAPT H1/H1 genotype as a risk factor for the related tauopathies PSP and CBD10, 11. Genetic association studies in FTD case-control series have long been hampered by the clinical heterogeneity of the condition. However, the increasing size of available FTD brain autopsy series now means that through collaborative efforts large case-control (~1000 vs 1000) series can be assembled that are restricted to specific sub-types of FTD based on histopathology (ie MAPT and TDP-43-positive inclusions).  Genome-wide association studies within these pathology-confirmed FTD patient series are now an obvious priority (see Specific Recommendations).

Suggestions for future resource development and initiatives

a. Creation of a DNA bank for FTD families at the Indiana or Coriell repositories
Establishment of such a bank will require the input of a panel of experts that will determine clinical and family structure inclusion criteria.  This is similar to the model established for the NCRAD repository for AD families at Indiana University. 

Prof. Bruce Miller (UCSF) has agreed to chair the panel and will focus on clinical criteria.  Prof. Mike Hutton (Mayo Clinic) will provide advice on genetic considerations.  Finally, the inclusion of an expert in MND was recommended since a significant proportion of the banked FTD families are expected to include cases of MND.

b. Performance of a Genome Wide Association study in FTD
In addition to loci for rare forms of familial FTD, there are likely to be common genetic factors that contribute to the risk of developing idiopathic disease.  The recent identification of TDP-43 as a major component of the ubiquitin-immunoreactive inclusions in patients with FTLD-U has allowed the sub-division of FTD patients on the basis of histopathological findings (ie tauopathy and TDP-43 immunoreactive types).  As a result genetic studies in FTD would now clearly benefit from the performance of suitably powered genome-wide association (GWA) studies.  GWA studies have already been performed in the related tauopathy Progressive Supranuclear Palsy (PSP)12 and as a result a major priority should be a GWA study in patients with FTLD-U (TDP-43).  This might then be followed at a later stage by a GWA performed specifically on FTD cases with MAPT histopathology (Pick’s, CBD and others).

Virginia Lee at the University of Pennsylvania has obtained funding to perform a GWA in FTLD-U through the Children’s Hospital of Pennsylvania (CHoP).  The aim is to perform this study with 1000 cases of pathology-confirmed FTLD-U (including FTD-MND cases) using the Illumina platform and Human Hapmap300 chips.  SNP genotype data from FTD cases will be compared with publicly available genotype data from control individuals obtained using the same platform.  In order to obtain the 1000 cases a multi-site consortium is being established by Dr Lee.  Data analysis will initially be performed in collaboration with geneticists at CHoP however genotype data will be available to all members of the consortium.

3. VCP and CHMP2b mutations in FTD
Mutations in the valosin-containing protein (VCP)13 and charged multivesicular body protein 2B (CHMP2B)14 have been shown to be rare causes of familial FTD.  VCP is an AAA-type ATPase likely involved in endoplasmic reticulum-associated protein degradation while CHMP2B is a component of the endosomal sorting complex (ESCRTIII) required for the translocation and turnover of cell surface receptor complexes13, 14.  Mutations in both genes are rare 13-15, nonetheless the pathogenic mechanisms involved may well have broader significance to idiopathic FTD.   Interestingly, despite the fact that mutations in VCP are associated with a rare variant of FTD, with patients also developing inclusion body myopathy and Paget disease of bone, the neuropathology in these cases is that of FTLD-U with TDP-43 positive neuronal intracytoplasmic and abundant intranuclear inclusions5.  This implies that the mutations in VCP initiate a common neurodegenerative cascade that also occurs in idiopathic FTLD-U, which is associated with the mislocalization and aggregation of TDP-43.  As a result, additional functional studies examining the pathogenic mechanism of VCP mutations and their link to TDP-43 are warranted. 

The histopathology in cases with mutations in CHMP2B remains unclear. At this stage further studies are needed to determine if these cases also develop significant TDP-43-positive inclusion pathology or whether these mutations are associated with a different form of tau-negative histopathology. Future studies will be critical to determine the extent to which mutations in this gene can be related to idiopathic forms of FTD.

4. MAPT mutations in FTDP-17 and H1 haplotypes as a genetic risk factor for tauopathy.
More than 40 different mutations have now been identified in MAPT that cause FTDP-17 presenting with a variety of clinical and pathological phenotypes. 6, 16,17.  The mutations in MAPT can be broadly split into two types6, 17: the first group disrupt the microtubule binding properties of tau and/or directly enhance the aggregation of tau18, 19.  These mutations almost exclusively affect the C-terminal region that contains the microtubule binding domains.  The second group of MAPT mutations disrupt the alternative splicing of exon 10 and thus alter the ratio of 4R to 3R tau isoforms which is sufficient to cause pathogenesis17, 20.  All of these splicing mutations occur within exon 10 or its flanking intronic sequences. 

>The mechanism by which the exon 10 splicing mutations cause disease is still uncertain, however it is striking that all but two mutations (16 of 18) increase the splicing-in of exon 10 and thus the level of 4R tau.  Moreover, the two remaining mutations (delK280 and E10+19) that have the opposite effect, increasing 3R tau, have not been convincingly shown to segregate with the disease in any given family21, 22.  These genetic results imply that increasing 4R tau has a stronger pathogenic impact than increasing 3R tau and indeed it cannot be excluded that only mutations which increase 4R tau are actually disease-causing. Because of this uncertainty it is clear that additional studies are required to determine the relative roles of 4R and 3R tau in the development of FTDP-17(MAPT) and other tauopathies including Alzheimer’s disease. In addition to the rare, highly penetrant mutations in MAPT that cause FTDP-17, common genetic variants within MAPT also influence risk for the development of the “sporadic” tauopathies PSP and CBD10, 11.  The neuropathology of both tauopathies is characterized by the selective deposition of 4R tau isoforms in filamentous inclusions within neurons and glia.  The chromosomal region around MAPT has a highly unusual structure in that an ancient inversion event between two Low Copy Repeats (LCR) (~750KDa) has resulted in two common MAPT haplotype groups (H1 and H2) that are inverted relative to each other, in European and admixed populations23.  H1 and H2 are defined by a large number of polymorphisms (>400) that are in complete linkage disequilibrium with each other, with no recombination observed between the two haplotypes within the inverted region10, 23, 24. 

>The H1/H1 genotype has been consistently associated with an increased risk for PSP and CBD with up to 95% of patients in some series having this genotype, compared to 50-65% in European and US control populations.10, 25. Recently, several groups have further identified H1 sub-haplotypes that account at least in part for the H1/H1 MAPT association with tauopathy26, 27.  Moreover, Myers and colleagues further demonstrated that one of these haplotypes, designated H1c, appears to be associated with higher levels of MAPT RNA and especially with exon 10 containing transcripts28.  As a result the increase in risk for PSP and CBD, associated with inheritance of the MAPT H1/H1 genotype, may simply reflect higher levels, in the brain, of both total MAPT protein and selectively the 4R isoforms which would be expected to create a permissive environment for the development of these tauopathies.  It is clear that this area will require further research both to better define the genetic association between the MAPT haplotypes and tauopathy and also to confirm the pathogenic mechanism of this association.

>Suggestions for future resource development and initiatives

Stimulate research into the role of 4R vs 3R tau isoforms in FTDP-17, AD, PSP and other tauopathies:
The mechanism by which MAPT splicing mutations, which disrupt the 4R:3R isoform ratio, cause neurodegeneration has remained uncertain since they were first reported in 1998.  However, mounting evidence suggests that only mutations that increase 4R isoforms are pathogenic in FTDP-17 families. Indeed, it is already clear that increasing MAPT 4R isoforms has a more severe pathogenic impact than increasing 3R isoforms.  As a result, additional focus on the relative roles of 4R vs 3R MAPT in both FTDP-17 and other tauopathies is warranted.  It was therefore suggested that the NIA/NINDS stimulate research in this area by issuing a specific “RFA” in this area of research.

5. Null mutations in PGRN cause MAPT-negative FTDP-17 with intraneuronal TDP-43 immunoreactive inclusions
The identification of mutations in Progranulin (PGRN), in 2006, in families with FTDP-17 that lacked MAPT mutations finally resolved a longstanding question about the underlying genetic cause of the disease in this group of patients29, 30.  Indeed, all previously known cases of FTDP-17 have now been explained by mutations in either MAPT or PGRN.  The two genes are less than 2Mb apart on chromosome 17q21 however thus far there is nothing to suggest that this is anything other than a simple coincidence29, 30. 

>To date 47 mutations have been identified in PGRN that all appear to create functional null alleles29-37, most commonly through the introduction of a premature termination codon that leads to degradation of the mutant PGRN RNA through nonsense mediated decay.  As a result it is almost certain that these mutations cause FTD through haploinsufficiency, thereby reducing the level of PGRN protein, sufficient to result in an adult-onset neurodegenerative disease29, 30.  Given the effects of these mutations it is critical to understand the normal function of PGRN within the brain (See Specific Recommendation). 

PGRN is known to be a pluripotent growth factor that is involved in multiple tissue remodeling processes including wound repair, development and tumorigenesis38, 39.  However its importance for neuronal survival and/or function has previously received little attention.  It is known that PGRN is expressed within neurons and is therefore likely to have neurotrophic effects, however it is also upregulated within activated microglia as part of the brain’s response to injury38.  Regardless of the precise effect of PGRN haploinsufficiency, it is known that, in FTD patients, this eventually results in the mislocalization of TDP-43 and its accumulation into neuronal cytoplasmic and intranuclear inclusions4.  As a result the PGRN mutations appear likely to activate the same neurodegenerative cascade that is associated with TDP-43 inclusion histopathology in all forms of FTLD-U and also with a significant proportion of cases of Amyotrophic Lateral Sclerosis (ALS).  Moreover, the haploinsufficiency mechanism of the PGRN mutations implies that therapies designed to replace PGRN (e.g. by stimulating the remaining normal allele to express PGRN at higher levels) represent a clear strategy to prevent FTD onset in mutation carriers and possibly to treat the disease in affected individuals.  The fact that PGRN is upregulated within activated microglia in many neurodegenerative diseases, including Alzheimer’s disease and ALS, also implies that PGRN is likely to perform a critical role in the repair or inflammatory response to brain injury and may thus have therapeutic relevance in these other conditions38.  Future studies will examine this question by a range of approaches, most immediately it will be interesting to determine if common genetic variants within PGRN influences risk for the development of idiopathic FTLD-U as well as ALS, AD and other neurodegenerative conditions that feature upregulation of PGRN expression.

>Suggestions for future resource development and initiatives

Workshop on the biology of PGRN/TDP-43
Over the past year loss of function mutations in PGRN have been shown to cause familial FTLD-U and TDP-43 has been identified as a major component of the ubiquitin-immunoreactive neuronal inclusions, in FTLD-U and idiopathic ALS.  Although these two findings will have a major impact within the field over the next few years, the role of these proteins in neuronal survival and function is currently uncertain.  As a result a major recommendation of the Genetics sub-group was that the NIA/NINDS should finance a workshop on the biology of PGRN and TDP-43.  This meeting would bring together researchers studying the role of TDP-43 and PGRN in neurodegeneration as well those who have previously examined these proteins in other contexts. 

6. References

  1. Knopman DS, Petersen RC, Edland SD et al. The incidence of frontotemporal lobar degeneration in Rochester, Minnesota, 1990 through 1994. Neurology. 2004;62:506-508
  2. Ratnavalli E, Brayne C, Dawson K, Hodges JR. The prevalence of frontotemporal dementia. Neurology. 2002;58:1615-1621.
  3. Rosso SM, Donker Kaat L, Baks T et al. Frontotemporal dementia in The Netherlands: patient characteristics and prevalence estimates from a population-based study. Brain. 2003;126:2016-2022
  4. Neumann M, Sampathu DM, Kwong LK et al. Ubiquitinated TDP-43 in frontotemporal lobar degeneration and amyotrophic lateral sclerosis. Science. 2006;314:130-133
  5. Neumann M, Mackenzie IR, Cairns NJ et al. TDP-43 in the ubiquitin pathology of frontotemporal dementia with VCP gene mutations. J Neuropathol Exp Neurol. 2007;66:152-157
  6. Ingram EM, Spillantini MG. Tau gene mutations: dissecting the pathogenesis of FTDP-17. Trends Mol Med. 2002;8:555-562.
  7. Reed LA, Wszolek ZK, Hutton M. Phenotypic correlations in FTDP-17. Neurobiol Aging. 2001;22:89-107.
  8. Morita M, Al-Chalabi A, Andersen PM et al. A locus on chromosome 9p confers susceptibility to ALS and frontotemporal dementia. Neurology. 2006;66:839-844
  9. Vance C, Al-Chalabi A, Ruddy D et al. Familial amyotrophic lateral sclerosis with frontotemporal dementia is linked to a locus on chromosome 9p13.2-21.3. Brain. 2006;129:868-876
  10. Baker M, Litvan I, Houlden H et al. Association of an extended haplotype in the tau gene with progressive supranuclear palsy. Hum Mol Genet. 1999;8:711-715.
  11. Houlden H, Baker M, Morris HR et al. Corticobasal degeneration and progressive supranuclear palsy share a common tau haplotype. Neurology. 2001;56:1702-1706.
  12. Melquist S, Craig D, Huentelman M et al. Identification of a novel risk locus for Progressive Supranuclear Palsy by a pooled genome-wide scan of 500,288 SNPs. Am J Hum Genet. in press .
  13. Watts GD, Wymer J, Kovach MJ et al. Inclusion body myopathy associated with Paget disease of bone and frontotemporal dementia is caused by mutant valosin-containing protein. Nat Genet. 2004;36:377-381
  14. Skibinski G, Parkinson NJ, Brown JM et al. Mutations in the endosomal ESCRTIII-complex subunit CHMP2B in frontotemporal dementia. Nat Genet. 2005;37:806-808
  15. Cannon A, Baker M, Boeve B et al. CHMP2B mutations are not a common cause of frontotemporal lobar degeneration. Neurosci Lett. 2006;398:83-84
  16. Wszolek ZK, Tsuboi Y, Ghetti B et al. Frontotemporal dementia and parkinsonism linked to chromosome 17 (FTDP-17). Orphanet J Rare Dis. 2006;1:30
  17. Hutton M, Lendon CL, Rizzu P et al. Association of missense and 5'-splice-site mutations in tau with the inherited dementia FTDP-17. Nature. 1998;393:702-705.
  18. Nacharaju P, Lewis J, Easson C et al. Accelerated filament formation from tau protein with specific FTDP-17 missense mutations. FEBS Lett. 1999;447:195-199.
  19. Hong M, Zhukareva V, Vogelsberg-Ragaglia V et al. Mutation-specific functional impairments in distinct tau isoforms of hereditary FTDP-17. Science. 1998;282:1914-1917.
  20. Spillantini MG, Murrell JR, Goedert M et al. Mutation in the tau gene in familial multiple system tauopathy with presenile dementia. Proc Natl Acad Sci U S A. 1998;95:7737-7741.
  21. Stanford PM, Shepherd CE, Halliday GM et al. Mutations in the tau gene that cause an increase in three repeat tau and frontotemporal dementia. Brain. 2003;126:814-826.
  22. Rizzu P, Van Swieten JC, Joosse M et al. High prevalence of mutations in the microtubule-associated protein tau in a population study of frontotemporal dementia in the Netherlands. Am J Hum Genet. 1999;64:414-421.
  23. Stefansson H, Helgason A, Thorleifsson G et al. A common inversion under selection in Europeans. Nat Genet. 2005;37:129-137
  24. Cruts M, Rademakers R, Gijselinck I et al. Genomic architecture of human 17q21 linked to frontotemporal dementia uncovers a highly homologous family of low-copy repeats in the tau region. Hum Mol Genet. 2005;14:1753-1762
  25. Pittman AM, Myers AJ, Duckworth J et al. The structure of the tau haplotype in controls and in progressive supranuclear palsy. Hum Mol Genet. 2004;13:1267-1274
  26. Rademakers R, Melquist S, Cruts M et al. High-density SNP haplotyping suggests altered regulation of tau gene expression in progressive supranuclear palsy. Hum Mol Genet. 2005;14:3281-3292
  27. Myers AJ, Kaleem M, Marlowe L et al. The H1c haplotype at the MAPT locus is associated with Alzheimer's disease. Hum Mol Genet. 2005;14:2399-2404
  28. Myers AJ, Pittman AM, Zhao AS et al. The MAPT H1c risk haplotype is associated with increased expression of tau and especially of 4 repeat containing transcripts. Neurobiol Dis. 2007;25:561-570
  29. Baker M, Mackenzie IR, Pickering-Brown SM et al. Mutations in progranulin cause tau-negative frontotemporal dementia linked to chromosome 17. Nature. 2006;442:916-919
  30. Cruts M, Gijselinck I, van der Zee J et al. Null mutations in progranulin cause ubiquitin-positive frontotemporal dementia linked to chromosome 17q21. Nature. 2006;442:920-924
  31. Gass J, Cannon A, Mackenzie IR et al. Mutations in progranulin are a major cause of ubiquitin-positive frontotemporal lobar degeneration. Hum Mol Genet. 2006;15:2988-3001
  32. Huey ED, Grafman J, Wassermann EM et al. Characteristics of frontotemporal dementia patients with a Progranulin mutation. Ann Neurol. 2006;60:374-380
  33. Masellis M, Momeni P, Meschino W et al. Novel splicing mutation in the progranulin gene causing familial corticobasal syndrome. Brain. 2006;129:3115-3123
  34. Mukherjee O, Pastor P, Cairns NJ et al. HDDD2 is a familial frontotemporal lobar degeneration with ubiquitin-positive, tau-negative inclusions caused by a missense mutation in the signal peptide of progranulin. Ann Neurol. 2006;60:314-322
  35. Snowden JS, Pickering-Brown SM, Mackenzie IR et al. Progranulin gene mutations associated with frontotemporal dementia and progressive non-fluent aphasia. Brain. 2006;129:3091-3102
  36. Mesulam M, Johnson N, Krefft TA et al. Progranulin mutations in primary progressive aphasia: the PPA1 and PPA3 families. Arch Neurol. 2007;64:43-47
  37. Spina S, Murrell JR, Huey ED et al. Clinicopathologic features of frontotemporal dementia with Progranulin sequence variation. Neurology. 2007
  38. Ahmed Z, Mackenzie IR, Hutton ML, Dickson DW. Progranulin in frontotemporal lobar degeneration and neuroinflammation. J Neuroinflammation. 2007;4:7
  39. He Z, Bateman A. Progranulin (granulin-epithelin precursor, PC-cell-derived growth factor, acrogranin) mediates tissue repair and tumorigenesis. J Mol Med. 2003;81:600-612

THE PATHOLOGY WORKING GROUP SUMMARY AND SUGGESTIONS FOR
FUTURE RESOURCE DEVELOPMENT AND RESEARCH INITIATIVES

Dennis W. Dickson, M.D., Mayo Clinic, Jacksonville, Fl (Chair)
Bernardino Ghetti, MD, Indiana University, Indianapolis, IN
Antony R. Horton, PhD, Alzheimer’s Drug Discovery Foundation, New York, NY
Virginia M-Y Lee, PhD, University of Pennsylvania, Philadelphia, PA
Ian R. A. Mackenzie, MD, University of British Columbia, Vancouver, Canada
Manuela Neumann, MD, PhD, Ludwig-Maximilians University, Munich, Germany

Introduction
         The neuropathology of FTLD reveals considerable heterogeneity, but most cases fall into one of two broad categories.  In one category are disorders associated with neuronal and glial lesions that are composed of abnormal conformers of the microtubule associated protein tau.  In the other category are a host of disorders that do not have tau pathology greater than that which might be attributed to aging.  The disorders with tau pathology are sometimes referred to as “tauopathies.”  Tauopathies can be further subdivided based upon the predominant isoform of tau protein that accumulates within the neuronal and glial lesions – either 3R or 4R tau – reflecting the presence of three or four conserved 30-33 amino acid residues in the microtubule binding domain, generated by alternative mRNA splicing of exon 10 of the tau gene, MAPT.  Tauopathies can also be classified based upon whether or not they are linked to mutations in MAPT, which is located on chromosome 17.  Cases with MAPT mutations are referred to as frontotemporal dementia and parkinsonism linked to chromosome 17 (FTDP-17).   The most common of the 3R tauopathies is Pick’s disease, and some cases of FTDP-17 have pathology that virtually indistinguishable from Pick’s disease.  The 4R tauopathies include corticobasal degeneration, progressive supranuclear palsy and argyrophilic grain disease.  These disorders share pathologic features, but are distinct clinicopathologic disorders.

         The non-tauopathies constitute a heterogeneous group of disorders with distinct neuropathologic and molecular features.  The most common of the non-tau FTLDs is characterized by ubiquitin-immunoreactive neuronal inclusions (FTLD-U).  Recent evidence suggests that the inclusions may be composed of the DNA and RNA binding protein TDP-43.  These disorders may or may not be associated with motor neuron disease (FTLD-MND) and overlapping pathology is detected in some cases of amyotrophic lateral sclerosis.  A list of the other non-tauopathies that account for less common causes of FTLD is shown in Table 1.

Non-tau and non-TDP-43 FTLDs

Disorder

Molecular composition

Neuronal intermediate filament disease (NIFID)

neuronal intermediate filaments (α-internexin)

Multiple system atrophy with lobar atrophy

α-synuclein

Nonspecific dementia due to mutations in PRND

prion protein

Neuroserpinopathy

Neuroserpin

Neuroferritinopathy

Neuroferritin

Hereditary leukoencephalopathy with spheroids

Unknown

Given the relative rarity of most of these FTLDs, particularly those not associated with tau or TDP-43 inclusions, the Pathology Working Group felt that at this time research objectives should primarily be focused on FTLD-U and to a lesser extent the tauopathies.  Less emphasis was placed on tauopathies since a great deal is already known about clinical and pathologic features of tauopathies.  Moreover, there are several animal models of tauopathies that permit detailed analysis of pathogenesis and development of treatment strategies. 

With respect to major unmet research questions on the tauopathies, additional research is needed on FTDP-17 that specifically addresses genotype-phenotype correlations and molecular pathogenesis given knowledge of MAPT mutations and the various mechanisms of disease.  Specifically, research is needed on how relative over expression of 4R relative to 3R tau leads to disease since this likely has relevance also to much more common sporadic 4R tauopathies, such as progressive supranuclear palsy and argyrophilic grain disease.

While research on FTLD-U has increased steadily over the last decade or more, two major research discoveries have given new impetus to making this the primary focus of the proposed research objectives.  Those breakthroughs are discovery that mutations in the gene for progranulin (PGRN) are a major cause of FTLD-U and that neuronal inclusions in FTLD-U are composed of TDP-43.

The Pathology Working Group has several major recommendations:

  1. Brain banking for FTLDs
  2. Standardization of methods – brain banking and diagnosis
  3. Developing and validating research criteria for FTLD-U
  4. Molecular neuropathology of TDP-43

Before recommendations for specific research objectives related to FTLD-U are addressed, there are several recommendations of a more general nature, specifically addressing needs for promoting brain resources.

Brain Banking for FTLDs
The Pathology Working Group recommends that efforts be made to promote postmortem studies of brain pathology through several initiatives, including education about importance of autopsies and facilitation of autopsies.  The latter needs to address both logistic issues, particularly related to deaths outside of a medical care setting, and fiscal issues, such as costs for transportation, tissue harvest and tissue banking.  Many of these issues are dealt with effectively in the setting on NIA-funded Alzheimer research centers or NINDS-funded Udall centers for research on Parkinson’s disease, but there are currently no NIH-funded FTLD centers.  While not a specific recommendation of the Pathology Working Group, the possibility of NIH funded FTD centers (P50 grants) is suggested.  There are also NIH sponsored brain banks (e.g., McLean and Maryland NICHD bank) that might address some of these issues.

For other uncommon brain disorders patient advocacy groups (e.g., Society for PSP) have made efforts to promote autopsies through web-based information and caregiver symposia.  The Frontotemporal Dementia Association (FTDA) may also be enlisted to contribute to fulfillment of some of these research objectives. 

While a centralized brain bank for FTLD would be ideal, there are other means for meeting the needs of the research community.  For example, a centralized database or clearinghouse can be established that provides a network of pre-existing brain banks that currently house FTLD specimens.  A prime example of this would be the National Alzheimer Coordinating Center (NACC), which is an NIA-sponsored initiative that links all of the ADRCs and ADCCs.  A comparable initiative is currently being entertained for the Udall Centers through the Parkinson Disease Data Organizing Center.  The NACC database contains information about brain banking procedures at each of the centers as well as pathologic diagnoses using a standardized reporting method.  It currently lacks an inventory database that would readily permit an interested party from finding out easily where particular types of brain specimens (e.g., frozen tissue from FTLD-U) could be found.

The NIH funded HIV clearing house is another paradigm for networking pre-existing brain banks, where a central office coordinates donations and withdrawals from several networked brain banks.  This central clearinghouse maintains an inventory of the number and types of specimens at various brain banks and facilitates retrieval of specimens for individuals who request material.

Regardless of how it is accomplished, brain banking at either a central FTLD brain bank funded by NIH or FTDA or a distributed network of brain banks, the Pathology Working Group places great importance on this research objective since it will provide resources for addressing some of the other research objectives to follow.

Develop FTD Consortia
Given the relative rarity of FTLD-U and FTLD-MND and the related disorder ALS, a research priority not limited to pathology, is the development of multicenter consortia to study FTLD.  Such an initiative currently exists for the mid-west consortium of centers in Dallas, St. Louis and Chicago.  Other consortia of this type should be developed for other regions of the United States as well as for international sites.  The NACC has in the past sponsored a consortium to look at clinical issues related to FTD, but there have been no funded consortia for pathology. 

In Europe the European Union has established a network of brain banks (BrainNet) to address neuropathology of degenerative disease.  The initial priorities of this program were to develop standardized methods for tissue processing and diagnostic criteria for the most common neurodegenerative disorders, Alzheimer’s disease and Lewy body disease.  The methods used in BrainNet build on past multicenter consortia to develop standardized practice parameters such as the Consortium to Establish a Registry for Alzheimer’s Disease (CERAD).

Brain Banking Methods
The Pathology Working Group discussed the importance of using modern brain banking methods for collecting brain samples that would permit the greatest number and types of investigations.  It was also recognized that specialized centers may deviate from any proposed ideal practice parameters, which often involve freezing tissue from at least half the brain.  While laterality is less of an issue for relatively symmetrical disease processes (e.g., Alzheimer’s disease), many FTLD patients have focal cortical syndromes where the issue of which side of the brain is processed in a given manner is a significant issue.  To some extent these issues have been previously addressed in brain banks that are devoted to studies of ischemic-vascular cognitive impairment.  In these particular special circumstances the entire brain may be preserved for histologic studies.  Given the need for molecular studies in FTLD, this is currently not recommended.  The Pathology Working Group felt that this and other issues related to methods of brain banking for FTLD should be addressed in an NIH- or FTDA-sponsored workshop.

FTLD Diagnostic Issues
As has been mentioned previously several multicenter initiatives have addressed standardization of neuropathologic evaluation of degenerative disorders.  Examples are CERAD, EuroDem and BrainNet.  Such an initiative is needed for FTLD.  This cannot easily be accomplished in other than a formal and funded initiative for it will require commitment from a principal investigator to organize the program and it will require willing participation from a number of pathologists to develop consensus on a variety of issues.  The place to start might be a small conference (e.g., teleconference) to develop an RFA for such an initiative.  The eventual funding should be open and competitive, with a peer review process determining how to proceed.  Among the various issues that need to be addressed by this multicenter investigation are the following:

    Tissue Processing

  • Fixation – should there be standard fixation solutions and fixation times
  • Documentation – should photodocumentation be required and standardized
  • Sampling – what regions of the brain should be sampled (e.g., CERAD methods or other); should small or large (e.g., hemisphere) sections be promoted; what about tissue microarrays
  • Processing – should tissue be processed for routine paraffin embedding; cryostat; microtome, etc.
  • Staining – what routine stains are needed; can the stains be standardized across centers; should a central laboratory be involved
  • Quality control – for histologic techniques

    Diagnostic methods

  • Other pathology – should concurrent pathologies (e.g., Alzheimer, vascular, Lewy) be standardized; what about tauopathies (e.g., 3R, 4R, immunohistochemistry; silver stains)
  • What methods should be used to detect FTLD – TDP-43, ubiquitin, p62
  • Quality assurance – interrater reliability for diagnosis

Consensus FTLD-U and FTLD-MND Pathologic Criteria
The discovery of TDP-43 as a marker for FTLD-U and FTLD-MND has generated research questions that are unique to this field.  In particular, biochemical and immunohistochemical studies have suggested that there may be subtypes of FTLD-U that can be recognized by specific TDP-43 conformers or by particular patterns of TDP-43 immunoreactivity.  With respect to histologic patterns, at least two different schemes have been proposed based upon appearance of lesions in particular cortical layers with or without stipulation of involvement in the hippocampal dentate fascia (see Table).

Other schemes classify FTLD based upon extent of involvement of subcortical areas, such as the striatum.  Still other schemes separate cases based upon the presence or absence of hippocampal sclerosis.

The importance of differentiating the various subtypes of FTLD-U will stem from clinical correlates.  Specifically, it will be of value to learn if there are particular clinical phenotypes (e.g., familial or specific FTD syndrome, such as semantic dementia) that are associated with the FTLD-U subtypes.

A consensus conference is needed to address diagnostic issues.  This conference would ideally be structured and might be patterned after initiatives to develop pathologic criteria for an equally rare and pathologically heterogenous disorder, corticobasal degeneration.  In this type of conference a two phased approach has been used.  The first phase involves convening experts in neuropathology of FTLD who propose essentials for diagnosis and submit cases that have been processed optimally and meet minimal standards.  Cases are selected that include not only typical cases, but also atypical cases of FTLD-U and FTLD-MND as well as disorders that would likely be in the differential diagnosis (e.g., hippocampal sclerosis due to tauopathies or ischemic vascular disease or nonspecific dementia due to Creutzfeldt-Jakob disease).  The second phase tests the proposed criteria in a structured setting where preferably a second independent group of evaluators examines the cases, generates a diagnosis and provides information on various parameters that went into the diagnostic process.

Expanding Knowledge of the Neuropathology of FTLD-U
While much is known about the neuropathology of FTLD-U, there is still need for further basic studies on FTLD-U using a range of methods, including silver staining histology, immunohistochemistry, electron microscopy and biochemistry.  Specific issues are the detailed description of the nature and distribution of neuronal pathology in multiple brain regions, the association of FTLD-U with other pathologies (e.g., white matter pathology, neuroinflammation, extrapyramidal pathology (e.g., basal ganglia, thalamus & substantia nigra), motor neuron pathology and hippocampal sclerosis).  Neuronal intranuclear inclusions (NII) have been described in most if not all cases of FTLD-U due to PGRN mutations.  They can also be detected in nonfamilial cases of FTLD-U.  Further investigation is needed into the value of this histopathologic feature for familial disease.

Additional studies (e.g., in situ hybridization, immunohistochemistry) are needed on the normal anatomical distribution of PGRN and TDP-43 within the nervous system.  What cell types express these molecules and does this contribute to understanding the selective vulnerability of FTLD-U.

The Pathology Working Group felt that there was a particular need to explore the nature of pathologic overlap between motor neurons disorders (e.g., ALS) and FTLD-U.  This might involve including motor neuron specialists in all of the research objectives.

Hippocampal sclerosis is found in over 70% of FTLD-U cases and further studies are needed into the pathogenesis of this selective vulnerability and its clinical and neuroradiologic correlates.  It is also a priority to determine the value of this histopathologic feature in the differential diagnosis of degenerative dementia.  Preliminary evidence suggests that presence of hippocampal sclerosis is associated with TDP-43 immunoreactive lesions consistent with FTLD?U even when it is detected in the setting of other pathologic processes (e.g., Alzheimer type pathology).

TDP-43 Related Research Objectives
The Pathology Working Group felt that a related research priority was to confirm and to extend molecular biologic studies of TDP-43 in FTLD-U with a focus on post-translational modifications (e.g., ubiquitination, phosphorylation and proteolytic cleavage).  There was also a need to learn if TDP-43 is the constituent of the fibrillar structures that are detected in the inclusions of FTLD-U or if it is a non-fibrillar granular constituent.  These studies include addressing the fibrillogenic potential of TPD-43 similar to studies that were developed to study fibrillogenesis of tau protein.  Further immunoelectron microscopic studies are also warranted into the subcellular distribution of TDP-43 in FTLD-U, with particular reference to inclusion bodies.

Other research efforts are needed to understand the specificity of the biochemical signature of TDP-43 alterations and to determine how they relate to clinical subtypes and possibility of using these biochemical variants of TDP-43 as a biomarker for FTLD-U.  For example, can one measure TDP-43 in cerebrospinal fluid?

Another issue that must be addressed is whether the different modifications in TDP-43 are representative of qualitatively different disease processes or merely reflection of disease severity.  As such, it would be important to study FTLD-U with varying degrees of disease severity or disease duration.  Although difficult, developing a staging scheme for the pathology of FTLD-U should be a research objective.  This will be especially important in evaluating effectiveness of future disease modifying therapies. 

It will also be important to study if biochemical variants have any relationship to agonal or postmortem factors (cf. dephosphorylation and proteolysis of tau as a postmortem artifact).

Preliminary evidence was presented that TDP-43 may be detected in the setting of other neurodegenerative disease processes, such as Alzheimer’s disease and Guam Parkinson dementia complex.  These results need to be confirmed and validated.  There need to be additional studies and consensus about the meaning of these phenomena.  Does it represent concurrent FTDL-U in AD and Guam PDC or does it indicate that TDP-43, like synuclein, may co-deposit with tau in some brains and sometimes within the same neuron?  The essential research objective is to learn the specificity of TDP-43 for FTLD-U.  Large scale screening of brain collections for TDP-43 immunoreactivity (e.g., wide age ranges, different ethnic groups, and a range of different pathologic processes) is needed.

Should evidence exist that TDP-43 may co-deposit with tau (or other proteins) then it will be essential to determine the molecular basis for this phenomenon.

State-Of-The-Art PGRN & TDP-43 Symposium
Another important research objective is to learn more about TDP-43 and progranulin.  In the past, NIH has convened state of the art symposia where experts form various fields gather to review the current state of knowledge about a particular topic.  The Pathology Working Group felt that such an NIH sponsored symposium should be convened and be open to the wider research community (e.g., web cast) whereby the sponsored symposium would bring together experts on both TDP-43 and PGRN from fields of cancer cell biology, inflammation and immunology, neurobiology, genetics, transcription control, and molecular pathology, as well as clinicians with insight into the neurologic, psychiatric and neuropathologic aspects of FTLD-U and ALS.

Development of Other Research Resources
Currently, there are a limited number of commercial antibodies to TDP-43.  The NIH should consider supporting efforts to develop antibodies through its resources (e.g., NINDS NeuroMab Facility) that are specific to various TDP-43 forms (e.g., specific cleavage products or post-translational modifications).

As animal models become available (e.g., TDP-43 transgenic, PGRN knock-outs, and TDP-43 transgenic crossed with PGRN knock-outs), these need to be made widely available to the research community.

THE CLINICAL DIAGNOSIS AND IMAGING/BIOMARKERS GROUP SUMMARY AND SUGGESTIONS FOR FUTURE RESOURCE DEVELOPMENT AND RESEARCH INITIATIVES

Brad Boeve, MD., Mayo Clinic College of Medicine, Rochester, MN
Murray Grossman, MD., U. of Penn. School of Medicine, Philadelphia, PA.
David Knopman, MD., Mayo Clinic College of Medicine, Rochester, MN.
M. Marsel Mesulam, MD., Northwestern University, Chicago, IL.
Bruce Miller, MD., UCSF School of Medicine, San Francisco, CA.
Peter Nestor, FRACP.,U. of Cambridge, Cambridge, UK.
Martin Rossor, MD, FRCP., Dementia Research Centre Institute of Neurology, London, UK.
Gary Small, MD, UCLA Semel Institute, Los Angeles, CA.

The Clinical Diagnosis, Imaging, and Biomarkers Group considered several issues pertaining to frontotemporal dementia (FTD) at the FTD Workshop held in January 2007 in Miami, Florida. The following summary describes the issues considered to be critical for further research in FTD, followed by a list of items considered of highest priority for funding.

Nomenclature of FTD spectrum disorders: How should research proceed to refine/update the nomenclature?
      Confusion and debate continues in the nomenclature of FTD, primary progressive aphasia (PPA), semantic dementia (SD) and related syndromes (henceforth considered collectively as “FTD spectrum disorders”). Inherent in this discussion is the underlying tenet that symptomatology, and hence syndromic terminology, reflects the topography of degeneration, which does not necessarily reflect the underlying disease. In other words, frontotemporal dementia reflects the topography of frontotemporal degeneration, but does not necessarily indicate underlying Pick’s disease. Ultimately what is needed is a terminology that predicts molecular pathogenesis and response to therapy—goals that need to be pursued aggressively in the next decade.
      The diagnostic criteria for PPA have been published (1). Criteria for PPA subtypes are being developed by a working group of aphasia experts who convened a Progressive Aphasia and Semantic Dementia meeting in San Francisco in April, 2006. To summarize the initial deliberations of this group and current practice, the FTD spectrum disorders can be labeled as follows:

  • behavioral variant FTD (bvFTD)/frontal variant of FTD (fvFTD)
  • primary progressive aphasia (PPA) with three variant forms

    a) nonfluent/agrammatic PPA, also known as progressive nonfluent aphasia (PNFA)
    b) semantic PPA, also known as aphasic type of semantic dementia (SD)
    c) logopenic PPA

  • visual agnosia/prosopagnosia/nondominant hemisphere temporal variant of FTD

Unresolved questions remain, such as how should features of parkinsonism, corticobasal syndrome, PSP, spasticity, lower motor neuron dysfunction not fulfilling ALS criteria, etc, be considered in each of these syndromes?

Clinical diagnosis of FTD in patients with cognitive/behavioral changes: How should research proceed to establish or negate the diagnosis of behavioral variant FTD?
      The so-called “Neary criteria” for the behavioral variant of FTD was the first attempt to describe the clinical features and findings on ancillary testing needed for the diagnosis of FTD (2). This was a significant move forward in the dementia field, but over time weaknesses have been appreciated, such as the lack of definitions for some core features (eg, social interpersonal conduct), difficulty with operationalizing items in the criteria, absence of specific requirements for the number and types of features that must be present for the diagnosis, and the sheer number of features that must be considered (>20). A different set of criteria was proposed as research criteria for the clinical diagnosis of FTD (developed by Bruce Miller and collaborators at UCSF):

Proposed research criteria for the clinical diagnosis of behavioral variant frontotemporal dementia:

Clinical Features

  • Early (initial 2-3 yrs) behavioral disinhibition
  • Early (initial 2-3 yrs) apathy or inertia
  • Early (initial 2-3 yrs) loss of emotional reactivity/sympathy and empathy
  • Perseverative, stereotyped or compulsive/ritualistic behavior
  • Hyperorality and dietary changes
  • Neuropsychology showing an FTD neuropsychological profile 

Imaging and Genetic Features - any or all of the following

  • Imaging: frontal and/or anterior temporal atrophy on MRI, frontal and/or anterior temporal hypoperfusion on SPECT, or frontal and/or anterior temporal hypometabolism on FDG PET)
  • Genetic: presence of known causative mutation (mutation in MAPT, PGRN, VCP, etc)

Clinically probable FTD = any 4 of the 6 clinical features, or at least 2 of clinical features along with meet imaging or genetic criteria

Clinically possible FTD = any 2 of the 6 clinical features without imaging or genetic features

Abbrreviations: FDG PET=fluorodeoxyglucose positron emission tomography, MAPT=gene encoding microtubule associated protein tau, MRI=magnetic resonance imaging, PGRN=gene encoding progranulin, SPECT=single photon emission computed tomography, VCP=gene encoding valosin-containing protein

The sensitivity and specificity of these criteria could be tested in a prospective fashion to determine their sensitivity/specificity for FTD-spectrum pathology versus AD pathology. One advantage of these criteria is that they can easily be adapted to the use of questionnaires or other scales that operationalize the core features.  This allows reliably delineating each feature as present or absent. Numerous research questions could be put to the test, such as:

  1. Whether to require normal or near normal memory and visuospatial functioning and/or clearly abnormal executive functioning on neuropsychological tests to meet the neuropsychological criteria.
  2. Whether focal or asymmetric frontal and/or anterior temporal cortical atrophy on MRI is sufficient as the radiological criterion.
  3. Whether MRI, SPECT or FDG PET is most valuable
  4. Whether the absence of uptake on PET using Pittsburgh Compound B (PIB) (3) adds value to separating FTD from AD
  5. Whether the absence of uptake, or at least presence of uptake in the medial temporal lobes yet absence of uptake in the lateral temporal lobes, using the 18FDDNP tracer (4) on PET is sufficient as the radiological criterion, etc.

These are all research questions that can be addressed using clinical/pathological approaches, and refinements of the proposed research criteria will be necessary. These refinements should be data-driven rather than based on consensus or clinical experience. 

Clinical diagnosis of PPA in patients with changes in speech articulation/language: How should research proceed to establish or negate the diagnosis of PPA?

The criteria proposed by Mesulam was discussed, and all agreed that the criteria published in 2003 (1) should be maintained:

Criteria for the clinical diagnosis of PPA

  • There is an insidious onset and gradual but progressive impairment of word finding, object naming, syntax, or word comprehension manifested during conversation or assessed with the use of standard neuropsychological tests of language.
  • All major limitations in activities of daily living can be attributed to the language impairment for at least two years after onset.
  • Premorbid language function (except for developmental dyslexia) is known to be intact.
  • Prominent apathy, disinhibition, loss of memory of recent events, visuospatial impairment, visual-recognition deficits, and sensory-motor dysfunction are absent during the initial two years of illness, as indicated by the history, evaluation of activities of daily living, or neuropsychological testing, so that the patient would not fulfill diagnostic criteria for any other dementia syndrome.
  • Acalculia (inability to perform simple mathematical calculations) and ideomotor apraxia (inability to pantomime movement as instructed by an examiner) can be present even in the first two years of illness, and deficits in copying simple drawings and perseveration may also be noted, but neither visuospatial deficits nor behavioral disinhibition substantially limits activities of daily living.
  • Other cognitive functions may be affected after the first two years of illness, but language remains the most impaired function throughout the course of the illness and deteriorates faster than other affected functions.
  • Specific causes of aphasia, such as stroke or tumor, as ascertained by neuroimaging, are absent.

From Mesulam MM, New Engl J Med 2003 (1)

The core deficits for all PPA patients are word-finding, naming or spelling impairments. Single word comprehension, fluency and syntax can help subtyping using the following flowchart:

 

                                                single word comprehension

                            bad                                                                    good      

          fluency and/or syntax                                                fluency and/or syntax    

      good                          bad                                          good or                   bad
                                                                                       variable

       PPA                         PPA                                            PPA                      PPA
   Semantic                 mixed or global                               logopenic               nonfluent     
     dementia                    variant                                         variant              aphasia variant

Several other issues relating to FTD, PPA, and other syndromes within the FTD spectrum include the following:

Challenges for FTD and PPA that can be addressed through collaborative research

  • Determining the causes/mechanisms in cell death
  • Develop mechanism-based treatments rather than using agents that are available and do not have a clear mechanism for potential improvement (the bromocriptine trial in PPA was been negative; studies on galantamine and memantine are in progress)
  • Expand the role of speech therapy in PPA, such as using melodic intonation therapy (MIT) in PPA, and cognitive/behavioral rehabilitation in FTD
  • Determine risk factors and epidemiology of FTD and PPA
  • Develop biomarkers to distinguish AD vs FTLD
  • Develop biomarkers to separate FTLD-U versus FTLD-tau pathology
  • Expand the use of a PPA-specific Clinical Dementia Rating (CDR) scale
  • Develop better standardized tests
  • Patient and family support and education
  • World-wide registry for large scale studies

Opportunities

  • What are the first cells targeted in FTD and PPA?
  • Define physiological networks that determine social cognition
  • Define physiological networks underlying frontal-executive function
  • What are the molecular fingerprints of the language network?
  • What is the computational architecture of human language? 

Characterization of patients with FTD spectrum disorders: What are the minimum bedside and neuropsychological tests that should be performed for characterizing patients with cognitive/behavioral/language changes in routine neurologic practice? in academic centers?
      This is an important issue – the Progressive Aphasia and Semantic Dementia conference participants as well as the FTD Working Group of the NIA-sponsored ADC/ADRCs are addressing this.

Predicting the underlying “proteinopathy” in FTD spectrum patients: How should research proceed to determine the underlying proteinopathy in the FTD spectrum disorders?
      A critical issue for future drug trials into FTD spectrum disorders is to enroll subjects who have the underlying disorder for which the therapy has been identified or developed. While many disorders can underlie FTD and PPA, they generally fall into two classes: the “tauopathies” [Pick’s disease, corticobasal degeneration (CBD), progressive supranuclear palsy (PSP), argyrophilic grain disease (AGD), multisystem tauopathy (MST), or FTDP-17 due to mutations in MAPT] and the “ubiquitinopathies” [frontotemporal lobar degeneration with ubiquitin-positive inclusions (FTLD-U), frontotemporal dementia with motor neuron disease (FTD-MND), and FTLD-U with neuronal intranuclear inclusions (NII) due to mutations in PGRN]. Now with the identification of TAR DNA binding protein 43 (TDP-43) as a protein that is ubiquitinated in FTLD-U, FTD-MND, and FTLD-U with NII (5), the ubiquitinopathies could more specifically be called “TDPopathies” or “tardopathies.” The tauopathies and tardopathies comprise >80% of the disorders that underlie FTD-spectrum syndromes, with the remaining cases being almost entirely Alzheimer’s disease with neuritic plaques, neurofibrillary tangles, and cortical degeneration occurring in a focal/asymmetric frontotemporal pattern of topography rather than the more usual bilateral mesiotemporoparietal > frontal distribution.
      It is still unknown whether a treatment for FTLD-U will have equal efficacy for FTLD-tau.  However, if a drug that affects tau or progranulin or TDP-43 pathophysiology is developed, it will be crucial to identify those patients most likely to benefit from the drug (presuming any drug of interest would affect one protein and not multiple proteins, although the latter scenario would make antemortem proteinopathy determination far less critical). A high priority research question is determining which single or combination of features/test findings are most sensitive and specific for identifying a tauopathy vs a tardopathy vs an amyloidopathy vs another disease in the setting of an FTD spectrum disorder (or any dementia syndrome), considering the following and possibly other features:

  • Clinical features
  • Neurobehavioral features
  • Neuropsychological testing
  • Blood, genetic (eg, haplotype or polymorphism)
  • CSF
  • MRI
  • SPECT
  • PET (PIB, FDDNP, other)

Yet some clues have already emerged as hopeful candidates – some of the pertinent published and unpublished data are summarized below: 

Clinical syndrome +/- imaging predicting the underlying proteinopathy
      The combination of the clinical syndrome plus the findings on MRI, SPECT, or FDG-PET are predictive of pathology, to an extent. In fluent PPA/semantic dementia, the largest clinical series published to date involving 18 subjects with antemortem and postmortem data (6) revealed 13 had FTLD-U pathology, 3 had Pick’s disease pathology, and 2 had AD pathology. Thus, based on this series, ubiquitin-positive inclusion pathology is by far the most common underlying proteinopathy followed by a tauopathy – specifically the 3 repeat tauopathy of Pick’s disease. Yet the authors emphasize that the 2 AD cases failed to have the typical focal or asymmetric anterior temporal lobe atrophy on imaging, or were otherwise slightly atypical in other ways. It appears that the amyloidopathy of AD is rare in the fluent PPA/SD syndrome.
      In some patients with behavioral variant FTD, atrophy on MRI is minimal. Functional neuroimaging studies tend to show frontal (particularly right frontal) hypoperfusion on SPECT (7) and hypometabolism on FDG-PET (8) in the FTD syndrome.
      In those with progressive nonfluent aphasia (PNFA), particularly those who have subtle apraxia of speech followed by PNFA features, CBD and PSP was the most frequently encountered pathology in one series (9).
      Also in the PNFA variant of PPA, voxel-based morphometry (VBM) on MRI and FDG-PET shows focal abnormalities in the anterior insula (10).  If FDG-PET shows focal insula or even more widespread frontotemporal hypometabolism plus hypometabolism in the parietal region, the pathology is more likely to be AD compared to a non-AD disorder; if no parietal hypometabolism is present, a non-AD disorder is more likely (10).
      To summarize, FTLD-U pathology tends to underlie fluent PPA/SD, FTLD-tau pathology tends to underlie PNFA, and approximately equal proportions of FTLD-U and FTLD-tau pathologies tend to underlie behavioral variant FTD.

PET imaging with protein radiotracers predicting the underlying proteinopathy
      Two radiotracers (3, 4)have been studied most rigorously in aged individuals using PET imaging, with most individuals being normal controls or patients with amnestic mild cognitive impairment (MCI) or probable AD (4). The PIB agent primarily binds to amyloid plaques (3), while the FNNDP agent binds amyloid plaques, neurofibrillary tangles, as well as other amyloid-like structures (eg, prion protein plaques) (4). FDDNP labeling shows high positive predictive value for birefringence in senile plaques and NFTs in AD, prion plaques and amyloid deposits in cerebral amyloid angiopathy.  FDDNP labeled structures have not been observed in Pick’s disease, progressive supranuclear palsy, multiple system atrophy, or cerebral hypertensive vascular (11). Each agent could be useful in the characterization of patients with FTD and PPA syndromes, at least to the extent of excluding underlying AD pathology. The absence of uptake of PIB in FTD/PPA patients would argue against AD, particularly if the patients are older (70 or greater). The absence of uptake of FNNDP in the setting of FTD/PPA would likely argue against some tauopathies (eg, Pick’s disease) but not all of them (eg, PSP). There is no data yet using FNNDP in multiple cases of CBD, AGD, FTDP associated with MAPT mutations, nor in ubiquitin/TDP-43 positive inclusion disorders such as FTLD-U, FTD-MND, and FTLD-U with NII. Further use of PIB and FNNDP, and other protein radiotracers as they are developed, in the disorders that underlie the FTD/PPA syndromes, Lewy body disorders, etc., are clearly needed. 

Clinical syndromes and neuropsychologic tests predicting the underlying proteinopathy
      Several interesting differences are being revealed by analyzing antemortem data in those who exhibited an FTD syndrome during life depending on whether they had an underlying tauopathy (tau+), a ubiquitinopathy/tardopathy or DLDH pathology (tau-), or so-called “frontal variant Alzheimer’s disease” (fvAD).  The rate of neurologic decline is faster and survival is shorter in tau+ cases compared to tau- cases (12, 13). Differential performances on neuropsychological measures have also been found.  Cases with fvAD pathology perform less well on delayed recall measures compared to the tau+ and tau- cases.  This is correlated with medial temporal atrophy on MRI studies and with the density of amyloid pathology burden in the mesial temporal lobe of fvAD cases.  By comparison, performance on confrontation naming (Boston Naming Test) is poorer in the tau- cases, and scores of BNT decline more rapidly in tau- and AD cases than tau+ cases.  This may reflect greater inferolateral temporal lobe pathology in the tau- and fvAD cases.  Tau- cases also have greater social comportment problems, perhaps reflecting right frontal and temporal disease (14).  Performance on constructional praxis and some executive measures is worse, and the frequency of extrapyramidal signs is higher, in the tau+ cases (14).  Figure construction difficulty correlates with frontal cortical atrophy on MRI and with the density of tau+ pathology in frontal cortex (14).

Clinical syndromes and CSF protein levels predicting the underlying proteinopathy
      Comparing CSF levels of total tau, phospho-tau, and beta-amyloid among the varying clinical syndromes (some with pathologic characterization) shows lower CSF total tau levels in the presumed and path-proven tau+ cases (15). Thus, lower CSF tau may reflect greater amounts of deposited tau in brain. 
      All of the data presented above – much of which is yet unpublished or only published over the past few years – underscore the value of comprehensive clinical, neuropsychological, neuroimaging, and laboratory studies being performed longitudinally, with eventual neuropathologic characterization, in order to tease out the most useful predictors of the underlying proteinopathy in FTD spectrum patients. These studies are laborious for patients, caregivers, and investigators; are very expensive; are highly reliant on the volunteer enthusiasm of patients/families and their willingness to plan for autopsy; and are also reliant on the solid infrastructure and continued funding of academic programs interested in FTD. Yet, the development of eventual disease-altering therapies cannot proceed in the clinical arena without all of these groups of individuals working in concert.

Longitudinal assessment for determining the efficacy/tolerability/safety of drugs in symptomatic FTD spectrum patients: How should research proceed to test which clinical, neurobehavioral, neuropsychological, blood, CSF, and imaging measures are most useful for the longitudinal assessment of FTD to determine the efficacy of drugs?

      Two classes of therapies can be considered here: 1) those that affect the underlying biology of the disorder and hence could slow down or halt the rate of progression of disease in already symptomatic FTD spectrum patients (ie, pathophysiology-altering therapies, with examples being kinase inhibitors, tau stabilizers, agents that increase progranulin secretion/production), and 2) those that affect the problem symptoms/behaviors of the disease but may not necessarily affect the rate of progression of the underlying disease process (ie, symptomatic therapies, with examples being memantine, atypical neuroleptics, mood stabilizers, anticonvulsants). Some of the issues pertaining to treatment trial design are different between these classes of potential therapies, although obviously identifying and testing pathophysiology-altering therapies is of paramount importance.
      With the advancements in the basic sciences of understanding the pathophysiology of the tauopathies, and likely the same will occur with progranulin/TDP-43, etc., it would be highly unfortunate if a drug that could impact the pathophysiology of any of these processes of neurodegeneration were identified, but the stage was not set for therapeutic trials to commence. Preliminary data from the NIA-sponsored “FTD Instrument Study: A Basis for Clinical Trials” (RO1 AG23195; PI: David Knopman) protocol, were reviewed and can be summarized as follows:
      In order to design clinical trials in the FTD spectrum disorders, knowledge of the progression of the disease must be determined in order to estimate power and choose optimal outcome measures. The objective of the FTD Instrument Study was to conduct a multicenter, one year replica of a clinical trial in patients with one of 4 FTLD syndromes, behavioral variant frontotemporal dementia (bvFTD), progressive nonfluent aphasia (PNFA), progressive anomic aphasia (PAA) and semantic dementia (SD). Subjects with one of these 4 syndromes were recruited from 5 academic medical centers over a 2 year period. Standard diagnostic criteria were used that were operationalized for clinical trial usage. In addition to clinical inclusion and exclusion criteria, subjects were required to exhibit focal frontal, temporal or insular brain atrophy or dysfunction by neuroimaging. Subjects underwent a neuropsychological, functional, behavioral, neurological and MR imaging assessment at baseline and 12 months later. Potential outcome measures were examined for their rates of floor and ceiling values at baseline and end of study, their mean changes and variances. As of October 27, 2006, 102 subjects underwent baseline assessment and 55 have completed the 12 month assessment. Two global measures, the FTD-modified CDR and the CGIC demonstrated change in the majority of subjects. Several cognitive measures showed negligible floor or ceiling scores either at baseline or follow-up, as well as showing decline in the majority of subjects. Functional instruments including the FBI and FAQ also might be used as outcome measures. However, the FBI’s use was complicated by apparent amelioration of behavioral disturbances in a sizable number of subjects. The FAQ was subject to ceiling effects at baseline in a number of subjects. These findings suggest that it is feasible to conduct clinical trials in FTLD, and there are several candidate outcome measures – both global and cognitive – that could be used across the spectrum of FTLD.
      For symptomatic therapies, one would desire scales that measure neuropsychiatric burden, or executive dysfunction burden, or aphasia burden, etc. Assessing the quality of life of patients and caregivers, and costs associated with disease, are also important. Some potentially viable scales exist (eg, Neuropsychiatric Inventory or NPI, modified Clinical Dementia Rating scale or CDR for PPA. The PPA-CDR (Johnson, Weintraub et al, unpublished data) has a better correlation than the standard CDR with the daily living activity scale of PPA patients and may therefore be more useful for drug studies. Yet few drug studies have been conducted in FTD/PPA patients(16). One could argue that the neuropsychiatric burden for caregivers and families is particularly challenging, and while most clinicians who specialize in the diagnosis of management of FTD patients have vast experience with agents that tend to ameliorate problems and those which tend not to help, there are no FDA-approved indications for any drug in FTD. Plus, the recent negative publications and press on the atypical neuroleptics in Alzheimer’s disease (17, 18) have dried the ink in the script pen for this class of drugs in FTD. While it is clearly important to recognize agents with no efficacy as well as those with increased morbidity and mortality, in the era of evidence-based medicine, properly constructed studies focused on FTD patients are needed.

Identification of at-risk, presymptomatic, and early symptomatic (ie, prodromal FTD) FTD spectrum patients: How should
research proceed to identify at-risk FTD spectrum individuals who would be interested in research participation?

Those individuals who are genealogically at-risk but asymptomatic are a critical population to characterize longitudinally. Thus, the younger generations and asymptomatic members of MAPT mutation families, PGRN mutation families, VCP mutation families, and familial FTD-ALS families are strongly encouraged to participate in research. The Association for Frontotemporal Dementias (AFTD) could assist with these research efforts in increasing interest and access to those who link to the AFTD website to centers involved in FTD research.

How should research proceed to identify presymptomatic FTD spectrum individuals in the population (ie, mass screening of possible FTD spectrum patients)? 

There are significant challenges in mass screening for FTD - a significant minority of the human population have mild executive and/or language dysfunction, are disinhibited, engage in risk-taking behaviors, lack empathy/sympathy, have dyslexia and other learning disabilities, and hence differentiating such individuals in the “normal population” from those with early disease is problematic. This is worthy of further research, with the challenge being to ensure that any executive, language, or behavioral symptom represents a significant change from the individual’s previous state and level of functioning. 

How should research proceed to identify the “MCI of FTD spectrum disorders”?

In most individuals with an evolving degenerative dementia, one does not traverse from a neurologically normal to unequivocally abnormal state over days or weeks, but rather a transitional phase between normal cognitive functioning and a dementia syndrome is passed through over months or years. The syndrome of mild cognitive impairment (MCI), particularly amnestic MCI, has clearly proven useful in the characterization of individuals at risk for developing AD (19, 20), and a similar prodromal state has been postulated to occur in DLB (21) and perhaps other dementia syndromes. The “MCI of FTD” is more problematic, since some individuals exhibit mild executive dysfunction but no language dysfunction or behavioral changes, while others exhibit language dysfunction but no changes in behavior or executive functioning, and still others exhibit behavioral changes without abnormalities in cognitive functioning. Some patients with early and mild FTD perform in the normal range on all standard neuropsychological tests, including those considered sensitive to executive dysfunction (eg, Wisconsin Card Sorting Test, Stoop, Trailmaking Test, Digit Symbol, Frontal Assessment Battery, etc). Recent data suggests deficits in social cognition and theory of mind tasks may be most sensitive in very early FTD (22, 23) Subtle apraxia of speech and nonverbal oral apraxia tends to precede nonfluent aphasia in the PNFA syndrome (9), and poor confrontation naming (eg, low score on the BNT) is an early feature of semantic dementia. The early features of right anterior temporal lobe dysfunction may be object and/or facial angosia (prosopagnosia) (24). Subtle limb apraxia has been the heralding sign in very early CBS. Thus, the term “prodromal FTD” could be considered more fitting to capture the essence of very early FTD spectrum disorders better than MCI per se. Detailed assessment of a sufficient number of patients with very early FTD spectrum disorders will be needed to adequately characterize prodromal FTD.

Longitudinal assessment for determining the efficacy of drugs in asymptomatic subjects at risk for FTD?: How should research proceed to characterize the natural history of presymptomatic FTD so that drugs can be tested for delaying the onset or preventing FTD?

The ultimate goal in any neurodegenerative disease is to prevent the development, or at least delay the onset, of symptoms/signs in those who are at risk. In order to design prevention or disease-delaying trials, the natural history must be characterized so that a treatment effect can be demonstrated. This will be a challenge in neurodegenerative disease, since phenotypic variability is the norm rather than the rule, and the age of onset even within families with the same pathogenic mutation is wide.
      Therefore, one line of research should address which single or combination of features/test findings change in a consistent/linear fashion in asymptomatic at-risk individuals:

  • Clinical features
  • Neurobehavioral fe