Heterogeneous Language Profiles in Patients with Primary Progressive Aphasia due to Alzheimer’s Disease

Abstract

Background:

The logopenic variant of Primary Progressive Aphasia (lvPPA) is associated with underlying Alzheimer’s disease (AD) pathology and characterized by impaired single word retrieval and repetition of phrases and sentences.

Objective:

We set out to study whether logopenic aphasia is indeed the prototypic language profile in PPA patients with biomarker evidence of underlying AD pathology and to correlate language profiles with cortical atrophy patterns on MRI.

Methods:

Inclusion criteria: (I) clinical diagnosis of PPA; (II) CSF profile and/or PiB-PET scan indicative for amyloid pathology; (III) availability of expert language evaluation. Based on language evaluation, patients were classified as lvPPA (fulfilling lvPPA core criteria), lvPPA extended (fulfilling core criteria plus other language disturbances), or PPA unclassifiable (not fulfilling lvPPA core criteria). Cortical atrophy patterns on MRI were visually rated and quantitative measurements of cortical thickness were performed using FreeSurfer.

Results:

We included 22 patients (age 67±7 years, 50% female, MMSE 21±6). 41% were classified as lvPPA, 36% as lvPPA extended with additional deficits in language comprehension and/or confrontation naming, and 23% as PPA unclassifiable. By both qualitative and quantitative measurements, patients with lvPPA showed mild global cortical atrophy on MRI, whereas patients with lvPPA extended showed more focal cortical atrophy, predominantly at the left tempo-parietal side. For PPA unclassifiable, qualitative measurements revealed a heterogeneous atrophy pattern.

Conclusion:

Although most patients fulfilled the lvPPA criteria, we found that their language profiles were heterogeneous. The clinical and radiological spectrum of PPA due to underlying AD pathology is broader than pure lvPPA.

Keywords

Atypical Alzheimer’s disease cerebral atrophy language disorders logopenic aphasia neurodegenerative disorder primary progressive aphasia

INTRODUCTION

The term progressive aphasia was first introduced in 1982, in a case-series describing six patients with isolated progressive language difficulties [1, 2]. Hence, the term primary progressive aphasia (PPA) refers to a slowly progressive, and initially isolated, language disorder of a neurodegenerative nature. In former classifications of PPA, the underlying pathology was leading. In the 1998 consensus criteria for diagnosing three prototypic syndromes associated with frontotemporal lobar degeneration (FTLD), the clinical diagnostic features of progressive non-fluent aphasia (PA) and Semantic Dementia (SD) were described [3]. Additionally, the logopenic variant of PPA was introduced as an independent clinical entity that is usually associated with underlying Alzheimer’s disease (AD) [4]. More recently, an integrated approach was chosen in the international PPA consensus criteria, whereby the syndrome of PPA is the starting point of classification, resulting in improved uniformity to define and type PPA [5]. This classification includes: (I) the non-fluent/agrammatic variant (nfvPPA), (II) the semantic variant (svPPA), and (III) the logopenic variant of PPA (lvPPA). In short, nfvPPA is characterized by a language pattern of predominant speech production errors, agrammatism, and effortful or halting speech; svPPA by impaired confrontation naming and single word comprehension; and lvPPA by predominant word finding difficulties in spontaneous speech and impaired repetition of sentences and phrases. The PPA prototypes are assumed to be associated with rather specific cerebral atrophy patterns, involving the left posterior fronto-insular cortex in nfvPPA, the left anterior temporal cortex in svPPA, and the left posterior perisylvian or parietal cortex in lvPPA.

A general view is that clinical classification of the PPA syndrome predicts its underlying pathology, i.e., FTLD or AD, which is important in terms of prognosis and patient management. Proof of concept was found by post-mortem studies with evidence of TDP-43 or tau pathology in patients with nfvPPA or svPPA and amyloid-β (Aβ) pathology in patients with lvPPA [6, 7], and by in vivo evidence for Aβ deposition on Pittsburgh compound (PiB)-positron emission tomography (PET) imaging [8, 9], or by cerebrospinal fluid (CSF) biomarker investigation in patients with lvPPA [10, 11]. Although the majority of lvPPA cases seems to have evidence of underlying AD pathology, several lvPPA cases have been described without biomarker evidence of AD pathology. In a recent review, the authors reported AD pathology atpost-mortem research in 60% of lvPPA cases, 60% of cases had an AD profile of biomarkers in CSF, and 90–100% of cases showed a positive PiB-PET scan [12]. The authors concluded that there exists a far from perfect correlation between the clinical diagnosis of PPA and underlying neuropathology. Several other studies have shown that the current classification criteria do not cover all PPA clinical presentations. Two prospective studies found that one third to one half of PPA patients was non-classifiable according to the current criteria [13, 14]. This has mainly been the case in patients with underlying AD pathology [13]. Most of the non-classifiable patients presented with a language pattern closely related to lvPPA. These reports support our clinical observation that the language spectrum of PPA in AD is heterogeneous.

Therefore, in this study we set out to systematically explore the language profile of PPA due to underlying AD pathology. Additionally, we investigated cortical atrophy patterns on Magnetic Resonance Imaging (MRI) in relation to the language profiles.

MATERIAL AND METHODS

Participants

We retrospectively included 22 PPA patients from the Amsterdam Dementia Cohort, VU University Medical Center (VUmc) and the memory clinic of the Erasmus University Medical Center (EMC) [15]. Patients had undergone an extensive standardized dementia assessment, including medical history, informant-based history, physical and neurological examination, routine blood and CSF laboratory tests, neuropsychological testing including Mini Mental State Examination (MMSE) [16], electroencephalography (EEG), MRI of the brain, and PET imaging for some patients. Because these patients had presented with language problems, they had undergone an examination by the speech and language pathologist or a clinical linguist. The clinical diagnosis was made in a multidisciplinary consensus meeting.

For inclusion in the present study, patients had to fulfill the following inclusion criteria: (I) in order to make the clinical diagnosis PPA, speech and language deficits had to have been the most prominent initial symptoms of disease, and although other cognitive functions could have been affected in later disease stages, language remained the most impaired domain involved from symptom onset to clinical evaluation [5]; (II) CSF profile supporting the diagnosis of underlying AD pathology, and/or availability of a PiB-PET scan positive for Aβ deposition; (III) a detailed report of examination by the speech-language pathologist, describing the language deficits and supporting the diagnosis of PPA.

Written informed consent was obtained from all participants or their legal guardian. The Medical Ethics Review Committee of the VUmc approved the study.

Language testing

For language testing, the Dutch version of the Aachen Aphasia test battery (AAT) was used, which is a standardized diagnostic language test battery for the assessment of aphasia according to type and severity [17, 18]. The AAT has subdivisions for all major language components: Spontaneous speech production, repetition, written language, naming, comprehension, and also includes the Token Test. For the present study, we focused on the testing results for the spontaneous speech production, repetition, naming, and comprehension (see Table 1).

The spontaneous speech production was evaluated at six features; communicative ability, articulation and prosody, language automatisms, semantics, phonology and syntax. The test performances were rated on a 0–5 point scale (0 = severe disturbances of spontaneous speech; 5 = normal speech). Testing for repetition, confrontation naming and language comprehension was assessed by 5 (repetition), or 4 (naming and comprehension) different subtests, each containing 10 items increasing in difficulty, which were separately evaluated on a 0–3 point scale. The total scores corresponded with severity scores ranging from 1 through 9 (1–3 = severely disturbed, 4–5 = moderately disturbed, 6–7 = mildly disturbed, 8–9 = normal). In the AAT, the confrontation naming task is considered to be relatively simple, with the total score being based on a combination of naming tasks including the naming of colors. For as color naming is a simple task, lvPPA patients in general do not fail on this task. Finally, the more widely used Progressive Aphasia Language Scale (PALS, ranging from 0 through 3 (0 = normal, 3 = severely impaired)) was applied on the AAT severity scores [8]. In two patients from the EMC, naming of objects was assessed using the Boston Naming Test (BNT) [19, 20].

Subsequently, the lvPPA core clinical criteria were retrospectively assessed based on the results of the different language tests, by which we identified three subtypes: lvPPA, lvPPA extended, and PPA unclassifiable. The patients in the two lvPPA subtypes showed deficits in the repetition of compound words and sentences, reflecting dysfunction of the phonological short term memory. All patients in the lvPPA group had word finding difficulties as reflected by the following scores on testing of the spontaneous speech production; a maximum score of 4 for communicative ability (reflecting reduced rate of speech due to word finding pauses, and some minor disturbances in content), and a maximum score of 4 for semantic structure (reflecting presence of moderate word finding difficulties). Furthermore, the lvPPA cases fulfilled the supporting clinical features, meaning that at least 3 out of 4 of the following features were present: (I) Speech (phonologic) errors in spontaneous speech and naming; (II) Spared single-word comprehension or object knowledge; (III) Spared motor speech; (IV) Absence of frank agrammatism. Consequently, if patients fulfilled features I, III and IV, they were allowed to have mild to moderate impairment in the comprehension task of the AAT (supporting feature (II), reflected by score 1 or 2 on the PALS).

Additionally, a subset of lvPPA cases showed mild to moderate deficits in word and sentence comprehension, and in naming of objects and situations, and was classified as lvPPA extended. The third subtype consisted of patients neither fulfilling the lvPPA core criteria, mainly due to absence of impaired repetition, nor the nfvPPA or svPPA criteria and were classified as PPA unclassifiable.

CSF analysis

CSF was obtained by lumbar puncture between the L3/L4 or L4/L5 intervertebral space, using a 25-gauge needle and collected in 10-ml polypropylene tubes. For all patients, CSF analyses were performed at the Neurochemistry Laboratory Department of Clinical Chemistry, VUmc [21]. Within two hours, CSF samples were centrifuged at 2100 g for 10 min at 4°C.A small amount of CSF was used for routine analysis, including total cells, total protein, and erythrocytes. Levels of CSF Aβ, tau, and ptau were determined using a commercially available sandwich enzyme-linked immunosorbent assay (ELISA) (Innogenetics, Gent, Belgium). The performance of these assays is monitored with two internal quality control pools of surplus CSF (high and low biomarker values). Inter-assay coefficients of variation (mean±SD) were 10.9±1.8% for Aβ, 9.9±2.1% for tau, and 9.1±1.8% for ptau, as analyzed in a high and low pool from 13 consecutive pool preparations of two surplus CSF specimens (one normal and one AD profile), used in total in 189–231 runs. CSF profiles were found to be supportive for underlying AD pathology, when fitting the following cut-off values: A β<550 ng/L and total tau > 375 ng/L and/or phosphorylated tau at threonine-181 (ptau) > 52 ng/L, or if Aβ< 650 ng/L and ptau > 52 ng/L [21].

MRI analysis

For neuroimaging analysis 20 MRI scans were available. In addition, one computed tomography (CT) scan had been performed due to claustrophobia, which was suited for the analysis of cortical atrophy and vascular co-morbidity [22]. As a result of the retrospective design of our study, several types of MR systems with diverging imaging protocols were used. The majority of patients had undergone a multi-sequence MRI protocol using either a 1.5 or 3.0 Tesla MR system, including the acquisition of 3 dimensional (D) T1-weighted images with multiplanar reconstructions and 2D FLAIR and T2-weighted images. For one patient, only scans made on a 1.0 Tesla MR system were available. We used visual rating and quantitative measurements to compare atrophy profiles between the left and right hemispheres and between the different language disorder subtypes. For the qualitative assessment, nine different regions of interest (see Table 4) were selected bilaterally, and visually rated by an experienced neuroradiologist, who was blinded to PPA subtype. Visual rating was performed on the coronal and sagittal planes of the T1-weighted and the axial FLAIR sequences. The severity of lobar atrophy was ranked on a 0–3 point rating scale, with 0 representing no atrophy and 3 representing severe atrophy. For the visual rating of the medial temporal lobe, the rating scale for medial temporal lobe atrophy was applied [23]. For the parietal region the posterior cortical atrophy scale, ranging from 0 to 3, was used [24]. Subsequently, cortical thickness was measured in the 14 patients having 3D T1-weighted images with sufficient image quality available, including 8 and 6 subjects for the lvPPA, and lvPPA extended groups, respectively. Due to low scan quality, the MRI scans of the patients in the PPA unclassifiable group (n = 5), as well as the scans of two patients from the lvPPA extended group were excluded for the cortical thickness analysis. The automated analysis was performed using the FreeSurfer 5.3 (http://surfer.nmr.mgh.harvard.edu/) processing stream [25]. In short, FreeSurfer uses the T1-weighted image to locate the pial and white matter surface of the cortex. The distance between these surfaces gives the vertex-wise cortical thickness (i.e., the perpendicular thickness at each location). Non-linear registration of the cortical thickness map to the ‘fsaverage_sym’ (for differences in laterality) [26] and ‘fsaverage’ (for group differences) subjects finally allows for vertex-wise statistics. Prior to running FreeSurfer, all images suitable for analysis were corrected for geometric distortions. Moreover, the resulting cortical segmentations were manually checked and re-run if errors occurred. Statistical maps of cortical thickness differences between hemispheres (i.e. the left versus right hemisphere within both the lvPPA as well as the lvPPA extended group), and between groups (lvPPA versus lvPPA extended group) were produced using vertex-wise general linear modelling. Prior to vertex-wise analysis, the cortical thickness maps were smoothed using a Gaussian kernel with an FWHM of 10 mm. Clusterwise correction for multiple comparisons was applied using Monte Carlo Z simulation while thresholding the statistical maps at p < 0.01, using 5000 iterations and setting the cluster level statistical threshold at p < 0.05.

PET analysis

PET scans were obtained using an ECAT EXACT HR+scanner (Siemens/CTI, Knoxville, TN, USA). The properties of this scanner [27] and 11 C-PiB scanning protocol have been reported elsewhere [28]. Briefly, 90-minute dynamic emission scans were started simultaneously with the IV injection of 382±28 (mean±SD) MBq 11 C-PiB. Acquired images were assessed visually and scored as either positive or negative by an experienced nuclear medicine physician.

Statistical analysis

For statistical analysis we used SPSS IBM Statistics version 20 (SPSS Inc. Chicago Ill., USA). Differences in demographic and clinical variables between clinical subtypes were assessed by the non-parametric Kruskall-Wallis test, p < 0.05. The mean results of the language tests and of the visual rating of the atrophy patterns on MRI were analyzed between subgroups by applying the Kruskall-Wallis test and the Mann-Whitney-U test, p < 0.05.

RESULTS

Demographic and clinical data are reported in Table 2. A CSF profile supporting the diagnosis of underlying AD pathology was found in 21 patients, a positive PiB-PET scan in one, and both biomarkers supporting AD in 6 patients. The three clinically defined subtypes showed no significant differences in mean gender, age, Clinical Dementia Rate (CDR), MMSE or reported duration of disease. Per patient scores of language testing are displayed in Supplementary Table 1, and per subtype mean scores are displayed in Table 3.

The lvPPA core clinical criteria, i.e. word finding difficulties in spontaneous speech and impaired repetition of compound words and sentences, were met in 17 patients (77%). Among these patients, language profiles varied considerably with half of patients (lvPPA extended, n = 8, 37% of total) also presenting mild to moderate deficits in the testing of verbal and reading comprehension of words and sentences, with predominant impairment of the latter in half of patients, and mild to moderate errors in the naming task (see Fig. 1, and Supplementary Table 2). The remaining patients (n = 5, 23%) presented with normal results on the repetition task, but with varying deficits in comprehension and naming. Since these subjects did not fulfill the PPA core criteria of any of the PPA subtypes, they were termed PPA unclassifiable.

On spontaneous speech production, all patients had some disturbances in communicative ability (see Table 3), and showed varying results for testing of the other features. Impairment in articulation and prosody, or in language automatisms, both features mainly associated with nfvPPA, were not at all or just mildly present. Likewise, the combination of severe disturbances on single-word comprehension and confrontation naming, both core features of svPPA, was absent in the majority of patients. These results indicate that the selected patients indeed showed language profiles neither fitting with nfvPPA nor withsvPPA.

In general, on MRI we found that atrophy throughout the left hemisphere was more prominent than throughout the right hemisphere using qualitative assessment (see Table 4 for mean values per subgroup, and Supplementary Table 3 for results per patient). More specifically, patients with lvPPA extended showed the most severe atrophy in the left parietal and left temporal lobe compared to both lvPPA and PPA unclassifiable. Even though the groups were quite small, the difference was borderline significant for the comparison with PPA unclassifiable for the left parietal lobe (p = 0.06), and nominally significant for the left temporal lateral lobe (p < 0.05). The PPA unclassifiable showed predominant left medial temporal atrophy, and less prominent atrophy in the left fronto-mesial lobe and the left parietal lobe. Finally, we compared cortical thickness in lvPPA and lvPPA extended. Patients with lvPPA extended showed more cortical atrophy in the left temporal pole and left insula when compared to their right counterparts (p < 0.05 after correcting for multiple comparisons), while such a laterality difference was not as clearly present in the lvPPA subgroup (see Fig. 2I). In the direct group comparison, lvPPA extended showed more extensive cortical atrophy when compared to lvPPA, with differences most pronounced in the left hemisphere (p < 0.05 after multiple comparisons correction (see Fig. 2II).

DISCUSSION

The main finding of the present study is that most PPA cases with biomarker evidence of underlying AD pathology fitted the lvPPA classification criteria, although with considerable variation in type and severity of language deficits, whereas a quarter of patients did not fulfill lvPPA core criteria at all. We showed that the type of language deficits varied between patients, in particular with regard to severity of impaired semantic processing and lexical retrieval, and the presence of phonematic and semantic paraphasias. Furthermore, the patients with more extensive language disturbances also displayed more left-sided cortical atrophy by qualitative and quantitative measurements on MRI.

Our observation that a substantial number of patients with lvPPA present with more extensive language deficits than reflected in the current core criteria is in line with two other studies. In the first of these studies, investigating 13 patients with lvPPA and CSF results available, the authors reported that two thirds of the patients had a CSF AD-profile and that lvPPA patients also presented with difficulty with phonemic sequencing and semantics [11]. In the second study, more extensive deficits in single word repetition and comprehension were associated with more widespread atrophy patterns, including left inferior and medial temporal atrophy, than usually described in the typical lvPPA cases (called cluster 2 & 3 by the authors, and resembling our lvPPA extended subgroup) [29]. An important difference with our study is that one third of their patients had no biomarker confirmation of clinical diagnosis. Specifically, in the PPA unclassifiable group language profiles were heterogeneous, not meeting the core criteria for lvPPA or of any other PPA variant. In this group, all patients had preserved repetition skills. Furthermore, one patient presented with very mild disturbances in language possibly due to a very early stage of lvPPA, and the other four patients presented with varying impairments in naming and language comprehension. When taking into account the clinical spectrum of language impairment in AD as described by Leyton and Hodges, this type of language profile would lie more closely to the atypical, amnesic AD than to the logopenic aphasia [30].

Other investigators also struggled with correct classification of PPA patients, reporting 31% as non-classifiable, and suggested broadening of lvPPA criteria in order to cover the more widespread clinical presentations of lvPPA [14, 31]. An even more profound solution for the issue of non-classifiable lvPPA patients has been proposed by investigators suggesting that all PPA patients neither fulfilling the criteria of nfvPPA nor of svPPA ought to be classified as lvPPA and as “probable Alzheimer-related PPA” when supported by Aβ biomarkers [32].

The finding of more severe left sided temporal and parietal atrophy (independent of disease duration) in lvPPA patients with more extensive language disturbances is in line with former studies [6 , 34] and has been reflected by glucose hypometabolism on FDG-PET [9]. Repetition, as one of the two core clinical criteria, depends on intact verbal short term memory and an intact phonological loop, which is maintained by the dorsal directed neural network, including the left posterior temporal gyrus. Word finding difficulties resulting from deficits in lexical retrieval, as the second clinical core criterion are associated with the left inferior parietal and the posterior superior temporal lobe areas [30 , 36]. Interestingly, in our study the patients with lvPPA extended seem to fit better the classical atrophy pattern as described in the original PPA classification criteria from 2011, than the patients in the lvPPA subgroup. Following our finding that lvPPA patients with more extensive language disturbances display more left-sided temporal atrophy, other language pathways which are anatomically and functionally connected, probably become involved as the disease progresses [36]. A statement supported by another study with 14 PPA patients with evidence of underlying AD pathology, at postmortem or by CSF profiles, reported more widespread atrophy patterns in lvPPA patients with more severe disease [37]. Although the atrophy was more widespread in both hemispheres, the atrophy remained predominant asymmetrical left-sided, leading to the conclusion of the authors that lvPPA could be considered as an unihemispheric presentation of AD. An argument against interpreting our subgroup lvPPA extended as a more advanced stage of lvPPA, is that the three subtypes did not differ in terms of disease duration measured by history, as well as by MMSE or CDR scores, although MMSE scores are not fully appropriate in patients with aphasia. Furthermore, two other studies describing clinical heterogeneity in lvPPA patients, also reported the lack of association between disease severity and duration of disease [29, 38]. On the other hand, more widespread language disturbances associated with more left-sided cerebral atrophy might be considered as one of the variations of PPA in AD, since an onset with ‘pure’ lvPPA cannot be verified.

The major strength of our study is that we tested the PPA classification in a homogeneous group of patients with biomarker proven AD pathology. To allow for a more specific typing of language disorders within the PPA spectrum and to improve prediction of underlying pathologic processes, it is crucial to match progressive language characteristics with biomarkers of disease supporting the diagnosis of underlying AD or FTLD pathology. A limitation of the study is that we could only make use of the AAT, which was originally designed for testing aphasia in patients with a cerebrovascular accident. Although the AAT is a valid, widely used, and thorough language test, for future studies it would be preferable to use aphasia tests specifically validated for application with PPA patients such as the Progressive Aphasia Severity Scale [39]. Other limitations of this study, resulting from the retrospective design, were the use of different type of MRI scanners for neuroimaging and the lack of longitudinal data. We partially solved the first problem by measuring asymmetric atrophy patterns using FreeSurfer, although we had to exclude some scans from patients of the PPA unclassifiable subgroup because of the low quality of scans. Recent insight has linked neurodevelopmental learning disabilities like dyslexia with lvPPA [40]. Unfortunately, due to the retrospective design of our study, we did not have any information regarding premorbid language abilities. Further research including longitudinal neuropsychological assessment of other cognitive domains, extensive language testing, as well as neuroimaging may help us to improve the understanding and the correct classification of PPA language variants. Additional use of biomarkers will help us in adequate prediction of underlying pathology, with its importance for prognosis and trial selection.

In conclusion, although it might be suggested by the clinical consensus criteria that lvPPA is the clinical prototype of PPA due to AD, it is not the only clinical presentation of PPA due to AD. Based on our results showing clinical heterogeneity in PPA patients with underlying AD pathology, we recommend that more often the diagnosis of underlying AD should be considered in PPA patients with atypical language presentation.

ACKNOWLEDGMENTS

Research of the VUmc Alzheimer center is part of the neurodegeneration research program of the Neuroscience Campus Amsterdam. The VUmc Alzheimer center is supported by Alzheimer Nederland and Stichting VUmc funds. The clinical database structure was developed with funding from Stichting Dioraphte. E. Louwersheimer receives research funding from Stichting Dioraphte and a travel fellowship from Alzheimer Nederland.

Authors’ disclosures available online (http://j-alz.com/manuscript-disclosures/15-0812r1).

Footnotes

Appendix

References

Mesulam

(1982) Slowly progressive aphasia without generalized dementia. Ann Neurol 11, 592–598.

Mesulam

, Rogalski

, Wieneke

, Hurley

, Geula

, Bigio

, Thompson

, Weintraub

(2014) Primary progressive aphasia and the evolving neurology of the language network. Nat Rev Neurol 10, 554–569.

Neary

, Snowden

, Freedman

, Kertesz

, Robert

, Albert

, Boone

, Miller

, Cummings

, Benson

(1998) Frontotemporal lobar degeneration: A consensus on clinical diagnostic criteria. Neurology 51, 1546–1554.

Gorno-Tempini

, Dronkers

, Rankin

, Ogar

, Phengrasamy

, Rosen

, Johnson

, Weiner

, Miller

(2004) Cognition and anatomy in three variants of primary progressive aphasia. Ann Neurol 55, 335–346.

Gorno-Tempini

, Hillis

, Weintraub

, Kertesz

, Mendez

, Cappa

, Ogar

, Rohrer

, Black

, Boeve

, Manes

, Dronkers

, Vandenberghe

, Rascovsky

, Patterson

, Miller

, Knopman

, Hodges

, Mesulam

, Grossman

(2011) Classification of primary progressive aphasia and its variants. Neurology 76, 1006–1014.

Josephs

, Whitwell

, Duffy

, Wendy

, Strand

, Hu

, Boeve

, Graff-Radford

, Lond

, Parisi

, Knopman

, Dickson

, Jr

CRJ

, Petersen

(2008) Progressive aphasia secondary to Alzheimer disease pathology: A clinicopathologic and MRI study. Neurology 70, 25–34.

Mesulam

, Weintraub

, Rogalski

, Wieneke

, Geula

, Bigio

(2014) Asymmetry and heterogeneity of Alzheimer’s and frontotemporal pathology in primary progressive aphasia. Brain 137, 1176–1192.

Leyton

, Villemagne

, Savage

, Pike

, Ballard

, Piguet

, Burrell

, Rowe

, Hodges

(2011) Subtypes of progressive aphasia: Application of the International Consensus Criteria and validation using β-amyloid imaging. Brain 134, 3030–3043.

Rabinovici

, Jagust

, Furst

, Ogar

, Racine

, Mormino

, O’Neil

, Lal

, Dronkers

, Miller

, Gorno-Tempini

(2008) Abeta amyloid and glucose metabolism in three variants of primary progressive aphasia. Ann Neurol 64, 388–401.

10.

Santangelo

, Coppi

, Ferrari

, Bernasconi

, Pinto

, Passerini

, Comi

, Magnani

(2015) Cerebrospinal fluid biomarkers can play a pivotal role in the diagnostic work up of primary progressive aphasia. J Alzheimers Dis 43, 1429–1440.

11.

Teichmann

, Kas

, Boutet

, Ferrieux

, Nogues

, Samri

, Rogan

, Dormont

, Dubois

, Migliaccio

(2013) Deciphering logopenic primary progressive aphasia: A clinical, imaging and biomarker investigation. Brain 136, 3474–3488.

12.

Harris

, Jones

(2014) Pathology in primary progressive aphasia syndromes. Curr Neurol Neurosci Rep 14, 466.

13.

Sajjadi

, Patterson

, Arnold

, Watson

, Nestor

(2012) Primary progressive aphasia: A tale of two syndromes and the rest. Neurology 78, 1670–1677.

14.

Wicklund

, Duffy

, Strand

, Machulda

, Whitwell

, Josephs

(2014) Quantitative application of the primary progressive aphasia consensus criteria. Neurology 82, 1119–1126.

15.

Van der Flier

, Pijnenburg

, Prins

, Lemstra

, Bouwman

, Teunissen

, van Berckel

, Stam

, Barkhof

, Visser

, van Egmond

, Scheltens

(2014) Optimizing patient care and research: The Amsterdam Dementia Cohort. J Alzheimers Dis 41, 313–327.

16.

Folstein

, Folstein

, McHugh

(1975) “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12, 189–198.

17.

Biniek

, Huber

, Glindemann

, Willmes

, Klumm

(1992) The Aachen Aphasia Bedside Test–criteria for validity of psychologic tests. Nervenarzt 63, 473–479.

18.

Greatz

, De Bleser

, Willmess

(1992) Akense Afasie Test. Swets & Zeitlinger, Lisse.

19.

Kaplan

, Goodglass

, Weintraub

(1983) The Boston Naming Test. Lea & Febiger, Philadelphia.

20.

Van Loon-Vervoorn

, Stumpel

(1996) De Boston Benoemingstest. een test voor woordvinding by afasie. Normering voor Nederland.

21.

Mulder

, Verwey

, van der Flier

, Bouwman

, Kok

, van Elk

, Scheltens

, Blankenstein

(2010) Amyloid-beta(1-42), total tau, and phosphorylated tau as cerebrospinal fluid biomarkers for the diagnosis of Alzheimer disease. Clin Chem 56, 248–253.

22.

Wattjes

, Henneman

, van der Flier

, de Vries

, Träber

, Geurts

, Scheltens

, Vrenken

, Barkhof

(2009) Diagnostic imaging of patients in a memory clinic: Comparison of MR imaging and 64-detector row CT. Neuroradiology 253, 174–183.

23.

Scheltens

, Launer

, Barkhof

, Weinstein

, van Gool

(1995) Visual assessment of medial temporal lobe atrophy on magnetic resonance imaging: Interobserver reliability. J Neurol 242, 557–560.

24.

Koedam

, Lehmann

, van der Flier

, Scheltens

, Pijnenburg

, Fox

, Barkhof

, Wattjes

(2011) Visual assessment of posterior atrophy development of a MRI rating scale. Eur Radiol 21, 2618–2625.

25.

Fischl

(2012) FreeSurfer. Neuroimage 62, 774–781.

26.

Greve

, Van der Haegen

, Cai

, Stufflebeam

, Sabuncu

, Fischl

, Brysbaert

(2013) A surface-based analysis of language lateralization and cortical asymmetry. J Cogn Neurosci 25, 1477–1492.

27.

Brix

, Zaers

, Adam

, Bellemann

, Ostertag

, Trojan

, Haberkorn

, Doll

, Oberdorfer

, Lorenz

(1997) Performance evaluation of a whole-body PET scanner using the NEMA protocol. National Electrical Manufacturers Association. J Nucl Med 38, 1614–1623.

28.

Ossenkoppele

, Zwan

, Tolboom

, van Assema

, Adriaanse

, Kloet

, Boellaard

, Windhorst

, Barkhof

, Lammertsma

, Scheltens

, van der Flier

, van Berckel

(2012) Amyloid burden and metabolic function in early-onset Alzheimer’s disease: Parietal lobe involvement. Brain 135, 2115–2125.

29.

Leyton

, Hodges

, McLean

, Kril

, Piguet

, Ballard

(2015) Is the logopenic-variant of primary progressive aphasia a unitary disorder?. Cortex 67, 122–133.

30.

Leyton

, Hodges

(2013) Towards a clearer definition of logopenic progressive aphasia. Curr Neurol Neurosci Rep 13, 396.

31.

Mesulam

, Weintraub

(2014) Is it time to revisit the classification guidelines for primary progressive aphasia?. Neurology 82, 1108–1109.

32.

Sajjadi

, Patterson

, Nestor

(2014) Logopenic, mixed, or Alzheimer-related aphasia?. Neurology 82, 1127–1131.

33.

Gorno-Tempini

, Brambati

, Ginex

, Ogar

, Dronkers

, Marcone

, Perani

, Garibotto

, Cappa

, Miller

(2008) The logopenic/phonological variant of primary progressive aphasia. Neurology 71, 1227–1234.

34.

Rogalski

, Cobia

, Martersteck

, Rademaker

, Wieneke

, Weintraub

, Mesulam

(2014) Asymmetry of cortical decline in subtypes of primary progressive aphasia. Neurology 83, 1184–1191.

35.

Majerus

(2013) Language repetition and short-term memory: An integrative framework. Front Hum Neurosci 7, 357.

36.

Rohrer

, Ridgway

, Crutch

, Hailstone

, Goll

, Clarkson

, Mead

, Beck

, Mummery

, Ourselin

, Warrington

, Rossor

, Warren

(2010) Progressive logopenic/phonological aphasia: Erosion of the languagenetwork. Neuroimage 49, 984–993.

37.

Rohrer

, Rossor

, Warren

(2012) Alzheimer’s pathology in primary progressive aphasia. Neurobiol Aging 33, 744–752.

38.

Machulda

, Whitwell

, Duffy

, Strand

, Dean

, Senjem

, Jack

Jr , Josephs

(2013) Identification of an atypical variant of logopenic progressive aphasia. Brain Lang 127, 139–144.

39.

Sapolsky

, Domoto-Reilly

, Dickerson

(2014) Use of the Progressive Aphasia Severity Scale (PASS) in monitoring speech and language status in PPA. Aphasiology 28, 993–1003.

40.

Miller

, Mandelli

, Rankin

, Henry

, Babiak

, Frazier

, Lobach

, Bettcher

, Wu

, Rabinovici

, Graff-Radford

, Miller

, Gorno-Tempini

ML.

(2013) Handedness and language learning disability differentially distribute in progressive aphasia variants. Brain 136, 3461–3473.