Rates of Amyloid Imaging Positivity in Patients With Primary Progressive Aphasia

Abstract

The ability to predict the pathology underlying different neurodegenerative syndromes is of critical importance owing to the advent of molecule-specific therapies.OBJECTIVE To determine the rates of positron emission tomography (PET) amyloid positivity in the main clinical variants of primary progressive aphasia (PPA). DESIGN, SETTING, AND PARTICIPANTSThis prospective clinical-pathologic case series was conducted at a tertiary research clinic specialized in cognitive disorders. Patients were evaluated as part of a prospective, longitudinal research study between January 2002 and December 2015. Inclusion criteria included clinical diagnosis of PPA; availability of complete speech, language, and cognitive testing; magnetic resonance imaging performed within 6 months of the cognitive evaluation; and PET carbon 11-labeled Pittsburgh Compound-B or florbetapir F 18 brain scan results. Of 109 patients referred for evaluation of language symptoms who underwent amyloid brain imaging, 3 were excluded because of incomplete language evaluations, 5 for absence of significant aphasia, and 12 for presenting with significant initial symptoms outside of the language domain, leaving a cohort of 89 patients with PPA. MAIN OUTCOMES AND MEASURESClinical, cognitive, neuroimaging, and pathology results. RESULTS Twenty-eight cases were classified as imaging-supported semantic variant PPA (11 women [39.3%]; mean [SD] age, 64 [7] years), 31 nonfluent/agrammatic variant PPA (22 women [71.0%]; mean [SD] age, 68 [7] years), 26 logopenic variant PPA (17 women [65.4%]; mean [SD] age, 63 [8] years), and 4 mixed PPA cases. Twenty-four of 28 patients with semantic variant PPA (86%) and 28 of 31 patients with nonfluent/agrammatic variant PPA (90%) had negative amyloid PET scan results, while 25 of 26 patients with logopenic variant PPA (96%) and 3 of 4 mixed PPA cases (75%) had positive scan results. The amyloid positive semantic variant PPA and nonfluent/agrammatic variant PPA cases with available autopsy data (2 of 4 and 2 of 3, respectively) all had a primary frontotemporal lobar degeneration and secondary Alzheimer disease pathologic diagnoses, whereas autopsy of 2 patients with amyloid PET-positive logopenic variant PPA confirmed Alzheimer disease. One mixed PPA patient with a negative amyloid PET scan had Pick disease at autopsy. CONCLUSIONS AND RELEVANCE Primary progressive aphasia variant diagnosis according to the current classification scheme is associated with Alzheimer disease biomarker status, with the logopenic variant being associated with carbon 11-labeled Pittsburgh Compound-B positivity in more than 95% of cases. Furthermore, in the presence of a clinical syndrome highly predictive of frontotemporal lobar degeneration pathology, biomarker positivity for Alzheimer disease may be associated more with mixed pathology rather than primary Alzheimer disease.

Prevalence of amyloid‐β pathology in distinct variants of primary progressive aphasia

“…Likewise, some patients who fit the semantic or agrammatic classification also fulfilled criteria for lvPPA. The prevalence of mixed/unclassifiable phenotypes is consistent with most large cohorts published to date, 19,22–24,33,38 but lower than others. 64–67 The heterogeneous pathologies found at autopsy of PPA-M/U subjects suggest that the ambiguities in the current PPA classification cannot be easily resolved through the addition of a fourth clinical variant.…”

“…19–26,38 This meta-analysis using individual participant data from 36 centers showed that Aβ pathology was present in the vast majority of patients with lvPPA (86%) and in a minority of nfvPPA (20%) and svPPA (16%) cases. Furthermore, autopsy data from 357 PPA patients revealed that AD pathology was the major driving force in lvPPA (76%), whereas FTLD TDP-43 (80%, mostly type C) and FTLD tau (64%) pathology where most prevalent in svPPA and nfvPPA, respectively.…”

“…In previous individual studies, the prevalence of Aβ pathology detected using in vivo biomarkers or neuropathological examination in patients with distinct variants of PPA differed widely: 57-100% in lvPPA, 0-46% in nfvPPA, and 0-33% in svPPA. [19][20][21][22][23][24][25][26]38 This meta-analysis using individual participant data from 36 centers showed that Aβ pathology was present in the vast majority of patients with lvPPA (86%) and in a minority of nfvPPA (20%) and svPPA (16%) cases. Furthermore, autopsy data from 357 PPA patients revealed that AD pathology was the major driving force in lvPPA (76%), whereas FTLD TDP-43 (80%, mostly type C) and FTLD tau (64%) pathology where most prevalent in svPPA and nfvPPA, respectively.…”

“…Uniformly applied, research-level phenotyping of PPA patients would likely have resulted in stronger clinicopathological correlation. 19,38 Nevertheless, our study was able to assess the current classification system across a diverse spectrum of clinical settings. Third, acquisition and interpretation methods for PET and CSF were not harmonized across cohorts (eg, different PET tracers, CSF analytical steps, neuropathological procedures).…”

Quantitative assessment of grammar in amyloid-negative logopenic aphasia

“…Because of this, lvPPA is often considered an atypical AD variant. However, some patients do not show Aβ deposition on PET scanning and are thus believed to have underlying frontotemporal lobar degeneration, likely with TDP-43 pathology, (Josephs et al, 2014; Matías-Guiu et al, 2015; M. M. Mesulam et al, 2014; Santos-Santos et al, 2018; Spinelli et al, 2017; Whitwell, Jones, et al, 2015). …”