This page is an English translation of the Korean original. If there is any discrepancy, the Korean version prevails.
Synonyms and related terms: Logopenic PPA, left temporal-parietal Alzheimer disease, lvPPA
Logopenic variant primary progressive aphasia (lvPPA) was first defined in 2004 when María Luisa Gorno-Tempini et al., in the Gorno-Tempini classification system, described a progressive language decline subtype whose main features are impaired naming and impaired sentence repetition. The term "logopenic" is a coinage reflecting slowed word retrieval and pauses in speech, derived from the Greek logos (speech) and penic (poor). In the 2011 international PPA classification criteria (Gorno-Tempini et al., 2011), it was formally defined as the third PPA variant, together with svPPA and nfvPPA.
The most important feature of lvPPA is that, unlike the other PPA variants (FTLD-based), the underlying neuropathology has been shown to be Alzheimer pathology (coexisting amyloid and tau) in about 60-80% of cases. This arises from the pattern in which lvPPA, an atypical language-predominant variant of Alzheimer disease (atypical/language-predominant AD), damages the left temporoparietal association cortex before the hippocampus. Therefore, although lvPPA is clinically classified as PPA, it is clinically important to understand that, on a biomarker basis, it is atypical AD belonging to the Alzheimer continuum.
The underlying neuropathology of lvPPA is Alzheimer pathology (coexisting Aβ plaques and tau NFTs) in about 60-80%, and in the remainder FTLD-TDP type A, FTLD-tau, or other pathologies are reported. In lvPPA based on Alzheimer pathology, tau NFTs begin early in the left temporoparietal association cortex (left angular gyrus, supramarginal gyrus, posterior superior temporal gyrus), unlike the early involvement of the hippocampus and entorhinal cortex in typical Alzheimer disease, and the hippocampus is relatively preserved. This atypical tau topography determines that language impairment, rather than memory impairment, is the initial symptom.
In lvPPA based on FTLD-TDP type A, there are cases associated with GRN variants, and it can be differentiated from Alzheimer pathology by tau PET and CSF biomarkers. When Alzheimer pathology is the basis, it is confirmed by positive amyloid PET, decreased CSF Aβ42, and elevated p-tau, whereas FTLD-based cases have negative amyloid PET and normal CSF Aβ42.
In lvPPA based on Alzheimer pathology, the frequency of the APOE ε4 allele tends to be reported as lower than in typical Alzheimer dementia, but most cases are sporadic. PSEN1 variants may rarely present as a language-predominant atypical Alzheimer phenotype. As a familial cause of FTLD-based lvPPA, GRN variants are reported in some. MAPT variants and C9orf72 repeat expansion are also rarely reported in lvPPA.
Genetic testing is performed when there is a family history or the age of onset is unusually young, and the AD gene panel (PSEN1, PSEN2, APP) and the FTD gene panel (GRN, MAPT, C9orf72) are carried out stepwise according to clinical judgment. Plasma biomarkers (p-tau217, Aβ42/Aβ40) are increasingly used for antemortem differentiation of Alzheimer-based lvPPA and FTLD-based lvPPA.
lvPPA accounts for about 30-40% of all PPA, and some studies report it as the most common of the three PPA variants. The age of onset is usually in the 50s-70s, and cases based on Alzheimer pathology can occur across a wider range of ages. The sex distribution tends to be slightly higher in women, but most studies show no statistically significant difference. The population-based prevalence of all PPA is about 5-10 per 100,000, and lvPPA accounts for a considerable proportion of this.
Because a considerable proportion of lvPPA is based on Alzheimer pathology, early identification of Alzheimer pathology in patients presenting with language impairment symptoms is becoming clinically important for judging the indication for anti-amyloid therapies (lecanemab, donanemab). Diagnostic delay is common in Alzheimer-based lvPPA, and the possibility of Alzheimer disease is often not considered at the onset of language impairment.
The core pathological site in lvPPA based on Alzheimer pathology is the left temporoparietal association cortex—particularly the left angular gyrus (BA 39), supramarginal gyrus (BA 40), posterior superior temporal gyrus, and the area around Wernicke's area. This region is central to the phonological buffer, responsible for storage and manipulation of phonological working memory, and to sentence-level processing, and early concentration of Alzheimer tau NFTs here causes non-fluent speech and impaired sentence comprehension and repetition.
In Braak tau staging, typical Alzheimer disease starts in the entorhinal cortex and progresses to the hippocampus and then the association cortex, but in lvPPA the left temporoparietal association cortex is involved early and intensively, showing an atypical Braak pattern with relatively mild hippocampal atrophy. Amyloid deposition appears widely, as in typical AD, and a mechanism has been proposed in which temporoparietal amyloid deposition promotes local tau pathology. As a result, episodic memory is relatively preserved early on, and phonological processing and language function are selectively impaired early.
The core clinical symptoms of lvPPA are word-finding difficulty and sentence repetition impairment. Word-finding difficulty appears as hesitation during spontaneous speech or speech interruption because words do not come to mind, and a decline is confirmed on confrontation naming. Semantic knowledge is relatively preserved early, so impairment of word comprehension is mild.
Impaired sentence repetition is a specific finding of lvPPA, with marked errors on repetition of long sentences or complex syntax. This reflects reduced buffer capacity of phonological working memory, and short-term phonological memory (digit span, sentence span) is decreased. Apraxia of speech and agrammatism are absent or mild early on, which distinguishes it from nfvPPA, and loss of word meaning is also absent early on, which distinguishes it from svPPA. Speech is fluent, but frequent pauses while searching for words are characteristic.
As the disease progresses, decline across language function accelerates, and in Alzheimer-based cases, episodic memory decline, visuospatial decline, and executive function decline are added, converting to a typical Alzheimer dementia pattern. Behavioral symptoms tend to be relatively mild.
On neurological examination in early lvPPA, speech fluency is maintained, but pauses during word retrieval are prominent and errors are found on naming. A decline appears on the Boston Naming Test (BNT), and a pattern of some improvement when a phonemic cue is given contrasts with svPPA, in which semantic cues are ineffective. On sentence repetition testing, errors on repeating long sentences appear specifically, while repetition of monosyllabic words or short sentences is relatively preserved.
Auditory language comprehension is preserved early at the single-word level, but declines appear in comprehension of complex sentences, especially in sentence structures with a heavy phonological working memory load. A marked decline is confirmed on short-term phonological memory tests (digit span forward, word span). Speech motor function (articulation) is normal early on, and the absence of apraxia of speech or grammatical errors is the key differentiating point from nfvPPA. On MRI, atrophy of the left temporoparietal cortex, especially of the angular gyrus and posterior superior temporal gyrus, is characteristic.
For screening of lvPPA, language testing including naming, sentence repetition, and short-term phonological memory is key. The naming test (BNT) shows a decline, and checking for improvement after a phonemic cue helps in differential diagnosis. Sentence repetition test: errors in repeating multisyllabic sentences such as "the dog chased the cat through the woods" are a specific finding of lvPPA. Short-term phonological memory (digit span, word span) is decreased, which is an indicator of phonological buffer impairment.
A decline may appear in the MoCA language subitems (naming, sentence repetition), but for the specificity of lvPPA, standardized PPA language assessment batteries (WAB, BDAE, AQ test) are recommended. Amyloid PET is essential for identifying Alzheimer pathology-based lvPPA, and CSF Aβ42/p-tau biomarkers are also useful. Plasma p-tau217 shows a high positive predictive value for Alzheimer-based lvPPA, and its clinical adoption as a screening test is spreading. Confirming a left temporoparietal atrophy pattern on MRI is important for supporting the diagnosis.
The diagnosis of lvPPA follows the 2011 international PPA classification criteria. For the core diagnostic criteria, both of the following must be met: ① impaired word retrieval in spontaneous speech and naming tests, and ② impaired sentence and phrase repetition. Supportive diagnostic features include ③ phonological errors in spontaneous speech and naming, ④ relatively preserved single-word comprehension and motor speech, and ⑤ relatively preserved grammar processing. In particular, the absence of apraxia of speech and marked agrammatism is the key differentiating criterion from nfvPPA.
Imaging supportive criteria include atrophy of the left temporoparietal association cortex (especially the angular gyrus and posterior superior temporal gyrus) on MRI and hypometabolism of the same region on FDG-PET. For confirmation of Alzheimer pathology, amyloid PET or CSF Aβ42/p-tau biomarkers are performed, and if positive it is classified as Alzheimer-based lvPPA. If negative, FTLD-based disease is considered and tau PET, plasma progranulin for GRN, and an FTD gene panel are performed stepwise. Histopathological confirmation is made at autopsy by Alzheimer criteria (NIA-AA ABC score) or FTLD criteria.
In Alzheimer pathology-based lvPPA, judging the indication for anti-amyloid monoclonal antibody (lecanemab, donanemab) treatment is important. Both drugs are indicated in the early Alzheimer continuum (A+ confirmed, CDR 0.5-1 stage), and lvPPA, as an atypical Alzheimer phenotype, may meet these criteria when A+ is confirmed. However, the treatment effect in atypical phenotypes needs further study compared with typical AD cases.
As symptomatic treatment, small studies suggest that cholinesterase inhibitors (donepezil, rivastigmine) may help stabilize language function in Alzheimer-based lvPPA, but evidence from large clinical trials is lacking. Speech-language therapy includes training in word-finding strategies, use of augmentative and alternative communication (AAC) aids, and education in compensatory strategies (notes, picture cues). Introduction of voice banking technology before speech ability declines can be considered. Caregiver communication strategy training, family support programs, and linkage to community resources are essential.
The prognosis of lvPPA differs by the type of underlying pathology. Cases based on Alzheimer pathology follow a course similar to typical Alzheimer dementia, with global cognitive decline progressing within several years after diagnosis and episodic memory and visuospatial function gradually becoming involved. Survival from symptom onset to death is about 8-12 years, but varies with age at onset and comorbid pathology. FTLD-based cases tend to follow a 6-12 year course similar to other PPA variants.
With the introduction of anti-amyloid therapies, the possibility of early intervention has opened even for Alzheimer-based lvPPA patients, but the treatment effect in the language phenotype has not yet been separately analyzed enough. For tracking functional level, the PPA severity scale, WAB, and the CDR plus NACC FTLD sum of boxes score are used. Elevated plasma NfL is associated with rapid progression, and follow-up of amyloid PET and plasma p-tau217 is used to monitor treatment response.
There is no established primary prevention for lvPPA. For Alzheimer pathology-based lvPPA, management of modifiable Alzheimer risk factors (hypertension, diabetes, obesity, smoking, physical inactivity, social isolation, etc.) has preventive significance. A preventive anti-amyloid intervention study in APOE ε4 carriers (the AHEAD 3-45 trial) is ongoing, exploring the possibility of preemptive treatment in the presymptomatic stage. Genetic counseling and predictive testing are offered when there is a family history of PSEN1/APP/PSEN2 variants.
After early diagnosis, determining the indication for anti-amyloid therapy by confirming whether the disease is Alzheimer-based, early initiation of speech-language therapy, AAC planning, and advance decision-making are the core roles of the multidisciplinary team immediately after diagnosis. Regular exercise, a healthy diet, and increased social participation are recommended for maintaining overall brain health.