Subtypes of dementia with Lewy bodies are associated with α-synuclein and tau distribution

Abstract

Objective: To determine whether Lewy body disease subgroups have different clinical profiles.

Methods: Participants had dementia, autopsy-confirmed transitional or diffuse Lewy body disease (TLBD or DLBD) (n = 244), or Alzheimer disease (AD) (n = 210), and were seen at least twice (mean follow-up 6.2 ± 3.8 years). TLBD and DLBD groups were partitioned based on the presence or absence of neocortical neurofibrillary tangles using Braak staging. Four Lewy body disease subgroups and AD were compared on clinical features, dementia trajectory, and onset latency of probable dementia with Lewy bodies (DLB) or a DLB syndrome defined as probable DLB or dementia with one core feature of parkinsonism or probable REM sleep behavior disorder.

Results: In TLBD and DLBD without neocortical tangles, diagnostic sensitivity was strong for probable DLB (87% TLBD, 96% DLBD) and the DLB syndrome (97% TLBD, 98% DLBD) with median latencies <1 year from cognitive onset, and worse baseline attention-visual processing but better memory-naming scores than AD. In DLBD with neocortical tangles, diagnostic sensitivity was 70% for probable DLB and 77% for the DLB syndrome with respective median latencies of 3.7 years and 2.7 years from cognitive onset, each associated with tangle distribution. This group had worse baseline attention-visual processing than AD, but comparable memory-naming impairment. TLBD with neocortical tangles had 48% diagnostic sensitivity for probable DLB and 52% for the DLB syndrome, with median latencies >6 years from cognitive onset, and were cognitively similar to AD. Dementia trajectory was slowest for TLBD without neocortical tangles, and fastest for DLBD with neocortical tangles.

Conclusions: The phenotypic expression of DLB was associated with the distribution of α-synuclein and tau pathology.

Figure 1. Cumulative incidence model for the latency of probable dementia with Lewy bodies (DLB) across pathologic subgroups

Lines represent time from the estimated onset of cognitive impairment to the emergence of 2 core DLB features sufficient for a diagnosis of probable DLB. The temporal onset of probable DLB did not differ between transitional Lewy body disease (TLBD)–L and diffuse Lewy body disease (DLBD)–L, but all other groups differed from each other and from TLBD-L and DLBD-L (p ≤ 0.001). AD = Alzheimer disease; H = presence of neocortical tangles, categorized as high; L = absence of neocortical tangles, categorized as low.

Figure 2. Rates of dementia progression

(A) Global Deterioration Scale. (B) Mini-Mental State Examination. (C) Dementia Rating Scale. AD = Alzheimer disease; DLBD = diffuse Lewy body disease; H = presence of neocortical tangles, categorized as high; L = absence of neocortical tangles, categorized as low; TLBD = transitional Lewy body disease.

Figure 3. Diagnostic latency in diffuse Lewy body disease (DLBD) with neocortical tangles

Comparisons of DLBD patients with Braak NFT stages IV, V, and VI in the percentage who developed the dementia with Lewy bodies syndrome by specified intervals from cognitive onset. *p < 0.05, **p < 0.01. H = presence of neocortical tangles, categorized as high.