LETTER TO THE EDITOR Striatal A-Amyloid Deposition in Parkinson Disease With Dementia

In a recent article, Kalaitzakis et al (1) reported a significantly greater amyloid-A peptide (AA) burden in the striatum of cases of Parkinson disease (PD) with dementia (PDD) than in nondemented PD patients, whereas >synuclein (>SN) and tau deposition were similarly rare in both groups. Because neuritic Braak stages were rather low in both groups, it was concluded that AA deposition in the striatum, which was type independent of Alzheimer disease (AD) changes, strongly correlates with dementia in PD. These data can, at least in part, be confirmed from the results of a personal autopsy series of 64 cases of Lewy body disease (21 PD, 23 PDD, and 20 dementia with Lewy bodies [DLB]). Neuropathologic assessment was performed according to standardized methods. Multiple paraffin-embedded blocks were examined using routine stains and immunohistochemistry for tau, AA (clone 4G8), and >SN. Each case was staged for the degree of LB pathology (2); cortical AA plaque density from frontal, temporal cortex, caudate nucleus, putamen, globus pallidus, and thalamus; and generalized and capillary cerebral angiopathy using 4 different scores (for methods, see References 3 and 4) for all items. Parkinson disease with dementia patients (final Mini-Mental State Examination, 0Y15; mean, 7.5) were significantly older at death than nondemented ones (final Mini-Mental State Examination, 20Y29; mean, 25; mean, 86.8 T 4.6 vs 78.7 T 5.4 years [standard deviation]), whereas DLB cases (final Mini-Mental State Examination, 0Y18; mean, 9) were slightly younger than PDD patients (mean, 84.4 T 4.8 years [standard deviation]). Duration of illness was shortest in DLB G PDD G PD nondemented (mean, 6.5 vs 6.6 vs 12.2 years). Lewy body Braak scores were highest in DLB but similar in PDD and nondemented PD (mean, 5.9 vs 4.2 and 3.8, respectively), which did not confirm findings indicating that cognitive decline correlates with neuropathologic LB stage in PD (5), whereas neuritic Braak scores in both DLB and PDD were higher than in nondemented PD (mean, 4.2 vs 3.8 and 2.5). Likewise, cortical AA plaque, capillary cerebral angiopathy, and general cerebral angiopathy scores were higher in DLB than in both PD and PDD cases (4). Dementia with Lewy body cases showed a significantly higher AA plaque burden in putamen, caudate nucleus, and thalamus than in PDD and nondemented PD cases (mean, 2.5 vs 0.8 and 0.1, respectively). Dementia with Lewy body brains showed severe and dense AA plaques in striatum in 65%, moderate in 25%, and none or only very mild AA plaques in 10%; in PDD cases, only 26% showed moderately severe, less than 20% mild, and almost 48% no definite AA plaques, whereas in nondemented PD cases, only 1 of 20 brains (5%) showed occasional AA plaques in striatum. The globus pallidus was virtually negative in all 3 groups, whereas few >SN deposits were seen in 76% of DLB and in approximately 30% of PDD brains, but only in 10% of nondemented PD cases. Striatal AA plaque burden in DLB brains showed association with neuritic Braak stages, whereas in both of the other groups, such correlations were not found except in 2 cases scoring neuritic Braak Stage 5 and AA Phase 4 with only few or moderate numbers of striatal AA plaques. In addition, tau pathology in striatum of our cohort was also more frequent in DLB than in both PDD and PD, the incidence of negative cases being 70% vs 82% and 100%, respectively. These findings support a morphologic distinction between DLB and PDD, but also between PDD and nondemented PD, which, however, does not seem to be restricted to different AA deposition in the striatum because, at least in our material, neuritic AD pathology was more severe in both DLB and PDD than in nondemented PD. Hence, significantly greater deposition of AA in the striatum of PDD patients that seems to be quite independent of the extent of cortical AD pathologic findings is not a definitely novel finding, as suggested by some authors (1). Both the molecular background and clinical/pathophysiologic impact of striatal pathologies in LB disease remain to be further elucidated, although some synergistic reactions between AA, >SN, and tau proteins have been suggested.