Amyloid is linked to cognitive decline in patients with Parkinson disease without dementia

Abstract

Objective: To determine whether amyloid burden, as indexed by Pittsburgh compound B (PiB) retention, identifies patients with Parkinson disease with mild cognitive impairment (PD-MCI) compared to those with normal cognition (PD-nl). A related aim is to determine whether amyloid burden predicts cognitive decline in a cohort of subjects with PD without dementia.

Methods: In this prospective cohort study, we examined 46 subjects with PD without dementia, of whom 35 had normal cognition and 11 met criteria for PD-MCI at study baseline. All subjects underwent standardized neurologic and neuropsychological examinations and PiB PET at baseline, and clinical examinations were conducted annually for up to 5 years.

Results: At baseline, precuneus PiB retention did not distinguish PD-MCI from PD-nl. Subjects with PD-MCI declined more rapidly than PD-nl subjects on cognitive tests of memory, executive function, and activation retrieval. Of the 35 PD-nl subjects, 8 progressed to PD-MCI and 1 to dementia; of the 11 PD-MCI subjects, 5 converted to dementia. Both higher PiB retention and a diagnosis of PD-MCI predicted a greater hazard of conversion to a more severe diagnosis. Baseline PiB retention predicted worsening in executive function over time. The APOE ε4 allele also related to worsening in executive function, as well as visuospatial function, activation retrieval, and performance on the Mini-Mental State Examination. In contrast to its relation to cognitive decline, PiB retention did not affect progression of motor impairment.

Conclusions: At baseline measurements, amyloid burden does not distinguish between cognitively impaired and unimpaired subjects with PD without dementia, but our data suggest that amyloid contributes to cognitive, but not motor, decline over time.

Figure 1. PiB retention did not differentiate PD-nl and PD-MCI subjects

The black line extending across the whole figure represents the grand mean. The green line extending beyond the box for each group represents its mean. The red line for each group within the box represents the group median, box edges show the interquartile range, and box whiskers extend to as much as 1.5 interquartiles above and below the edges. Box width is proportional to group sample size. Subjects who subsequently progressed to a worse diagnosis are depicted as blue triangles, while those who remained stable are shown as red circles. PD-MCI = Parkinson disease with mild cognitive impairment; PD-nl = Parkinson disease with normal cognition; PiB = Pittsburgh compound B.

Figure 2. Kaplan-Meier survival curves

Subjects with Pittsburgh compound B (PiB) retention above the median for the sample converted to a more severe diagnosis sooner than those with values below the median. Parkinson disease with normal cognition and Parkinson disease with mild cognitive impairment groups were pooled together for this analysis. Shading around survival curves indicates 95% confidence bands. “Censored” indicates the last visit for subjects who had not transitioned to a more severe diagnosis.