Advertisement

Do cerebellar plaques influence 18F-florbetaben amyloid PET scan quantification?

      Background

      Standardized uptake value ratios (SUVR) are commonly used for quantification of 18F-Florbetaben (FBB) scans. Cerebellar gray matter is used as the reference region for quantification. However, cerebellar plaques may be present in Alzheimer disease (AD). The aim of this study was to assess the influence of cerebellar plaques in FBB SUVR, when using cerebellar gray matter as the reference.

      Methods

      Neuropathological assessment of cerebral (frontal, occipital, anterior and posterior cingulate) cortex and cerebellar cortex tissue from 87 end of life patients (64 AD, 14 other dementia, 9 non-demented aged volunteers; 80.4±10.2 yrs) who underwent a FBB PET before death was performed using the Bielschowsky silver stain and Amyloid β (Aβ) immunohistochemistry to quantify neuritic/cored and diffuse plaques, as absent, sparse, moderate and frequent. Mean cortical SUVRs were compared among brains with different cerebellar plaque loads.

      Results

      None from the 83 evaluable cerebellar samples showed frequent cerebellar plaques. Only 1 sample showed both sparse neuritic/cored and sparse diffuse plaques. Sparse diffuse plaques were found in 33 samples, and moderate diffuse plaques in 5. Subjects with higher cerebellar plaque loads showed higher cortical Aβ loads and standardized uptake values. Thus, cortical SUVRs significantly increased with cerebellar plaque load (table, figure 1). However, in cortical regions with moderate or frequent Aβ plaques no significant SUVR differences were found among brains showing different cerebellar plaque loads (table, figure 2).

      Conclusions

      In brains with higher cerebral cortical Aβ loads, cerebellar plaques were found in 47% of cases, mostly as sparse diffuse plaques (40%). However, the presence of cerebellar plaques did not influence the SUVRs in these subjects with moderate or frequent cortical Aβ. Therefore, the effect of cerebellar plaques in FBB SUVR appears to be negligible even in advanced stages of AD with a high cortical Aβ load.
      Table
      Region (n)Cerebellar plaquesANOVA
      AbsentSparseModeratep-value
      SUVR (full sample)Frontal (83)1.36±0.351.70±0.331.82±0.29<10-4
      Occipital (82)1.44±0.211.66±0.241.69±0.260.0001
      Ant. Cing (82)1.38±0.401.74±0.381.86±0.310.0001
      Post. Cing (82)1.59±0.381.87±0.371.86±0.240.004
      SUVR (moderate or frequent Aβ plaques)Frontal (47)1.64±0.301.77±0.281.73±0.240.39
      Occipital (44)1.64±0.201.68±0.231.69±0.260.84
      Ant. Cing (26)1.73±0.391.77±0.361.74±0.180.97
      Post. Cing (36)1.93±0.371.95±0.341.79±0.200.69
      Figure thumbnail fx1
      Figure thumbnail fx2
      Figure 2Moderate or frequent cortical AB plaques.