Alzheimer's disease (AD) and Creutzfeldt-Jakob disease (CJD) are prototypes of non-transmissible and transmissible cerebral amyloidoses, respectively.
There is clinical, neuropathological, and pathogenetic overlap between AD and CJD.
Several studies on single patients have reported combinations of AD and CJD, and speculations have been raised about some pathogenetic connection between both neurodegenerative conditions.
We address this question here by determining the frequency of coexisting AD pathology in a large series of neuropathologically diagnosed CJD cases as compared with appropriately matched control brains, and by analysing possible local combination of tissue lesions of both types of cerebral pathologies.
A definite diagnosis of CJD and AD requires neuropathological analysis.
In CJD, the neuropathological hallmark is accumulation of immunocytochemically detectable PrP, a pathological, amyloidogenic isoform of the constitutively expressed cellular PrP, in addition to spongiform change.
In AD, the major component of plaques is amyloid β/A4 peptide, which is proteolytically cleaved from a cell membrane spanning precursor glycoprotein.
For the assessment of AD, current neuropathological protocols are those of Khatchaturian, CERAD and Tierney, which are all based on assessment of plaques and/or neurofibrillary tangles in the cerebral cortex.
The CERAD neuropathology protocol for AD requires the semiquantitative assessment of maximum cortical involvement with neuritic plaques only.
Most recent consensus recommendations on the neuropathological assessment of AD endorsed in general the semiquantitative CERAD analysis of neuritic pathology.
Assessment of coexisting Alzheimer-type changes in our CJD series was performed on a single neocortical block of each case selected for prominent CJD pathology and immunocytochemically detectable PrP.
This selection procedure was chosen to address the major aim of this study, the assessment of any local correlation with Alzheimer- type pathology (including local deposition patterns of both PrP and β/A4).
Moreover, this allows simple and precise comparison with controls.
However, the selected area might not be that most severely affected by Alzheimer-type pathology in the whole brain; the complete CERAD protocol involves assessment of senile plaque density in the most severely affected region of the neocortex based upon examination of sections of frontal, temporal, and parietal cortex.
Thus more extensive screening of the neocortex might even increase the incidence and/or prominence of Alzheimer-type pathology in CJD and control groups.
PrP deposits surround β/A4 plaque cores, creating a "compound" β/A4-PrP amyloid plaque structure.
Double immunocytochemistry for β/A4 (brown) and PrP (blue) (PrP prion protein).
In this CJD series, concomitant CERAD AD occurred in 10.9% of the cases and primarily in aged patients.
CERAD AD in CJD is, thus, not an exceptional finding. In the matched control group, however, changes would have met CERAD criteria for AD if the patients had been demented; this occurred at an even higher frequency (19.1%) and slightly younger age.
Testing for a positive or negative correlation between CJD and AD was not statistically significant.
Thus, these results do not support previous speculations about a pathogenetic connection between AD and CJD.
Moreover, the pathological findings in our control group confirm previous findings that Alzheimer-type pathology is a frequent finding in the non-demented elderly population; non-demented individuals may harbour considerable numbers of plaques in the cerebral cortex.
A possible explanation of why Alzheimer-type pathology was more frequently found in controls than in CJD is the lack of prospective assessment for dementia in this retrospective investigation.
Therefore, it cannot be excluded that the control group includes patients in whom dementia was overlooked.
CERAD criteria require the presence of dementia for diagnosis of definite and probable AD; it does not take into account other neuropathological lesions likely to cause dementia.
In the CJD/AD group, the short duration of disease (which does not significantly differ from that in the non-AD/CJD group) argues for CJD as the predominant cause of dementia.
Clinicopathological data in the control group suggest that Alzheimer-type changes in the CJD/AD group are most likely an age-related coincidental pre- or subclinical finding.
Therefore, the inclusion of "other neuropathological lesions likely to cause dementia" among CERAD criteria results in (falsely) positive classification of presumable subclinical Alzheimer-type changes as AD in our CJD series.
A similar situation also might occur in the setting of other types of neurodegenerative conditions with more subtle neuropathological lesions.
The rationale to include other lesions likely to cause dementia among CERAD AD diagnostic criteria is debatable.
So far, co-localization of β/A4 peptide and PrP in the same amyloid plaque has been reported only in aged patients with Gerstmann-Straussler-Scheinker syndrome.
We show here for the first time the co-localization of β/A4 peptide and PrP in the same plaques in CJD.
We found β/A4-PrP compound plaques in most of our CJD patients with concomitant Alzheimer-type pathology (in 11/12 CJD/AD cases).
In most compound plaques, PrP accumulates at the periphery of β/A4 plaques.
Thus, apparently pre-existing β/A4 amyloid might act in CJD as a micro environmental factor influencing PrP morphogenesis by fostering the aggregation of one amyloidogenic protein onto a core composed by the other.
These morphological observations, however, cannot exclude the opposite possibility of amyloid co-deposition, i.e., β/A4 deposition triggered in the center of PrP plaques, although this is more difficult to imagine.