Cerebral and CSF amyloid load, and recovery of semantic material in Alzheimer disease patients

According to the new diagnostic criteria, the term “Alzheimer Disease” (AD) refers to a set of neuropathological changes that can be evaluated in vivo, rather than to a specific clinical symptomatology1. It is now widely accepted that β-Amyloid (Aβ42) in the cerebrospinal fluid is a valid indicator of alteration of the pathophysiological state, associated with fibrillar deposits of cerebral β-amyloid2. Comparative studies between imaging and autopsy findings have established that amyloid PET images are a valid in vivo surrogate for deposition of fibrillar β-amyloid3-10. We analyzed the values of cerebral spinal cord (CSF) biomarkers in 40 patients with clinical diagnosis of AD. Two groups emerged: the first with both clinical and liquor biomarkers consistent with AD; the second was clinically in line with AD, but it was missing its pathognomic CSF biomarkers. At this point we asked ourselves about the nosological entity of the second group. All patients underwent flutemetamol-PET and the second group was dichotomically divided into two more groups based on the PET report. Schematically: Group 1: CSF + / PET +; Group 2: CSF- / PET +; Group 3: CSF- / PET-. Our study then correlated the PET images through statistical software (spm12) in order to highlight any differences in cerebral β-amyloid accumulation. The first comparison was conducted between Group 1 and 2 revealing a significant accumulation of β-amyloid in the regions of the posterior cingulate gyrus. The posterior cingulate gyrus is involved in maintaining spatio-temporal orientation and memory functions, thanks to the connections with the parahippocampal cortex11. Involvement of the posterior cingulate gyrus is classic in patients with a typical clinical presentation of AD11. This result is also in agreement with the typical cerebral distribution of Aβ in AD (Braak and Braak stages)12 and highlights instead the possibility of non-typical deposits in the second group, for which a different etiopathogenetic mechanism from "ordinary” AD is hypothesized13. In support of this assumption, the results of the second comparison conducted between Group 2 versus Group 1, which showed a pattern of regional accumulation of cerebral β-amyloid in the regions of the frontal lobe, are explanatory. On the basis of this evidence, we hypothesized that the CSF- / PET + condition represents a clinical variant of the AD pathology defined in the literature as "frontal variant of the AD"13. Several studies have found that in the frontal variant of AD, the neuro-fibrillary tangle load (NFT) is about 10 times higher in the frontal cortex13 than in the typical AD group. On the other hand, patients with typical AD showed a greater accumulation of NFT in the entorhinal cortex, cingulate gyrus and temporal cortex13. Both evidences are consistent with the results of our study. Starting from the analysis of the neuropsychological tests carried out in AP patients behavioral and language alterations to the onset of illness have emerged, in addition to the memory impairment which is a pathognomonic sign of the typical AD. The typical AD refers to a pattern characterized by an early episodic memory loss followed by various combinations of deficits including attentional-executive deficit, language and visuospatial capacity deficits, which reflect the spread of the disease from the medial temporal lobe to other neocortical areas14-17. In contrast to this typical profile, the focal cortical variants of AD18 present an atypical symptomatological picture (executive dysfunctions19-20, deficits in design skills, behavioral abnormalities, impulsiveness, inattention to detail, inability to plan and language deficit21) . Despite the serious alterations to the tests that investigate the functioning of the frontal lobe, the performance of neuropsychological tests were similar to the typical AD group. This suggests that severe frontal deficiency is the main neuropsychological feature on top of an otherwise typical AD profile13. Several studies suggest that the deposition of fibrillar Aβ explains at most, a small part of the clinical-anatomical heterogeneity of AD13. In fact, in the frontal AD variant an increase in tangles of tau fibrils but not of amyloid plaques has been observed22-23. It is now widely accepted that in AD the neurofibrillary lesions begin to accumulate in the limbic and temporo-parietal regions and only then would they progress to the frontal and occipital cortex. Thus the frontal lobes would be affected by neurodegenerative lesions typical of AD in a subsequent temporal sequence12. It is therefore possible that in the AD variants there is a focal deficit that is indicative of a selective, early and prominent vulnerability of some regions of the brain that are normally involved in pato This result is also in agreement with the cerebral distribution typical of Aβ in AD (Braak and Braak stages)12 and highlights instead the possibility of non-typical deposits in the second group, for which a different etiopathogenetic mechanism is hypothesized from that "typical" of AD13. In support of this hypothesis, the results of the second comparison conducted between Group 2 versus Group 1, which showed a pattern of regional accumulation of cerebral β-amyloid in the regions of the frontal lobe (Fig.2), are explanatory. On the basis of this evidence, we hypothesized that the CSF-/PET + condition represents a clinical variant of the AD pathology defined in the literature as "frontal variant of the AD"13. Several studies have found that in the frontal variant of AD, the neuro-fibrillary tangle load (NFT) is about 10 times higher in the frontal cortex13 than in the typical AD group. On the other hand, patients with typical AD showed a greater accumulation of NFT in the entorhinal cortex, cingulate gyrus and temporal cortex13. Both evidences are consistent with the results of our study. Starting from the analysis of the neuropsychological tests carried out in AP patients, behavioral and language alterations to the onset of illness have emerged, in addition to the memory impairment which is a pathognomonic sign of the typical AD. Typical AD refers to a pattern characterized by an early episodic memory loss followed by various combinations of deficits including attentional-executive deficit, language and visuospatial capacity deficits, which reflect the spread of the disease from the medial temporal lobe to other neocortical areas14-17. In contrast to this typical profile, the focal cortical variants of AD18 present an atypical symptomatological picture (including executive dysfunctions19-20, deficits in planing skills, behavioral abnormalities, impulsiveness, inattention to detail, inability to plan and language deficit21) . Despite the serious alterations emerged at the tests investigating the functioning of the frontal lobe, the performance of neuropsychological tests were similar to the typical AD group. This suggests that severe frontal deficiency is the main neuropsychological feature on top of an otherwise typical AD profile13. Several studies suggest that the deposition of fibrillar Aβ explains at most a small part of the clinical-anatomical heterogeneity of AD13. Indeed, an increase in tau fibril tangles but not in amyloid plaques was observed in the frontal variant of AD.22-23. It is now widely accepted that in AD the neurofibrillary lesions begin to accumulate in the limbic and temporo-parietal regions and only afterwards they would progress towards the frontal and occipital cortex. Thus the frontal lobes would be affected by neurodegenerative lesions typical of AD in a subsequent temporal sequence12. It is therefore possible that in the AD variants there is a focal deficit that is indicative of a selective, early and prominent vulnerability of some brain regions which generally, as mentioned, will normally be involved in the AD pathology in a subsequent time sequence. This vulnerability would be caused by the primary deposition of tau at the frontal level24-28. On the other hand, the frontal variant of AD is characterized by a pathological process that does not seem to remain limited to the frontal lobes for a long time18. Aggregation of Aβ would be driven by the total flux of neuronal activity while tau aggregation would depend on trans-neuronal diffusion, generating neurodegeneration models that coincide with specific functional networks that eventually lead to specific clinical phenotypes (AD variants)13. A better understanding of the factors that drive the heterogeneity of these clinical phenotypes can provide important insights into the mechanisms of the disease and have direct implications on the diagnosis and management of patients with emerging disease-specific therapies18. Finally, in our study, the third and fourth comparisons were conducted between Group 1 and Group 3 and between Group 2 and Group 3 respectively. Both groups showed a significant pattern of accumulation of cerebral β-amyloid widespread almost in all brain areas. This result is not surprising, in light of the fact that Group 3 probably configures the SNAP Group (suspected non-Alzheimer's pathophysiology), or a syndrome defined by normal levels of amyloid biomarkers (CSF- / PET-) but neurodegeneration patterns evident at MRI or FDG-PET29 imaging study. In fact, from 10% to 30% of clinically diagnosed ADs do not show neuropathological alterations of AD during an autopsy30 and a similar proportion has Aβ31 or CSF Aβ42 levels normal31-40. Thus the multi-domain anamnestic phenotype of dementia is not specific. it may be the product of other diseases as well as AD31. To date, SNAP remains a not yet well-defined nosological entity. The clinical diagnosis of AD is often "incorrect" but there are significant differences with regard to clinical progress, genetic susceptibility and progression of the pathology, which have crucial implications for a precise and correct diagnosis, for clinical management and effectiveness of clinical trials on drugs29. SNAP is a very frequent condition in clinically normal subjects > 65 years and appears to be age-related. A study found that the frequency of SNAP was 0% in the age group between 50-60 years while it reached 24% around the age of 8929. However, the literature does not agree. The main controversy in the literature is whether SNAP is an indipendent pathological entity or can evolve into AD41. Some researchers believe that SNAP should be included as an integral part of the AD spectrum; if so, the pathogenetic explanation of the amyloid-centric models of AD and the concept of preclinical AD42 are wrong and should therefore be reviewed. On the contrary, if SNAP is a different entity from AD, the amyloid-centric models of AD and preclinical AD42 are completely consistent with current knowledge. In both cases, multiple studies have shown that the pathogenesis of SNAP is linked to the deposition of tau fibrils, which justify cerebral neurodegeneration; it would then be Aβ, even in small quantities, to act as the biological driver of taupathy, and cause the "spread" of tau in a widespread manner throughout the brain43,44. Therefore a better understanding of the factors that guide the clinical and etiopathogenetic heterogeneity of AD studied thanks to methods such as flutemetamol-PET can provide direct implications on correct diagnosis and prognostic precision in clinical practice. Furthermore, understanding the different nosological entities in study allows a better stratification of the patients in the future trials and the management of emerging specific therapies for this disease.