Autopsy-confirmed Alzheimer's Disease Patients Research Articles

Increased Inflammation and Unchanged Density of Synaptic Vesicle Glycoprotein 2A (SV2A) in the Postmortem Frontal Cortex of Alzheimer's Disease Patients.

Sections from the middle frontal gyrus (Brodmann area 46) of autopsy-confirmed Alzheimer’s disease (AD) patients and non-demented subjects were examined for the prevalence of hallmark AD pathology, including amyloid-β (Aβ) plaques, phosphorylated tau (pTau) tangles, neuroinflammation and synaptic loss (n = 7 subjects/group). Dense-core deposits of Aβ were present in all AD patients (7/7) and some non-demented subjects (3/7), as evidenced by 6E10 immunohistochemistry. Levels of Aβ immunoreactivity were higher in AD vs. non-AD cases. For pTau, AT8-positive neurofibrillary tangles and threads were exclusively observed in AD patient tissue. Levels of [3H]PK11195 binding to the translocator protein (TSPO), a marker of inflammatory processes, were elevated in the gray matter of AD patients compared to non-demented subjects. Levels of [3H]UCB-J binding to synaptic vesicle glycoprotein 2A (SV2A), a marker of synaptic density, were not different between groups. In AD patients, pTau immunoreactivity was positively correlated with [3H]PK11195, and negatively correlated with [3H]UCB-J binding levels. No correlation was observed between Aβ immunoreactivity and markers of neuroinflammation or synaptic density. These data demonstrate a close interplay between tau pathology, inflammation and SV2A density in AD, and provide useful information on the ability of neuroimaging biomarkers to diagnose AD dementia.

Loss in PKC Epsilon Causes Downregulation of MnSOD and BDNF Expression in Neurons of Alzheimer's Disease Hippocampus.

Oxidative stress and amyloid-β (Aβ) oligomers have been implicated in Alzheimer's disease (AD). The growth and maintenance of neuronal networks are influenced by brain derived neurotrophic factor (BDNF) expression, which is promoted by protein kinase C epsilon (PKCɛ). We investigated the reciprocal interaction among oxidative stress, Aβ, and PKCɛ levels and subsequent PKCɛ-dependent MnSOD and BDNF expression in hippocampal pyramidal neurons. Reduced levels of PKCɛ, MnSOD, and BDNF and an increased level of Aβ were also found in hippocampal neurons from autopsy-confirmed AD patients. In cultured human primary hippocampal neurons, spherical aggregation of Aβ (amylospheroids) decreased PKCɛ and MnSOD. Treatment with t-butyl hydroperoxide (TBHP) increased superoxide, the oxidative DNA/RNA damage marker, 8-OHG, and Aβ levels, but reduced PKCɛ, MnSOD, BDNF, and cultured neuron density. These changes were reversed with the PKCɛ activators, bryostatin and DCPLA-ME. PKCɛ knockdown suppressed PKCɛ, MnSOD, and BDNF but increased Aβ. In cultured neurons, the increase in reactive oxygen species (ROS) associated with reduced PKCɛ during neurodegeneration was inhibited by the SOD mimetic MnTMPyP and the ROS scavenger NAc, indicating that strong oxidative stress suppresses PKCɛ level. Reduction of PKCɛ and MnSOD was prevented with the PKCɛ activator bryostatin in 5-6-month-old Tg2576 AD transgenic mice. In conclusion, oxidative stress and Aβ decrease PKCɛ expression. Reciprocally, a depression of PKCɛ reduces BDNF and MnSOD, resulting in oxidative stress. These changes can be prevented with the PKCɛ-specific activators.

Cox regression model with doubly truncated data.

Truncation is a well-known phenomenon that may be present in observational studies of time-to-event data. While many methods exist to adjust for either left or right truncation, there are very few methods that adjust for simultaneous left and right truncation, also known as double truncation. We propose a Cox regression model to adjust for this double truncation using a weighted estimating equation approach, where the weights are estimated from the data both parametrically and nonparametrically, and are inversely proportional to the probability that a subject is observed. The resulting weighted estimators of the hazard ratio are consistent. The parametric weighted estimator is asymptotically normal and a consistent estimator of the asymptotic variance is provided. For the nonparametric weighted estimator, we apply the bootstrap technique to estimate the variance and confidence intervals. We demonstrate through extensive simulations that the proposed estimators greatly reduce the bias compared to the unweighted Cox regression estimator which ignores truncation. We illustrate our approach in an analysis of autopsy-confirmed Alzheimer's disease patients to assess the effect of education on survival.

Diagnostic Impact of Cerebrospinal Fluid Biomarker (Pre-)Analytical Variability in Alzheimer's Disease.

Intra- and inter-laboratory variability of cerebrospinal fluid (CSF) biomarker analyses remains an important issue. We investigated the clinical-diagnostic impact of CSF biomarker concentration shifts in mild cognitive impairment (MCI) and autopsy-confirmed Alzheimer’s disease (AD) dementia patients. MCI patients (n = 85), autopsy-confirmed AD dementia patients (n = 72), and cognitively healthy controls (n = 100) were included in this prospective, longitudinal study. AD dementia patients were followed up until death, and controls were included from 1992 until 2003. In-house validated cutoff values of biomarkers were applied: Aβ1-42 <638.5 pg/mL, T-tau>296.5 pg/mL, P-tau181P >56.5 pg/mL. Both increments and decrements (from ± 5% to ± 40% ) were added to the true (=observed) CSF biomarker values, imitating the anticipated differences in biomarker concentrations. Within certain limits, the clinical diagnostic performance of AD CSF biomarkers remains largely unchanged and clinical diagnostic accuracy deviated less than 8.2% from the reference when concentration shifts ranging between –20% and +20% were added to one of the three CSF biomarkers in MCI and autopsy-confirmed AD patients. Notwithstanding the fact that (pre- and post-)analytical parameters can affect the clinical classification, the present exploratory study provides evidence that for a specific context of use, the impact on clinical accuracy of biomarker concentration shifts might be lower than originally expected. In conclusion, induced shifts of ±20% in only one of the three biomarkers has limited impact on the clinical accuracy of AD CSF biomarkers in MCI and autopsy-confirmed AD patients when using the IWG-2 criteria.

PKCε deficits in Alzheimer's disease brains and skin fibroblasts.

In Alzheimer's disease (AD) transgenic mice, activation of synaptogenic protein kinase C ε (PKCε) was found to prevent synaptotoxic amyloid-β (Aβ)-oligomer elevation, PKCε deficits, early synaptic loss, cognitive deficits, and amyloid plaque formation. In humans, to study the role of PKCε in the pathophysiology of AD and to evaluate its possible use as an early AD-biomarker, we examined PKCε and Aβ in the brains of autopsy-confirmed AD patients (n = 20) and age-matched controls (AC, n = 19), and in skin fibroblast samples from AD (n = 14), non-AD dementia patients (n = 14), and AC (n = 22). Intraneuronal Aβ levels were measured immunohistochemically (using an Aβ-specific antibody) in hippocampal pyramidal cells of human autopsy brains. PKCε was significantly lower in the hippocampus and temporal pole areas of AD brains, whereas Aβ levels were significantly higher. The ratio of PKCε to Aβ in individual CA1 pyramidal cells was markedly lower in the autopsy AD brains versus controls. PKCε was inversely correlated with Aβ levels in controls, whereas in AD patients, PKCε showed no significant correlation with Aβ. In autopsy brains, PKCε decreased as the Braak score increased. Skin fibroblast samples from AD patients also demonstrated a deficit in PKCε compared to controls and an AD-specific change in the Aβ-oligomer effects on PKCε. Together, these data demonstrate that the relationship between Aβ levels and PKCε is markedly altered in AD patients' brains and skin fibroblasts, reflecting a loss of protective effect of PKCε against toxic Aβ accumulation. These changes of PKCε levels in human skin fibroblasts may provide an accurate, non-invasive peripheral AD biomarker.

Systematic literature review and meta-analysis of diagnostic test accuracy in Alzheimer's disease and other dementia using autopsy as standard of truth.

Early diagnosis of Alzheimer's disease (AD) is crucial to implement the latest treatment strategies and management of AD symptoms. Diagnostic procedures play a major role in this detection process but evidence on their respective accuracy is still limited. To conduct a systematic literature on the sensitivity and specificity of different test modalities to identify AD patients and perform meta-analyses on the test accuracy values of studies focusing on autopsy-confirmation as the standard of truth. The systematic review identified all English papers published between 1984 and 2011 on diagnostic imaging tests and cerebrospinal fluid biomarkers including results on the newest technologies currently investigated in this area. Meta-analyses using bivariate fixed and random-effect models and hierarchical summary receiver operating curve (HSROC) random-effect model were applied. Out of the 1,189 records, 20 publications were identified to report the accuracy of diagnostic tests in distinguishing autopsy-confirmed AD patients from other dementia types and healthy controls. Looking at all tests and comparator populations together, sensitivity was calculated at 85.4% (95% confidence interval [CI]: 80.9%-90.0%) and specificity at 77.7% (95% CI: 70.2%-85.1%). The area under the HSROC curve was 0.88. Sensitivity and specificity values were higher for imaging procedures, and slightly lower for CSF biomarkers. Test-specific random-effect models could not be calculated due to the small number of studies. The review and meta-analysis point to a slight advantage of imaging procedures in correctly detecting AD patients but also highlight the limited evidence on autopsy-confirmations and heterogeneity in study designs.

Validation of the AD-CSF-Index in Autopsy-Confirmed Alzheimer's Disease Patients and Healthy Controls

The cerebrospinal fluid (CSF) biomarkers amyloid-β peptide of 42 amino acids (Aβ1-42), total tau-protein (T-tau), and tau phosphorylated at threonine 181 (P-tau181P) are used to diagnose Alzheimer's disease (AD). In order to increase diagnostic power, several biomarker combinations have been proposed. In that sense, a new CSF biomarker index was developed, the AD-CSF-index, which has been validated in clinically diagnosed AD patients using electrochemoluminescence based Meso Scale Discovery and single-analyte ELISA kits. This study validated the AD-CSF-index in neuropathologically diagnosed AD patients, using both single-analyte ELISA and multi-analyte Luminex assays. CSF of 51 neuropathologically diagnosed AD patients and of 95 controls was analyzed by commercially available single-analyte ELISA-kits (INNOTEST, Innogenetics) and by a Research Use Only version of the multi-analyte Luminex xMAP assay (INNO-BIA AlzBio3, Innogenetics). Subsequently the AD-CSF-indices were calculated. Both T-tau and P-tau181P AD-CSF-indices were significantly increased in AD patients when compared to controls (p < 0.001). The diagnostic power of the indices was calculated using ROC analyses, resulting in excellent sensitivity and specificity values that systematically exceeded the 80% threshold for discriminating autopsy-confirmed AD patients from controls, independent of the analytical platform. The power to discriminate between AD and non-AD dementias was not included in this study and should be validated in the future. In conclusion, this study validated the AD-CSF-index in autopsy-confirmed AD patients and has shown that its excellent diagnostic accuracy is independent of the analytical platform.

Increased CSF α-synuclein levels in Alzheimer's disease: Correlation with tau levels

BackgroundGiven the difficult clinical differential diagnosis between Alzheimer's disease (AD) and dementia with Lewy bodies (DLB), growing interest resulted in research on α-synuclein as a potential cerebrospinal fluid biomarker (CSF) for synucleinopathies. MethodsCSF α-synuclein-140 concentrations were determined by a prototype xMAP™ bead-based assay (Innogenetics NV, Belgium). In addition, CSF amyloid β1-42 (Aβ1-42), total tau (T-tau), and phosphorylated tau (P-tau181P) levels were determined. ResultsCSF α-synuclein levels were higher in AD patients as compared with cognitively healthy controls (P = .019) and patients with synucleinopathies (P < .001). CSF α-synuclein levels were correlated with T-tau (P < .001) and P-tau181P (P < .001) levels in autopsy-confirmed AD patients. A diagnostic algorithm using α-synuclein and P-tau181P discriminated neuropathologically confirmed AD from DLB patients, resulting in sensitivity and specificity values of 85% and 81%, respectively. ConclusionBecause CSF α-synuclein levels were significantly higher in AD as compared with synucleinopathies, α-synuclein might have a value as a biomarker for differential dementia diagnosis.

Neuropsychiatric Features of Frontal Lobe Dysfunction in Autopsy-Confirmed Patients with Lewy Bodies and “Pure” Alzheimer Disease

To compare patients with autopsy-confirmed Alzheimer disease (AD) and dementia with Lewy bodies (DLB) on the frequency of behaviors related to frontal system dysfunction and the association of these behaviors with dementia severity. We performed a cross-sectional survey of a longitudinal cohort at a university research center for AD on a volunteer sample of 19 DLB and 38 AD participants with autopsy-confirmed diagnoses, similar in age (DLB: 77.3, AD: 77.5), education (DLB: 15.2, AD: 14.7), and Mini-Mental State Examination (MMSE) score (DLB: 20.6, AD: 20.5), with impairment ranging from mild deficits to moderate dementia. The Frontal Systems Behavior Scale (FrSBe)-Family Rating Form assessing patient apathy, disinhibition, and executive dysfunction by a knowledgeable informant was used. A two-way analysis of variance with the FrSBe total as the dependent variable revealed a significant MMSE by diagnosis interaction (F(1,53)= 9.34, p= 0.004). Mean FrSBe total for AD patients showed significant impairment across the range of dementia severity, whereas it was relatively preserved for DLB patients in the early stage of disease. The interaction term showed the same pattern for the executive dysfunction (F(1,53)= 7.62, p= 0.008), disinhibition (F(1,53)= 4.90, p= 0.031), and apathy (F(1,53)= 9.77, p= 0.003) subscales. Although frontal behavioral symptoms in AD patients were present regardless of stage of dementia, DLB patients showed significant frontal dysfunction only in later stages. Results suggest that frontal subcortical circuits associated with behaviors assessed by the FrSBe are affected early in AD but not until later stages in DLB. Assessing specific behaviors related to frontal systems, coupled with stage of cognitive decline, may aid in clinical differentiation of AD and DLB.

Redox-reactive antiphospholipid antibody differences between serum from Alzheimer's patients and age-matched controls

There is no universally acceptable inclusive laboratory biomarker for the diagnosis and staging of neurodegenerative diseases, for example, Alzheimer's. There is an abnormal increase of oxidative stress in the central nervous system (CNS) of Alzheimer's patients that causes oxidation of proteins, lipids and DNA. We have published that the antiphospholipid (aPL) autoantibodies that are members of the redox-reactive autoantibody (R-RAA) family, are significantly decreased or absent in the cerebrospinal fluids of autopsy-confirmed Alzheimer's disease (AD) patients. Because of the known elevation of oxidation-induced damage in the CNS and the abnormal enrichment of redox reactive metals in postmortem AD brains, we questioned if the R-RAA in the blood of AD patients might also show a departure from the normal aPL levels. We compared 16 AD serum samples to 17 serum samples, from age-matched volunteer blood donors. Each serum was tested before and after oxidation for four aPL specificities by using an in-house ELISA. Comparisons between the AD and normal populations revealed highly significant differences. In-sample Fisher's linear discriminate analysis found a sensitivity of 88% and a specificity of 94%. In-sample Classification and Regression Tree analysis (CART) found a sensitivity of 84% and a specificity of 100%. This study is the first to indicate that blood tests for R-RAA may be used as a laboratory criterion for an Alzheimer's diagnosis.

No association of CSF biomarkers with APOE 4, plaque and tangle burden in definite Alzheimer's disease

The CSF biomarkers beta-amyloid peptide (Abeta(1-42)), total tau protein (T-tau) and tau phosphorylated at threonine 181 (P-tau(181P)) were determined in autopsy-confirmed Alzheimer's disease patients in order to study possible associations with the epsilon4 allele of APOE and density and spread of plaques (SP) and tangles (NFT). CSF levels of Abeta(1-42), T-tau and P-tau(181P) were determined in 50 Alzheimer's disease patients using commercially available single parameter ELISA kits (INNOTEST(R)). Genomic DNA was extracted from whole blood and the APOE genotype was determined using standard methods. Tangle burden was assessed by means of Braak's NFT stages (I-VI), whereas the plaque burden was assessed by means of Braak's SP stages (A-C). CSF biomarker levels were not different when comparing epsilon4 carriers (n = 21) and non-carriers (n = 29) (P > 0.05 for all comparisons). No significant correlations between the number of epsilon4 alleles (0, 1 or 2) and CSF levels of Abeta(1-42) (Spearman Rank Order: r = -0.057, P = 0.695), T-tau (r = 0.104, P = 0.472) and P-tau(181P) (r = 0.062, P = 0.668) were found. Braak's SP (Abeta(1-42): r = -0.155, P = 0.280; T-tau: r = -0.044, P = 0.763; P-tau(181P): r = -0.010, P = 0.947) and NFT (Abeta(1-42): r = -0.145, P = 0.315; T-tau: r = 0.117, P = 0.415; P-tau(181P): r = 0.150, P = 0.296) stages were not significantly correlated with CSF biomarker levels. In conclusion, CSF levels of Abeta(1-42), T-tau and P-tau(181P) were not associated with epsilon4, tangle or plaque burden in 50 autopsy-confirmed Alzheimer's disease patients. In the light of future biomarker applications like monitoring of disease progression and as allocortical neuropathological changes significantly contribute to clinical symptoms, the concept of in vivo surrogate biomarkers should be further explored.

In Reply

Drs. Boersma and Eefsting are correct in stating that clinical diagnosis was made at an undefined moment during the course of the disease in our study of 100 autopsy-confirmed Alzheimer's disease patients.1 The insidious onset of this disease is but one reason for the observed differences in the time from symptom onset to clinical diagnosis among the patients in this cohort. Other reasons include varying caregiver thresholds for seeking medical help and the wide range of primary care physicians and specialists contacted by this cohort and their caregivers. It is noteworthy that, in most instances, the caregiver was a spouse or child living with the patient. Drs. Boersma and Eefsting believe the gender differences reported in our study for age at diagnosis are most likely attributable to social, i.e., marital, status rather than to pathological factors. We agree that marital status does indeed contribute to the explanation for these observed differences. Information pertaining to the patients' marital status at the time of clinical diagnosis was collected. This data was available for 91% of the sample and is shown in Table 1, stratified by gender. Patients who were married at the time of clinical diagnosis were indeed diagnosed earlier than their nonmarried counterparts.