Objective Neuropsychiatric symptoms affect many patients with Alzheimer's disease (AD). (11C)Pittsburgh Compound-B (PIB) positron emission tomography (PET) has enabled the in vivo visualisation of brain amyloid-β (Aβ) deposition. This study exploratively investigated the correlation between brain Aβ deposition measured by (11C)PIB PET and neuropsychiatric symptoms in AD.
Methods Participants were 28 patients (15 women, 13 men) with PIB-positive AD. Clinical assessments included Mini-Mental State Examination, Clinical Dementia Rating scale, neuropsychiatry inventory (NPI) and frontal assessment battery. All patients underwent three-dimensional T1-weighted MRI and (11C)PIB PET. The distribution volume ratio (DVR), an index of (11C)PIB retention and, thus, Aβ deposition, was estimated voxel by voxel from (11C)PIB PET data with partial volume correction. Voxel-based correlation analysis was performed to assess the relationships between DVR and each NPI subscale. Additionally, voxel-based analysis of covariance (ANCOVA) of the DVR images was performed between Patients with AD with and without each neuropsychiatric symptom. Voxel-based morphometry analysis of MRI was also performed.
Results Apathy subscale was correlated with (11C)PIB retention in the bilateral frontal and right anterior cingulate. (11C)PIB retention was greater in the bilateral frontal cortex of patients with AD with apathy than those of without apathy. Overlapping areas between the two analyses were the bilateral orbitofrontal gyrus and left superior frontal gyrus. Other NPI subscales were not correlated with (11C)PIB retention. Voxel-based morphometry analysis of MRI showed no significant cluster of correlation between grey matter volume and NPI subscales.
Conclusions This study revealed that prefrontal Aβ deposition correlates with apathy.
- ALZHEIMER'S DISEASE
- FRONTAL LOBE
Statistics from Altmetric.com
Alzheimer's disease (AD) is the most common form of dementia, accounting for approximately 60% of all dementia cases.1 AD is characterised by gradual deterioration of cognitive functions including memory, and neuropsychiatric symptoms affect as many as 88% of patients with AD.2 These neuropsychiatric symptoms, including delusions, hallucinations, agitation, depression, anxiety, euphoria, apathy, disinhibition, irritability and aberrant motor behaviour,3 have serious adverse consequences for patients and caregivers, such as greater impairment of daily living activities and deteriorating quality of life.4
In the last decades, neuroimaging measures, such as MRI, single-photon emission CT (SPECT) and 18F-flurodeoxyglucose (FDG) positron emission tomography (PET) have revealed some correlations between neuropsychiatric symptoms and specific neural networks.5 An MRI study revealed negative correlations between delusions and grey matter (GM) volumes in the left frontal lobe, right frontoparietal cortex and left claustrum, between apathy and GM volumes in the anterior cingulate and bilateral frontal, and agitation and GM volumes in the left insula and bilateral anterior cingulate cortex.5 Other FDG-PET studies demonstrated a correlation between apathy and the left orbitofrontal region6 and bilateral anterior cingulate region.7 Some SPECT studies reported that apathy was positively correlated with perfusion in prefrontal and anterior temporal cortices,8 or the right amygdala, bilateral temporal, right posterior cingulate, right superior frontal, postcentral and left superior temporal gyrus.9
Meanwhile, the development of amyloid imaging radiotracers for PET, represented by (11C)Pittsburgh Compound-B (PIB), has enabled the in vivo visualisation of brain amyloid-β (Aβ) deposition,10 one of the core pathological features of AD. Aβ oligomers are the principal effectors of synaptic dysfunction and loss.11 Postmortem study of AD has demonstrated strong positive correlation between the load of Aβ oligomers and cognitive decline.12 Taken together, neuropsychiatric symptoms are correlated with regional brain damage, which could be related to neuronal toxicity of focal Aβ oligomers. Although (11C)PIB PET does not detect Aβ oligomers, oligomers aggregate to form (11C)PIB-detectable Aβ sheets and Aβ plaques. It has been demonstrated using (11C)PIB PET that episodic memory impairment is related to Aβ deposition in the temporal cortex of patients with AD in the predementia stage.13 However, to the best of our knowledge, there has been no study concerning the correlation between neuropsychiatric symptoms and brain Aβ deposition in patients with AD.
The aim of the present study was to investigate the correlation between brain Aβ deposition and neuropsychiatric symptoms in patients with AD exploratively by using (11C)PIB PET with voxel-based whole-brain quantitative analysis.
Thirty-one persons clinically diagnosed with AD based on the National Institute of Neurological and Communicative Disease and Stroke/Alzheimer's Disease and Related Disorders Association (NINCDS-ADRDA) criteria were recruited.14 Three of the patients with AD (9.7%) were diagnosed as PIB-negative by visual assessment of distribution volume ratio (DVR) images of (11C)PIB PET (see method for creating DVR images) and were excluded. PIB-negative was defined as lower PIB uptake of any GM region than that of white matter (WM) by visual assessment of the DVR image. The 28 PIB-positive AD subjects were enrolled for further analyses.
Neuropsychiatric symptoms were assessed by neuropsychiatry inventory (NPI),15 which is composed of 10 subscales: delusions, hallucinations, agitation, depression, anxiety, euphoria, apathy, disinhibition, irritability and aberrant motor behaviour. Subjects were also assessed by Mini-Mental State Examination (MMSE)16 and Clinical Dementia Rating scale (CDR)17 to measure the severity of global cognitive impairment, and by frontal assessment battery (FAB)18 to measure the possible frontal dysfunction related to AD.19
The PET study was approved by the institutional review board of the National Institute of Radiological Sciences, Japan. Written informed consent was obtained from all subjects or from their spouses or other close family members.
(11C)PIB was synthesised according to a previous method, by reaction of 2-(4′-aminophenyl)-6-hydroxy-benzothiazole and (11C)methyl triflate.20 The product had radiochemical purity greater than 96.2%. Specific activity ranged from 35.4 to 221.8 GBq/µ mol at the time of injection.
PET images were acquired by Siemens ECAT EXACT HR+ scanner (CTI PET Systems, Knoxville, Tennessee, USA) with an axial field of view of 155 mm, providing 63 contiguous 2.46 mm slices, with 5.6 mm transaxial and 5.4 mm axial resolution. A 10 min transmission scan was performed to measure tissue attenuation. Dynamic emission scan data were acquired in three-dimensional mode for a period of 90 min. Subjects were examined with their eyes closed and their ears unplugged in a quiet room, and their heads were restrained with a band extending across the forehead attached to the headrest. An examiner carefully monitored head movement with laser beams during each scan, and corrections were made when necessary. A dose (370±34 MBq in 5 mL) of (11C)PIB was intravenously injected for 60 s by infusion pump into the right cubital vein. The PET measurement protocol was based on a previously reported method,20 ,21 with some modifications. Briefly, a sequence of 19 scans was acquired during 90 min (3×20 s, 3×40 s, 1×1 min, 2×3 min, 5×6 min, 5×10 min) after (11C)PIB injection. All data processing and image reconstruction were performed using standard Siemens software including scatter correction.
MR images were obtained on 1.5 T Intera (Phillips Medical Systems, Best, The Netherlands) on the same day as the PET study. Subjects were scanned with a 3D T1-weighted turbo gradient echo sequence (repetition time (TR) range/echo time (TE) range, 16 msec/5.2 ms; field of view, 256 mm; matrix, 256×256; 196 contiguous axial slices of 1.0 mm thickness).
PET data analysis
Correction of (11C)PIB PET data for partial volume effects was performed with an algorithm implemented in the PMOD software package (PMOD V.3.2; Technologies, Adliswil, Switzerland). This correction is based on the assumption that WM uptake is homogeneous. All brain pixels are classified as WM or GM and sorted into respective segments. Based on these segments and the assumed PET resolution, the spill-out from WM to GM can be estimated and subtracted. Similarly, the spill-out from GM to the surroundings can be estimated and compensated for. The result is a GM image with corrected activity values in all pixels. This method was introduced by Müller–Gärtner et al.22 The parameters were performed as Point Spread Function FWHM: 2.0×2.0×2.0 mm, Correction mode: GM spill-out and spill-in. WM estimation: regression 0.95, and GM threshold: 0.3. All imaging data were then preprocessed and analysed with statistical parametric mapping software (SPM5, Wellcome Department of Cognitive Neurology, London, UK), operating in the Matlab software environment (V.7.10; MathWorks, Natick, Massachusetts, USA). Each T1-weighted MRI scan was coregistered to each PET image, and the spatial normalisation of the MRI images to the SPM5 T1 MRI template was applied to PET images.
A voxel-based DVR was estimated using Logan plot graphical analysis with the cerebellum as reference region by custom software designed by IDL (V.6.0; Jicoux Datasystems, Tokyo, Japan) from (11C)PIB PET data.
First, a voxel-based correlation analysis was performed to assess the relationships between DVR and each NPI subscale using SPM5. Age, gender, education years, MMSE and FAB scores were also included in the model as covariates for all symptoms. Second, we performed voxel-based analysis of covariance (ANCOVA) of the DVR images between patients with AD with and without each neuropsychiatric symptom, adjusting for differences in age, gender, years of education, MMSE and FAB scores. FAB score was used as a covariate to remove the effects of frontal dysfunction, such as executive function and specific brain regions which are solely related to NPI symptoms.19 ,23 Patients with AD with neuropsychiatric symptoms were defined with a cut-off score of 1 for each NPI subscale.
We performed voxel-based morphometry to clarify the effect of GM atrophy, as a past study reported that GM atrophy was associated with some neuropsychiatric symptoms.6 T1-weighted images were segmented into GM, WM and cerebrospinal fluid. In SPM5, spatial normalisation, segmentation and modulation are processed by means of a unified segmentation algorithm using SPM5.24 ,25 The segmented and modulated normalised GM images were smoothed with a 12 mm FWHM Gaussian kernel. Total intracranial volume (TIV) was computed using the native-space tissue maps of each subject. TIV was included in the model as covariate in the voxel-based morphometry (VBM) analyses. The correlations between GM volumes and each NPI subscale were also analysed. Additionally, ANCOVA of GM volumes between patients with AD with and without neuropsychiatric symptoms was performed. In these analyses, besides TIV, age, gender and education years, also included in the model as covariates were the MMSE and FAB scores.
Group comparisons in demographic variables were performed by Fisher's exact test for gender ratio and CDR, and Student t test for the others. Statistical analyses were performed using Statistical Package for the Social Sciences software (SPSS V.19, SPSS, Chicago, Illinois, USA). In SPM voxel-based correlation analyses and ANCOVA analyses, false discovery rate (FDR) corrected, p<0.05 was considered significant. The extent thresholds were defined as >500 in correlation analyses, and >200 in ANCOVA.
Twenty-eight patients with AD (15 women, 13 men), aged from 56 to 85 years, were the participants in this study. Their demographic and clinical data are presented in table 1, and their neuropsychiatric symptoms are shown in table 2. Apathy was the most frequent neuropsychiatric symptom (39.2%), while hallucinations and euphoria were rare (7.1%).
Voxel-based correlation analysis of (11C)PIB DVR images showed that apathy subscale was significantly correlated with (11C)PIB retention in the bilateral middle frontal gyrus, bilateral orbitofrontal gyrus, bilateral medial frontal gyrus, bilateral inferior frontal gyrus, bilateral superior fontal gyrus, bilateral insula, and right anterior cingulate gyrus (FDR, corrected p<0.05, cluster extent >500 voxels, voxel size: 2×2×2 mm) (table 3, figure 1). Additionally, the correlation between NPI apathy scale and prefrontal. (11C)PIB DVR was confirmed in the volume of interest analysis without partial volume correction (r2=0.374, p<0.05, see online supplementary figure E-1). SPM analysis did not show any significant cluster of correlation between any other neuropsychiatric symptom subscale and (11C)PIB retention except apathy.
Voxel-based ANCOVA of the DVR images between patients with AD with and without each of the neuropsychiatric symptoms showed greater (11C)PIB retention in 11 patients with AD with apathy in the bilateral superior frontal gyri, bilateral orbitofrontal gyri, bilateral medial frontal gyri, right middle frontal gyrus, and right middle temporal gyrus than in the 17 without apathy (FDR, corrected p<0.05, cluster extent >200 voxels, voxel size: 2×2×2 mm) (figure 2). There were no significant differences in gender, MMSE, education, FAB, NPI, disease duration and CDR (all p>0.05), and only age of the patients with AD with apathy was significantly lower compared with those without (p<0.01) (table 4). Voxel-based ANCOVA did not show any significant (11C)PIB retention cluster in other neuropsychiatric symptoms except apathy. The overlapping areas between correlation and two-sample analyses were the bilateral orbitofrontal gyri and left superior frontal gyrus (figure 3).
SPM analysis of T1-weighted MRI did not show any significant cluster of correlation between GM volume and any NPI subscale score. Voxel-by-voxel ANCOVA of GM volume between patients with AD with and without neuropsychiatric symptoms did not show any significant cluster in any neuropsychiatric subscale.
We performed an exploratory study of the correlation between (11C)PIB retention and neuropsychiatric symptoms. We introduced a cross-validation approach to detect any robust relationship between them. A positive correlation was found between apathy severity and (11C)PIB retention in the bilateral superior, middle and inferior frontal gyri; bilateral orbitofrontal gyri, bilateral medial frontal gyri, bilateral insula and right anterior cingulate gyrus. Moreover, patients with AD with apathy showed increased (11C)PIB retention in the bilateral superior frontal gyri, bilateral orbitofrontal gyri, bilateral medial frontal gyri, right middle frontal gyrus, and right middle temporal gyrus, when compared with those without apathy. Overlapping areas were the bilateral orbitofrontal gyri and left superior frontal gyrus. We did not find any significant correlation between GM volume and neuropsychiatric symptoms. This is the first study to reveal the correlation between brain Aβ deposition and neuropsychiatric symptoms in patients with AD by the use of (11C)PIB PET with voxel-based whole-brain quantitative analysis.
Several neuroimaging studies have reported the specific brain region responsible for apathy in patients with AD. Previous FDG PET studies reported reduced glucose metabolism in the bilateral anterior cingulate region extending inferiorly to the medial orbitofrontal region and bilateral medial thalamus,7 or hypometabolism in left orbitofrontal regions6 in patients with AD with apathy compared with those without. SPECT studies reported that patients with AD with apathy were associated with hypoperfusion in prefrontal and anterior temporal cortices,8 or the right amygdala, bilateral temporal, right posterior cingulate, right superior frontal, postcentral and left superior temporal gyri.9 An MRI study reported greater cortical thinning in the left caudal anterior cingulate cortex and left lateral orbitofrontal cortex, and left superior and ventrolateral frontal regions in patients with AD with apathy than in those without.26 Another MRI study revealed that apathy was associated with GM density loss in the bilateral anterior cingulate and bilateral frontal cortex, head of the left caudate nucleus and the bilateral putamen in patients with AD.5 There is evidence that regional patterns of amyloid deposition may not match the patterns of glucose metabolism or GM loss in the brain of AD.27 ,28 However, we found a relationship between apathy and (11C)PIB retention in the bilateral orbitofrontal gyri and right superior frontal gyrus of patients with AD, which was in agreement with these prior neuroimaging findings in terms of brain regions.5–9
A neuropathological study demonstrated that chronic apathy and total NPI composite scores were correlated with anterior cingulate neurofibrillary tangles, but not neuritic plaques in AD brains,29 a result discordant with our findings. This discrepancy could be explained by the fact that AD brains were obtained at an advanced disease stage in the neuropathological study, while we measured Aβ deposition in patients with AD in the early disease stage in vivo.
In vitro and in vivo studies have shown that Aβ oligomers reduce glutamatergic synaptic transmission strength and plasticity,30 and in mice transgenic for human amyloid precursor protein, pathologically elevated levels of Aβ promote the formation of pathogenic Aβ oligomers and cause wide fluctuations in the neuronal expression of synaptic activity-regulated genes,31 epileptiform activity and non-convulsive seizures.31 Aβ deposition also increases the proportion of abnormally hyperactive neurons in cortical circuits.32 (11C)PIB has been demonstrated to bind to fibrillar Aβ in diffuse, cored and neuritic plaques, but not to Aβ oligomers,33 whereas, Aβ oligomers aggregate to form Aβ sheets and Aβ plaques. Therefore, brain regions with high (11C)PIB retention could have greater amounts of Aβ oligomers than brain regions with modest (11C)PIB retention at least in the early or preclinical stage of AD. It has been demonstrated that healthy elderly individuals with high (11C)PIB retention show accelerated cortical atrophy, suggesting that the Aβ load is toxic to local brain regions at the preclinical stage of AD. Taken together, prominent Aβ deposition in the orbitofrontal and superior frontal gyri is thought to play a role in the neuropsychiatric symptom of apathy in patients with AD in the present study.
In this study, we demonstrated that Aβ deposition in the orbitofrontal and superior frontal gyri is related to apathy in patients with AD, while GM volume in those areas was not correlated with apathy in patients with AD, or differed between patients with AD with and without apathy. The current hypothetical model suggests that Aβ peptide accumulation is a key early event in the pathophysiological process of AD, followed by synaptic dysfunction identified by FDG PET, and subsequent neural loss demonstrated by structural MRI progresses.34 ,35 Increased (11C)PIB retention in the frontal cortex may be an early marker of cortical damage in AD, and apparent atrophy may follow as the disease progresses. Another possible cause for the absent correlation between GM volume and apathy in this study may be that the AD group without apathy was older than the AD group with apathy.
It has been established that damage of the anterior cingulate circuit presents with apathy,36 and that the neurons project to the ventral striatum that includes the olfactory tubercle.36 The function of the anterior cingulate cortex relates to the initiation and motivational drivers for goal-directed activities, and therefore, damage to this cortical structure would likely lead to a degree of behavioural and cognitive inertia.37 ,38 The orbitofrontal afferents provide information regarding the emotional relevance of external stimuli to the anterior cingulate circuit from internal and external environments,36 and the medial orbitofrontal region has mutual connections with the anterior cingulate and medial dorsal nucleus of the thalamus.39 Cholinergic neurons originate from the basal nucleus of Meynert and project to frontal limbic cortical regions.39 Cholinergic dysfunction may also contribute to apathy, because some studies have demonstrated that cholinesterase inhibitors improved apathy in patients with AD.40
A previous study using MRI demonstrated that delusions were associated with decreased GM density in the left frontal lobe, right frontoparietal cortex and left claustrum, and agitation was associated with decreased GM volumes in the left insula and bilateral anterior cingulate cortex.5 Another study using FDG-PET reported that depression was associated with hypometabolism in bilateral dorsolateral prefrontal regions.6 We did not find any correlation between (11C)PIB retention and other neuropsychiatric symptoms except apathy, which probably resulted from the lack of, or only mild symptoms of, delusion, agitation and depression in the patients with AD in the present study.
The limitations of this exploratory study are its small sample size, and mild NPI symptoms including apathy. Although apathy was the most prevalent symptom, less than half the subjects exhibited apathy in this study. PIB retention is well known to have a predilection for the frontal lobes. Taken together, a correlation observed with overall mild NPI apathy symptoms might suggest that frontal PIB retention is a sensitive marker of apathy symptoms, but not necessarily a very specific one. Further studies in a larger number of patients with AD with NPI symptoms will be needed to confirm the correlation between prefrontal (11C)PIB retention and the severity of apathy, and reveal a correlation with (11C)PIB retention and other neuropsychiatric symptoms besides apathy, which may help to clarify the involved mechanisms and brain circuits, and lead to new insights into the pathophysiology of patients with AD.
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online supplement
TM and HS are joint first authors.
Contributors Study concept/design: TM, HShim, HShin, SHira. Data analysis/interpretation: TM, HShim, HShin, SHira, YE. Manuscript preparation: TM, HShim, HShin, SHira, YE, MY, RF, ST, M-RZ, SK, S-cU, TS.
Funding Dr Kuwabara is funded by Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan (23591267, 20590988, 20591018, 23591269), and a Grant-in-Aid for Scientific Research from the Ministry of Health, Labour and Welfare of Japan (JMA-IIA00046), and serves as Associate Editor of ‘Journal of Neurology, Neurosurgery, and Psychiatry’, Editor of ‘Internal Medicine’, and Editorial Board member of ‘Clinical Neurophysiology’. Dr Ueno is funded by a Health and Labour Science Research Grant from the Japanese Ministry of Health, Labour and Welfare, and a Grant-in-Aid for Scientific Research from the Japanese Ministry of Education, Culture, Sports, Science and Technology (23659567). Dr Suhara is funded by the Ministry of Education, Culture, Sports, Science and Technology of Japan, the Strategic Research Program for Brain Sciences and Molecular Imaging Program, and serves as an Editorial Board member of ‘International Journal of Neuropsychopharmacology’, ‘Psychogeriatrics’ and ‘Current Psychiatry Review’.
Competing interests A part of this study is the result of ‘Integrated research on neuropsychiatric disorders’ carried out under the Strategic Research Program for Brain Sciences by the Ministry of Education, Culture, Sports, Science and Technology of Japan, ‘Japan Advanced Molecular Imaging Program (J-AMP)’ of the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan, a Grant-in-Aid for Scientific Research on Innovative Areas from the Ministry of Education, Culture, Sports, Science and Technology, Japan, and a Grant-in-Aid for Comprehensive Research on Dementia (No. 11103404) from the Ministry of Health, Labour and Welfare, Japan.
Patient consent Obtained.
Ethics approval The Institutional Review Board of the National Institute of Radiological Sciences.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.