Next Article in Journal
Real-World Clinical Outcomes and Replacement Factor VIII Consumption in Patients with Haemophilia A in Italy: A Comparison between Prophylaxis Pre and Post Octocog Alfa (BAY 81-8973)
Previous Article in Journal
White Matter Correlates of Early-Onset Bipolar Illness and Predictors of One-Year Recurrence of Depression in Adults with Bipolar Disorder
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
JCM
Volume 11
Issue 12
10.3390/jcm11123433
jcm-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Predictive Scale for Amyloid PET Positivity Based on Clinical and MRI Variables in Patients with Amnestic Mild Cognitive Impairment

by
Min Young Chun
1,
Geon Ha Kim
2,
Hee Kyung Park
2,3,
Dong Won Yang
4,
SangYun Kim
5,
Seong Hye Choi
6,*,† and
Jee Hyang Jeong
7,*,†
1
Department of Neurology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul 06351, Korea
2
Department of Neurology, Ewha Womans University Mokdong Hospital, Ewha Womans University College of Medicine, Seoul 07985, Korea
3
Division of Psychiatry, Department of Mental Health Care of Older People, University College London, London W1T 7NF, UK
4
Department of Neurology, St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul 06591, Korea
5
Department of Neurology, Seoul National University College of Medicine & Seoul National University Bundang Hospital, Seongnam 13620, Korea
6
Department of Neurology, Inha University School of Medicine, Inchon 22232, Korea
7
Department of Neurology, Ewha Womans University Seoul Hospital, Ewha Womans University College of Medicine, Seoul 07804, Korea
*
Authors to whom correspondence should be addressed.
†
These authors contributed equally to this work.
J. Clin. Med. 2022, 11(12), 3433; https://doi.org/10.3390/jcm11123433
Submission received: 11 April 2022 / Revised: 11 June 2022 / Accepted: 12 June 2022 / Published: 15 June 2022
(This article belongs to the Topic Age-Related Neurodegenerative Diseases and Stroke)
Browse Figures
Versions Notes

Abstract

:
The presence of amyloid-β (Aβ) deposition is considered important in patients with amnestic mild cognitive impairment (aMCI), since they can progress to Alzheimer’s disease dementia. Amyloid positron emission tomography (PET) has been used for detecting Aβ deposition, but its high cost is a significant barrier for clinical usage. Therefore, we aimed to develop a new predictive scale for amyloid PET positivity using easily accessible tools. Overall, 161 aMCI patients were recruited from six memory clinics and underwent neuropsychological tests, brain magnetic resonance imaging (MRI), apolipoprotein E (APOE) genotype testing, and amyloid PET. Among the potential predictors, verbal and visual memory tests, medial temporal lobe atrophy, APOE genotype, and age showed significant differences between the Aβ-positive and Aβ-negative groups and were combined to make a model for predicting amyloid PET positivity with the area under the curve (AUC) of 0.856. Based on the best model, we developed the new predictive scale comprising integers, which had an optimal cutoff score ≥ 3. The new predictive scale was validated in another cohort of 98 participants and showed a good performance with AUC of 0.835. This new predictive scale with accessible variables may be useful for predicting Aβ positivity in aMCI patients in clinical practice.

1. Introduction

Mild cognitive impairment (MCI) has been considered as a transitional state between normal aging and dementia [1]. Since the growth of the elderly population has led to an increase in the prevalence and socioeconomic burden of dementia patients [2], early intervention at the MCI stage and prevention of progression to dementia have become important. In particular, amnestic MCI (aMCI) has a high probability of progression to dementia due to Alzheimer’s disease (AD) [3,4].
AD is neuropathologically diagnosed based on the deposition of amyloid-β (Aβ) plaques and pathologic tau proteins [5]. As the pathologic changes occur in the brain before the symptoms of AD appear [6], modalities for examining the presence of AD pathology, such as cerebrospinal fluid (CSF) analysis and amyloid positron emission tomography (PET) scans, have been developed to detect AD in the preclinical or prodromal stages. The CSF assay has the advantages of being able to measure not only Aβ but also tau protein [7]. On the other hand, amyloid PET scans are noninvasive and have high reliability in longitudinal studies and center-to-center examinations [7].
Although these AD confirmatory diagnostic modalities have been used as enrollment criteria for research, their application in primary care clinical settings is still difficult. CSF testing is an invasive test that the patients may be reluctant to undergo. Amyloid PET scans have the issue of high cost, availability of PET scanners, and radiation exposure [8,9]. Thus, their widespread use is limited to the first-line approach in clinical trials.
Although several longitudinal studies have developed models for predicting the conversion of MCI to AD dementia, AD was estimated clinically, not pathologically [10,11,12,13]. In addition, a few Aβ-positive (Aβ+) AD prediction models that could replace the confirmatory tests have been suggested using cortical thickness and brain volumetry [14,15,16], or using several blood-based markers, including prostatic acid phosphatase, transthyretin, matrix metalloproteinase 10 [17], and plasma Aβ42/Aβ40 [18]. However, these models used variables that are still difficult to apply in clinical practice. Therefore, it will be useful to develop a predictive scale of Aβ positivity using commonly available variables in aMCI patients.
This study aimed to develop a predictive scale to distinguish amyloid PET positivity among aMCI patients using easily accessible variables for routine clinical practice. These variables included demographic and clinical characteristics of aMCI patients such as age, sex, education, detailed neuropsychological tests including verbal and visual memory tests [11,19,20,21,22], medial temporal lobe atrophy (MTA) [23,24,25,26], and apolipoprotein E (APOE) epsilon 4 (ε4) allele [27,28].

2. Materials and Methods

2.1. Participants

A total of 161 participants with aMCI were recruited from six different memory clinics in South Korea.
Diagnosis of aMCI was based on the Petersen’s criteria for MCI [3]: (1) subjective memory complaint by the subject or an informant; (2) preserved general cognitive function; (3) objective memory impairment below −1.0 standard deviation (SD) of age, sex, and education matched norms on either verbal or visual delayed recall tests; and (4) independent performance of daily life activities. The participants had a global Clinical Dementia Rating (CDR) score of 0.5. All participants underwent standard assessments including clinical history, neurological examination, detailed neuropsychological tests, brain magnetic resonance imaging (MRI), amyloid PET, APOE genotype, and laboratory tests. Patients who had a critical illness, stroke, brain tumors, psychiatric diseases, and head trauma with loss of consciousness were excluded from the study population.
The participants were categorized into two groups based on whether the amyloid PET results were Aβ+ or Aβ-negative (Aβ−). For the APOE genotype, the participants were classified as carriers or noncarriers of the APOE ε4 allele.
The present study was approved by the Institutional Review Board for Human Research of the institution in each center and written informed consent was obtained from all participants (ClinicalTrials.gov identifier: NCT02656498).

2.2. Neuropsychological Tests

All participants underwent the Korean-Mini Mental Status Examination (K-MMSE), global CDR, Sum of Boxes of CDR (CDR-SB), and Seoul Neuropsychological Screening Battery (SNSB) [29]. SNSB, a detailed neuropsychological test, is subdivided into the following domains. Memory function domain included Seoul Verbal Learning Test-delayed recall (SVLT-DR) and Rey Complex Figure Test-delayed recall (RCFT-DR). Attention domain included Digit Span Test-forward (DST-forward) and language function domain included Korean-Boston Naming Test (K-BNT). Visuospatial function domain was composed of the RCFT-copy. Frontal/executive function domain was evaluated by Controlled Oral Word Association Test-phonemic (COWAT-phonemic), Korean-Color Word Stroop Test-color reading (K-CWST-CR), Digit Symbol Coding (DSC), and Korean-Trail Making Test-part B (K-TMT-B).
All SNSB variables were converted to z-scores standardized for age, sex, and years of education. Each SNSB variable was considered abnormal when the z-score was lower than −1.0 SD (16th percentiles) of the age, sex, and education-standardized norms. Subjects were classified as aMCI, if either SVLT-DR or RCFT-DR was less than −1.0 SD with independent activity of daily living.

2.3. Axial Visual Rating Scale of Medial Temporal Lobe Atrophy

All participants were evaluated with the axial visual rating scale (aVRS) of MTA according to the method developed by Kim et al. [30] as the inter-rater and intra-rater reliability of aVRS were higher than coronal VRS proposed by Scheltens et al. [31] The MTA of each subject was rated twice by two trained neurologists, who were unaware of the clinical information of the participants. The left and right sides of the MTA were evaluated separately, and the larger value was chosen if the left and right values were different [30,32,33]. According to previous studies which found that an MTA score of 2 or higher was associated with progression from aMCI to dementia [24,33,34], MTA variables were classified as presence or absence of MTA on the basis of aVRS of 2. The kappa value for intra-rater reliability of MTA aVRS was 0.820 (95% confidence interval (CI) 0.779–0.862) and inter-rater reliability was 0.752 (95% CI 0.651–0.845).

2.4. MRI and PET Imaging Acquisition

MRI scans were obtained using 1.5-Tesla (T) MRI (Siemens Magnetom Avanto, Siemens, Erlangen, Germany), 3.0-T MRI (Philips Achieva, Philips, Best, The Netherlands), 3.0-T MRI (Siemens Magnetom VIDA, Siemens, Erlangen, Germany), 3.0-T MRI (GE Signa Architect, Waukesha, WI, USA), 3.0-T MRI (Philips Ingenia, Philips, Best, The Netherlands), or 3.0-T MRI (Verio, Siemens with a Siemens matrix coil, Erlangen, Germany) machines at six different memory clinics. All MR images included T1-weighted axial, T2-weighted axial, and fluid attenuated inversion recovery (FLAIR) axial images. For axial MR images, the images were taken parallel to the line from the anterior commissure to the posterior commissure.
All participants underwent one of the following two amyloid PET tests: 11C-Pittsburgh compound B ([11C] PiB) or 18F-florbetaben ([18F] FBB). Nine subjects underwent [11C] PiB PET (Siemens) and 138 subjects underwent [18F] FBB PET (Biograph mCT, Siemens) scans. The standard uptake value ratio (SUVR) was obtained for [11C] PiB images to determine amyloid PET positivity [35]. [18F] FBB PET was evaluated visually by a trained nuclear medicine specialist in each center using brain amyloid plaque load (BAPL) score.

2.5. Validation of the Predictive Scale

For the validation cohort, 98 aMCI subjects were recruited from the Inha University hospital in South Korea. The inclusion criteria of the validation cohort were the same as those of the original group. To validate the new predictive scale of amyloid PET positivity, the scale was applied to the validation cohort set. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the predictive performance. Area under the curve (AUC) value, sensitivity, and specificity were calculated at the optimal cutoff score of the newly developed scale.

2.6. Statistical Analyses

In this cross-sectional study, Student’s t-test was used to analyze the continuous variables, while the chi-square test was used for the dichotomous variables. Based on the results of the comparison between Aβ+ and Aβ− subjects, potential predictors were selected. Pearson’s correlation analysis was conducted to determine the association between the potential predictors as well as to exclude the correlates.
The selected significant continuous variables were converted to categorical variables to maximize their utility in clinical practice. Since the two groups of Aβ+ and Aβ− subjects were divided based on the average age of 70 years, 1 point was given to those aged 70 or older and 0 point to those under the age of 70. For SNSB, subjects with higher than −1.0 SD were scored as 0, while subjects with lower than −1.0 SD were scored as 1. The presence of MTA was determined and scored as 1 if MTA aVRS was ≥2. If the subject was an APOE-ε4 carrier, a score of 1 was given, while if the subject was an APOE-ε4 noncarrier, a score of 0 was given.
Univariate logistic regression analysis was applied to predict amyloid PET positivity with each potential predictor. Subsequently, these predictors were combined to develop several models using multivariate logistic regression analysis. The ROC curves for all models were analyzed to assess their predictive ability for amyloid PET positivity. The model with the highest AUC value was selected as the most appropriate model for predicting amyloid PET positivity. To detect multicollinearity between the predictors in the model, variance inflation factor (VIF) statistics were calculated. The calibration of each prediction model was assessed using the Hosmer–Lemeshow goodness-of-fit test.
The most appropriate model could generate a new predictive scale based on the regression β coefficients from the multivariate logistic regression analysis. The β coefficients values were adjusted to the nearest integer to develop a simple integer-based modeling scale. ROC curves were constructed to determine the diagnostic ability of the scale. The optimal cutoff score of the scale was calculated using the Youden’s index.
For statistical analysis, Statistical Package for the Social Sciences was used (version 25.0, IBM Corp., Armonk, NY, USA). Statistical significance was defined as a probability value of p < 0.05.

3. Results

3.1. Participants’ Characteristics

The average age of the 161 subjects was 69.88 ± 7.23 years (mean ± SD). Of the subjects, 90 (55.9%) were women and 71 (44.1%) were men; also, the subjects consisted of 70 APOE-ε4 carriers (43.5%). A total of 78 participants (48.4%) showed Aβ positivity in amyloid PET.
Table 1 shows the demographic findings of the subjects according to the amyloid status. There was a significant difference in the age (p = 0.034), but not in sex (p = 0.086) and education (p = 0.685) between the Aβ+ and Aβ− groups. However, there were no significant differences in the cardiovascular risk factors, such as hypertension, diabetes mellitus, hyperlipidemia, and heart disease, as well as stroke and cancer.
Table 2 demonstrates the clinical characteristics such as the cognition status, MTA aVRS, and APOE genotypes of the subjects according to the amyloid status. Among neuropsychological tests, Aβ+ subjects showed significantly lower z-scores of SVLT-DR, RCFT-DR, and K-BNT than Aβ− subjects (SVLT-DR: p < 0.001; RCFT-DR: p < 0.001; K-BNT: p = 0.029). Compared to the Aβ− group, the Aβ+ group was more likely to have both verbal and visual memory impairments rather than either visual or verbal memory alone (p < 0.001). The K-MMSE score of the Aβ+ group was lower than that of the Aβ− group (p < 0.001). Aβ+ subjects had higher global CDR and CDR-SB scores than Aβ− subjects (p = 0.002; p < 0.001). Aβ+ subjects had lower MTA aVRS than Aβ− subjects (p < 0.001). The proportion of APOE-ε4 carriers among Aβ+ subjects was significantly higher than that among Aβ− subjects (p < 0.001).

3.2. Development of a New Predictive Scale for Amyloid PET Positivity

Potential predictors that showed significant differences between the Aβ+ and Aβ−groups were extracted, including SVLT-DR, RCFT-DR, K-BNT, K-MMSE, global CDR, CDR-SB, MTA aVRS, APOE genotype, and age. Correlation analysis of the variables demonstrated that K-MMSE had a correlation with SVLT-DR (p < 0.001), RCFT-DR (p < 0.001), K-BNT (p = 0.030), global CDR (p < 0.001), and CDR-SB (p < 0.001). Global CDR and CDR-SB also showed significant correlation with SVLT-DR, RCFT-DR, K-BNT, and K-MMSE. Thus, K-MMSE, global CDR, and CDR-SB were excluded when constructing the models.
Table 3 shows the results of univariate logistic regression analyses to assess the association between each potential predictor and amyloid PET positivity after adjusting for age. SVLT-DR (adjusted odds ratio (aOR) 3.237, 95% CI 1.407–7.448), RCFT-DR (aOR 2.997, 95% CI 1.414–6.351), MTA aVRS (aOR 3.301, 95% CI 1.611–6.766), APOE-ε4 carriers (aOR 12.253, 95% CI 5.410–27.753), and age (odds ratio (OR) 2.340, 95% CI 1.207–4.538) were associated with amyloid PET positivity. However, K-BNT (p = 0.207) did not demonstrate significance in the univariate analysis. Since all aMCI patients subjects already had either impaired visual or verbal memory, the aOR of both aMCI with both visual and verbal memory impairment was calculated and compared to single aMCI with only visual or verbal memory impairment (aOR 7.579, 95% CI 3.754–15.299).
Both aMCI with both visual and verbal memory impairment, MTA aVRS, and APOE genotype were included in the multivariate analyses after adjustment for age. These predictors were combined to develop the following models:
  • Model 1: included age, both aMCI.
  • Model 2: included age, both aMCI, MTA aVRS.
  • Model 3: included age, both aMCI, APOE genotype.
  • Model 4: included age, both aMCI, MTA aVRS, APOE genotype.
Multivariate analyses showed that predictors in all models had significant ORs (Table 4).
Figure 1 shows the ROC curves and AUC values of the four models. Among the models, model 4 had the highest AUC value (0.856, 95% CI 0.796–0.917) and could be selected as the most appropriate model for predicting amyloid PET positivity (sensitivity: 73.1%, specificity: 86.7%). Model 4 included the following predictors: (1) age older than 70 years (aOR 1.443, 95% CI 0.600–3.473); (2) both aMCI compared to single aMCI (aOR 3.485, 95% CI 1.530–7.940); (3) MTA aVRS value greater than or equal to 2 (aOR 2.668, 95% CI 1.106–6.436); and (5) APOE-ε4 carrier (aOR 9.254, 95% CI 3.971–21.566). VIF values of all variables were <4 (age: 1.293; both aMCI: 1.305; MTA aVRS: 1.381; APOE genotype: 1.241), indicating that there was no multicollinearity among the predictors. The Hosmer−Lemeshow test for model 4 verified that the model fitted well with the data (χ2 = 4.787, p-value = 0.686).
To develop a new predictive scale based on a simple integer scoring system, the regression β coefficients generated from the multivariable analysis of model 4 were simplified. The β coefficients were divided by the β coefficient of MTA aVRS, which was the lowest at 0.981 among all predictors. They were rounded to the nearest integer for the scoring system. Table 5 demonstrates the new predictive scale, configured to score 1 point each for age 70 or older, both aMCI compared to single aMCI, MTA aVRS above 2, and 2 points for APOE-ε4 carrier, based on the value of the β coefficients. In the scale, a higher score had a higher probability of Aβ positivity (score range: 0–5).
ROC curve analysis (Figure 2) shows the performance of the β coefficient-based scoring modeling score to differentiate Aβ+ from Aβ− with an AUC value of 0.848 (95% CI 0.786–0.909). The optimal cutoff point of the modeling score was ≥3 according to the highest Youden’s index with a sensitivity of 72.1% and specificity of 84.3%.

3.3. Validation Analysis

Among the 98 aMCI patients subjects in the validation cohort, 34 (34.7%) were amyloid PET positive. Figure 3 shows the prediction performance of the newly developed scale in the validation cohort with an AUC value of 0.835 (95% CI 0.752–0.917). At the optimal cutoff point of the scale, which was ≥3, sensitivity was 82.4% and specificity was 62.5%.

4. Discussion

Our study developed a new predictive scale for detecting amyloid PET positivity with easily accessible variables in aMCI subjects. To our knowledge, this is the first predictive scale consisting of integers to identify Aβ+ aMCI subjects by using a combination of the neuropsychological memory domain, APOE genotype, and the neurodegenerative marker of MTA VRS. As the prevalence of MCI due to AD increases, this practical scale will be valuable for recognizing Aβ positivity in the prodromal stage.
This predictive scale has the following distinctive strong points. Firstly, the scale can be applied quickly and cost-effectively in clinical practice where amyloid PET evaluation is difficult to access as amyloid PET is a costly and uncommonly used equipment. Several studies have attempted to predict amyloid PET positivity in aMCI participants. These studies required advanced testing tools such as MRI volumetry [14,15,36,37,38] and plasma-based markers [17,39,40], and they have not yet been implemented in clinical trials. However, the predictive scale in the present study comprises readily obtainable variables and is composed of integers. With this new scale, Aβ positivity among aMCI subjects can be predicted easily in clinical settings.
Few studies have attempted to detect Aβ positivity among aMCI subjects with commonly available variables. Kim et al. developed a nomogram for predicting Aβ+ aMCI [41]. This nomogram helps precision medicine at the individual level. However, on the other hand, our predictive scale has the advantage of less variation because each variable is scored as an integer such as 1 or 2 points, regardless of the change in the research subject. Another Aβ positivity prediction model proposed by Lee et al. requires a logit value for each subject to screen, which is challenging to use intuitively in clinical practice [42]. In addition, MTA VRS was not included in the above two studies, although MTA is a characteristic neurodegenerative biomarker of AD [5]. Pekkala et al. also created a model by combining the APOE genotype, cognitive status, and MTA [43]. However, their study analyzed a small number of subjects at 48 and recruited not only aMCI but also non-amnestic MCI (naMCI) subjects. The sensitivity and specificity of the model were not as high as the one in the present study. Hence, the comprehensive predictive scale in our study, consisting of integers, may be more useful and convenient for identifying Aβ positive aMCI patients with a higher statistical power than previous studies.
Secondly, this new scale can predict amyloid PET positivity with high specificity in aMCI patients. In the present study, the specificity was high at 86.7% for the optimal cutoff value of the scale. This study was not intended to screen aMCI patients but to develop a scale for predicting Aβ positivity among those diagnosed as aMCI. It will be helpful not only for implementing clinical care but also in the research field. Studies on AD treatment drugs have not been successful, probably due to the difficulty recruiting Aβ+ aMCI patients. The proportion of Aβ+ patients among MCI patients is known to be only 40–60% [44,45]; the present study also had 47.6% amyloid PET-positive subjects in the entire study group. This predictive scale with high specificity may be helpful in precisely selecting patients with Aβ+ aMCI for future clinical studies.
Third, all subjects underwent amyloid PET scans in this study. There have been several observational longitudinal studies on conversion from MCI to AD dementia, but the subjects were diagnosed clinically with probable AD dementia [10,46,47,48]. The present study differs from other studies since all participants were pathologically checked for Aβ load in the brain by PET. Fourth, since this study was based on multi-centers, the results are high generalizable.
Moreover, the results of this study suggest that all predictors in the scale influenced Aβ accumulation independently of each other. Therefore, it is worth looking into each predictor in this study, which is consistent with the results of previous studies. Above all, the APOE ε4 allele was found to be an essential factor for the development of Aβ positivity, in line with the results of previous studies [27,49,50]. The APOE genotype variable showed the highest aOR in univariate and multivariate analyses (Table 3 and Table 4). The APOE genotype also had the highest regression β coefficient generated from model 4. It had twice the points compared to other variables in the predictive scale (Table 5), suggesting that the APOE ε4 allele is the most influential predictor of Aβ positivity.
In this study, memory domain impairment in neuropsychological tests was also found to be related to Aβ positivity, concurrent with previous studies [11,20,51]. SVLT-DR and RCFT-DR z-scores showed significant associations with amyloid PET positivity in univariate analysis (Table 3), suggesting that verbal and visual memory impairment were associated with amyloid PET positivity in patients with aMCI. Impairment of both verbal and visual memory modalities was also significantly more predictive of Aβ positivity than an impairment of either modality among the patients with aMCI, consistent with the results of previous studies [41,47,52,53,54,55]. Thus, verbal and visual memory scores, especially combined visual and verbal impaired features, can be considered major predictors of Aβ positivity.
Lastly, MTA acted as a useful predictor in this study. MTA can be found not only in normal aging [56], but also in other neurodegenerative diseases such as frontotemporal dementia, hippocampal sclerosis, Lewy body-related pathology, and tau protein aggregation, including argyrophilic grain disease [57,58,59]. There has been a controversy regarding whether the MTA can predict AD. Ten Kate et al. reported that MTA measured by quantitative MRI had a limited added value in detecting Aβ pathology in MCI groups [15]. However, as their study included aMCI and naMCI subjects, the result could be different from those of other studies. In contrast, the present study showed that MTA was significantly associated with Aβ+ aMCI (Table 3 and Table 4). Although another study suggested no significant difference in hippocampal volume atrophy between Alzheimer’s disease and behavioral variant frontotemporal dementia [59], our results showed that MTA could be an essential predictor of Aβ positivity in aMCI patients, if the clinical phenotypes such as age, memory impairment, and the presence of APOE ε4 allele were combined for prediction. Our results suggest that MTA as a neuronal injury biomarker of AD [5] can help to recognize amyloid PET positivity in aMCI subjects, consistent with those of previous studies [23,60,61].
This study used a visual scale rather than quantitative analysis regarding MTA measurement. VRS is known to be as accurate as quantitative analysis, and the intra-rater reliability is high [24,43,62,63]. In addition, the aVRS proposed by Kim et al. was applied because it has higher intra-rater reliability than coronal VRS and axial MR images are easier to obtain in clinical settings [30]. Thus, the VRS based on axial MR images was used in this study as it is readily applicable in clinical practice.
Our study has the following limitations. First, when this predictive scale was validated, the specificity was not very high at 62.5%. It may be because the positive rate of amyloid PET in the validation cohort was 34.7%, which is less than the commonly known 40–60% among aMCI patients [44,45]. Nevertheless, the AUC was high at 0.835, indicating the high predictive ability of this scale. Second, since this scale used the SNSB variable, it applies only when SNSB is performed. Further research is required to assess if this scale can be applied by using the z-score of other neuropsychological tests’ verbal and visual memory tests. Third, subjects underwent MRI and PET scans with different equipment, reducing the consistency of results. The MTA aVRS value could have been measured differently, depending on whether it was 1.5-T MRI or 3.0-T MRI. Nevertheless, the fact that the six centers used a variety of machines may be helpful to further generalize the results of this study. Fourth, amyloid positive and negative results of [18F] FBB PET were evaluated by visual interpretation, not by standardized uptake value-based quantification. However, the diagnostic accuracy of BAPL and SUVR in [18F] FBB PET was relatively consistent in previous studies [64,65]. A nuclear medicine specialist evaluated all [18F] FBB PET images at each center. For these reasons, the accuracy of [18F] FBB PET results was expected to be high. Furthermore, our additional analysis showed that the BAPL score was correlated with the scores of the new predictive scale (Pearson’s correlation analysis: r = 0.678, p < 0.001). Further study of the associations between the BAPL score and the scale score can be investigated. Finally, although memory impairment in MCI is the most distinguishing feature of Aβ positivity [20,66], Aβ positivity in aMCI patients is not necessarily associated with AD. The presence of Aβ positivity reflects only Alzheimer’s pathologic change, and the possibility of other comorbidities such as cerebrovascular disease, frontotemporal lobar degeneration, or α-synucleinopathies cannot be excluded [67].
Despite these limitations, the advantages of this study are as follows: (1) both aMCI, MTA aVRS, and APOE genotype were revealed as the critical predictors of Aβ positivity in subjects with aMCI, and (2) the new predictive scale comprising these factors can be easily applied in clinical practice.

5. Conclusions

This study shows that verbal and visual memory impairment, MTA, and the APOE ε4 allele are associated with Aβ load in the brain of aMCI patients. Integration of these predictors was used to develop a new predictive scale of integers to detect Aβ positivity in aMCI patients with a cutoff score of ≥3. The scale can be used conveniently to predict AD and prognosis of aMCI patients in clinical practice.

Author Contributions

Conceptualization: M.Y.C., G.H.K. and J.H.J.; methodology: M.Y.C. and G.H.K.; formal analysis and investigation: M.Y.C. and H.K.P.; writing—original draft preparation: M.Y.C. and G.H.K.; writing—review and editing: J.H.J., H.K.P., D.W.Y., S.K., J.H.J. and S.H.C. Funding acquisition: J.H.J., D.W.Y. and S.H.C.; Supervision: J.H.J. and S.H.C. All authors contributed to manuscript revision, and read and approved the submitted version. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Ministry of Health & Welfare, Republic of Korea (the grants of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute, grant number: HU20C0271 and HU21C0016) and the Ministry of Science and ICT (the Original Technology Research Program for Brain Science through the National Research Foundation of Korea, grant number: NRF-2018M3C7A1057137 and NRF-2020M3E5D2A01084721).

Institutional Review Board Statement

The present study was approved by the Institutional Review Board for Human Research at three institutions: Ewha Womans University Mokdong Hospital (IRB no. 2018-08-005), Ewha Womans University Seoul Hospital (IRB No. 2020-09-006), and Inha University Hospital (IRB No. 2020-10-014). Written informed consent was obtained from all the participants.

Informed Consent Statement

Informed written consent was obtained from all participants at each center in this study.

Data Availability Statement

The raw data supporting the conclusion of this article will be made available by the authors without undue reservation.

Conflicts of Interest

The authors have no conflict of interest to report.

References

  1. Petersen, R.C.; Doody, R.; Kurz, A.; Mohs, R.C.; Morris, J.C.; Rabins, P.V.; Ritchie, K.; Rossor, M.; Thal, L.; Winblad, B. Current concepts in mild cognitive impairment. Arch. Neurol. 2001, 58, 1985–1992. [Google Scholar] [CrossRef] [PubMed]
  2. GBD Dementia Collaborators. Global, regional, and national burden of Alzheimer’s disease and other dementias, 1990–2016: A systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol. 2019, 18, 88–106. [Google Scholar] [CrossRef] [Green Version]
  3. Petersen, R.C. Mild cognitive impairment as a diagnostic entity. J. Intern. Med. 2004, 256, 183–194. [Google Scholar] [CrossRef] [PubMed]
  4. Sarazin, M.; Berr, C.; De Rotrou, J.; Fabrigoule, C.; Pasquier, F.; Legrain, S.; Michel, B.; Puel, M.; Volteau, M.; Touchon, J.; et al. Amnestic syndrome of the medial temporal type identifies prodromal AD: A longitudinal study. Neurology 2007, 69, 1859–1867. [Google Scholar] [CrossRef] [PubMed]
  5. Jack, C.R., Jr.; Bennett, D.A.; Blennow, K.; Carrillo, M.C.; Dunn, B.; Haeberlein, S.B.; Holtzman, D.M.; Jagust, W.; Jessen, F.; Karlawish, J.; et al. NIA-AA Research Framework: Toward a biological definition of Alzheimer’s disease. Alzheimer’s Dement. 2018, 14, 535–562. [Google Scholar] [CrossRef]
  6. Dubois, B.; Hampel, H.; Feldman, H.H.; Scheltens, P.; Aisen, P.; Andrieu, S.; Bakardjian, H.; Benali, H.; Bertram, L.; Blennow, K.; et al. Preclinical Alzheimer’s disease: Definition, natural history, and diagnostic criteria. Alzheimer’s Dement 2016, 12, 292–323. [Google Scholar] [CrossRef]
  7. Palmqvist, S.; Zetterberg, H.; Mattsson, N.; Johansson, P.; Minthon, L.; Blennow, K.; Olsson, M.; Hansson, O.; Initiative, A.s.D.N.; Group, S.B.S. Detailed comparison of amyloid PET and CSF biomarkers for identifying early Alzheimer disease. Neurology 2015, 85, 1240–1249. [Google Scholar] [CrossRef]
  8. Johnson, K.A.; Minoshima, S.; Bohnen, N.I.; Donohoe, K.J.; Foster, N.L.; Herscovitch, P.; Karlawish, J.H.; Rowe, C.C.; Carrillo, M.C.; Hartley, D.M.; et al. Appropriate use criteria for amyloid PET: A report of the Amyloid Imaging Task Force, the Society of Nuclear Medicine and Molecular Imaging, and the Alzheimer’s Association. J. Nucl. Med. 2013, 54, 476–490. [Google Scholar] [CrossRef] [Green Version]
  9. Teipel, S.; Drzezga, A.; Grothe, M.J.; Barthel, H.; Chételat, G.; Schuff, N.; Skudlarski, P.; Cavedo, E.; Frisoni, G.B.; Hoffmann, W. Multimodal imaging in Alzheimer’s disease: Validity and usefulness for early detection. Lancet Neurol. 2015, 14, 1037–1053. [Google Scholar] [CrossRef]
  10. Daly, E.; Zaitchik, D.; Copeland, M.; Schmahmann, J.; Gunther, J.; Albert, M. Predicting conversion to Alzheimer disease using standardized clinical information. Arch. Neurol. 2000, 57, 675–680. [Google Scholar] [CrossRef] [Green Version]
  11. De Simone, M.S.; Perri, R.; Fadda, L.; Caltagirone, C.; Carlesimo, G.A. Predicting progression to Alzheimer’s disease in subjects with amnestic mild cognitive impairment using performance on recall and recognition tests. J. Neurol 2019, 266, 102–111. [Google Scholar] [CrossRef] [PubMed]
  12. Hong, Y.J.; Yoon, B.; Shim, Y.S.; Kim, S.O.; Kim, H.J.; Choi, S.H.; Jeong, J.H.; Yoon, S.J.; Yang, D.W.; Lee, J.H. Predictors of Clinical Progression of Subjective Memory Impairment in Elderly Subjects: Data from the Clinical Research Centers for Dementia of South Korea (CREDOS). Dement. Geriatr. Cogn. Disord. 2015, 40, 158–165. [Google Scholar] [CrossRef]
  13. Tabert, M.H.; Manly, J.J.; Liu, X.; Pelton, G.H.; Rosenblum, S.; Jacobs, M.; Zamora, D.; Goodkind, M.; Bell, K.; Stern, Y.; et al. Neuropsychological prediction of conversion to Alzheimer disease in patients with mild cognitive impairment. Arch. Gen. Psychiatry 2006, 63, 916–924. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  14. Moradi, E.; Pepe, A.; Gaser, C.; Huttunen, H.; Tohka, J.; Alzheimer’s Disease Neuroimaging Initiative. Machine learning framework for early MRI-based Alzheimer’s conversion prediction in MCI subjects. Neuroimage 2015, 104, 398–412. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  15. Ten Kate, M.; Redolfi, A.; Peira, E.; Bos, I.; Vos, S.J.; Vandenberghe, R.; Gabel, S.; Schaeverbeke, J.; Scheltens, P.; Blin, O.; et al. MRI predictors of amyloid pathology: Results from the EMIF-AD Multimodal Biomarker Discovery study. Alzheimer’s Res. Ther. 2018, 10, 100. [Google Scholar] [CrossRef]
  16. Ye, B.S.; Seo, S.W.; Kim, C.H.; Jeon, S.; Kim, G.H.; Noh, Y.; Cho, H.; Yoon, C.W.; Kim, H.J.; Jang, E.Y.; et al. Hippocampal and cortical atrophy in amyloid-negative mild cognitive impairments: Comparison with amyloid-positive mild cognitive impairment. Neurobiol. Aging 2014, 35, 291–300. [Google Scholar] [CrossRef]
  17. Haghighi, M.; Smith, A.; Morgan, D.; Small, B.; Huang, S. Identifying cost-effective predictive rules of amyloid-beta level by integrating neuropsychological tests and plasma-based markers. J. Alzheimer’s Dis. 2015, 43, 1261–1270. [Google Scholar] [CrossRef]
  18. Palmqvist, S.; Insel, P.S.; Zetterberg, H.; Blennow, K.; Brix, B.; Stomrud, E.; Alzheimer’s Disease Neuroimaging Initiative; Swedish BioFINDER Study; Mattsson, N.; Hansson, O. Accurate risk estimation of beta-amyloid positivity to identify prodromal Alzheimer’s disease: Cross-validation study of practical algorithms. Alzheimer’s Dement. 2019, 15, 194–204. [Google Scholar] [CrossRef]
  19. Alladi, S.; Arnold, R.; Mitchell, J.; Nestor, P.J.; Hodges, J.R. Mild cognitive impairment: Applicability of research criteria in a memory clinic and characterization of cognitive profile. Psychol. Med. 2006, 36, 507–515. [Google Scholar] [CrossRef]
  20. Bahar-Fuchs, A.; Villemagne, V.; Ong, K.; Chetelat, G.; Lamb, F.; Reininger, C.B.; Woodward, M.; Rowe, C.C. Prediction of amyloid-beta pathology in amnestic mild cognitive impairment with neuropsychological tests. J. Alzheimer’s Dis. 2013, 33, 451–462. [Google Scholar] [CrossRef]
  21. Lim, Y.Y.; Maruff, P.; Pietrzak, R.H.; Ames, D.; Ellis, K.A.; Harrington, K.; Lautenschlager, N.T.; Szoeke, C.; Martins, R.N.; Masters, C.L.; et al. Effect of amyloid on memory and non-memory decline from preclinical to clinical Alzheimer’s disease. Brain 2014, 137, 221–231. [Google Scholar] [CrossRef]
  22. Summers, M.J.; Saunders, N.L. Neuropsychological measures predict decline to Alzheimer’s dementia from mild cognitive impairment. Neuropsychology 2012, 26, 498–508. [Google Scholar] [CrossRef]
  23. Clerx, L.; van Rossum, I.A.; Burns, L.; Knol, D.L.; Scheltens, P.; Verhey, F.; Aalten, P.; Lapuerta, P.; van de Pol, L.; van Schijndel, R.; et al. Measurements of medial temporal lobe atrophy for prediction of Alzheimer’s disease in subjects with mild cognitive impairment. Neurobiol. Aging 2013, 34, 2003–2013. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  24. DeCarli, C.; Frisoni, G.B.; Clark, C.M.; Harvey, D.; Grundman, M.; Petersen, R.C.; Thal, L.J.; Jin, S.; Jack, C.R., Jr.; Scheltens, P.; et al. Qualitative estimates of medial temporal atrophy as a predictor of progression from mild cognitive impairment to dementia. Arch. Neurol. 2007, 64, 108–115. [Google Scholar] [CrossRef] [PubMed]
  25. Risacher, S.L.; Saykin, A.J.; West, J.D.; Shen, L.; Firpi, H.A.; McDonald, B.C.; Alzheimer’s Disease Neuroimaging Initiative. Baseline MRI predictors of conversion from MCI to probable AD in the ADNI cohort. Curr. Alzheimer Res. 2009, 6, 347–361. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  26. Visser, P.J.; Verhey, F.R.; Hofman, P.A.; Scheltens, P.; Jolles, J. Medial temporal lobe atrophy predicts Alzheimer’s disease in patients with minor cognitive impairment. J. Neurol. Neurosurg. Psychiatry 2002, 72, 491–497. [Google Scholar]
  27. Petersen, R.C.; Smith, G.E.; Ivnik, R.J.; Tangalos, E.G.; Schaid, D.J.; Thibodeau, S.N.; Kokmen, E.; Waring, S.C.; Kurland, L.T. Apolipoprotein E status as a predictor of the development of Alzheimer’s disease in memory-impaired individuals. JAMA 1995, 273, 1274–1278. [Google Scholar] [CrossRef]
  28. Risacher, S.L.; Kim, S.; Shen, L.; Nho, K.; Foroud, T.; Green, R.C.; Petersen, R.C.; Jack, C.R., Jr.; Aisen, P.S.; Koeppe, R.A.; et al. The role of apolipoprotein E (APOE) genotype in early mild cognitive impairment (E-MCI). Front. Aging Neurosci. 2013, 5, 11. [Google Scholar] [CrossRef] [Green Version]
  29. Ahn, H.J.; Chin, J.; Park, A.; Lee, B.H.; Suh, M.K.; Seo, S.W.; Na, D.L. Seoul Neuropsychological Screening Battery-dementia version (SNSB-D): A useful tool for assessing and monitoring cognitive impairments in dementia patients. J. Korean Med. Sci 2010, 25, 1071–1076. [Google Scholar] [CrossRef] [Green Version]
  30. Kim, G.H.; Kim, J.E.; Choi, K.G.; Lim, S.M.; Lee, J.M.; Na, D.L.; Jeong, J.H. T1-weighted axial visual rating scale for an assessment of medial temporal atrophy in Alzheimer’s disease. J. Alzheimer’s Dis. 2014, 41, 169–178. [Google Scholar] [CrossRef] [Green Version]
  31. Scheltens, P.; Leys, D.; Barkhof, F.; Huglo, D.; Weinstein, H.C.; Vermersch, P.; Kuiper, M.; Steinling, M.; Wolters, E.C.; Valk, J. Atrophy of medial temporal lobes on MRI in “probable” Alzheimer’s disease and normal ageing: Diagnostic value and neuropsychological correlates. J. Neurol. Neurosurg. Psychiatry 1992, 55, 967–972. [Google Scholar] [CrossRef] [PubMed]
  32. Choi, G.S.; Kim, G.H.; Choi, J.H.; Hwang, J.; Kwon, E.; Lee, S.A.; Kong, K.A.; Kang, H.J.; Yoon, B.; Kim, B.C.; et al. Age-Specific Cutoff Scores on a T1-Weighted Axial Medial Temporal-Lobe Atrophy Visual Rating Scale in Alzheimer’s Disease Using Clinical Research Center for Dementia of South Korea Data. J. Clin. Neurol. 2018, 14, 275–282. [Google Scholar] [CrossRef] [PubMed]
  33. Harper, L.; Barkhof, F.; Fox, N.C.; Schott, J.M. Using visual rating to diagnose dementia: A critical evaluation of MRI atrophy scales. J. Neurol. Neurosurg. Psychiatry 2015, 86, 1225–1233. [Google Scholar] [CrossRef] [PubMed]
  34. Pereira, J.B.; Cavallin, L.; Spulber, G.; Aguilar, C.; Mecocci, P.; Vellas, B.; Tsolaki, M.; Kloszewska, I.; Soininen, H.; Spenger, C.; et al. Influence of age, disease onset and ApoE4 on visual medial temporal lobe atrophy cut-offs. J. Intern. Med. 2014, 275, 317–330. [Google Scholar] [CrossRef]
  35. Hwang, J.; Jeong, J.H.; Yoon, S.J.; Park, K.W.; Kim, E.J.; Yoon, B.; Jang, J.W.; Kim, H.J.; Hong, J.Y.; Lee, J.M.; et al. Clinical and Biomarker Characteristics According to Clinical Spectrum of Alzheimer’s Disease (AD) in the Validation Cohort of Korean Brain Aging Study for the Early Diagnosis and Prediction of AD. J. Clin. Med. 2019, 8, 341. [Google Scholar] [CrossRef] [Green Version]
  36. Gupta, Y.; Lama, R.K.; Kwon, G.R.; Alzheimer’s Disease Neuroimaging Initiative. Prediction and Classification of Alzheimer’s Disease Based on Combined Features From Apolipoprotein-E Genotype, Cerebrospinal Fluid, MR, and FDG-PET Imaging Biomarkers. Front. Comput. Neurosci. 2019, 13, 72. [Google Scholar] [CrossRef] [Green Version]
  37. Long, X.; Chen, L.; Jiang, C.; Zhang, L.; Alzheimer’s Disease Neuroimaging Initiative. Prediction and classification of Alzheimer disease based on quantification of MRI deformation. PLoS ONE 2017, 12, e0173372. [Google Scholar] [CrossRef]
  38. Tosun, D.; Joshi, S.; Weiner, M.W.; Alzheimer’s Disease Neuroimaging Initiative. Neuroimaging predictors of brain amyloidosis in mild cognitive impairment. Ann. Neurol. 2013, 74, 188–198. [Google Scholar] [CrossRef] [Green Version]
  39. Apostolova, L.G.; Hwang, K.S.; Avila, D.; Elashoff, D.; Kohannim, O.; Teng, E.; Sokolow, S.; Jack, C.R.; Jagust, W.J.; Shaw, L.; et al. Brain amyloidosis ascertainment from cognitive, imaging, and peripheral blood protein measures. Neurology 2015, 84, 729–737. [Google Scholar] [CrossRef] [Green Version]
  40. Nakamura, A.; Kaneko, N.; Villemagne, V.L.; Kato, T.; Doecke, J.; Dore, V.; Fowler, C.; Li, Q.X.; Martins, R.; Rowe, C.; et al. High performance plasma amyloid-beta biomarkers for Alzheimer’s disease. Nature 2018, 554, 249–254. [Google Scholar] [CrossRef]
  41. Kim, S.E.; Woo, S.; Kim, S.W.; Chin, J.; Kim, H.J.; Lee, B.I.; Park, J.; Park, K.W.; Kang, D.Y.; Noh, Y.; et al. A Nomogram for Predicting Amyloid PET Positivity in Amnestic Mild Cognitive Impairment. J. Alzheimer’s Dis. 2018, 66, 681–691. [Google Scholar] [CrossRef] [PubMed]
  42. Lee, J.H.; Byun, M.S.; Yi, D.; Sohn, B.K.; Jeon, S.Y.; Lee, Y.; Lee, J.Y.; Kim, Y.K.; Lee, Y.S.; Lee, D.Y. Prediction of Cerebral Amyloid With Common Information Obtained From Memory Clinic Practice. Front. Aging Neurosci. 2018, 10, 309. [Google Scholar] [CrossRef] [PubMed]
  43. Pekkala, T.; Hall, A.; Ngandu, T.; van Gils, M.; Helisalmi, S.; Hanninen, T.; Kemppainen, N.; Liu, Y.; Lotjonen, J.; Paajanen, T.; et al. Detecting Amyloid Positivity in Elderly With Increased Risk of Cognitive Decline. Front. Aging Neurosci. 2020, 12, 228. [Google Scholar] [CrossRef]
  44. Landau, S.M.; Mintun, M.A.; Joshi, A.D.; Koeppe, R.A.; Petersen, R.C.; Aisen, P.S.; Weiner, M.W.; Jagust, W.J.; Alzheimer’s Disease Neuroimaging Initiative. Amyloid deposition, hypometabolism, and longitudinal cognitive decline. Ann. Neurol. 2012, 72, 578–586. [Google Scholar] [CrossRef] [PubMed]
  45. Roberts, R.O.; Aakre, J.A.; Kremers, W.K.; Vassilaki, M.; Knopman, D.S.; Mielke, M.M.; Alhurani, R.; Geda, Y.E.; Machulda, M.M.; Coloma, P.; et al. Prevalence and Outcomes of Amyloid Positivity Among Persons Without Dementia in a Longitudinal, Population-Based Setting. JAMA Neurol. 2018, 75, 970–979. [Google Scholar] [CrossRef] [PubMed]
  46. Bruscoli, M.; Lovestone, S. Is MCI really just early dementia? A systematic review of conversion studies. Int. Psychogeriatr. 2004, 16, 129–140. [Google Scholar] [CrossRef]
  47. Jang, H.; Ye, B.S.; Woo, S.; Kim, S.W.; Chin, J.; Choi, S.H.; Jeong, J.H.; Yoon, S.J.; Yoon, B.; Park, K.W.; et al. Prediction Model of Conversion to Dementia Risk in Subjects with Amnestic Mild Cognitive Impairment: A Longitudinal, Multi-Center Clinic-Based Study. J. Alzheimer’s Dis. 2017, 60, 1579–1587. [Google Scholar] [CrossRef]
  48. Rabin, L.A.; Pare, N.; Saykin, A.J.; Brown, M.J.; Wishart, H.A.; Flashman, L.A.; Santulli, R.B. Differential memory test sensitivity for diagnosing amnestic mild cognitive impairment and predicting conversion to Alzheimer’s disease. Aging Neuropsychol. Cogn. 2009, 16, 357–376. [Google Scholar] [CrossRef] [Green Version]
  49. Liu, C.C.; Liu, C.C.; Kanekiyo, T.; Xu, H.; Bu, G. Apolipoprotein E and Alzheimer disease: Risk, mechanisms and therapy. Nat. Rev. Neurol. 2013, 9, 106–118. [Google Scholar] [CrossRef] [Green Version]
  50. Lopresti, B.J.; Campbell, E.M.; Yu, Z.; Anderson, S.J.; Cohen, A.D.; Minhas, D.S.; Snitz, B.E.; Royse, S.K.; Becker, C.R.; Aizenstein, H.J.; et al. Influence of apolipoprotein-E genotype on brain amyloid load and longitudinal trajectories. Neurobiol. Aging 2020, 94, 111–120. [Google Scholar] [CrossRef]
  51. Mormino, E.C.; Betensky, R.A.; Hedden, T.; Schultz, A.P.; Amariglio, R.E.; Rentz, D.M.; Johnson, K.A.; Sperling, R.A. Synergistic effect of beta-amyloid and neurodegeneration on cognitive decline in clinically normal individuals. JAMA Neurol. 2014, 71, 1379–1385. [Google Scholar] [CrossRef] [PubMed]
  52. Bilgel, M.; An, Y.; Helphrey, J.; Elkins, W.; Gomez, G.; Wong, D.F.; Davatzikos, C.; Ferrucci, L.; Resnick, S.M. Effects of amyloid pathology and neurodegeneration on cognitive change in cognitively normal adults. Brain 2018, 141, 2475–2485. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  53. Kim, J.Y.; Lim, J.H.; Jeong, Y.J.; Kang, D.Y.; Park, K.W. The Effect of Clinical Characteristics and Subtypes on Amyloid Positivity in Patients with Amnestic Mild Cognitive Impairment. Dement. Neurocogn. Disord. 2019, 18, 130–137. [Google Scholar] [CrossRef] [PubMed]
  54. Kim, M.J.; Im, K.; Lee, J.M.; Park, A.; Chin, J.; Kim, G.H.; Kim, J.H.; Roh, J.H.; Seo, S.W.; Na, D.L. Cortical thinning in verbal, visual, and both memory-predominant mild cognitive impairment. Alzheimer Dis. Assoc. Disord. 2011, 25, 242–249. [Google Scholar] [CrossRef] [PubMed]
  55. Oltra-Cucarella, J.; Sanchez-SanSegundo, M.; Lipnicki, D.M.; Crawford, J.D.; Lipton, R.B.; Katz, M.J.; Zammit, A.R.; Scarmeas, N.; Dardiotis, E.; Kosmidis, M.H.; et al. Visual memory tests enhance the identification of amnestic MCI cases at greater risk of Alzheimer’s disease. Int. Psychogeriatr. 2019, 31, 997–1006. [Google Scholar] [CrossRef]
  56. van de Pol, L.A.; Hensel, A.; Barkhof, F.; Gertz, H.J.; Scheltens, P.; van der Flier, W.M. Hippocampal atrophy in Alzheimer disease: Age matters. Neurology 2006, 66, 236–238. [Google Scholar] [CrossRef]
  57. Barkhof, F.; Polvikoski, T.M.; van Straaten, E.C.; Kalaria, R.N.; Sulkava, R.; Aronen, H.J.; Niinisto, L.; Rastas, S.; Oinas, M.; Scheltens, P.; et al. The significance of medial temporal lobe atrophy: A postmortem MRI study in the very old. Neurology 2007, 69, 1521–1527. [Google Scholar] [CrossRef]
  58. Marks, S.M.; Lockhart, S.N.; Baker, S.L.; Jagust, W.J. Tau and beta-Amyloid Are Associated with Medial Temporal Lobe Structure, Function, and Memory Encoding in Normal Aging. J. Neurosci. 2017, 37, 3192–3201. [Google Scholar] [CrossRef]
  59. De Souza, L.C.; Chupin, M.; Bertoux, M.; Lehéricy, S.; Dubois, B.; Lamari, F.; Le Ber, I.; Bottlaender, M.; Colliot, O.; Sarazin, M. Is hippocampal volume a good marker to differentiate Alzheimer’s disease from frontotemporal dementia? J. Alzheimer’s Dis. 2013, 36, 57–66. [Google Scholar] [CrossRef]
  60. Burton, E.J.; Barber, R.; Mukaetova-Ladinska, E.B.; Robson, J.; Perry, R.H.; Jaros, E.; Kalaria, R.N.; O’Brien, J.T. Medial temporal lobe atrophy on MRI differentiates Alzheimer’s disease from dementia with Lewy bodies and vascular cognitive impairment: A prospective study with pathological verification of diagnosis. Brain 2009, 132, 195–203. [Google Scholar] [CrossRef] [Green Version]
  61. Harper, L.; Fumagalli, G.G.; Barkhof, F.; Scheltens, P.; O’Brien, J.T.; Bouwman, F.; Burton, E.J.; Rohrer, J.D.; Fox, N.C.; Ridgway, G.R.; et al. MRI visual rating scales in the diagnosis of dementia: Evaluation in 184 post-mortem confirmed cases. Brain 2016, 139, 1211–1225. [Google Scholar] [CrossRef] [PubMed]
  62. Varon, D.; Barker, W.; Loewenstein, D.; Greig, M.; Bohorquez, A.; Santos, I.; Shen, Q.; Harper, M.; Vallejo-Luces, T.; Duara, R.; et al. Visual rating and volumetric measurement of medial temporal atrophy in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort: Baseline diagnosis and the prediction of MCI outcome. Int. J. Geriatr. Psychiatry 2015, 30, 192–200. [Google Scholar] [CrossRef] [PubMed]
  63. Wahlund, L.O.; Julin, P.; Johansson, S.E.; Scheltens, P. Visual rating and volumetry of the medial temporal lobe on magnetic resonance imaging in dementia: A comparative study. J. Neurol. Neurosurg. Psychiatry 2000, 69, 630–635. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  64. Bullich, S.; Seibyl, J.; Catafau, A.M.; Jovalekic, A.; Koglin, N.; Barthel, H.; Sabri, O.; De Santi, S. Optimized classification of (18)F-Florbetaben PET scans as positive and negative using an SUVR quantitative approach and comparison to visual assessment. NeuroImage Clin. 2017, 15, 325–332. [Google Scholar] [CrossRef] [PubMed]
  65. Seibyl, J.; Catafau, A.M.; Barthel, H.; Ishii, K.; Rowe, C.C.; Leverenz, J.B.; Ghetti, B.; Ironside, J.W.; Takao, M.; Akatsu, H.; et al. Impact of Training Method on the Robustness of the Visual Assessment of 18F-Florbetaben PET Scans: Results from a Phase-3 Study. J. Nucl. Med. 2016, 57, 900–906. [Google Scholar] [CrossRef] [Green Version]
  66. Kandel, B.M.; Avants, B.B.; Gee, J.C.; Arnold, S.E.; Wolk, D.A.; Initiative, A.s.D.N. Neuropsychological testing predicts cerebrospinal fluid amyloid-β in mild cognitive impairment. J. Alzheimer’s Dis. 2015, 46, 901–912. [Google Scholar] [CrossRef] [Green Version]
  67. Toledo, J.B.; Arnold, S.E.; Raible, K.; Brettschneider, J.; Xie, S.X.; Grossman, M.; Monsell, S.E.; Kukull, W.A.; Trojanowski, J.Q. Contribution of cerebrovascular disease in autopsy confirmed neurodegenerative disease cases in the National Alzheimer’s Coordinating Centre. Brain 2013, 136, 2697–2706. [Google Scholar] [CrossRef]
Jcm 11 03433 g001 550
Figure 1. Receiver operating characteristic (ROC) curves of the models to predict amyloid positron emission tomography (PET) positivity. The ROC curves and the corresponding area under the curve (AUC) with 95% confidence interval (CI), sensitivity (SN), and specificity (SP) are shown for all the models. Among the four models, model 4 demonstrates the highest AUC value of 0.856, along with a sensitivity of 73.1% and specificity of 86.7%.
Figure 1. Receiver operating characteristic (ROC) curves of the models to predict amyloid positron emission tomography (PET) positivity. The ROC curves and the corresponding area under the curve (AUC) with 95% confidence interval (CI), sensitivity (SN), and specificity (SP) are shown for all the models. Among the four models, model 4 demonstrates the highest AUC value of 0.856, along with a sensitivity of 73.1% and specificity of 86.7%.
Jcm 11 03433 g001
Jcm 11 03433 g002 550
Figure 2. Receiver operating characteristic (ROC) curve of the new predictive scale for amyloid PET positivity. The ROC curve and corresponding area under the curve (AUC) with 95% confidence interval (CI), sensitivity (SN), and specificity (SP) are shown. The AUC value of the scale was high at 0.847, and the sensitivity and specificity were also high at 72.1% and 84.3%, respectively.
Figure 2. Receiver operating characteristic (ROC) curve of the new predictive scale for amyloid PET positivity. The ROC curve and corresponding area under the curve (AUC) with 95% confidence interval (CI), sensitivity (SN), and specificity (SP) are shown. The AUC value of the scale was high at 0.847, and the sensitivity and specificity were also high at 72.1% and 84.3%, respectively.
Jcm 11 03433 g002
Jcm 11 03433 g003 550
Figure 3. Receiver operating characteristic (ROC) curve of the new predictive scale in the validation cohort. The ROC curve and the corresponding area under the curve (AUC) with 95% confidence interval (CI), sensitivity (SN), and specificity (SP) are shown. In the validation cohort, the AUC value was as high as 0.835, and the sensitivity and specificity were 82.4% and 62.5%, respectively.
Figure 3. Receiver operating characteristic (ROC) curve of the new predictive scale in the validation cohort. The ROC curve and the corresponding area under the curve (AUC) with 95% confidence interval (CI), sensitivity (SN), and specificity (SP) are shown. In the validation cohort, the AUC value was as high as 0.835, and the sensitivity and specificity were 82.4% and 62.5%, respectively.
Jcm 11 03433 g003
Table
Table 1. Participants’ demographic findings according to the amyloid status.
Table 1. Participants’ demographic findings according to the amyloid status.
Total
(n = 161)		Aβ−
(n = 83)		Aβ+
(n = 78)		p-Value
Age (years)69.88 ± 7.2368.71 ± 7.5771.12 ± 6.670.034
Sex (men:women)71:9042:4129:490.086
Education (years)11.09 ± 4.3611.23 ± 4.3810.95 ± 4.360.685
Family history of dementia48/153 (31.4%)27/81 (33.3%)21/72 (29.2%)0.579
Hypertension 74/152 (49.0%)40/76 (53.3%)34/76 (44.7%)0.291
Diabetes mellitus33/152 (21.9%)20/76 (26.7%)13/76 (17.1%)0.155
Hyperlipidemia63/152 (41.4%)34/76 (44.7%)29/76 (38.2%)0.410
Heart disease4/151 (2.6%)3/75 (4.0%)1/76 (1.3%)0.367
Stroke2/151 (1.3%)2/75 (2.7%)0/76 (0.0%)0.245
Cancer13/137 (9.5%)3/65 (4.6%)10/72 (13.9%)0.064
Abbreviations: Aβ, amyloid-β; Aβ−, amyloid PET negative, Aβ+, amyloid PET positive.
Table
Table 2. Participants’ cognition, MTA, and APOE-ε4 according to the amyloid status.
Table 2. Participants’ cognition, MTA, and APOE-ε4 according to the amyloid status.
Total
(n = 161)		Aβ−
(n = 83)		Aβ+
(n = 78)		p-Value
SVLT-DR z-score−1.51 ± 1.06−1.18 ± 0.96−1.86 ± 1.06<0.001
RCFT-DR z-score−1.19 ± 1.02−0.78 ± 1.03−1.63 ± 0.82<0.001
DST-forward z-score0.03 ± 1.050.17 ± 1.12−0.12 ± 0.960.090
K-BNT z-score−0.39 ± 1.22−0.18 ± 1.27−0.12 ± 0.960.029
RCFT-copy−0.65 ± 1.65−0.58 ± 1.35−0.72 ± 1.930.576
COWAT-phonemic z-score−0.29 ± 0.82−0.23 ± 0.84−0.35 ± 0.790.341
K-CWST-CR z-score−0.41 ± 1.11−0.33 ± 1.01−0.49 ± 1.220.366
DSC z-score0.09 ± 0.980.10 ± 1.010.08 ± 0.960.930
K-TMT-B z-score−0.63 ± 1.84−0.54 ± 1.78−0.75 ± 1.920.493
Visual aMCI38 (23.6%)28 (33.7%)10 (12.8%)<0.001
Verbal aMCI50 (31.1%)36 (43.4%)14 (17.9%)
Both aMCI73 (45.3%)19 (22.9%)54 (69.2%)
K-MMSE 25.80 ± 2.7926.88 ± 2.2024.64 ± 2.90<0.001
Global CDR 0.42 ± 0.250.36 ± 0.240.48 ± 0.250.002
CDR-SB 1.76 ± 1.501.31 ± 1.242.22 ± 1.61<0.001
MTA aVRS ≥ 281 (50.3%)29 (34.9%)52 (66.7%)<0.001
APOE-ε4 carrier (%)70 (43.5%)14 (16.9%)56 (71.8%)<0.001
Abbreviations: Aβ, amyloid-β; Aβ−, amyloid PET negative, Aβ+, amyloid PET positive; SVLT-DR, Seoul Verbal Learning Test-delayed recall; RCFT-DR, Rey Complex Figure Test-delayed recall; DST-forward, Digit Span Test-forward; K-BNT, Korean-Boston Naming Test; COWAT-phonemic, Controlled Oral Word Association Test-phonemic; K-CWST-CR, Korean-Color Word Stroop Test-color reading; DSC, Digit Symbol Coding; K-TMT-B, Korean-Trail Making Test-part B; aMCI, amnestic mild cognitive impairment; K-MMSE, Korean version-Mini-Mental State Examination; CDR, Clinical Dementia Rating scale; CDR-SB, sum of boxes of CDR; MTA, medial temporal lobe atrophy; aVRS, axial visual rating scale; APOE, apolipoprotein E.
Table
Table 3. Univariate models to predict for amyloid PET positivity.
Table 3. Univariate models to predict for amyloid PET positivity.
VariablesOR (95% CI)p-Value
SVLT-DR (Ref. > −1 SD) *3.652 (1.637–8.149)0.002
RCFT-DR (Ref. > −1 SD) *3.333 (1.616–6.876)0.001
Both aMCI
(Ref. single aMCI)		7.579 (3.754–15.299)<0.001
K-BNT (Ref. > −1 SD) *1.684 (0.840–3.377)0.142
MTA aVRS (Ref. < 2) *3.724 (1.940–7.149)<0.001
APOE-ε4 carrier (Ref. noncarrier) *12.545 (5.884–26.749)<0.001
Age (Ref. < 70)1.916 (1.022–3.594)0.043
* Age-adjusted; Abbreviations: OR, odds ratio; CI, confidence interval; SVLT-DR, Seoul Verbal Learning Test-delayed recall; SD, standard deviation; RCFT-DR, Rey Complex Figure Test-delayed recall; aMCI, amnestic mild cognitive impairment; DST-forward, Digit Span Test-forward; K-BNT, Korean-Boston Naming Test; MTA, medial temporal lobe atrophy; aVRS, axial visual rating scale; APOE, apolipoprotein E.
Table
Table 4. Multivariate analyses of models combining predictors for amyloid PET positivity.
Table 4. Multivariate analyses of models combining predictors for amyloid PET positivity.
Model 1
OR (95% CI)		Model 2
OR (95% CI)		Model 3
OR (95% CI)		Model 4
OR (95% CI)
Both aMCI
(Ref. single aMCI)		7.368 ***
(3.633–14.943)		6.348 ***
(3.080–13.083)		4.114 **
(1.850–9.150)		3.485 **
(1.530–7.940)
MTA aVRS
(Ref. < 2)		NI2.533 *
(1.177–5.449)		NI2.668 *
(1.106–6.436)
APOE-ε4 carrier
(Ref. noncarrier)		NINI9.090 ***
(3.988–20.722)		9.254 ***
(3.971–21.566)
Age
(Ref. < 70)		1.684
(0.833–3.405)		1.228
(0.571–2.638)		2.082
(0.933–4.645)		1.443
(0.600–3.473)
* p < 0.05, ** p < 0.01, *** p < 0.001, difference between Aβ+ and Aβ−; Abbreviations: OR, odds ratio; CI, confidence interval; aMCI, amnestic mild cognitive impairment; MTA, medial temporal lobe atrophy; aVRS, axial visual rating scale; APOE, apolipoprotein E; NI, not included.
Table
Table 5. A new predictive scale for predicting amyloid PET positivity.
Table 5. A new predictive scale for predicting amyloid PET positivity.
Predictors β CoefficientScoring System
Age<7000
≥700.3671
Memory impairmentSingle (verbal or visual)00
Both (verbal and visual)1.2491
MTA aVRS <200
≥20.9811
APOE-ε4 alleleNoncarrier00
Carrier2.2252
Total 0~5
Abbreviations: SVLT-DR, Seoul Verbal Learning Test-delayed recall; RCFT-DR, Rey Complex Figure Test-delayed recall; MTA, medial temporal lobe atrophy; aVRS, axial visual rating scale; APOE, apolipoprotein E.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Chun, M.Y.; Kim, G.H.; Park, H.K.; Yang, D.W.; Kim, S.; Choi, S.H.; Jeong, J.H. Predictive Scale for Amyloid PET Positivity Based on Clinical and MRI Variables in Patients with Amnestic Mild Cognitive Impairment. J. Clin. Med. 2022, 11, 3433. https://doi.org/10.3390/jcm11123433

AMA Style

Chun MY, Kim GH, Park HK, Yang DW, Kim S, Choi SH, Jeong JH. Predictive Scale for Amyloid PET Positivity Based on Clinical and MRI Variables in Patients with Amnestic Mild Cognitive Impairment. Journal of Clinical Medicine. 2022; 11(12):3433. https://doi.org/10.3390/jcm11123433

Chicago/Turabian Style

Chun, Min Young, Geon Ha Kim, Hee Kyung Park, Dong Won Yang, SangYun Kim, Seong Hye Choi, and Jee Hyang Jeong. 2022. "Predictive Scale for Amyloid PET Positivity Based on Clinical and MRI Variables in Patients with Amnestic Mild Cognitive Impairment" Journal of Clinical Medicine 11, no. 12: 3433. https://doi.org/10.3390/jcm11123433

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop