Sparse learning and stability selection for predicting MCI to AD conversion using baseline ADNI data

ABSTRACT: BACKGROUND: Patients with Mild Cognitive Impairment (MCI) are at high risk of progression to Alzheimer's dementia. Identifying MCI individuals with high likelihood of conversion to dementia and the associated biosignatures has recently received increasing attention in AD research. Different biosignatures for AD (neuroimaging, demographic, genetic and cognitive measures) may contain complementary information for diagnosis and prognosis of AD. METHODS: We have conducted a comprehensive study using a large number of samples from the Alzheimer's Disease Neuroimaging Initiative (ADNI) to test the power of integrating various baseline data for predicting the conversion from MCI to probable AD and identifying a small subset of biosignatures for the prediction and assess the relative importance of different modalities in predicting MCI to AD conversion. We have employed sparse logistic regression with stability selection for the integration and selection of potential predictors. Our study differs from many of the other ones in three important respects: (1) we use a large cohort of MCI samples that are unbiased with respect to age or education status between case and controls (2) we integrate and test various types of baseline data available in ADNI including MRI, demographic, genetic and cognitive measures and (3) we apply sparse logistic regression with stability selection to ADNI data for robust feature selection. RESULTS: We have used 319 MCI subjects from ADNI that had MRI measurements at the baseline and passed quality control, including 177 MCI Non-converters and 142 MCI Converters. Conversion was considered over the course of a 4-year follow-up period. A combination of 15 features (predictors) including those from MRI scans, APOE genotyping, and cognitive measures achieves the best prediction with an AUC score of 0.8587. These results also demonstrate the effectiveness of stability selection for feature selection in the context of sparse logistic regression.     

A Subset of Cerebrospinal Fluid Proteins from a Multi-Analyte Panel Associated with Brain Atrophy,Disease Classification and Prediction in Alzheimer’s Disease

In this exploratory neuroimaging-proteomic study, we aimed to identify CSF proteins associated with AD and test their prognostic ability for disease classification and MCI to AD conversion prediction. Our study sample consisted of 295 subjects with CSF multi-analyte panel data and MRI at baseline downloaded from ADNI. Firstly, we tested the statistical effects of CSF proteins (n = 83) to measures of brain atrophy, CSF biomarkers, ApoE genotype and cognitive decline. We found that several proteins (primarily CgA and FABP) were related to either brain atrophy or CSF biomarkers. In relation to ApoE genotype, a unique biochemical profile characterised by low CSF levels of Apo E was evident in ε4 carriers compared to ε3 carriers. In an exploratory analysis, 3/83 proteins (SGOT, MCP-1, IL6r) were also found to be mildly associated with cognitive decline in MCI subjects over a 4-year period. Future studies are warranted to establish the validity of these proteins as prognostic factors for cognitive decline. For disease classification, a subset of proteins (n = 24) combined with MRI measurements and CSF biomarkers achieved an accuracy of 95.1% (Sensitivity 87.7%; Specificity 94.3%; AUC 0.95) and accurately detected 94.1% of MCI subjects progressing to AD at 12 months. The subset of proteins included FABP, CgA, MMP-2, and PPP as strong predictors in the model. Our findings suggest that the marker of panel of proteins identified here may be important candidates for improving the earlier detection of AD. Further targeted proteomic and longitudinal studies would be required to validate these findings with more generalisability.     

Diagnosed Mild Cognitive Impairment Due to Alzheimer’s Disease with PET Biomarkers of Beta Amyloid and Neuronal Dysfunction

The aim of this study is to identify mild cognitive impairment (MCI) due to Alzheimer’s disease (AD) using amyloid imaging of beta amyloid (Aβ) deposition and FDG imaging of reflecting neuronal dysfunction as PET biomarkers. Sixty-eight MCI patients underwent cognitive testing, [11C]-PIB PET and [18F]-FDG PET at baseline and follow-up. Regions of interest were defined on co-registered MRI. PIB distribution volume ratio (DVR) was calculated using Logan graphical analysis, and the standardized uptake value ratio (SUVR) on the same regions was used as quantitative analysis for [18F]-FDG. Thirty (44.1%) of all 68 MCI patients converted to AD over 19.2±7.1 months. The annual rate of MCI conversion was 23.4%. A positive Aβ PET biomarker significantly identified MCI due to AD in individual MCI subjects with a sensitivity (SS) of 96.6% and specificity (SP) of 42.1%. The positive predictive value (PPV) was 56.8%. A positive Aβ biomarker in APOE ε4/4 carriers distinguished with a SS of 100%. In individual MCI subjects who had a prominent impairment in episodic memory and aged older than 75 years, an Aβ biomarker identified MCI due to AD with a greater SS of 100%, SP of 66.6% and PPV of 80%, compared to FDG biomarker alone or both PET biomarkers combined. In contrast, when assessed in precuneus, both Aβ and FDG biomarkers had the greatest level of certainty for MCI due to AD with a PPV of 87.8%. The Aβ PET biomarker primarily defines MCI due to AD in individual MCI subjects. Furthermore, combined FDG biomarker in a cortical region of precuneus provides an added diagnostic value in predicting AD over a short period.     

Hierarchical Interactions Model for Predicting Mild Cognitive Impairment (MCI) to Alzheimer's Disease (AD) Conversion

Identifying patients with Mild Cognitive Impairment (MCI) who are likely to convert to dementia has recently attracted increasing attention in Alzheimer''s disease (AD) research. An accurate prediction of conversion from MCI to AD can aid clinicians to initiate treatments at early stage and monitor their effectiveness. However, existing prediction systems based on the original biosignatures are not satisfactory. In this paper, we propose to fit the prediction models using pairwise biosignature interactions, thus capturing higher-order relationship among biosignatures. Specifically, we employ hierarchical constraints and sparsity regularization to prune the high-dimensional input features. Based on the significant biosignatures and underlying interactions identified, we build classifiers to predict the conversion probability based on the selected features. We further analyze the underlying interaction effects of different biosignatures based on the so-called stable expectation scores. We have used 293 MCI subjects from Alzheimer''s Disease Neuroimaging Initiative (ADNI) database that have MRI measurements at the baseline to evaluate the effectiveness of the proposed method. Our proposed method achieves better classification performance than state-of-the-art methods. Moreover, we discover several significant interactions predictive of MCI-to-AD conversion. These results shed light on improving the prediction performance using interaction features.     

BrainAGE in Mild Cognitive Impaired Patients: Predicting the Conversion to Alzheimer’s Disease

Alzheimer’s disease (AD), the most common form of dementia, shares many aspects of abnormal brain aging. We present a novel magnetic resonance imaging (MRI)-based biomarker that predicts the individual progression of mild cognitive impairment (MCI) to AD on the basis of pathological brain aging patterns. By employing kernel regression methods, the expression of normal brain-aging patterns forms the basis to estimate the brain age of a given new subject. If the estimated age is higher than the chronological age, a positive brain age gap estimation (BrainAGE) score indicates accelerated atrophy and is considered a risk factor for conversion to AD. Here, the BrainAGE framework was applied to predict the individual brain ages of 195 subjects with MCI at baseline, of which a total of 133 developed AD during 36 months of follow-up (corresponding to a pre-test probability of 68%). The ability of the BrainAGE framework to correctly identify MCI-converters was compared with the performance of commonly used cognitive scales, hippocampus volume, and state-of-the-art biomarkers derived from cerebrospinal fluid (CSF). With accuracy rates of up to 81%, BrainAGE outperformed all cognitive scales and CSF biomarkers in predicting conversion of MCI to AD within 3 years of follow-up. Each additional year in the BrainAGE score was associated with a 10% greater risk of developing AD (hazard rate: 1.10 [CI: 1.07–1.13]). Furthermore, the post-test probability was increased to 90% when using baseline BrainAGE scores to predict conversion to AD. The presented framework allows an accurate prediction even with multicenter data. Its fast and fully automated nature facilitates the integration into the clinical workflow. It can be exploited as a tool for screening as well as for monitoring treatment options.     

CSF and Brain Structural Imaging Markers of the Alzheimer's Pathological Cascade

Cerebral spinal fluid (CSF) and structural imaging markers are suggested as biomarkers amended to existing diagnostic criteria of mild cognitive impairment (MCI) and Alzheimer''s disease (AD). But there is no clear instruction on which markers should be used at which stage of dementia. This study aimed to first investigate associations of the CSF markers as well as volumes and shapes of the hippocampus and lateral ventricles with MCI and AD at the baseline and secondly apply these baseline markers to predict MCI conversion in a two-year time using the Alzheimer''s Disease Neuroimaging Initiative (ADNI) cohort. Our results suggested that the CSF markers, including Aβ42, t-tau, and p-tau, distinguished MCI or AD from NC, while the Aβ42 CSF marker contributed to the differentiation between MCI and AD. The hippocampal shapes performed better than the hippocampal volumes in classifying NC and MCI, NC and AD, as well as MCI and AD. Interestingly, the ventricular volumes were better than the ventricular shapes to distinguish MCI or AD from NC, while the ventricular shapes showed better accuracy than the ventricular volumes in classifying MCI and AD. As the CSF markers and the structural markers are complementary, the combination of them showed great improvements in the classification accuracies of MCI and AD. Moreover, the combination of these markers showed high sensitivity but low specificity for predicting conversion from MCI to AD in two years. Hence, it is feasible to employ a cross-sectional sample to investigate dynamic associations of the CSF and imaging markers with MCI and AD and to predict future MCI conversion. In particular, the volumetric information may be good for the early stage of AD, while morphological shapes should be considered as markers in the prediction of MCI conversion to AD together with the CSF markers.     

Predicting future clinical changes of MCI patients using longitudinal and multimodal biomarkers

Accurate prediction of clinical changes of mild cognitive impairment (MCI) patients, including both qualitative change (i.e., conversion to Alzheimer's disease (AD)) and quantitative change (i.e., cognitive scores) at future time points, is important for early diagnosis of AD and for monitoring the disease progression. In this paper, we propose to predict future clinical changes of MCI patients by using both baseline and longitudinal multimodality data. To do this, we first develop a longitudinal feature selection method to jointly select brain regions across multiple time points for each modality. Specifically, for each time point, we train a sparse linear regression model by using the imaging data and the corresponding clinical scores, with an extra 'group regularization' to group the weights corresponding to the same brain region across multiple time points together and to allow for selection of brain regions based on the strength of multiple time points jointly. Then, to further reflect the longitudinal changes on the selected brain regions, we extract a set of longitudinal features from the original baseline and longitudinal data. Finally, we combine all features on the selected brain regions, from different modalities, for prediction by using our previously proposed multi-kernel SVM. We validate our method on 88 ADNI MCI subjects, with both MRI and FDG-PET data and the corresponding clinical scores (i.e., MMSE and ADAS-Cog) at 5 different time points. We first predict the clinical scores (MMSE and ADAS-Cog) at 24-month by using the multimodality data at previous time points, and then predict the conversion of MCI to AD by using the multimodality data at time points which are at least 6-month ahead of the conversion. The results on both sets of experiments show that our proposed method can achieve better performance in predicting future clinical changes of MCI patients than the conventional methods.     

Multi-method analysis of MRI images in early diagnostics of Alzheimer's disease

The role of structural brain magnetic resonance imaging (MRI) is becoming more and more emphasized in the early diagnostics of Alzheimer's disease (AD). This study aimed to assess the improvement in classification accuracy that can be achieved by combining features from different structural MRI analysis techniques. Automatically estimated MR features used are hippocampal volume, tensor-based morphometry, cortical thickness and a novel technique based on manifold learning. Baseline MRIs acquired from all 834 subjects (231 healthy controls (HC), 238 stable mild cognitive impairment (S-MCI), 167 MCI to AD progressors (P-MCI), 198 AD) from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database were used for evaluation. We compared the classification accuracy achieved with linear discriminant analysis (LDA) and support vector machines (SVM). The best results achieved with individual features are 90% sensitivity and 84% specificity (HC/AD classification), 64%/66% (S-MCI/P-MCI) and 82%/76% (HC/P-MCI) with the LDA classifier. The combination of all features improved these results to 93% sensitivity and 85% specificity (HC/AD), 67%/69% (S-MCI/P-MCI) and 86%/82% (HC/P-MCI). Compared with previously published results in the ADNI database using individual MR-based features, the presented results show that a comprehensive analysis of MRI images combining multiple features improves classification accuracy and predictive power in detecting early AD. The most stable and reliable classification was achieved when combining all available features.     

Selective disruption of the cerebral neocortex in Alzheimer's disease

Background

Alzheimer''s disease (AD) and its transitional state mild cognitive impairment (MCI) are characterized by amyloid plaque and tau neurofibrillary tangle (NFT) deposition within the cerebral neocortex and neuronal loss within the hippocampal formation. However, the precise relationship between pathologic changes in neocortical regions and hippocampal atrophy is largely unknown.

Methodology/Principal Findings

In this study, combining structural MRI scans and automated image analysis tools with reduced cerebrospinal fluid (CSF) Aß levels, a surrogate for intra-cranial amyloid plaques and elevated CSF phosphorylated tau (p-tau) levels, a surrogate for neocortical NFTs, we examined the relationship between the presence of Alzheimer''s pathology, gray matter thickness of select neocortical regions, and hippocampal volume in cognitively normal older participants and individuals with MCI and AD (n = 724). Amongst all 3 groups, only select heteromodal cortical regions significantly correlated with hippocampal volume. Amongst MCI and AD individuals, gray matter thickness of the entorhinal cortex and inferior temporal gyrus significantly predicted longitudinal hippocampal volume loss in both amyloid positive and p-tau positive individuals. Amongst cognitively normal older adults, thinning only within the medial portion of the orbital frontal cortex significantly differentiated amyloid positive from amyloid negative individuals whereas thinning only within the entorhinal cortex significantly discriminated p-tau positive from p-tau negative individuals.

Conclusions/Significance

Cortical Aβ and tau pathology affects gray matter thinning within select neocortical regions and potentially contributes to downstream hippocampal degeneration. Neocortical Alzheimer''s pathology is evident even amongst older asymptomatic individuals suggesting the existence of a preclinical phase of dementia.     

Predicting AD Conversion: Comparison between Prodromal AD Guidelines and Computer Assisted PredictAD Tool

Purpose

To compare the accuracies of predicting AD conversion by using a decision support system (PredictAD tool) and current research criteria of prodromal AD as identified by combinations of episodic memory impairment of hippocampal type and visual assessment of medial temporal lobe atrophy (MTA) on MRI and CSF biomarkers.

Methods

Altogether 391 MCI cases (158 AD converters) were selected from the ADNI cohort. All the cases had baseline cognitive tests, MRI and/or CSF levels of Aβ1–42 and Tau. Using baseline data, the status of MCI patients (AD or MCI) three years later was predicted using current diagnostic research guidelines and the PredictAD software tool designed for supporting clinical diagnostics. The data used were 1) clinical criteria for episodic memory loss of the hippocampal type, 2) visual MTA, 3) positive CSF markers, 4) their combinations, and 5) when the PredictAD tool was applied, automatically computed MRI measures were used instead of the visual MTA results. The accuracies of diagnosis were evaluated with the diagnosis made 3 years later.

Results

The PredictAD tool achieved the overall accuracy of 72% (sensitivity 73%, specificity 71%) in predicting the AD diagnosis. The corresponding number for a clinician’s prediction with the assistance of the PredictAD tool was 71% (sensitivity 75%, specificity 68%). Diagnosis with the PredictAD tool was significantly better than diagnosis by biomarkers alone or the combinations of clinical diagnosis of hippocampal pattern for the memory loss and biomarkers (p≤0.037).

Conclusion

With the assistance of PredictAD tool, the clinician can predict AD conversion more accurately than the current diagnostic criteria.     

Multi-Task Linear Programming Discriminant Analysis for the Identification of Progressive MCI Individuals

Accurately identifying mild cognitive impairment (MCI) individuals who will progress to Alzheimer''s disease (AD) is very important for making early interventions. Many classification methods focus on integrating multiple imaging modalities such as magnetic resonance imaging (MRI) and fluorodeoxyglucose positron emission tomography (FDG-PET). However, the main challenge for MCI classification using multiple imaging modalities is the existence of a lot of missing data in many subjects. For example, in the Alzheimer''s Disease Neuroimaging Initiative (ADNI) study, almost half of the subjects do not have PET images. In this paper, we propose a new and flexible binary classification method, namely Multi-task Linear Programming Discriminant (MLPD) analysis, for the incomplete multi-source feature learning. Specifically, we decompose the classification problem into different classification tasks, i.e., one for each combination of available data sources. To solve all different classification tasks jointly, our proposed MLPD method links them together by constraining them to achieve the similar estimated mean difference between the two classes (under classification) for those shared features. Compared with the state-of-the-art incomplete Multi-Source Feature (iMSF) learning method, instead of constraining different classification tasks to choose a common feature subset for those shared features, MLPD can flexibly and adaptively choose different feature subsets for different classification tasks. Furthermore, our proposed MLPD method can be efficiently implemented by linear programming. To validate our MLPD method, we perform experiments on the ADNI baseline dataset with the incomplete MRI and PET images from 167 progressive MCI (pMCI) subjects and 226 stable MCI (sMCI) subjects. We further compared our method with the iMSF method (using incomplete MRI and PET images) and also the single-task classification method (using only MRI or only subjects with both MRI and PET images). Experimental results show very promising performance of our proposed MLPD method.     

Dkk-3 is elevated in CSF and plasma of Alzheimer's disease patients

Biomarkers in CSF can offer improved diagnostic accuracy for Alzheimer's disease (AD). The present study investigated whether the glycoprotein and putative tumor suppressor Dickkopf homolog 3 (Dkk-3) is secreted into CSF and evaluated its applicability as a diagnostic marker for AD. Using our highly specific immunoenzymometric assay, Dkk-3 levels were measured in plasma and/or CSF of patients suffering from depression, mild cognitive impairment (MCI), or AD and compared with healthy subjects. Dkk-3 identity was verified by western blot and matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry (MS)/MS. High concentrations of Dkk-3 were detected in CSF compared with plasma (28.2 ± 1.3 vs. 1.22 ± 0.04 nmol/L, respectively). Consistently Dkk-3 expression was demonstrated in neurons of the cortex and epithelial cells of the choroid plexus, the major source of CSF. Significantly increased Dkk-3 levels in plasma and CSF were observed for AD patients compared with healthy subjects but not patients suffering from MCI or depression. In summary, our data indicate that elevated Dkk-3 levels are specifically associated with AD and might serve as a potential non-invasive AD biomarker in plasma.     

Cingulate cortex hypoperfusion predicts Alzheimer's disease in mild cognitive impairment

Background  

Mild cognitive impairment (MCI) was recently described as a heterogeneous group with a variety of clinical outcomes and high risk to develop Alzheimer's disease (AD). Regional cerebral blood flow (rCBF) as measured by single photon emission computed tomography (SPECT) was used to study the heterogeneity of MCI and to look for predictors of future development of AD.     

Déficit cognitif léger : mythe ou réalité ?

The concept of mild cognitive impairment, MCI, has been proposed by Petersen and described like a state between the cognitive changes of normal aging and very early dementia. However, MCI appears to be a heterogeneous clinical syndrome in term of etiological factors, clinical patterns or clinical course. New criteria of MCI are proposed for use in clinical research. Identification of patients at risk for Alzheimer disease, AD, is an important goal. Ongoing clinical and neuroimaging (magnetic resonance imaging (MRI), single photon emission computed tomography (SPECT),¹⁸F flouorodeoxyglucose-photo emission tomography (FDG-PET)) studies are focusing on the identification of those individuals with mild cognitive impairment (MCI) who are most likely to convert to AD.     

Enrichment and Stratification for Predementia Alzheimer Disease Clinical Trials

The tau and amyloid pathobiological processes underlying Alzheimer disease (AD) progresses slowly over periods of decades before clinical manifestation as mild cognitive impairment (MCI), then more rapidly to dementia, and eventually to end-stage organ failure. The failure of clinical trials of candidate disease modifying therapies to slow disease progression in patients already diagnosed with early AD has led to increased interest in exploring the possibility of early intervention and prevention trials, targeting MCI and cognitively healthy (HC) populations. Here, we stratify MCI individuals based on cerebrospinal fluid (CSF) biomarkers and structural atrophy risk factors for the disease. We also stratify HC individuals into risk groups on the basis of CSF biomarkers for the two hallmark AD pathologies. Results show that the broad category of MCI can be decomposed into subsets of individuals with significantly different average regional atrophy rates. By thus selectively identifying individuals, combinations of these biomarkers and risk factors could enable significant reductions in sample size requirements for clinical trials of investigational AD-modifying therapies, and provide stratification mechanisms to more finely assess response to therapy. Power is sufficiently high that detecting efficacy in MCI cohorts should not be a limiting factor in AD therapeutics research. In contrast, we show that sample size estimates for clinical trials aimed at the preclinical stage of the disorder (HCs with evidence of AD pathology) are prohibitively large. Longer natural history studies are needed to inform design of trials aimed at the presymptomatic stage.     

Fibrinogen gamma-A chain precursor in CSF: a candidate biomarker for Alzheimer's disease

Background  

Cerebrospinal fluid (CSF) may be valuable for exploring protein markers for the diagnosis of Alzheimer's disease (AD). The prospect of early detection and treatment, to slow progression, holds hope for aging populations with increased average lifespan. The aim of the present study was to investigate candidate CSF biological markers in patients with mild cognitive impairment (MCI) and AD and compare them with age-matched normal control subjects.     

Mild cognitive impairment: a prodromal phase of dementia?

Cognitive decline without dementia is common among older persons. A variety of clinical concepts have been introduced in the past 30 years, in order to describe these cognitive deficits arising in older persons. The most frequently used concept is Mild Cognitive Impairment (MCI). MCI is generally seen as a prodromal phase of Alzheimer disease (AD). Several concepts are described, with the neuropsychiatric features and predictors of conversion to dementia c.q. AD. Finally, consequences of preclinically diagnoses for health care are clarified.     

CCL2 is associated with a faster rate of cognitive decline during early stages of Alzheimer's disease

Chemokine (C-C motif) receptor 2 (CCR2)-signaling can mediate accumulation of microglia at sites affected by neuroinflammation. CCR2 and its main ligand CCL2 (MCP-1) might also be involved in the altered metabolism of beta-amyloid (Aβ) underlying Alzheimer''s disease (AD). We therefore measured the levels of CCL2 and three other CCR2 ligands, i.e. CCL11 (eotaxin), CCL13 (MCP-4) and CCL26 (eotaxin-3), in the cerebrospinal fluid (CSF) and plasma of 30 controls and 119 patients with mild cognitive impairment (MCI) at baseline. During clinical follow-up 52 MCI patients were clinically stable for five years, 47 developed AD (i.e. cases with prodromal AD at baseline) and 20 developed other dementias. Only CSF CCL26 was statistically significantly elevated in patients with prodromal AD when compared to controls (p = 0.002). However, in patients with prodromal AD, the CCL2 levels in CSF at baseline correlated with a faster cognitive decline during follow-up (r _s = 0.42, p = 0.004). Furthermore, prodromal AD patients in the highest tertile of CSF CCL2 exhibited a significantly faster cognitive decline (p<0.001) and developed AD dementia within a shorter time period (p<0.003) compared to those in the lowest tertile. Finally, in the entire MCI cohort, CSF CCL2 could be combined with CSF Tau, P-tau and Aβ42 to predict both future conversion to AD and the rate of cognitive decline. If these results are corroborated in future studies, CCL2 in CSF could be a candidate biomarker for prediction of future disease progression rate in prodromal AD. Moreover, CCR2-related signaling pathways might be new therapeutic targets for therapies aiming at slowing down the disease progression rate of AD.