Background: Diabetes mellitus (DM) is a risk factor for Alzheimer's disease (AD), but its stage-dependent effects on the ATN framework remain unclear. Objective: To determine whether DM differentially affects clinical progression from cognitively normal (CN) to mild cognitive impairment (MCI) versus MCI to AD, how DM modulates the ATN cascade, and the optimal follow-up for detecting structural-cognitive changes. Methods: We analyzed longitudinal data from 1638 ADNI participants (765 CN, 71 with DM; 873 MCI, 91 with DM). Clinical progression was assessed by Cox regression. Annualized slopes of ATN biomarkers and cognition were estimated. Moderation analyses tested DM & times; predictor interactions on A -> T (amyloid-tau) and N -> C (neurodegeneration-cognition) pathways with FDR correction. Piecewise linear mixed models identified the optimal knot for structural-cognitive coupling. Results: DM increased CN-to-MCI risk (adjusted HR = 1.73, p = 0.028) but not MCI-to-AD progression (adjusted HR = 1.10, p = 0.574). In CN, DM amplified amyloid-driven tau accumulation in entorhinal cortex (A-T interaction beta = 0.77, FDR p =0.043) and attenuated the protective effect of hippocampal volume on cognition (e.g., hippocampal volume slope & times; DM on MMSE slope: beta = -0.19, FDR p = 0.031). These effects were absent in MCI (all FDR p > 0.05). A 36-month follow-up was the optimal window for detecting robust structural-cognitive coupling in the CN cohort, independent of DM status. Conclusions: Diabetes accelerates preclinical AD by amplifying A-T and eroding structural-cognitive resilience, but does not hasten progression from established MCI to AD. A minimum 36-month observation period is recommended for early-phase prevention trials.
Background: Mitochondrial dysfunction and lipid metabolism dysregulation represent early key events in Alzheimer's disease (AD), yet integrated transcriptomic signatures combining these two biological processes for AD diagnosis remain poorly defined. Objective: This study aimed to construct and validate a mitochondrial-lipid metabolism (MLM) gene-based diagnostic classifier for AD using machine-learning approaches and explore its underlying biological mechanisms. Methods: Two public GEO datasets GSE5281 (training) and GSE28146 (independent validation) were included. Candidate MLM genes were retrieved from MSigDB. Differentially expressed MLM genes were screened by limma, followed by two-step feature selection with LASSO regression and random forest. Four machine-learning algorithms were compared. Model performance was evaluated by AUC, calibration, decision-curve analysis and subgroup analysis across AD severity. GSVA and hypergeometric test were applied for mechanistic exploration. Results: A five-gene signature (SDR16C5, EGR1, ASAH2B, HIP1R, CKMT1B) was finally identified. Logistic regression model achieved 5-fold cross-validated AUC of 0.953 in training cohort and AUC = 0.892 (95%CI:0.768-1.000) in validation cohort. The signature maintained high diagnostic accuracy across incipient, moderate and severe AD stages. Hypergeometric test demonstrated significant MLM-set enrichment (p = 3.85 & times; 10(-5)). GSVA revealed close correlations between this signature with oxidative stress, synaptic dysfunction and apoptosis pathways. Conclusions: This validated five-gene MLM signature exhibits favorable diagnostic performance for AD and reflects disturbed mitochondrial-lipid metabolic network. It provides a promising transcriptomic biomarker for AD diagnosis.
Background Odor identification impairment is an early feature of Alzheimer's disease and predicts memory decline, but its microstructural basis remains unclear. Objective To evaluate FD, a semi-quantitative synthetic magnetic resonance imaging (MRI)-derived Fluid-Attenuated Inversion Recovery (FLAIR)-Double Inversion Recovery (DIR) contrast, against myelin volume fraction (MVF) in olfactory-limbic brain regions of patients diagnosed with mild cognitive impairment (MCI). Methods Sixteen healthy controls (HC) and 17 individuals diagnosed with MCI completed olfactory and cognitive testing and underwent 3T MRI using a QALAS sequence. MVF maps were generated and FD was computed as (FLAIR - DIR)/FLAIR. Group differences were examined in olfactory-limbic gray matter and associated white-matter tracts, and voxel-wise regression assessed FD-odor identification associations, including models adjusted for voxel-wise MVF. Region of interest and voxel-wise analyses were exploratory and largely uncorrected for multiple comparisons. Results Compared with HC, MCI showed lower FD across olfactory-limbic gray matter and white-matter pathways, including hippocampus, amygdala, orbitofrontal cortex, thalamus and corpus callosum, whereas MVF differences were more limited. Better odor identification was associated with higher FD in hippocampal/parahippocampal regions and insula, persisting after MVF adjustment. MVF showed positive associations in the right parahippocampal gyrus, left thalamus and genu of the corpus callosum. In an exploratory secondary analysis reported in the Supplemental Material, FD-based classification exceeded chance but was not separable from age alone. Conclusions FD is a semi-quantitative relaxation-contrast metric showing regional patterns partially distinct from MVF. These exploratory findings require replication in larger, demographically matched, biomarker-confirmed cohorts before FD can be regarded as a marker of early olfactory-limbic microstructural change in MCI.
Background Differentiating tau-related dementias remains challenging, particularly during early disease stages when clinical symptoms overlap. Existing biomarkers mainly reflect downstream pathology and provide limited insight into disease-specific molecular processes. Objective We examined whether cerebrospinal fluid (CSF)-derived cell-free DNA (cfDNA) methylation could distinguish Alzheimer's disease (AD), behavioral variant frontotemporal dementia (bvFTD), and progressive supranuclear palsy (PSP). Methods Promoter methylation of 20 genes associated with AD risk, neuroinflammation, cellular senescence, and glymphatic or lymphatic regulation was analyzed in CSF and matched plasma cfDNA from controls (n = 15), AD (n = 21), bvFTD (n = 14), and PSP (n = 17). cfDNA concentration and fragmentomic profiles were assessed. Diagnostic performance was evaluated using logistic regression and receiver operating characteristic (ROC) analysis. Findings were validated in an independent cohort. Results CSF cfDNA concentrations were increased in AD, whereas fragment size distributions were similar across diagnostic groups. Disease-associated methylation signatures were prominent in CSF but weak or absent in plasma. AD demonstrated widespread hypomethylation of inflammatory and senescence-associated genes together with hypermethylation of glymphatic and lymphatic regulators. bvFTD showed intermediate glial-and stress-related methylation changes, while PSP retained relatively preserved inflammatory and clearance-associated profiles. CSF methylation panels consistently outperformed plasma-derived markers for disease classification. ROC analyses demonstrated strong discrimination for AD, with AUC values reaching 0.84-0.85, while plasma-based markers showed limited diagnostic utility. Conclusions CSF cfDNA methylation captures disease-specific epigenetic states linked to core pathogenic pathways in tau-related dementias. This approach may provide a biologically informative and clinically useful platform for molecular classification and diagnostic stratification.
Background This study tested ALZT-OP1 (inhaled cromolyn 17.1 mg and oral ibuprofen 10 mg) in early Alzheimer's disease (AD).Objective To evaluate the efficacy and safety of ALZT-OP1 in early AD. Methods Cognite was a Phase III, double-blind, placebo-controlled, four-arm factorial trial in 619 participants with early AD (cerebrospinal fluid [CSF] amyloid-beta 42 (A beta 42) 180-690 pg/mL; Clinical Dementia Rating-Global 0.5). Randomization was 1:1:1:1 to cromolyn, ibuprofen, combination therapy, or placebo for 72 weeks. The primary endpoint was change from baseline to Week 72 in Clinical Dementia Rating-Sum of Boxes (CDR-SB), comparing combination versus each monotherapy. Secondary and exploratory endpoints included comparisons with placebo, subgroup analyses by apolipoprotein E4 (APOE4) status, Mini-Mental State Examination scores, CSF biomarkers, and safety. Results In the modified intent-to-treat population (n = 594), the primary endpoint was not met. No statistically significant difference was found between combination and cromolyn (difference 0.16, 95% CI -0.58 to 0.90; p = 0.675) or ibuprofen (0.40, 95% CI -0.32 to 1.13; p = 0.276). APOE4 non-carriers receiving combination therapy showed a numerical benefit versus placebo (least-squares mean difference -0.96; 95% CI -2.00 to 0.08; p = 0.069), although this was not statistically significant. ALZT-OP1 was safe and well tolerated. Conclusions ALZT-OP1 did not demonstrate efficacy in an unselected early AD population. The numerical benefit observed in APOE4 non-carriers warrants prospective evaluation in a genotype-stratified trial. ALZT-OP1 was safe and well tolerated over 72 weeks, and the study provides insights for future AD trial design. The study is registered at ClinicalTrials.gov (; NCT02547818).Conclusions ALZT-OP1 did not demonstrate efficacy in an unselected early AD population. The numerical benefit observed in APOE4 non-carriers warrants prospective evaluation in a genotype-stratified trial. ALZT-OP1 was safe and well tolerated over 72 weeks, and the study provides insights for future AD trial design. The study is registered at ClinicalTrials.gov (; NCT02547818).
