Alzheimer's Research & Therapy

Alzheimer's disease (AD) patients suffer from consequential diagnostic delay due to the lack of accessible biomarkers. They also show different responses to treatments due to disease heterogeneity and progression. Here, we developed a novel framework to identify disease progression and subtypes by using geometric brain signatures derived from multiple neuroimaging modalities, including [18F]-Florbetapir (AV45) Positron Emission Tomography (PET), [18F]-Fludeoxyglucose (FDG) PET, and structural Magnetic Resonance Imaging (MRI). These signatures were derived by decomposing corresponding maps of amyloid-beta levels, metabolic activity, and cortical thickness in terms of the fundamental, resonant modes-eigenmodes-of cortical geometry, each tied to a specific spatial resolution scale. Our results showed that geometric eigenmode-based features identified trajectories of disease progression, quantified as pseudotime, in distinct subtypes. The disease progression trajectories and subtypes are identified with high stability and are highly related to biological and cognitive measures. These performances are superior to those obtained using conventional localised features and remain robust across datasets, indicating that geometric signatures of brain structure and function can be used to uncover new markers of AD diagnosis and prognosis that are missed by conventional localisation approaches.

Alzheimers disease (AD) is a neurodegenerative disorder affecting more than 55 million people worldwide, with a diagnosis that remains predominantly clinical and frequently delayed. The electroretinogram (ERG) offers a non-invasive electrophysiological method for detecting retinal dysfunction associated with neurodegeneration; however, it remains unclear whether robust and reliable candidate biomarkers can be extracted from ERG signals beyond conventional amplitude- and latency-based parameters. Here we present a pilot study of a multi-domain signal processing framework applied to ERGs recorded from 46 participants (20 AD patients, 26 controls) with a handheld device (RETeval, LKC Technologies) using sinusoidal (1-50 Hz) and photopic ISCEV protocols. Five complementary techniques were implemented: (i) multiscale fuzzy entropy (MSFuzzyEn); (ii) FFT harmonic analysis; (iii) stimulus-response wavelet time-frequency coherence (WTC); (iv) a novel inter-cycle lag variant of sample entropy (SampEnT), introduced to isolate cycle-to-cycle retinal response consistency independently of stimulus periodicity; and (v) discrete wavelet transform (DWT) for energetic extraction of oscillatory potentials (OPs). Univariate comparisons (Mann-Whitney, Cliffs{delta} , Benjamini-Hochberg FDR) identified seven significant candidate biomarkers (q < 0.05), five with large effect size: AUCfast (|{delta}| = 0.546, q = 0.009), Slopevery-slow (|{delta}| = 0.554, q = 0.007), R14f (|{delta}| = 0.515, q = 0.031), SampEnT (|{delta}| = 0.504, q = 0.019) and WTCR,mean (|{delta}| = 0.531, q = 0.023); and two with medium effect size (OP_amp_sum, band_snr). A logistic regression classifier combining three candidate biomarkers, validated by leave-one-out cross-validation, achieved ROC-AUC = 0.858, sensitivity = 70.0% and specificity = 88.5% (n = 46). These proof-of-concept results demonstrate that multi-domain ERG analysis captures retinal temporal dysfunction signatures in AD that are inaccessible to standard clinical analysis, supporting further investigation of portable ERG devices as a source of non-invasive candidate biomarkers for early AD detection.

PURPOSE: Alzheimer disease (AD) is associated with cognitive impairment, brain atrophy, and elevated amyloid-beta and tau. The study aimed to characterize regional atrophy associated with elevated amyloid-beta and tau, as measured by [18F]florbetapir (FBP) and [18F]flortaucipir (FTP) positron emission tomography (PET), respectively, and determine whether combining PET and atrophy data improves the prediction of cognitive impairment. METHODS: Alzheimer Disease Neuroimaging Initiative data (n = 381) were retrospectively analyzed. PET results were correlated with cortical thickness, gray matter (GM) volumes, Mini-Mental State Examination, and Montreal Cognitive Assessment. Linear/logistic regression and area under the curve (AUC) were used to evaluate for significant correlations and compare performances in distinguishing cognitive impairment, respectively. RESULTS: Incremental loss of cortical thickness and GM volume was observed from FBP-/FTP- (n = 205) to single PET-positive (FBP+/FTP-, n = 133; FBP-/FTP+, n = 5) and FBP+/FTP+ (n = 38) groups, particularly in the temporal and parietal lobes. FBP+/FTP+ showed the most severe cortical thickness loss in the entorhinal cortex, temporal lobe GM atrophy, and cognitive impairment. Adding brain atrophy as the third variable resulted in higher odds ratios and improved AUCs for cognitive impairment, with FBP+/FTP+/temporal GM or entorhinal cortical atrophy+ demonstrating the strongest associations with cognitive impairment. CONCLUSION: A multimodal approach combining PET and MRI may help improve the assessment of cognitive impairment in AD.

Digital health technologies (DHT) offer a promising solution to the timely identification of early Alzheimer's disease (eAD) to enable early treatment. This study evaluated the feasibility, acceptability, adherence, reliability, and preliminary clinical and content validity of the novel AD Digital Assessment Suite (AD-DAS). 123 individuals (32 healthy controls (HC), 31 amyloid-PET negative (SCDn), 30 amyloid-PET positive (SCDp) with subjective cognitive decline, and 30 early AD (eAD)) participated. AD-DAS was remotely deployed for 28 days. Remote testing was feasible (97.6% completers), acceptable (>85% ''good''), and associated with high adherence (96%). Metrics showed moderate to excellent test-retest reliability (ICC 0.53-0.91), associations with clinical comparators (adjusted R2 0.01-0.24), differentiated eAD from other known groups (absolute log odds differences 0.6-3.28), and correlated with brain atrophy in expected regions. Episodic and working memory AD-DAS metrics differentiated SCDp from SCDn participants. These preliminary findings suggest that AD-DAS may be a promising tool for detecting cognitive impairments in early AD stages.

Diabetes mellitus (DM) is a major risk factor contributing to the development of Alzheimers disease-related dementias (ADRD). While one of the early symptoms of both Alzheimers disease (AD) and DM-related ADRD is a reduction in cerebral blood flow, the underlying biological mechanisms driving this decline remain to be fully elucidated. Genome-wide association studies have linked AD/ADRD to single-nucleotide polymorphisms in the gene encoding soluble epoxide hydrolase (sEH), an enzyme we previously reported to be upregulated in the brains of an AD rat model. Our previous work also demonstrated that chronic inhibition of sEH with 1-trifluoromethoxyphenyl-3-(1-propionylpiperidin-4-yl) urea (TPPU) preserves hippocampal-dependent spatial learning and memory and improves cerebral hemodynamics in both AD and DM-ADRD models. In the present study, we found that chronic TPPU treatment (1 mg/kg/day for 9 weeks) reduced brain sEH expression, improved cortical-based long-term non-spatial recognition memory involving both cortical and hippocampal networks, and reduced anxiety in DM-ADRD rats. TPPU improved brain perfusion and normalized impaired whisker-evoked functional hyperemia, an effect linked to upregulation of Kir2.1 expression in cerebral capillaries. Furthermore, TPPU restored tight junction proteins (ZO-1 and OCLN), mitigated capillary rarefaction, and suppressed astrocyte and microglial activation. At the cellular level, TPPU attenuated hippocampal neurodegeneration, restored the expression of synaptic proteins (PSD95 and SY38), and reduced levels of key pro-inflammatory chemokines, including MCP-1, RANTES, and MIP-1, in DM-ADRD. In conclusion, TPPU preserves cognitive function in DM-ADRD by mitigating cerebrovascular dysfunction, neuroinflammation, and gliosis while protecting synaptic integrity and neuronal survival, representing a promising therapeutic strategy for DM-ADRD.

The TREM2 R47H variant increases the risk of Alzheimers disease (AD), yet its functional impact in aged mouse models remains incompletely understood. We generated a humanized Trem2 R47H knock-in (KI) line on the AppNL-F background and compared it with a Trem2 knockout (KO) line to assess the degree of TREM2 functional impairment. Accumulation of amyloid {beta} 42 and formation of dystrophic neurites were increased in Trem2 KO mice but not in Trem2 R47H KI mice at 18 or 24 months. qPCR and transcriptomic analyses revealed Trem2 KO mice showed deficits in upregulation of microglial genes while Trem2 R47H KI mice showed a response similar to control mice. Differential gene expression analysis identified altered expressions of genes responsible for ER stress/unfolded protein response and intracellular signalling in Trem2 R47H KI mice. Among the differentially expressed genes, Pmel and Gpnmb were or tended to be downregulated in Trem2 R47H KI as well as in Trem2 KO mice indicating their involvement in AD pathogenesis. These results clearly indicate that the TREM2 R47H variant confers a mild, rather than null, effect on microglial alterations during AD development and that Trem2 R47H KI mice should be used to understand pathological mechanism elicited by TREM2. Further identification and characterization of genes differentially expressed in Trem2 R47H KI mice will provide important insights into how the TREM2 risk variant modulates Alzheimers disease-related pathology. HighlightsO_LIExon2-humanized Trem2 R47H knock-in mice are established, which will serve as a platform to study the role of TREM2 in Alzheimers disease development. C_LIO_LITrem2 knockout mice exhibit deficits in clearance of highly aggregated A{beta}42, suppression of dystrophic neurites and regulation of microglial genes in AppNL-F mice, whereas Trem2 R47H knock-in mice do not. C_LIO_LIRNA-seq reveals transcriptional profiles of Trem2 R47H knock-in mice C_LIO_LIqPCR confirms that Gpnmb and Pmel are or tended to be downregulated in Trem2 R47H knock-in mice. C_LIO_LIFindings demonstrate that TREM2 R47H is hypomorphic rather than loss of function. C_LI

Background: A recent Lancet Commission estimated that up to 45% of Alzheimers Disease and Related Dementias (ADRD) cases could be prevented by addressing modifiable lifestyle risk factors. Meanwhile, genome wide association studies (GWAS) have shown that common genetic variants also account for substantial ADRD risk. Whether a favorable lifestyle can offset risk in genetically predisposed individuals remains unclear. Methods: We conducted a retrospective cohort study of 105,886 participants from the All of Us Research Program enrolled between 2018 and 2023. Participants were over age 49, assigned male or female at birth, of European ancestry, and without ADRD at baseline. ADRD diagnoses were identified via electronic health records (EHR). Fourteen potentially modifiable risk factors for ADRD were assessed using surveys, EHR records, and wearable data. Genetic risk was quantified as a polygenic risk score (PRS) based on 81 independent GWAS loci and APOE E4 genotype. Results: Overall, 967 incident ADRD events occurred over a median follow-up of 3.7 years. Ten out of 13 modifiable risk factors were significantly associated with ADRD. When grouped into risk factor profiles, intermediate and unfavorable modifiable risk factor scores were associated with substantially higher ADRD risk (HR 3.07, 95% CI 2.47, 3.83; HR 8.01, 95% CI 6.39, 10.05, respectively) compared to a favorable lifestyle; APOE E4 dosage and polygenic risk score were also independently associated with ADRD risk. Among individuals in the highest polygenic risk group, a favorable lifestyle reduced ADRD risk from HR 18.63 (95% CI 10.25, 33.86) to 1.90 (95% CI 0.94, 3.81), whereas APOE E4 homozygotes remained at elevated risk even with a favorable lifestyle (HR 6.52, 95% CI 2.97, 14.33). Conclusions: Our data suggest ADRD risk is driven more by modifiable risk factors and APOE genotype than polygenic risk score. Future genomic informed risk assessments for ADRD should calibrate their findings to accurately identify high risk individuals.

High placebo response is an obstacle in developing drugs to treat agitation in Alzheimer's disease (AAD), a prevalent and burdensome symptom. However, it has proved challenging to develop actionable models of placebo response that 1) can be applied prospectively, requiring only information available at screening or baseline, 2) yield strategies for reducing placebo response without equally depressing drug response, and 3) show generalizability across trials. Here, we first investigated placebo response in AAD at the trial level using meta-regression applied to 23 clinical trials. Meta-regression identified several factors associated with increased placebo response, but most of these factors were non-specific such that they predicted improvements in drug response as well. We therefore turned to individual level clinical trial datasets and applied causal modeling to predict which participants would have high placebo response relative to predicted drug response. We successfully built and validated the causal model across two independent clinical trials of risperidone and haloperidol at the level of individual patients (ability to predict subsequent improvement on drug or placebo). Crucially, we also found efficacy improvements in the overall trial through in silico exclusion/screen failing of high placebo-predicted subjects. We further characterized features most associated with placebo response to improve explainability and, lastly, validated the effect of these features at the trial level in clinical trials of galantamine, an acetylcholinesterase inhibitor (hence in a different class of drugs than those in the other two trials used). Taken together, we have developed and applied a causal modeling framework for reducing placebo response and increasing trial-level efficacy in neuropsychiatry clinical trials using historical trial datasets.

Immune inhibitory signaling in microglia contributes to impaired amyloid clearance and neuroinflammation in Alzheimers disease (AD), yet small molecule modulators targeting these pathways remain largely unexplored. Here, we report the development of a high-throughput cellular thermal shift assay (HT-CETSA) platform for identification of small molecule binders targeting the inhibitory immune receptor ILT3 (LILRB4). Screening of [~]40,000 compounds yielded multiple validated hits, including IB15C, a submicromolar ILT3 binder identified through preliminary structure-activity relationship optimization. Orthogonal validation by microscale thermophoresis, surface plasmon resonance, docking, and site-directed mutagenesis confirmed direct and target-specific ILT3 engagement. Functionally, IB15C disrupted the ILT3-ApoE interaction and restored microglial activity in human iPSC-derived microglia, reducing SHP1/2, suppressing cytokine secretion, and enhancing amyloid uptake. IB15C also demonstrated favorable in vitro pharmacokinetic and safety properties, supporting further development of ILT3-targeted neuroimmune therapeutics.

Alzheimer's disease (AD) is characterized by complex alterations in synaptic, glial, neuronal and inflammatory markers. Given its emerging role at the interface of synaptic dysfunction and inflammation, the astrocytic marker GFAP may represent a cross-domain hub linking synaptic, neuronal and inflammatory alterations. Using multivariate and network-based analyses we examined the relationships among cerebrospinal fluid (CSF) biomarkers of astrocytic activation and synaptic failure, inflammation, and neurodegeneration in biologically confirmed AD patients and healthy controls (HC). We studied 60 AD patients and 40 HC. CSF concentrations of Neurogranin, SNAP-25, CAP2, NfL, GFAP, IL-1 , IL-1{beta}, IL-8, MCP-1, TNF were measured. Associations were assessed using Spearman correlations, LASSO regression, and network analysis to characterize multivariate dependency structures. Compared with controls, AD patients showed significantly higher CSF levels of Neurogranin, SNAP-25, CAP2, NfL, GFAP, IL-1{beta}, TNF- .. In AD, synaptic biomarkers were strongly intercorrelated and associated with astroglial activation, inflammatory markers, and tau-related pathology. Network analysis identified GFAP as a cross-domain hub linking synaptic, inflammatory, and neurodegenerative domains in AD. In controls, GFAP was mainly associated with neuronal injury markers. Network-based modelling revealed a disease-related reorganization of biomarker connectivity in AD, with GFAP occupying a central cross-domain position, supporting a systems-level view of AD pathophysiology.

BACKGROUND: Neuropsychiatric symptoms (NPS) are prevalent and debilitating in Alzheimer's disease (AD). Existing pharmacologic treatments are often ineffective and associated with serious adverse events. Identifying modifiable risk factors (MRFs) is critical for prevention and treatment. METHODS: Capitalizing on data from 14,497 individuals with AD from the National Alzheimer's Coordinating Center (NACC) database, we examined longitudinal associations between modifiable risk factors, APOE genotype and NPI-Q-assessed NPS using Cox proportional hazards models adjusted for demographics. RESULTS: Diabetes, alcohol consumption, smoking, and TBI were associated with an increased risk of specific NPS in AD. APOE{varepsilon}4 carrier status was linked to multiple NPS, showing a dose-response relationship. Education, LDL-C, and corrective lenses were protective; hypertension showed no associations. CONCLUSION: These findings strongly suggest that individual MRFs are associated with specific NPS in line with a complex etiology underlying these symptoms. Early detection and management of vascular, lifestyle and sensory factors could reduce NPS.

4-Methylumbelliferone (4-MU) inhibits hyaluronic acid (HA) synthesis and is currently approved in Europe for biliary spasm. 4-MU administration reduces perineuronal nets (PNNs), and enzymatic degradation of PNNs in mouse models of Alzheimers disease (AD) attenuates memory impairment. Although 4-MU has therapeutic efficacy in rodent models of fibrosis and cancer, it has not been examined in an Alzheimers model. Here, we evaluated the impact of long-term 4-MU treatment in the APP/PS1 amyloid mouse model. From three months of age, mice were on either a vehicle or 4-MU-supplemented diet for 70 days or 52 weeks. Short and long-term 4-MU treatment decreased the soluble parenchymal A{beta}1-42/A{beta}1-40 ratio. Reductions in insoluble amyloid plaque were observed following 52 weeks of treatment. Extended 4-MU administration also reduced PNN intensity and ameliorated spatial memory deficits in APP/PS1 mice. These findings provide support for targeting brain extracellular matrix (ECM) as a therapeutic strategy for AD.

We trained a self-configuring nnU-Net model for CMB segmentation in a heterogeneous multicenter sample (n=264), including 1.5T and 3T field strengths, SWI and T2*-GRE sequences, and community and clinical cohorts. Model performance was evaluated using 5-fold cross-validation with a focus on object-level detection metrics. Real-world performance was evaluated on scans from an unseen dataset of people with cerebrovascular disease (n=20). The model achieved 0.82 cluster Dice, 0.88 precision, and 0.77 sensitivity on hold-out test data. Notably, the model demonstrated a low false-positive rate, averaging 0.58 false positives (FPs) per scan, an improvement on existing publicly available models. The model achieved high performance in dataset of those with Alzheimer's disease and mild cognitive impairment (0.89 cluster Dice, 0.94 sensitivity), supporting its utility in clinical settings where ARIA-H monitoring is critical. In external validation, the model maintained high robustness with 0.79 sensitivity and 0.95 FPs per scan. By leveraging a heterogenous training strategy and a self-adapting architecture, we demonstrate that deep learning can achieve high-precision CMB detection that is robust to domain shifts. The low FP rate suggests this publicly available pipeline is suitable for automated screening and lesion counting in heterogenous large-scale clinical trials, reducing the burden of manual quantification.

To resolve discrepancies in the literature regarding the association between Alzheimers disease (AD) and Biondi body (BB) amyloid in choroid plexus epithelial cells (CPECs), we investigated postmortem hippocampal paraffin blocks with and without a neuropathological diagnosis of AD (n=26-27 each). Similar to previous studies, age was associated with an increased fraction of hippocampal-associated CPECs bearing thioflavin S-positive BBs (p=0.004). In addition, we found that paraffin block storage time was associated with decreased BB detectability (p=0.038) while sex had no effect (p=0.577). Controlling for age, sex, and storage time, AD was associated with a near-significant increase in the BB-containing CPEC fraction (p=0.066) and a significantly greater load of BB-like amyloid in hippocampal-associated ependymal cells (p=0.032). The AD-BB association contrasts with our findings on choroid plexus from the atrium of the lateral ventricle, which lacked this association. We discuss potential explanations for the apparent discrepancy such as regional amyloid cross-seeding.

Spousal caregivers of individuals with Alzheimers disease and related dementias frequently experience elevated perceived stress, caregiver burden, and loneliness, which are associated with adverse health outcomes. Early identification is therefore critical for timely intervention. Existing approaches commonly rely on wearable sensor data and standardized psychological questionnaires, while recent multimodal methods aim to improve prediction by integrating behavioral and linguistic information. In this study, we explored three modality configurations, wearable-derived features, interview-based text, and their combination, to classify caregiver psychological risk using the Perceived Stress Scale (PSS), Zarit Burden Interview, and UCLA Loneliness Scale. We compared traditional machine learning models and large language models (LLMs) (Gemini 2.0, Llama 4, and GPT-4o) under psychometrician-centered and caregiver-centered prompting strategies. Traditional machine learning models performed better under multimodal settings, while LLMs achieved stronger performance with Interview-Only input. We further demonstrate that PSS was the most predictable construct and prompting strategies substantially influenced LLM performance.

INTRODUCTION: Amyloid-related imaging abnormalities with microhaemorrhage (ARIA-H) are an important safety consideration with anti-amyloid therapies, yet evidence exploring short-term risk prediction remains limited. METHODS: We analysed longitudinal MRI data from the A4 study, constructing a discrete person-interval dataset for modelling the short-term risk of ARIA-H. Models incorporated baseline covariates, alongside dynamic variables of recent microhaemorrhage accumulation and current microhaemorrhage burden. Incident ARIA-H was defined as 2 or more new microhaemorrhages or 1 or more new superficial siderosis between consecutive MRI scans. RESULTS: Among 1,069 participants (3,647 intervals), 171 ARIA-H events occurred. Both current burden (time-to-event: OR 1.37, 95% CI 1.09-1.73; all-event: OR 1.24, 95% CI: 1.04-1.49) and recent microhaemorrhage accumulation (time-to-event: OR 1.83, 95% CI 1.01-3.30; all-event: OR 1.43, 95% CI: 1.00-2.04) were independently associated with an increased risk of ARIA-H. DISCUSSION: Temporal imaging variables may provide independent prognostic information beyond baseline risk, supporting a dynamic model of haemorrhagic risk in Alzheimer's disease.

INTRODUCTION: Postmenopausal estrogen decline may contribute to Alzheimer's disease (AD) risk, but longitudinal evidence linking circulating estrogens to cerebrospinal fluid (CSF) biomarkers is lacking. METHODS: We analyzed 866 female participants from the European Prevention of AD Longitudinal Cohort Study with baseline serum estradiol and estrone measured by liquid chromatography tandem mass spectrometry and repeated CSF measurements of amyloid-beta (A{beta})42, phosphorylated (p) Tau181, and total (t) Tau. RESULTS: Neither estradiol nor estrone was associated with longitudinal A{beta}42. Higher estradiol was associated with lower baseline tau and slower tau increases over time. Baseline estradiol-tau associations were stronger in apolipoprotein E (APOE) {epsilon}4 carriers, though APOE{epsilon}4 did not modify longitudinal associations. Amyloid positivity did not moderate hormone-tau associations but was associated with steeper tau increases over time. Estrone showed no significant associations. DISCUSSION: These findings suggest a more consistent relationship between estradiol and tau-related rather than amyloid-related pathology.

STRUCTURED ABSTRACTO_ST_ABSINTRODUCTIONC_ST_ABSA higher incidence of dementia, including Alzheimer s-like pathology, is observed in aged people living with HIV-1. However, mechanisms linking HIV-1 to Alzheimers disease (AD) pathology remain unclear, due to the lack of animal models that allow for concurrent study. METHODSWe created a novel APP knock-in (KI) AD mouse, NOG/APPKM670,671NL/IL-34 (hNAIL) that permits study of progressive brain HIV-1 replication. The mice harbor human microglia-like cells. Four-month-old CD34+ human cell reconstituted mice infected with the HIV-1ADA strain facilitated studies of HIV-1 replication on AD pathologies. RESULTSHIV-1 replication increased A{beta} levels and reduced synaptic and neuronal integrity. Spatial transcriptomics demonstrated distinct A{beta} and HIV-1 transcriptional patterns, whereas dual diseased combinations amplified AD pathology. Neurons showed highest transcriptional change, with genes linked to neuroinflammation, protein trafficking, and synaptic dysfunction. DISCUSSIONThe hNAIL mice enable interrogation of HIV-AD comorbidities, with a future potential for the development of novel therapeutic interventions.

Clinical Dementia Rating (CDR) scores are used to classify the cognitive state of patients and are provided within neuroimaging datasets. This is achieved through a standardized clinical assessment that evaluates participants cognitive and functional abilities in everyday life, after which they are given a score ranging from 0 to 3. Where 0 represents no signs of dementia and three represents severe dementia1. These scores are then used to track the progression of dementia over time2. This study explored if these CDR labels within the OASIS-1 dataset produced consistent volumetric separation across the hippocampus, amygdala, and cortex.

Using data from a randomized clinical trial, we examined whether daily biofeedback training that modulates heart rate oscillations is associated with changes in microstructural brain texture in Alzheimer's disease signature cortical (ADSC) and hippocampal regions. Younger and older adults were randomly assigned to one of two daily biofeedback practices for five weeks: slow-paced breathing designed to increase heart rate oscillations (Osc+) or self-selected strategies aimed at decreasing oscillations (Osc-). Intervention effects were observed in both ADSC and hippocampus regions and were confined to a composite texture factor dominated by uniformity and entropy. Across regions, effects were expressed primarily as Time x Condition interactions, indicating differential texture trajectories between Osc+ and Osc-. In the hippocampus, this pattern was further qualified by a Time x Condition x Age Group interaction, reflecting more pronounced effects in older adults, whereas younger adults showed no reliable texture modulation. Partial least squares correlation analyses further demonstrated that training-related texture changes in the left hippocampus, right fusiform gyrus, and right entorhinal cortex covaried with concurrent changes in plasma AD-related biomarkers, with tau- and p-tau related measures contributing most strongly to the multivariate association. Together, these findings suggest that HRV biofeedback may selectively influence specific dimensions of brain microstructural texture and that such changes are meaningfully coupled with plasma AD-related biomarker profiles.