Hypertension and hyperhomocysteinemia as modifiable risk factors for Alzheimer's disease and dementia: New evidence, potential therapeutic strategies, and biomarkers

Trends in the Prevalence of Multiple Chronic Conditions Among US Adults With Hypertension From 1999-2000 Through 2017-2020

BACKGROUND

The prevalence of many chronic conditions has increased among US adults. Many adults with hypertension have other chronic conditions.

METHODS

We estimated changes in the age-adjusted prevalence of multiple (≥3) chronic conditions, not including hypertension, using data from the National Health and Nutrition Examination Survey, from 1999-2000 to 2017-2020, among US adults with (n = 24,851) and without (n = 24,337 hypertension. Hypertension included systolic blood pressure (BP) ≥130 mm Hg, diastolic BP ≥80 mm Hg, or antihypertensive medication use. We studied 14 chronic conditions: arthritis, asthma, cancer, coronary heart disease, chronic kidney disease, depression, diabetes, dyslipidemia, hepatitis B, hepatitis C, heart failure, lung disease, obesity, and stroke.

RESULTS

From 1999-2000 to 2017-2020, the age-adjusted mean number of chronic conditions increased more among US adults with vs. without hypertension (2.2 to 2.8 vs. 1.7 to 2.0; P-interaction <0.001). Also, the age-adjusted prevalence of multiple chronic conditions increased from 39.0% to 52.0% among US adults with hypertension and from 26.0% to 30.0% among US adults without hypertension (P-interaction = 0.022). In 2017-2020, after age, gender, and race/ethnicity adjustment, US adults with hypertension were 1.94 (95% confidence interval: 1.72-2.18) times as likely to have multiple chronic conditions compared to those without hypertension. In 2017-2020, dyslipidemia, obesity, and arthritis were the most common 3 co-occurring chronic conditions among US adults with and without hypertension (age-adjusted prevalence 16.5% and 3.1%, respectively).

CONCLUSIONS

In 2017-2020, more than half of US adults with hypertension had ≥3 additional chronic conditions, a substantial increase from 20 years ago.

Blood biomarkers of Alzheimer's disease in the community: Variation by chronic diseases and inflammatory status

INTRODUCTION

We explored the variations of blood biomarkers of Alzheimer's disease (AD) by chronic diseases and systemic inflammation.

METHODS

We explored the association of AD blood biomarkers with chronic diseases and systemic inflammation (interleukin-6 [IL-6]), in 2366 dementia-free participants of the Swedish National Study on Aging and Care-in Kungsholmen, using quantile regression models.

RESULTS

A greater number of co-occurring chronic diseases was associated with higher concentrations of phosphorylated-tau 181 (p-tau181), total-tau (t-tau), neurofilament light chain (NfL), and glial fibrillary acidic protein (GFAP) (p < 0.01). Anemia, kidney, cerebrovascular, and heart diseases were associated with variations in the levels of AD blood biomarkers. Participants in the highest (vs. lowest) interleukin-6 (IL-6) tertile had higher NfL concentration. Systemic inflammation amplified the associations between several chronic diseases and p-tau181, t-tau, NfL, and GFAP.

DISCUSSION

In the community, the concentration of AD blood biomarkers varies in relation to medical conditions and systemic inflammation. Recognizing these influences is crucial for the accurate interpretation and clinical implementation of blood biomarkers.

HIGHLIGHTS

Participants with a complex clinical profile (i.e., multiple co-occurring diseases or specific disease combinations) display elevated levels of AD blood-biomarkers. Anemia, heart, cerebrovascular, and kidney diseases are associated with variations is the levels of AD blood biomarkers in cognitively intact older adults. Systemic inflammation amplifies the association between several chronic diseases and AD blood biomarkers.

Bidirectional two-sample Mendelian randomization analyses support causal relationships between structural and diffusion imaging-derived phenotypes and the risk of major neurodegenerative diseases

Previous observational investigations suggest that structural and diffusion imaging-derived phenotypes (IDPs) are associated with major neurodegenerative diseases; however, whether these associations are causal remains largely uncertain. Herein we conducted bidirectional two-sample Mendelian randomization analyses to infer the causal relationships between structural and diffusion IDPs and major neurodegenerative diseases using common genetic variants-single nucleotide polymorphism (SNPs) as instrumental variables. Summary statistics of genome-wide association study (GWAS) for structural and diffusion IDPs were obtained from 33,224 individuals in the UK Biobank cohort. Summary statistics of GWAS for seven major neurodegenerative diseases were obtained from the largest GWAS for each disease to date. The forward MR analyses identified significant or suggestively statistical causal effects of genetically predicted three structural IDPs on Alzheimer's disease (AD), frontotemporal dementia (FTD), and multiple sclerosis. For example, the reduction in the surface area of the left superior temporal gyrus was associated with a higher risk of AD. The reverse MR analyses identified significantly or suggestively statistical causal effects of genetically predicted AD, Lewy body dementia (LBD), and FTD on nine structural and diffusion IDPs. For example, LBD was associated with increased mean diffusivity in the right superior longitudinal fasciculus and AD was associated with decreased gray matter volume in the right ventral striatum. Our findings might contribute to shedding light on the prediction and therapeutic intervention for the major neurodegenerative diseases at the neuroimaging level.

Is serum homocysteine level a biomarker of suicide attempts: A preliminary study

Suicide is a global public health concern, and understanding its multifaceted determinants is crucial for effective prevention. This study was designed to find an answer to the question of whether serum homocysteine level can be a biomarker of suicide attempts. This preliminary study involving 90 participants (45 suicide attempt cases and 45 controls) was conducted at Elazig Fethi Sekin City Hospital. Biochemical analyses were performed to assess serum homocysteine, vitamin B12, and folic acid levels. Statistical analyses, including t-tests, Mann-Whitney U tests, and ROC analysis, were employed to explore differences between groups and assess the diagnostic potential of homocysteine. Elevated homocysteine levels were found in individuals who attempted suicide compared to the control group (p= <0.001). Additionally, lower levels of vitamin B12 (p=<0.001) and folic acid (p=<0.001) were observed in the suicide attempt group. ROC analysis indicated a significant diagnostic potential for homocysteine in predicting suicide attempts (AUC = 0.845, sensitivity = 91%, specificity = 71%). This study establishes a significant association between high homocysteine levels and suicide attempts, accompanied by lower vitamin B12 and folic acid levels. The findings suggest a potential link between disturbances in homocysteine metabolism and suicidal tendencies, urging further research to establish causation and explore therapeutic implications. Consideration of the study's limitations and directions for future research are warranted.

Homocysteine potentiates amyloid β -induced death receptor 4- and 5-mediated cerebral endothelial cell apoptosis, blood brain barrier dysfunction and angiogenic impairment

Cerebrovascular dysfunction has been implicated as a major contributor to Alzheimer's Disease (AD) pathology, with cerebral endothelial cell (cEC) stress promoting ischemia, cerebral-blood flow impairments and blood-brain barrier (BBB) permeability. Recent evidence suggests that cardiovascular (CV)/cerebrovascular risk factors, including hyperhomocysteinemia (Hhcy), exacerbate AD pathology and risk. Yet, the underlying molecular mechanisms for this interaction remain unclear. Our lab has demonstrated that amyloid beta 40 (Aβ40) species, and particularly Aβ40-E22Q (AβQ22; vasculotropic Dutch mutant), promote death receptor 4 and 5 (DR4/DR5)-mediated apoptosis in human cECs, barrier permeability, and angiogenic impairment. Previous studies show that Hhcy also induces EC dysfunction, but it remains unknown whether Aβ and homocysteine function through common molecular mechanisms. We tested the hypotheses that Hhcy exacerbates Aβ-induced cEC DR4/5-mediated apoptosis, barrier dysfunction, and angiogenesis defects. This study was the first to demonstrate that Hhcy specifically potentiates AβQ22-mediated activation of the DR4/5-mediated extrinsic apoptotic pathway in cECs, including DR4/5 expression, caspase 8/9/3 activation, cytochrome-c release and DNA fragmentation. Additionally, we revealed that Hhcy intensifies the deregulation of the same cEC junction proteins mediated by Aβ, precipitating BBB permeability. Furthermore, Hhcy and AβQ22, impairing VEGF-A/VEGFR2 signaling and VEGFR2 endosomal trafficking, additively decrease cEC angiogenic capabilities. Overall, these results show that the presence of the CV risk factor Hhcy exacerbates Aβ-induced cEC apoptosis, barrier dysfunction, and angiogenic impairment. This study reveals specific mechanisms through which amyloidosis and Hhcy jointly operate to produce brain EC dysfunction and death, highlighting new potential molecular targets against vascular pathology in comorbid AD/CAA and Hhcy conditions.

Evaluating p-tau217 and p-tau231 as Biomarkers for Early Diagnosis and Differentiation of Alzheimer's Disease: A Narrative Review

The escalating prevalence of Alzheimer's disease (AD) highlights the urgent need to develop reliable biomarkers for early diagnosis and intervention. AD is characterized by the pathological accumulation of amyloid-beta plaques and tau neurofibrillary tangles. Phosphorylated tau (p-tau) proteins, particularly p-tau217 and p-tau231, have been identified as promising biomarker candidates to differentiate the disease progression from preclinical stages. This narrative review is devoted to a critical evaluation of the diagnostic accuracy, sensitivity, and specificity of p-tau217 and p-tau231 levels in the detection of AD, measured in plasma, serum, and cerebrospinal fluid, compared to established biomarkers. Additionally, the efficacy of these markers in distinguishing AD from other neurodegenerative disorders is examined. The significant advances offered by p-tau217 and p-tau231 in AD diagnostics are highlighted, demonstrating their unique utility in early detection and differential diagnosis. This comprehensive analysis not only confirms the excellent diagnostic capabilities of these markers, but also deepens the understanding of the molecular dynamics of AD, contributing to the broader scientific discourse on neurodegenerative diseases. This review is aimed to provide key information for researchers and clinicians across disciplines, filling interdisciplinary gaps and highlighting the role of p-tau proteins in revolutionizing AD research and clinical practice.

Genomic Insights into Dementia: Precision Medicine and the Impact of Gene-Environment Interaction

The diagnosis, treatment, and management of dementia provide significant challenges due to its chronic cognitive impairment. The complexity of this condition is further highlighted by the impact of gene-environment interactions. A recent strategy combines advanced genomics and precision medicine methods to explore the complex genetic foundations of dementia. Utilizing the most recent research in the field of neurogenetics, the importance of precise genetic data in explaining the variation seen in dementia patients can be investigated. Gene-environment interactions are important because they influence genetic susceptibilities and aid in the development and progression of dementia. Modified to each patient's genetic profile, precision medicine has the potential to detect groups at risk and make previously unheard-of predictions about the course of diseases. Precision medicine techniques have the potential to completely transform treatment and diagnosis methods. Targeted medications that target genetic abnormalities will probably appear, providing the possibility for more efficient and customized medical interventions. Investigating the relationship between genes and the environment may lead to preventive measures that would enable people to change their surroundings and minimize the risk of dementia, leading to the improved lifestyle of affected people. This paper provides a comprehensive overview of the genomic insights into dementia, emphasizing the pivotal role of precision medicine, and gene-environment interactions.

Engineering Exosomes to Specifically Target the Mitochondria of Brain Cells

Mitochondrial dysfunction is associated with various health conditions, including cardiovascular and neurodegenerative diseases. Mitochondrial-targeting therapy aims to restore or enhance mitochondrial function to treat or alleviate these conditions. Exosomes, small vesicles that cells secrete, containing a variety of biomolecules, are critical in cell-to-cell communication and have been studied as potential therapeutic agents. Exosome-based therapy has the potential to treat both cardiovascular and neurodegenerative diseases. Combining these two approaches involves using exosomes as carriers to transport mitochondrial-targeting agents to dysfunctional or damaged mitochondria within target cells. This article presents a new technique for engineering brain-derived exosomes that target mitochondria and has demonstrated promise in initial tests with primary neuron cells and healthy rats. This promising development represents a significant step forward in treating these debilitating conditions.

The Role of Methionine-Rich Diet in Unhealthy Cerebrovascular and Brain Aging: Mechanisms and Implications for Cognitive Impairment

As aging societies in the western world face a growing prevalence of vascular cognitive impairment and Alzheimer's disease (AD), understanding their underlying causes and associated risk factors becomes increasingly critical. A salient concern in the western dietary context is the high consumption of methionine-rich foods such as red meat. The present review delves into the impact of this methionine-heavy diet and the resultant hyperhomocysteinemia on accelerated cerebrovascular and brain aging, emphasizing their potential roles in cognitive impairment. Through a comprehensive exploration of existing evidence, a link between high methionine intake and hyperhomocysteinemia and oxidative stress, mitochondrial dysfunction, inflammation, and accelerated epigenetic aging is drawn. Moreover, the microvascular determinants of cognitive deterioration, including endothelial dysfunction, reduced cerebral blood flow, microvascular rarefaction, impaired neurovascular coupling, and blood-brain barrier (BBB) disruption, are explored. The mechanisms by which excessive methionine consumption and hyperhomocysteinemia might drive cerebromicrovascular and brain aging processes are elucidated. By presenting an intricate understanding of the relationships among methionine-rich diets, hyperhomocysteinemia, cerebrovascular and brain aging, and cognitive impairment, avenues for future research and potential therapeutic interventions are suggested.

Association analysis of the gut microbiota in predicting outcomes for patients with acute ischemic stroke and H-type hypertension

Introduction

H-type hypertension (HHTN) is a subtype of hypertension that tends to worsen the prognosis of acute ischemic stroke (AIS). Recent studies have highlighted the vital role of gut microbiota in both hypertension and AIS, but there is little available data on the relationship between gut microbiota and the progression of AIS patients with HHTN. In this study, we investigated the microbial signature of AIS patients with HHTN and identified characteristic bacteria as biomarkers for predicting prognosis.

Methods

AIS patients with HHTN (n = 150) and without HHTN (n = 50) were enrolled. All patients received a modified Rankin Scale (mRS) assessment at 3 months after discharge. Fecal samples were collected from the participants upon admission, including 150 AIS patients with HHTN, 50 AIS patients with non-HHTN, and 90 healthy subjects with HHTN. These samples were analyzed using 16S rRNA sequencing to characterize the bacterial taxa, predict functions, and conduct correlation analysis between specific taxa and clinical features.

Results

Our results showed that the composition of the gut microbiota in HHTN patients differed significantly from that in non-HHTN patients. The abundance of the genera Bacteroides, Escherichia-Shigella, Lactobacillus, Bifidobacterium, and Prevotella in AIS patients with HHTN was significantly increased compared to AIS patients without HHTN, while the genus Streptococcus, Faecalibacterium, and Klebsiella were significantly decreased. Moreover, Bacteroides, Lactobacillus, Bifidobacterium, and Klebsiella in AIS patients with HHTN were more abundant than healthy subjects with HHTN, while Escherichia-Shigella, Blautia, and Faecalibacterium were less abundant. Moreover, the genera Butyricicoccus, Rothia, and Family_XIII_UCG-001 were negatively connected with the NIHSS score, and the genera Butyricicoccus and Rothia were observed to be negatively associated with the mRS score. The genera Butyricicoccus, Romboutsia, and Terrisporobacter were associated with a poor prognosis, whereas the increase in Butyricimonas and Odoribacter was correlated with good outcomes. Generated by eight genera and clinical indexes, the area under the curve (AUC) value of the receiver operating characteristic (ROC) curve achieved 0.739 to effectively predict the prognosis of AIS patients with HHTN.

Conclusion

These findings revealed the microbial signature of AIS patients with HHTN and further provided potential microbial biomarkers for the clinical diagnosis of AIS patients with HHTN.

Serum PAI-1/BDNF Ratio Is Increased in Alzheimer's Disease and Correlates with Disease Severity

We previously demonstrated that serum levels of plasminogen activator inhibitor-1 (PAI-1), which inhibits both the tissue plasminogen activator (tPA) and plasmin activity, are increased in patients with Alzheimer's disease. tPA/plasmin not only prevents the accumulation of β-amyloid in the brain but also is involved in the synthesis of the brain-derived neurotrophic factor (BDNF), a neurotrophin whose levels are reduced in Alzheimer. In the present study, we compared BDNF serum levels in Alzheimer patients with dementia to those in Alzheimer patients with amnestic mild cognitive impairment and to cognitively healthy controls. Moreover, we examined whether the PAI-1/BDNF ratio correlates with disease severity, as measured by Mini-Mental State Examination. Our results showed that BDNF serum levels are lower (13.7% less) and PAI-1 levels are higher in Alzheimer patients with dementia than in Alzheimer patients with amnestic mild cognitive impairment patients (23% more) or controls (36% more). Furthermore, the PAI-1/BDNF ratio was significantly increased in Alzheimer patients as compared to amnestic mild cognitive impairment (36.4% more) and controls (40% more). Lastly, the PAI-1/BDNF ratio negatively correlated with the Mini-Mental score. Our results suggest that increased PAI-1 levels in Alzheimer, by impairing the production of the BDNF, are implicated in disease progression. They also indicate that the PAI-1/BDNF ratio could be used as a marker of Alzheimer. In support of this hypothesis, a strong negative correlation between the PAI-1/BDNF ratio and the Mini-Mental score was observed.

Variety and Duration of Different Sedentary Behaviors, Inflammation, Genetic Susceptibility, and New-Onset Dementia in the Older Population

OBJECTIVES

We aimed to evaluate the relationship of the variety and duration of different sedentary behaviors (TV-watching, driving, and nonoccupational computer use) with the risk of dementia in older participants, and examine whether inflammation and genetic susceptibility may modify the relationship.

DESIGN

A prospective cohort study.

SETTING AND PARTICIPANTS

173,829 older participants (≥60 years) without prior dementia in the UK Biobank were enrolled.

METHODS

A healthy sedentary behavior score was calculated as the number of the 3 major sedentary behaviors with a duration associated with the lowest risk of dementia. The primary outcome was new-onset all-cause dementia.

RESULTS

During a median follow-up of 12.4 years, 4965 (2.9%) participants developed new-onset dementia. There were U-shaped associations for TV-watching and driving time, and a reversed J-shaped association for nonoccupational computer use time with new-onset all-cause dementia, with the lowest dementia risk at >0-<2 hours/day for all the 3 sedentary behaviors. Moreover, a higher healthy sedentary behavior score was significantly associated with a lower risk of all-cause dementia (per 1 score increment: hazard ratio 0.78, 95% CI 0.75-0.81), with a stronger inverse association in those with higher levels of high-sensitivity C-reactive protein and monocytes (both P-interactions <.05). Genetic risks of dementia did not significantly modify the association. Similar trends were found for new-onset Alzheimer's disease and vascular dementia.

CONCLUSIONS AND IMPLICATIONS

The associations between the duration of different sedentary behaviors and new-onset dementia were different in the older population. Moreover, the variety of sedentary behavior was inversely associated with new-onset dementia, especially among those with higher levels of inflammation.

The MOG antibody associated encephalitis preceded by COVID-19 infection; a case study and systematic review of the literature

Background

New neurological complications of COVID-19 infection have been reported in recent research. Among them, the spectrum of anti-MOG positive diseases, defined as anti-MOG antibody associated disease (MOGAD), is distinguished, which can manifest as optic neuritis, myelitis, or various forms of encephalitis (MOGAE).

Materials and methods

This study reports a new case of MOGAE following SARS-CoV-2 infection. A literature review of other MOGAE cases associated with COVID-19 infection was conducted and summarized.

Results

A 60-year-old male patient, who had previously been infected with COVID-19, was admitted to the Neurology Department with a rapidly progressive deterioration of his cognitive functions that lasted for about 3 months. On neurological examination, the Mini-Mental State Examination (MMSE) score was 17, which further deteriorated to 13. In addition, central paresis of the right VIIth nerve and pyramidal hemiparesis on the right side were noted. The MRI of the brain showed multiple hyperintense lesions. The CSF examination revealed an elevated total protein level with a normal cell count, and serum showed a positive finding of anti-MOG antibodies. Taking into account all the information, the diagnosis of MOGAE, following COVID-19 infection, was made. A total of 9 similar cases of MOGAE associated with SARS-CoV-2 infection were identified in the available literature. Among them 2 cases presented progressive cognitive dysfunction and another 5 altered mental status. The most frequently described MRI changes were hyperintense lesions located cortically and/or subcortically. Anti-MOG antibodies were positive in all patients. In 5 cases they were detected only in serum, in 2 cases in serum and CSF, and in 2 cases the origin was not reported.

Conclusion

The reported cases of MOGAE following COVID-19 infection suggest an increasing new clinical problem, and show an association between COVID-19 and MOGADs.

Novel Strategy for Alzheimer’s Disease Treatment through Oral Vaccine Therapy with Amyloid Beta

Alzheimer’s disease (AD) is a neuropathology characterized by progressive cognitive impairment and dementia. The disease is attributed to senile plaques, which are aggregates of amyloid beta (Aβ) outside nerve cells; neurofibrillary tangles, which are filamentous accumulations of phosphorylated tau in nerve cells; and loss of neurons in the brain tissue. Immunization of an AD mouse model with Aβ-eliminated pre-existing senile plaque amyloids and prevented new accumulation. Furthermore, its effect showed that cognitive function can be improved by passive immunity without side effects, such as lymphocyte infiltration in AD model mice treated with vaccine therapy, indicating the possibility of vaccine therapy for AD. Further, considering the possibility of side effects due to direct administration of Aβ, the practical use of the safe oral vaccine, which expressed Aβ in plants, is expected. Indeed, administration of this oral vaccine to Alzheimer’s model mice reduced Aβ accumulation in the brain. Moreover, almost no expression of inflammatory IgG was observed. Therefore, vaccination prior to Aβ accumulation or at an early stage of accumulation may prevent Aβ from causing AD.

Autonomic nervous system and cardiac neuro-signaling pathway modulation in cardiovascular disorders and Alzheimer's disease

The heart is a functional syncytium controlled by a delicate and sophisticated balance ensured by the tight coordination of its several cell subpopulations. Accordingly, cardiomyocytes together with the surrounding microenvironment participate in the heart tissue homeostasis. In the right atrium, the sinoatrial nodal cells regulate the cardiac impulse propagation through cardiomyocytes, thus ensuring the maintenance of the electric network in the heart tissue. Notably, the central nervous system (CNS) modulates the cardiac rhythm through the two limbs of the autonomic nervous system (ANS): the parasympathetic and sympathetic compartments. The autonomic nervous system exerts non-voluntary effects on different peripheral organs. The main neuromodulator of the Sympathetic Nervous System (SNS) is norepinephrine, while the principal neurotransmitter of the Parasympathetic Nervous System (PNS) is acetylcholine. Through these two main neurohormones, the ANS can gradually regulate cardiac, vascular, visceral, and glandular functions by turning on one of its two branches (adrenergic and/or cholinergic), which exert opposite effects on targeted organs. Besides these neuromodulators, the cardiac nervous system is ruled by specific neuropeptides (neurotrophic factors) that help to preserve innervation homeostasis through the myocardial layers (from epicardium to endocardium). Interestingly, the dysregulation of this neuro-signaling pathway may expose the cardiac tissue to severe disorders of different etiology and nature. Specifically, a maladaptive remodeling of the cardiac nervous system may culminate in a progressive loss of neurotrophins, thus leading to severe myocardial denervation, as observed in different cardiometabolic and neurodegenerative diseases (myocardial infarction, heart failure, Alzheimer's disease). This review analyzes the current knowledge on the pathophysiological processes involved in cardiac nervous system impairment from the perspectives of both cardiac disorders and a widely diffused and devastating neurodegenerative disorder, Alzheimer's disease, proposing a relationship between neurodegeneration, loss of neurotrophic factors, and cardiac nervous system impairment. This overview is conducive to a more comprehensive understanding of the process of cardiac neuro-signaling dysfunction, while bringing to light potential therapeutic scenarios to correct or delay the adverse cardiovascular remodeling, thus improving the cardiac prognosis and quality of life in patients with heart or neurodegenerative disorders.

Comorbidities Incorporated to Improve Prediction for Prevalent Mild Cognitive Impairment and Alzheimer's Disease in the HABS-HD Study

BACKGROUND

Blood biomarkers have the potential to transform Alzheimer's disease (AD) diagnosis and monitoring, yet their integration with common medical comorbidities remains insufficiently explored.

OBJECTIVE

This study aims to enhance blood biomarkers' sensitivity, specificity, and predictive performance by incorporating comorbidities. We assess this integration's efficacy in diagnostic classification using machine learning, hypothesizing that it can identify a confident set of predictive features.

METHODS

We analyzed data from 1,705 participants in the Health and Aging Brain Study-Health Disparities, including 116 AD patients, 261 with mild cognitive impairment, and 1,328 cognitively normal controls. Blood samples were assayed using electrochemiluminescence and single molecule array technology, alongside comorbidity data gathered through clinical interviews and medical records. We visually explored blood biomarker and comorbidity characteristics, developed a Feature Importance and SVM-based Leave-One-Out Recursive Feature Elimination (FI-SVM-RFE-LOO) method to optimize feature selection, and compared four models: Biomarker Only, Comorbidity Only, Biomarker and Comorbidity, and Feature-Selected Biomarker and Comorbidity.

RESULTS

The combination model incorporating 17 blood biomarkers and 12 comorbidity variables outperformed single-modal models, with NPV12 at 92.78%, AUC at 67.59%, and Sensitivity at 65.70%. Feature selection led to 22 chosen features, resulting in the highest performance, with NPV12 at 93.76%, AUC at 69.22%, and Sensitivity at 70.69%. Additionally, interpretative machine learning highlighted factors contributing to improved prediction performance.

CONCLUSIONS

In conclusion, combining feature-selected biomarkers and comorbidities enhances prediction performance, while feature selection optimizes their integration. These findings hold promise for understanding AD pathophysiology and advancing preventive treatments.