What this is
- This research examines the relationship between and APOE genotype on () integrity in Alzheimer's disease (AD).
- was assessed using the () in a cohort of 196 non-diabetic AD patients.
- Findings indicate a significant interaction between levels and the APOE ε4/ε4 genotype, affecting permeability.
Essence
- , measured by the index, correlates with altered integrity in Alzheimer's disease, particularly in APOE ε4 carriers.
Key takeaways
- 64% of AD patients exhibited abnormal index values, indicating prevalent among the cohort.
- A significant interaction was found between high levels and the APOE ε4/ε4 genotype, leading to increased permeability.
- Elevated CSF/serum albumin and free light chain levels were associated with high index and APOE ε4/ε4 genotype, suggesting compromised integrity.
Caveats
- The study's cross-sectional design limits causal inference regarding the relationship between and integrity.
- Using indirect measures for permeability, such as CSF/serum albumin ratios, may not fully capture the complexity of function.
- The sample consisted only of biologically confirmed AD patients, which may limit the generalizability of the findings.
Definitions
- insulin resistance: A state of decreased responsiveness of target tissues to insulin, affecting glucose uptake and metabolism.
- blood-brain barrier (BBB): A selective permeability barrier that protects the brain from harmful substances while allowing essential nutrients to pass.
- triglyceride-glucose index (TyG): A clinical marker used to assess insulin resistance, calculated from triglyceride and glucose levels.
AI simplified
BACKGROUND
Converging evidence from epidemiology, clinical, and biological studies supports a strong relationship between insulin resistance, diabetes, and Alzheimer's Disease (AD).1, 2
Insulin resistance is defined as a state of decreased responsiveness of target tissues to insulin.3 Several lines of evidence suggest that insulin resistance disrupts insulin‐signaling pathways in the brain, leading to impairment in glucose uptake and utilization by neurons. Insulin resistance in the brain can lead to amyloid accumulation, tau hyperphosphorylation, increased oxidative stress, and increased protein glycation.4, 5 Moreover, insulin resistance may play a significant role in cognitive dysfunction and in the pathogenesis and progression of AD and other neurodegenerative disorders.6, 7, 8, 9, 10
Several studies have shown that AD is characterized by an increased blood–brain barrier (BBB) permeability,11, 12, 13, 14, 15, 16, 17, 18, 19 and that compromised integrity of cerebrovascular BBB precedes cognitive decline in AD, indicating its potential causal association.20, 21, 22 In fact, the assessment of BBB permeability in vivo poses a significant challenge due to the protected location and complex structure of the brain. Nonetheless, indirect markers such as the cerebrospinal fluid (CSF)/serum albumin ratio have been employed as proxy for BBB integrity.23 Likewise, both kappa and lambda free light chains (FLCs), components of immunoglobulins, are synthesized by plasma cells, and a change in their CSF/serum ratio may also reflect changes in BBB permeability.24
Among others, the apolipoprotein E (APOE) ε4 isoform is a known promoter of BBB dysfunction.25 Recently, we highlighted the role of APOE in modulating BBB permeability, namely the CSF/serum albumin ratio and kappa and lambda FLCs.26 Moreover, insulin regulates the integrity and permeability of BBB through increasing endothelial cell proliferation and expression of tight junction proteins.27 However, although in vitro studies have shed valuable light on the impact of diabetes on BBB permeability, there is a paucity of in vivo evidence supporting a role for insulin resistance.1, 4, 28 According to these data and the large evidence of the role of metabolic syndrome in AD, we hypothesised that insulin resistance might affect BBB integrity synergistically with APOE genotype. To explore this hypothesis, we measured insulin resistance by using the triglyceride–glucose (TyG) index, which has been shown to be a reliable clinical surrogate marker of insulin resistance. The TyG index has shown good performance in the estimation of insulin resistance compared with the homeostasis model assessment of insulin resistance (HOMA‐IR) index in individuals with and without diabetes, while it does not require insulin quantification and it is independent of insulin treatment status.29, 30, 31, 32, 33
In this study, we explored the potential role of insulin resistance, assessed by TyG index, in BBB integrity. Furthermore, we explored the relationship between APOE genotype and insulin resistance–related damage with respect to BBB integrity, adjusting for vascular and metabolic covariates in a sample of biologically confirmed AD patients.
METHODS
Participants
This cohort study involved a consecutive sample of participants recruited from the Neurology Unit and the Center for Brain Health of the Department of Clinical and Experimental Sciences at the University of Brescia, Italy.
During the same visit, all patients underwent an extensive standardized evaluation, following standard procedures, and CSF and blood collection. This assessment encompassed a standardized clinical, cognitive, behavioral, and functional protocol, including the Montreal Cognitive Assessment (MoCA) and the Clinical Dementia Rating (CDR‐SB) Sum of Boxes scores34 to stratify severity and monitor progression. The presence of neuropsychiatric symptoms was assessed using Neuropsychiatric Inventory (NPI). Brain magnetic resonance imaging (MRI) scans were performed on all patients using either a 1.5 or 3 Tesla scanner to exclude cortical infarcts/hemorrhage or brain tumors. APOE genotype was evaluated as reported previously.26 Somatic comorbidities were evaluated using the Cumulative Illness Rating Scale (CIRS).35
Vascular risk factors, comorbidities, and medication data were evaluated during the clinical assessment. Diabetes was defined as a fasting glucose greater than or equal to 126 mg/dL or the use of diabetes medications. Lifetime diagnosis of hypertension and dyslipidemia and use of antihypertension or hypolipidemic medications were determined by interview. Body mass index (BMI) was collected for all patients. All patients underwent blood collection for standard screening including blood counts, creatinine, folate, thyroid function, fasting glucose, triglyceride, and APOE genotype. TyG Index was calculated according to the following formula: TyGindex=ln[(trygliceride∗glucose)]/2, and a cut‐off of 4.55 was used to classify patients with insulin resistance.32
Participants satisfying current clinical criteria for probable AD36 with positive CSF markers were included (see the Section 2.2 for CSF and cut‐offs further details). Full written informed consent was obtained from all subjects according to the Declaration of Helsinki. The Brescia Ethics Committee approved the study protocol (NP 1471, DMA, Brescia).
CSF standard analyses
CSF was obtained during routine diagnostic lumbar puncture according to a standardized protocol, in the outpatient clinic, at fasting, from 09:30 to 10:30 h. CSF was collected in sterile polypropylene tubes and gently mixed to avoid gradient effects. Routine chemical measures were determined. The remaining CSF was centrifuged for 3 min at 3000 rpm and aliquots (500 mm3) were immediately stored at 193.15 K* or in liquid nitrogen for subsequent analysis. CSF total tau (t‐tau), phosphorylated tau‐181 (p‐tau181), amyloid beta (Aβ1‐42 and Aβ1‐40) concentrations were measured by Lumipulse (Fujirebio) by a single experienced technician who was blinded to diagnosis. The internal cut‐off values for AD diagnosis were Aβ42/ p‐tau181 ratio <1.1.37
Kappa and λ FLCs, and albumin concentrations in CSF and serum samples were analyzed using the turbidimetric analyzer SPAplus (The Binding Site Group Ltd, Birmingham, UK) with the serum free light chain immunoassay Freelite (The Binding Site Group Ltd, Birmingham, UK) according to the manufacturer's instructions. Only for a subset of patients, intrathecal synthesis of kappa and λ FLCs was determined as published previously,38 by the following formulas considering serum FLC concentrations and blood–CSF barrier function:
The following cut‐offs were used to define the presence of an intrathecal kappa FLC (≥6.39) and λ synthesis (≥5.5)38; CSF/serum albumin ratio ≥9 was rated as pathologic and positive for BBB damage.39
Statistical analysis
Continuous variables are reported as median (interquartile range [IQR]), and categorical variables are reported as numbers and percentages (n, %). Normality distribution of all variables was tested using the Shapiro–Wilk test. Tertile cut‐off values for the TyG distribution were calculated to obtain specific cut‐off values. Thus patients were stratified in three groups with low, intermediate, or high TyG index. Between‐group differences in clinical features were assessed using the Kruskal–Wallis test and chi‐square test for categorical variables, as appropriate. Due to the non‐normality of data, Box‐Cox power transformations of the BBB integrity markers variables were used to correct for the skewness of the residuals, and the appropriate back‐transformation of model β coefficients was performed. Between‐group differences in CSF variables were assessed using univariate models, adjusting for sex, age, and BMI. The interaction between APOE and TyG index on BBB integrity was tested using the analysis of covariance (ANCOVA) two‐factor interaction model, with patients being classified according to the number of APOE ε4 alleles (0, 1, or 2). Furthermore, to explore the association between insulin resistance, APOE genotype, and clinical variables, patients were categorized according to the CDR, MoCA, and NPI scores. Univariate models were employed to test main effects of clinical symptoms and the interaction effect between them, TyG index, and APOE ε4 alleles. The correlation between BMI and TyG index, as well as other CSF measures, was explored using Spearman's correlation.
Statistical significance was set at p < 0.05 for all tests. Data analyses were performed using JASP version 0.18.1 and R version 4.3.1.
RESULTS
Participant demographics
The study enrolled 196 biologically confirmed AD subjects (mean age ± SD, 71.4 ± 7.2 years; 76 male [38.8%]) (Figure). Mean TyG index was 4.570 ± 0.22 in the whole sample, with 64% of the sample showing an abnormal TyG index (i.e., ≥4.55). S1
TyG index was independent of patients’ age (Spearman's ρ = 0.090, p = 0.208) and gender (t = −0.139, p = 0.890), whereas it was correlated with BMI (ρ = 0.313, p < 0.001). There was a trend of a correlation between TyG index and APOE ε4/ε4 genotype (t = −1.445, p = 0.075). The only two patients carrying the APOE ε2 allele were excluded from further analyses, being not representative of this rare subgroup.
Participants were categorized into three TyG groups based on tertiles of the distribution. Demographic, cardiovascular, and metabolic characteristics of included participants are reported in Table 1. Significant differences between groups were observed for BMI and triglyceride and glucose baseline levels, whereas no cognitive/behavioral differences were detected at baseline (Table S1). Thus, age, sex, and BMI were included as covariates of nuisance for further analyses.
| Normal | Intermediate | High | ||
|---|---|---|---|---|
| = 62n | = 72n | = 62n | ‐valuep | |
| Demographics | ||||
| Age | 71.68 (65.3–77.0) | 72.00 (68.0–75.1) | 74.00 (69.0–78.0) | 0.15 |
| Sex (F/M) | 25/13 | 47/26 | 18/13 | 0.765 |
| CIRS, total score | 6.00 (3.0–10.0) | 6.00 (3.0–10.0) | 6.50 (4.0–10.8) | 0.546 |
| Vascular risk factors | ||||
| Hypertension (%) | 34 (55%) | 31 (43%) | 29 (46%) | 0.164 |
| Dyslipidemia (%) | 41 (66%) | 48 (66%) | 24 (39%) | 0.424 |
| Statin Intake (%) | 19 (30%) | 28 (40%) | 30 (48%) | 0.15 |
| Heart disease | 0 (0%) | 4 (6%) | 2 (3%) | 0.298 |
| BMI | 22.10 (19.8–24.8) | 24.00 (22.0–26.4) | 24.90 (22.9–26.6) | <0.001 14556 |
| Biological variables | ||||
| ε3/ε4 (%)APOE | 28 (45%) | 37 (51%) | 29 (47%) | 0.736 |
| ε4/ε4 (%)APOE | 5 (8%) | 8 (11%) | 6 (10%) | 0.563 |
| Triglycerides | 64.50 (56.0–74.8) | 91.00 (83.0‐–98.0) | 140.50 (112.0–165.5) | <0.001, 14556 14556 |
| Glucose | 93.50 (86.3–97.8) | 97.00 (90.0–105.0) | 105.00 (95.0–114.0) | <0.001, 14556 14556 |
| Creatinine | 0.80 (0.7–0.9) | 0.84 (0.8–0.9) | 0.885 (0.8–1.1) | 0.098 |
| TyG index | 4.35 (4.3–4.4) | 4.55 (4.5–4.6) | 4.76 (4.7–4.9) | <0.001, 14556 14556 |
TyG associates with CSF core AD and BBB integrity markers
Fluid biomarker levels were categorized by the three TyG groups (Table 2). Insulin resistance severity was not associated with CSF AD pathological hallmarks.
Regarding BBB integrity markers, we found a significant relationship between TyG index values and CSF/serum albumin levels (F = 4.658, p = 0.032). The post hoc analysis revealed that albumin was significantly higher in AD patients with high TyG values compared to the other subgroups (p = 0.007).
In a subset of 142 patients with kappa and λ FLCs available, we found a significant association between λ, but not kappa, FLCs and insulin resistance severity measured using TyG index (F = 4.607, p = 0.038). CSF/serum λ FLC levels were higher in AD patients with high TyG values compared to other subgroups (p = 0.025) according to the post hoc analysis. No effect of BMI on albumin CSF/serum ratio or kappa or λ FLCs was found.
APOE ε4/ε4 genotype was associated with higher CSF/serum albumin levels and CSF/serum λ FLC levels (Table S2). Figure 1 shows the combined effect of APOE genotype and insulin resistance on BBB integrity. Specifically, AD patients with APOE ε4/ε4 and high TyG index showed significantly higher CSF/serum albumin levels (F = 4.753, p = 0.001) and CSF/serum λ FLCs levels (F = 3.689; p = 0.005) as compared to the other AD subgroups.
A significant interaction effect was observed between a more severe cognitive impairment (i.e., MoCA <24) and CSF/serum albumin levels (F = 3.464, p = 0.034). Moreover, a significant interaction was found between a more severe cognitive impairment and APOE ε4/ε4 genotype on CSF/serum λ FLC levels (F = 4.168, p = 0.004). No significant effect was observed considering disease severity (i.e., CDR) or neuropsychiatric symptoms (i.e., NPI). (See Tables 3–5.)
Interaction betweengenotype and insulin resistance on markers of BBB integrity, namely CSF serum/albumin ratio (A) and CSF/serum lambda FLC (B). Analyses were adjusted for age, sex, BMI, and hypertension., apolipoprotein E; BBB, blood–brain barrier; BMI, body mass index; CSF, cerebrospinal fluid; FLC, free light chain. APOE APOE
| Normal | Intermediate | High | ||
|---|---|---|---|---|
| = 62n | = 72n | = 62n | ‐valuep | |
| CSF core biomarkers | ||||
| Total tau | 728.00 (468.8–989.3) | 575.50 (452.3–858.0) | 658.90 (459.8–763.3) | 0.121 |
| Phosphorylated tau | 113.00 (73.3–135.9) | 95.50 (79.8–137.2) | 91.50 (70.9–131.8) | 0.481 |
| Aβ42 | 494.80 (414.0–611.0) | 504.50 (377.0–590.8) | 496.50 (410.5–626.3) | 0.965 |
| Aβ40 | 9536.50 (8206.5–13478.3) | 10206.50 (7561.8–12977.8) | 9256.00 (7232.8–1259.3) | 0.749 |
| Aβ42/Aβ40 | 0.049 (0.04–0.05) | 0.046 (0.04–0.05) | 0.047 (0.04–0.06) | 0.488 |
| Aβ42/p‐tau | 4.98 (3.5–6.8) | 4.93 (3.5–6.9) | 5.24 (3.6–7.1) | 0.372 |
| BBB integrity markers | ||||
| Cells | 1.00 (1.0–2.8) | 1.00 (1.0–2.0) | 1.00 (1.0–3.0) | 0.71 |
| Protein | 463.50 (332.8–534.5) | 430.50 (348.3–537.8) | 462.50 (333.8–561.7) | 0.774 |
| Albumin CSF/serum | 6.40 (5.0–7.9) | 6.90 (4.9–8.1) | 7.30 (5.3–9.3) | 0.032 14556 |
| Altered albumin CSF/serum | 19% | 19% | 46% | 0.017 |
| Kappa FLCs CSF/serum 14556 | 1.65 (1.4–2.4) | 1.55 (1.2–2.2) | 1.85 (1.3–2.2) | 0.305 |
| Altered kappa FLCs CSF/serum | 0% | 3% | 6% | 0.668 |
| Lambda FLCs CSF/serum 14556 | 2.43 (1.9–3.1) | 2.54 (1.9–3.4) | 2.76 (2.1–3.5) | 0.038 14556 |
| Altered lambda FLCs CSF/serum | 0% | 0% | 10% | 0.003 |
DISCUSSION
The present study aimed to investigate the role of insulin resistance measured by TyG index on BBB permeability CSF markers and on AD core–related CSF biomarkers. Furthermore, we explored the interaction of TyG index and APOE genotype on either CSF markers of BBB permeability or AD core–related CSF biomarkers. The study was carried out on a consecutive sample of patients with AD and showed that a large proportion of patients displayed pathological TyG index values. Furthermore, our findings revealed a significant correlation between elevated TyG index and increased BBB permeability, exhibiting a clear relationship with APOE ε4/ε4 genotype. Notably, although CSF AD biomarkers demonstrated no association with TyG index and APOE genotype, there was a strong interaction between APOE ε4/ε4 genotype and high TyG index on BBB integrity.
Traditionally, BBB integrity has been assessed in vivo using the CSF/serum albumin ratio, as it is a reliable indicator of BBB permeability because albumin, a relatively large protein (≈67 kDa) synthesized by the liver, does not readily cross the intact BBB.40 Under normal physiological conditions, the concentration of albumin in the CSF is much lower than in the serum. However, in instances where the BBB is compromised, the barrier's permeability to substances like albumin increases, leading to a higher CSF concentration relative to serum, thus elevating the CSF/serum albumin ratio. This elevation serves as a clear biomarker of increased BBB permeability and as an indirect, yet effective measure of BBB integrity.
Accordingly, we also observed an increase in the CSF/serum FLCs ratio in individuals with high TyG index, as well as a significant interaction with APOE ε4 dosing suggesting that the increased BBB permeability extends to other large molecules, including FLCs. The association with λ but not the kappa FLC is probably due to their different molecular weight, as recently observed also for APOE and BBB integrity association.26
Previous studies have shown that AD is characterized by an increased BBB permeability, even in the preclinical and prodromal stages of disease.11, 13, 14, 15, 16, 17, 18 The relationship between AD and BBB integrity is likely mediated by the APOE ε4 allele.25 In this regard, it has been reported that the APOE ε4 allele might impair BBB integrity through several mechanisms, including the interaction between APOE ε4 and low‐density lipoprotein receptor–related protein 1 (LRP1) on pericytes, key cells in maintaining BBB stability. The reduction in LRP1 in endothelial cells, caused by the APOE ε4 allele, leads to a loss of important endothelial tight junction proteins, further compromising BBB integrity.41, 42 In line with these findings, a significant increase in BBB permeability (measured by CSF/serum albumin ratio and kappa and λ FLCs) in relation with APOE genotype has been reported.26
Moreover, several other factors have been associated with BBB integrity, such as aging,43, 44 chronic vascular risk factors, type 2 diabetes mellitus (T2DM),12, 45, 46, 47 arterial hypertension, dyslipidemia, and hyperhomocysteinemia.48, 49
Accordingly, a higher BBB permeability was associated with higher levels of glycated hemoglobin and fasting blood glucose levels after adjusting for all confounders in dementia cases.28 In fact, a higher BBB permeability was found in individuals with T2DM compared with subjects without T2DM in two different large cohorts.11 Furthermore, T2DM was associated with high CSF levels of intercellular adhesion molecule‐1, vascular cellular adhesion molecule‐1, and vascular endothelial‐derived growth factor—CSF biomarkers of angiogenesis and endothelial dysfunction. In animal models, physiological levels of insulin regulate the integrity and permeability of BBB through increasing endothelial cell proliferation and expression of tight junction proteins.41, 42 Using a murine model of prediabetes, it was shown that the breakdown of BBB integrity precedes the development of cognitive decline and neurodegeneration.50 These findings are coherent, with some studies suggesting the pivotal involvement of BBB dysfunction during the onset and early progression of AD.19 The mechanisms by which prediabetes compromises BBB integrity and triggers neurodegeneration include heightened inflammation, oxidative stress, pericyte dysfunction, and leukocyte recruitment,51, 52 possibly affecting amyloid clearence.53 Of interest, systemic inflammation induced by LPS injection results in increased permeability of BBB through the loss of tight junction expression and compromised behavior.54 In fact, it has been documented that insulin insufficiency and hyperglycemia may alter LDL Receptor Related Protein 1 (LRP1) function, thus decreasing Aβ clearance by modulating tight junction proteins, endothelial cells, and the remodeling of extracellular matrices.55 Thus, LRP1‐ induced BBB integrity damage might represent the possible mechanism linking APOE genotype and insulin resistance.56
Growing evidence supports a role of low chronic inflammation in the pathogenesis of insulin resistance. Clinical conditions of overweight and obesity are in fact characterized by release of free fatty acids and proinflammatory cytokines that eventually might contribute to reduce insulin sensitivity. Of interest, vascular homeostasis is impaired in obesity, a condition in which perivascular adipose tissue (PVAT) releases adipo‐cytokines, leading to oxidation of low‐density lipoprotein and endothelial dysfunction, precisely by promoting disruption of inter‐endothelial junctions, increasing reactive oxygen species and a variety of inflammatory mediators.57, 58
Thus we postulate that insulin resistance measured by TyG index might contribute to BBB damage in APOE ε4 carriers as a result of a low chronic inflammation status, which will need to be tested in larger studies. In this regard, it is of interest to notice that APOE produced by microglia seems to be the primary source of APOE deposition into Aβ plaques.59
In an AD mouse model, APOE isoforms in the brain have been reported to impact both Aβ degradation and glucose uptake, particularly by affecting brain insulin/insulin growth factor (IGF) metabolism. Similarly, APOE ε4 astrocytes had a poorer glucose metabolism in vitro.60, 61, 62
From a clinical perspective, these findings may have significant implications. First, the variation in BBB permeability, which we found to be related to insulin resistance and APOE ε4/ε4, might be relevant for a deeper understanding of the individual response to treatments and side effects of treatment with monoclonal antibodies, known to cause amyloid‐related imaging abnormalities (ARIAs). Second, this study highlights the need to evaluate insulin resistance and prediabetic status in patients with AD, in order to possibly include treatment of this relevant risk factor, particularly in APOE ε4 carriers. Homozygous APOE ε4 carriers appeared to have an even higher vulnerability, in line with recent reports, highlighting their rarity but unique biological risk for AD.63, 64 To the best of our knowledge, this is the first study investigating the role of insulin resistance in AD patients by using the TyG index and the first report describing significantly higher TyG among AD patient samples. Previous experimental studies have shown that antidiabetic drugs such as probucol and metformin prevent cognitive deficits by attenuating the neuroinflammation and neurodegeneration mediated through BBB protective properties in a dietary‐induced prediabetic insulin‐resistant mouse model. Thus drugs currently approved to treat metabolic dysfunction hold promise to improve BBB function and reduce the pace of cognitive impairment, as demonstrated in both animal models65, 66 and clinical trials.67 Of note, the glucagon‐like peptide‐1 (GLP‐1) analogue liraglutide crosses the BBB and reduces intrahippocampal amyloid toxicity, significantly increasing memory retention and hippocampal pyramidal neuron numbers in animal models. Moreover, an exploratory trial of the GLP‐1 analog dulaglutide found potential for slowing cognitive decline in patients with T2DM.68 Several mechanisms have been advanced to explain the effect of GLP‐1 analogs on dementia, namely the reduction of dementia‐related vascular risk factors68 and neuroinflammation.69
Finally, the results presented in this study should be interpreted considering certain limitations. First, BBB permeability was inferred using CSF/serum albumin and FLCs ratios which, while informative, are indirect measures. In addition, the study's cross‐sectional nature limits our ability to determine causal relationships or the directionality of the observed associations. Another limitation may be that the present study included only biologically confirmed AD patients, potentially leading to selection bias. Future studies on more diverse neurodegenerative cohorts might be valuable for generalizability of results and to evaluate the potential association between APOE, BBB integrity, prediabetes, and diabetes. Moreover, future studies with a longitudinal design might help to understand the temporal dynamics of BBB permeability changes in relation to neurodegenerative disease progression.
In conclusion, our study provides evidence of the role of insulin resistance measured by the TyG in modulating BBB permeability in Alzheimer's disease, in relation to APOE genotype, with APOE e4/e4 displaying a higher BBB permeability. These findings not only advance our understanding of the pathophysiological mechanisms underlying these diseases but also open new avenues for diagnostic and therapeutic strategies. As the field moves forward, integrating genetic, molecular, and clinical data will be crucial in developing a holistic approach to managing AD. Our study represents a significant step in this direction, offering a new perspective on the interplay between insulin resistance and BBB integrity in the context of AD.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest. Author disclosures are available in the. Supporting Information
CONSENT STATEMENT
Full written informed consent was obtained from all subjects according to the Declaration of Helsinki. The Brescia Ethics Committee approved the study protocol (NP 1471, DMA, Brescia).
Supporting information
ACKNOWLEDGMENTS
The authors wish to thank all patients for their participation in this research. Alessandro Padovani received grant support from the Italian Ministry of University and Research PRIN COCOON (2017MYJ5TH) and PRIN 2021 RePlast (PRIN202039WMFP), the H2020 IMI IDEA‐FAST (ID853981); from the Italian Ministry of Health, Grant/Award Number: RF‐2018‐12366209, PNRR‐Health PNRR‐MAD‐2022‐12376110, and from CARIPLO Foundation. Andrea Pilotto has been supported by grants of AIRALZH Foundation AGYR2021 Life‐Bio Grant and the LIMPE‐DISMOV Foundation Segala Grant 2021. He received grant support from the Italian Ministry of University and Research PRIN COCOON (2017MYJ5TH) and PRIN 2021 RePlast (PRIN202039WMFP), the H2020 IMI IDEA‐FAST(ID853981); from the Italian Ministry of Health, Grant/Award Number: RF‐2018‐12366209 and PNRR‐Health PNRR‐MAD‐2022‐12376110.
Padovani A, Galli A, Bazzoli E, et al. The role of insulin resistance and APOE genotype on blood–brain barrier integrity in Alzheimer's disease. Alzheimer's Dement. 2025;21:e14556. 10.1002/alz.14556