Impact of Apolipoprotein E4 on Blood-Brain Barrier Integrity in Target Replacement Murine Models: A Systematic Review and Meta-Analysis

Impact of Apolipoprotein E4 on Blood-Brain Barrier Integrity in Target Replacement Murine Models: A Systematic Review and Meta-Analysis | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Systematic Review Impact of Apolipoprotein E4 on Blood-Brain Barrier Integrity in Target Replacement Murine Models: A Systematic Review and Meta-Analysis Krystal Laing, Nela Fialova, Joanna Wardlaw, Axel Montagne This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8011437/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Background The E4 variant of Apolipoprotein E (APOE) is a primary genetic susceptibility risk factor for late-onset Alzheimer’s disease and has been implicated in cerebrovascular dysfunction. Preclinical mouse models are widely used to study APOE4 , but cohesive understanding of APOE’s role is still inconsistent and lacking. The aim of this study was to systematically review and synthesise evidence from preclinical mouse studies assessing APOE4 related effects on blood-brain barrier (BBB) integrity, vascular morphology and cerebral blood flow (CBF). Main A systematic search of MEDLINE, Embase, Scopus, and Web of Science was conducted (March-April 2025). Eligible studies included transgenic APOE -targeted replacement or knock-in mice reporting vascular outcomes (cerebral blood flow, blood brain barrier permeability, vascular measures). Risk of bias was assessed using SYRCLE and reporting quality with CAMARADES. Random-effects meta-analyses were conducted (where sufficient data was available), otherwise findings were narratively synthesised. Eighteen studies met inclusion. Outcome measures varied widely, including diverse approaches to CBF measurement (e.g. arterial spin labelling, autoradiography, DSC-MRI), immunohistochemical measures (e.g. collagen-IV, laminin, CD31), and diverse approaches to measurement of BBB leakage (e.g. fibronectin, fibrinogen, gadolinium-based ktrans). Seven studies contributed to meta-analysis: APOE4 mice showed a consistent reduction in CBF associated with APOE4 genotype (SMD = -2.87, 95% CI: -5.14 to -0.604, df = 2.66), and a negative non-significant trend towards reduced vascular morphology expression. Narrative synthesis identified three key mechanistic pathways linking APOE4 to vascular dysfunction: (i) insulin resistance and PI3K/AKT-mTOR signalling, (ii) Cyclophilin A–NFκB–MMP9 activation, and (iii) occludin/ECM remodelling. Risk of bias assessment revealed frequent shortcomings in randomisation, blinding, and sample size justification. Conclusions Preclinical evidence demonstrates that APOE4 drives alterations in vascular functioning primarily through involvement with pathways related to vascular metabolism, ECM remodelling and BBB leakage. However, heterogeneity in the model (e.g. age, sex, techniques), restricts direct comparability across studies. As such, standardisation or clarification of methodological approaches are necessary for rigorous assessment in the future. Apolipoprotein E4 blood–brain barrier cerebral blood flow vascular dysfunction Alzheimer’s disease systematic review meta-analysis Figures Figure 1 Figure 2 Figure 3 Figure 4 Background Apolipoprotein E (APOE) is a multifunctional lipid transporter involved in lipoprotein metabolism and cholesterol homeostasis ( 1 , 2 ). The ε4 variant is the strongest genetic risk factor for late-onset Alzheimer’s disease (AD) ( 3 – 5 ) and has also been linked to other disorders, including vascular dementia, atherosclerosis, and cardiovascular disease, through mechanisms such as amyloid-β (Aβ) accumulation, neurodegeneration, and neuroinflammation ( 6 – 8 ). Heterozygosity for APOE4 increases AD risk 3–4-fold, while homozygosity raises it by ~ 15-fold ( 9 , 10 ). Although only 20–25% of the general population carry an ɛ4 allele, 40–65% of AD patients are ɛ4 carriers. As of 2020, nearly 50 million people were affected globally by AD or related dementias ( 10 – 13 ). The mechanisms by which ε4 increases dementia risk remain incompletely understood despite extensive study. Recently, several Aβ-targeting therapies have demonstrated robust biomarker outcomes but carry an elevated risk of Amyloid-Related Imaging Abnormalities (ARIA) in APOE4 carriers ( 6 ). ARIA manifests as cerebral oedema or haemorrhage on MRI following Aβ immunotherapy and can, in severe cases, cause morbidity or death ( 7 ). This underscores the need for a deeper understanding of the molecular and cellular mechanisms driving APOE4-related pathology to inform safer, more precise interventions. Human studies of APOE4 pathophysiology are constrained by limited ɛ4 homozygosity (~ 9.6% in AD patients ( 8 )) and the invasiveness of outcome measures. Rodent models, however, permit controlled investigation of cellular and molecular mechanisms using transgenic manipulations that mimic human disease. These models enable access to tissue-level endpoints, high reproducibility, and defined experimental conditions. Several humanised APOE mouse lines are available, some incorporating AD-related mutations in amyloid precursor protein (APP), microtubule-associated protein tau (MAPT), or familial AD (FAD) genes ( 9 ). Beyond central nervous system alterations, APOE4 also contributes to vascular dysfunction during disease progression. A key mechanism involves blood–brain barrier (BBB) breakdown ( 10 – 13 ), which can be assessed through non-invasive measures such as cerebral blood flow and contrast-enhanced permeability. Additional tools—including immunohistochemistry, in vivo MRI, and behavioural testing—provide complementary insight into vascular morphology, structural integrity, and cognitive outcomes. This review synthesises current evidence on APOE-driven vascular alterations in mouse models, focusing on genotype-related differences in BBB integrity, vascular morphology, and cerebral blood flow. Given the heterogeneity of study designs, outcomes, and ages assessed, these comparisons are crucial for building a coherent picture of APOE4-related vascular pathology and its contribution to disease progression. Methods Identification and Search Strategy This review followed the PRISMA 2020 guidelines (14) and was registered in PROSPERO (CRD420251010665), and the detailed search strategy was separately registered with SearchRxiv (DOI: 10.1079/searchRxiv.2025.01124) to enhance methodological transparency and reproducibility. Literature searches were conducted in Scopus, MEDLINE, Embase, and Web of Science between 25 March and 3 April 2025. The MEDLINE search strategy was: (Apolipoproteins E/ OR Apolipoprotein E3/ OR Apolipoprotein E4/) AND (mice/ OR mice, transgenic/) AND ((brain/ OR blood–brain barrier/) OR (perfusion/ OR cerebrovascular circulation/)) ; full strategies for other databases are provided in Supplementary Table 1 . Additional manuscripts identified by manual search were included. Unpublished data and conference abstracts were excluded. Study Selection and Data Extraction Screening and data extraction were performed using the Systematic Review Facility (SyRF) platform (15). Two reviewers (K.L. and N.F.) independently screened titles, abstracts, and full texts, resolving disagreements by consensus with a third reviewer (A.M.). Eligible studies: Used humanised targeted replacement APOE mice (APOE3-TR, APOE4-TR) (16) or APOE-TR;5xFAD (EFAD) lines. Provided comparative analysis between APOE3 and APOE4 homozygotes; APOE4-only studies without APOE3 controls were also accepted (17–19). Assessed cerebrovascular structure or function (e.g., cerebral blood flow, vascular density, permeability, pericyte coverage). Exclusion criteria included non-mice models, human studies, irrelevant outcomes (e.g., amyloid pathology only), non-vascular interventions, or descriptive designs. Additional criteria appear in Supplementary Table 2. Key study characteristics (strain, age, design) were extracted by one reviewer (K.L.) and validated by another (N.F.) (Table 1). Outcome data were extracted independently and consolidated into a consensus dataset. Graphical data were digitised using WebPlotDigitizer (20). No major discrepancies were identified. Risk of Bias (RoB) and Quality Assessment Study quality and risk of bias were evaluated using established preclinical tools: the CAMARADES checklist (21) for overall study quality and the SYRCLE Risk of Bias tool (22) for individual bias domains. Assessments were conducted by one reviewer (K.L.) and cross-checked on a subset of studies by a second reviewer (N.F.) to ensure consistency. Synthesis Methods Meta-analyses assessed the impact of APOE genotype on cerebral blood flow (CBF) and vascular morphology in humanised APOE knock-in or targeted replacement (APOE-KI/TR) mice using RStudio (version 2024.12.1). Key study characteristics—experimental design, outcomes, age, and region analysed—were tabulated, and studies grouped by outcome domain (CBF or vascular morphology). Only those providing sufficient quantitative data (mean, SEM, and sample size) for APOE3 versus APOE4 comparisons were included in meta-analyses; others were summarised narratively. SEMs were converted to standard deviations using: Multiple studies compared different ages and brain subregions, resulting in several layers of analysis within a single study(11,23–26). Studies which contributed multiple subgroup comparisons (e.g., data from different brain regions or age groups) were treated as separate data points and analysed independently. Each subgroup was assigned an alphabetical suffix (e.g., a, b, c) and specific comparisons were detailed in the captions of forest plots. Given expected heterogeneity in design, imaging methods, and antibodies, all quantitative syntheses employed random-effects models. When studies contributed multiple within-study comparisons, a hierarchical random-effects model using robust variance estimation (RVE) accounted for non-independence of effect sizes. Between-group differences (APOE3 vs APOE4) were expressed as standardised mean differences (SMDs) with 95% confidence intervals (CIs). Where multiple measures were reported for the same group, pooled values were calculated following CAMARADES guidelines (27). As traditional heterogeneity metrics (Cochran’s Q, I²) are not available with RVE, τ² was reported to represent between-study variance. For analyses with a single effect size per study, standard random-effects models (metafor package, R) were used, providing SMDs, 95% CIs, and heterogeneity indices (τ², I², Q). Formal meta-analysis was not undertaken for outcomes with fewer than three studies. In these cases, results were summarised narratively, and consistency across subgroups and study characteristics was assessed qualitatively. Results Literature Retrieval The database search identified 1,493 records, with three additional studies found through manual searching. After removal of five duplicates, 1,488 titles and abstracts were screened. Of these, 1,438 were excluded: 711 were irrelevant; 17 involved the wrong population (e.g., human studies (28)); 109 used inappropriate designs (e.g., amyloid clearance (29,30), gene editing, nanomedicine); 384 employed unsuitable models (e.g., ApoE−/−, APP, or Lrp1-KO mice (31)); 79 used unrelated interventions (e.g., traumatic brain injury, artery occlusion, chronic inflammation (32)); and 138 reported unrelated outcomes (e.g., amyloid burden, behaviour, transcriptomics). Fifty studies underwent full-text review, and 18 met inclusion criteria for the systematic review ( Figure 1 ). Of these, seven were eligible for meta-analysis: six reported vascular morphology (four unique, two overlapping with CBF outcomes), and three examined genotype-dependent differences in cerebral blood flow. The remaining 11 were synthesised narratively. Study characteristics and exclusion details are provided in Supplementary Table 3 . Study Characteristics Characteristics of the 18 included studies are summarised in Table 1 . All used transgenic murine models expressing human APOE alleles, with variation in strain, background, supplier, group size, age, and sex. Nine studies employed C57BL/6 background mice, while others used outbred strains, EFAD lines (11,33–35), or did not specify background. Mouse ages ranged from 2 to 24 months, encompassing early life (2–4 months), midlife (6–12 months), and later life (up to 24 months). Methodologies also varied, including assessments of BBB permeability (e.g., gadolinium-based Ktrans, fibrinogen, fibronectin), vascular morphology, and cerebral blood flow. Detailed PICO (Population, Intervention, Comparator, Outcome) information for each study is provided in Table 2 . [Table 1, Table 2 here] Quality Assessment and Risk of Bias Study quality and reporting were evaluated using the CAMARADES checklist and SYRCLE Risk of Bias tool across all 18 studies. The mean CAMARADES score was 7.2/10 (72%), ranging from 4 to 9.5 ( Table 3 ). All studies used appropriate transgenic mouse models and adhered to animal welfare regulations; all but one (Bhattarai et al., 2025, bioRxiv ) were peer-reviewed. Owing to the transgenic design, all studies met the criterion for blinded induction. However, several domains showed incomplete reporting, and only four studies explicitly described sample size calculations. [ Table 3 here ] Several domains exhibited unclear risk of bias, particularly performance bias, selection bias, and detection bias, due to insufficient reporting ( Supplementary Table 4 ). Randomised housing was poorly reported in most studies. In contrast, attrition bias, reporting bias, and other sources of bias (e.g., conflict of interest declarations) were most consistently rated as low risk. Only Onos et al., (2024) was rated low risk across all SYRCLE domains. Study Outcomes Included studies assessed diverse yet overlapping aspects of blood–brain barrier breakdown (BBB-b). Five studies reported cerebral blood flow changes (11,23–25,33), and three examined tight junction integrity (24,36,37). Ten studies identified cerebrovascular morphological differences by APOE genotype, including vessel density, pericyte coverage, vessel diameter, and vascular volume (11,23–25,34–36,38–40). Mechanistic investigations varied: one study conducted multi-omic profiling (transcriptomic, lipidomic, proteomic) (41), and two evaluated neurovascular coupling in the visual cortex (42,43). Three studies explored Cyclophilin-A–NFκB–MMP9 signalling (23,24,30), while seven addressed pathways involving insulin resistance, metabolism, and mTOR signalling (23,26,33,38,44–46). Additional studies examined EGF-related resilience (34), peripheral inflammation (35), and glial contributions to BBB-b (25,39,40,47). Given the limited number of studies per outcome domain, formal assessments of reporting bias (e.g., funnel plot asymmetry, Egger’s test) were not performed. APOE4 Contributions to BBB Destabilisation and Meta-Analysis Cerebral Blood Flow Three studies assessed APOE-dependent differences in cerebral blood flow (CBF) across distinct age groups and brain regions, using dynamic susceptibility-contrast MRI (DSC-MRI) (11), arterial spin labelling MRI (ASL-MRI) (25), or autoradiography with radiolabelled tracers (24). Despite methodological variation, RVE meta-analysis indicated a consistent reduction in CBF in APOE4 mice relative to APOE3 (SMD = −2.87, 95% CI −5.14 to −0.60, df = 2.66; Figure 2 ). Between-study heterogeneity was moderate to substantial (τ² = 2.25), reflecting differences in imaging modality. All three studies were of moderate-to-high quality (mean CAMARADES score 8.3/10). However, risk of bias may have arisen from incomplete reporting of sample size calculations and uncertainty in performance bias domains (e.g., randomised housing). Overall, 10–60% of SYRCLE items were rated as unclear risk. Vascular Morphology Six studies investigated APOE genotype–dependent differences in vascular morphology using immunohistochemical markers. Two assessed laminin coverage as cortical percent area in 5- and 8-month-old EFAD mice (34,35). Three evaluated collagen-IV expression as a marker of vessel density—quantified as hippocampal capillary area in 9- and 12-month-old mice (6,43) or cortical signal intensity in 22-month-old mice (40). One measured neocortical microvessel density via CD31 immunostaining in 9-month-old mice (25). A random-effects meta-analysis (k = 6) showed a negative but non-significant trend toward reduced vascular morphology markers in APOE4 versus APOE3 mice (SMD = −0.59; 95% CI −1.39 to 0.20; p = 0.14; Figure 3 ). Heterogeneity was moderate to substantial (τ² = 0.64; I² = 66.3%; Q = 13.35, p = 0.020). Included studies had a mean CAMARADES score of 6.9/10. Risk of bias reflected incomplete reporting of temperature control, randomisation, and sample size calculation. According to SYRCLE, 30–60% of domains were rated unclear, primarily for selection and performance bias ( Supplementary Table 4 ). Narrative Summary of APOE4-Associated Mechanistic Pathways Remaining studies that could not be quantitatively synthesised were analysed narratively to provide complementary mechanistic insights ( Table 4 ). Table 4. Summary of key mechanistic axes implicated in APOE4-related blood–brain barrier (BBB) dysfunction. Insulin/mTOR Signalling Several studies have investigated the role of insulin resistance in brain metabolism and endothelial dysfunction in APOE4 mice, often focusing on the mammalian target of rapamycin (mTOR) signalling pathway. Johnson et al. (2019)(44) reported that high-fat diet (HFD)-induced insulin resistance was associated with impaired cognition, reduced cerebral blood volume (CBV), and decreased glucose uptake, with more pronounced effects in 15-month-old ε4 than ε3 mice. Administration of an acute exogenous glucose bolus produced a paradoxical improvement in CBV and cognition for APOE4 mice only, demonstrating a significant genotype effect (p=0.028). Onos et al. (2024)(46) further confirmed genotype-linked alterations in glucose metabolism. Longitudinal FDG-PET imaging revealed widespread reductions in glucose uptake in APOE4 mice across multiple brain regions, with complex age- and sex-specific dynamics. Fluctuating uptake patterns suggested a Type 1 neurovascular uncoupling process - where decreased glucose uptake co-occurs with increased perfusion at 8 and 12 months. Rhea et al. (2021)(38) examined insulin pharmacokinetics in 15-month-old APOE -TR mice, identifying genotype- and sex-specific differences in vascular insulin binding. APOE4 females exhibited higher amounts of reversible binding of 125 I-Insulin in the frontal cortex, than APOE3 females, proposed to reflect differences in insulin receptor availability or binding site accessibility on the luminal surface of brain endothelial cells. However, downstream dynamics regarding insulin transport across the BBB remained unclear. Lin et al. (2020)(33) further linked APOE4 -associated vascular dysfunction to overactivation of mTOR. In 7-month-old E4FAD mice, mTOR hyperactivity was associated with reduced P-glycoprotein (P-gp) transport at the BBB (p<0.001), impaired CBF, disrupted lipid metabolism, and elevated free fatty acids (FFAs). Sixteen weeks of rapamycin treatment restored BBB function and lipid homeostasis in APOE4 mice. Interestingly, rapamycin administration in APOE3 mice slowed glucose-associated metabolism – reducing glycolytic and TCA cycle intermediates – whereas APOE4 metabolism was normalised. This effect was suggested to mirror caloric restriction-like outcomes, where mTOR inhibition in relatively healthy animals reduces glucose uptake. Together, these findings support a model in which the APOE4 genotype is associated with dysregulated brain insulin signalling, reduced brain glucose uptake, and a paradoxical relationship with mTOR activation processes further shaped by diet, age, and sex. Cyclophilin A – NFκB – MMP9 Four studies demonstrate that the APOE4 genotype contributes to blood-brain barrier (BBB) breakdown via activation of the Cyclophilin A (CypA) – Nuclear Factor kappa B (NFkB) – matrix metalloproteinase 9 (MMP9) pathway. Bell et al. (2012)(24) reported a 5-6-fold increase in CypA levels within the cerebral microvessels of 6-month-old APOE4 mice under the GFAP promoter, primarily localised to pericytes. Further experiments demonstrated that this increase drives NFkB and MMP9 expression, resulting in pericyte-mediated ECM remodelling. Pharmacological inhibition of CypA with cyclosporine A, an immunosuppressant, attenuated these effects on BBB integrity. Lin et al. (2017)(23) corroborated CypA’s role in APOE4 -related vascular dysfunction. Here, treatment with rapamycin in 12-month-old APOE4 mice under the GFAP promoter normalised CypA and NFkB levels in cortical and microvascular compartments, improved BBB integrity, and restored CBF (ASL-MRI). Although total brain CypA levels were not significantly different between genotypes, APOE4 control mice showed reduced parenchymal and elevated vascular CypA levels, indicating regional heterogeneity in CypA expression. Ringland et al. (2020)(30) found impaired regulation of MMP9 in 6-month-old E4FAD mice, reporting a 56% increase in MMP9 immunoreactivity in cortical endothelial cells compared to E3FAD. At 70 weeks, E4FAD mice continued to show elevated MMP9 levels relative to E3FAD mice (p=0.0533), suggesting chronic dysregulation. In vivo experiments by Jackson et al. (2022) confirmed BBB disruption in APOE4 mice by 9-months-old. Using fluorescein-labelled 40 kDa dextran in tracer extravasation studies, increased permeability was observed. Quantitative PCR revealed a ~30% increase in MMP9 transcript levels in APOE4 mice, accompanied by elevated MMP9 protein levels. Transmission electron microscopy showed irregular gaps in tight junction membranes, suggesting focal openings rather than a global loss of structure. Additionally, measurement of astrocytic end-foot coverage around vessels demonstrated a significant reduction. Together, these studies implicate APOE4 in activating a proinflammatory and extracellular matrix remodelling cascade involving pericyte-endothelial interactions. This pathway contributes to BBB breakdown, including localised tight junction dysregulation, as early as 6 months of age. Occludin & ECM The APOE4 genotype has been strongly associated with compromised BBB integrity, driven by both structural and functional disruptions of endothelial and perivascular components. Several studies highlight specific markers of BBB dysregulation, including tight junction protein, fibrinogen, fibronectin and extracellular matrix components. Nishitsuji et al. (2011)(37) employed an in vitro BBB model incorporating primary astrocytes from APOE3 - or APOE4 -knock-in mice and found significantly reduced trans endothelial electrical resistance (TEER) in the APOE4 condition. While findings revealed no change in total tight junction protein levels, phosphorylation of occludin at threonine (Thr) residues – a modification necessary for occluding assembly into tight junctions – was significantly reduced. This was accompanied by diminished PKCη activation, suggesting a pathological role for APOE4 in impaired LRP1–PKCη–occludin signalling, ultimately compromising tight junction integrity. Yanckello et al. (2022) further examined tight junction protein expression in brain capillaries isolated from 7-month-old E3FAD and E4FAD mice, with or without dietary inulin (a prebiotic fibre). While inulin did not significantly alter occludin protein levels, Claudin-1 expression differed at baseline, with E4FAD control mice exhibiting elevated levels ( p = 0.0012). Yamazaki et al. (2020)(40) provided further evidence of APOE genotype-dependent BBB permeability using both in vitro and in vivo approaches. TEER was again significantly lower in BBB models containing APOE4 -expressing pericytes, despite no significant differences in tight junction protein levels. However, collagen IV expression was significantly reduced in APOE4 -pericyte models. Immunofluorescence confirmed diminished collagen IV in 8-month-old APOE4 mice. ELISA assays revealed increased cortical IgG and fibrinogen, supporting a role for extracellular matrix impairment in BBB permeability and an inverse relationship between extravasating markers and basement membrane structural integrity. Thomas et al. (2017)(34) identified that 8-month-old female E4FAD mice exhibited ~65% higher cortical fibrinogen levels and increased sodium fluorescein (NaFl) leakage compared o to other groups, suggesting a sex-specific vulnerability of the BBB. Treatment with epidermal growth factor (EGF) attenuated fibrinogen extravasation by about 40%. Barisano et al. (2022) used dynamic contrast-enhanced MRI and histological analysis to investigate BBB integrity in APOE -targeted replacement mice. Their findings revealed progressive BBB leakage in APOE4 mice, evidenced by pericapillary fibrinogen accumulation. Tissue analysis also showed a progressive loss of pericyte coverage (CD13 + ), which strongly inversely correlated with fibrinogen deposition, linking structural loss to BBB leakiness. Transcriptomic analysis of endothelial cells revealed upregulation of genes involved in cell adhesion (e.g., cadherins, contactins, catenins), interpreted as a possible compensatory response to vascular compromise. Finally, Bhattarai et al. (2025)(47) reported significantly elevated fibronectin accumulation in APOE4 mice - a hallmark of extracellular matrix remodelling - without changes in the CD31 endothelial marker, further demonstrating enhanced susceptibility to BBB permeability in the APOE4 genotype. Altogether, these studies emphasise the relationship between increased BBB permeability and reduced vascular stability in the APOE4 genotype. These three mechanistic axes provide compelling evidence of pervasive and progressive BBB structural breakdown in APOE4 contexts ( Figure 4 ). This disruption is marked by increased permeability - evidenced by tracer extravasation, blood product accumulation, and tight junction widening - and is accompanied by chronic inflammation, possibly driven by sustained extracellular matrix remodelling and perivascular dysfunction. Discussion and Future Perspectives This systematic review examined associations between the Apolipoprotein E4 ( APOE4 ) genotype and BBB integrity in humanised APOE -target replacement (TR) and knock-in (KI) mouse models. Recognising that methodological quality and bias assessment are critical for translational validity, this review began by evaluating the quality of included studies before synthesising trends in BBB structure, function, and molecular mechanisms. Using the CAMARADES checklist and SYRCLE Risk of Bias (RoB) tool(22), we assessed 22 studies that met eligibility criteria for reporting quality. The average CAMARADES score was 7.2/10 (72%), suggesting overall satisfactory experimental design; however, several areas were consistently underreported, particularly random allocation to treatment, blinding of outcome assessment, and transparent reporting of sample size calculation. These items are often assumed or implied within the field, but lack of - or insufficient - reporting of these critical design elements could compromise experimental validity or increase the risk of bias in outcomes. This concern was further underscored by the SYRCLE RoB assessment, which identified high rates of unclear reporting in selection and performance bias domains (e.g., allocation concealment, randomised housing, and caregiver blinding), highlighting the need for more rigorous and standardised reporting practices in preclinical BBB research. To assess genotype-related differences, meta-analyses were performed examining CBF and vascular morphology in APOE4 mice vs. APOE3 mice. Despite heterogeneity in imaging modalities (DSC- and ASL-MRI, and 14 C-autoradiography), studies consistently reported reduced CBF in APOE4 compared with other APOE genotypes across a broad age range (0–24 months). Analysis of vascular morphology, evaluating markers of endothelial and basement membrane integrity (laminin, collagen-IV, CD31), revealed a less straightforward trend with middle-aged mice (5-11 months) demonstrated higher vascular expression, compared to older (12-24 months) though the standard mean difference for APOE4 relative to APOE3 was not statistically significant. This finding, although not statistically significant, is notable and aligns with broader mechanistic data, namely the strong presence of Cyclophilin A (CypA) – NFkB – MMP9 and Occludin/ECM remodelling axes. Studies within the CypA axis consistently reported elevated levels of CypA, NFkB, and MMP9 – key mediators of inflammatory signalling and ECM remodelling. Although MMP9 can cleave glycoproteins such as laminin, it demonstrates higher binding affinity and substrate specificity to collagens, particularly collagen-IV(49,50). This supports a plausible link between ECM protein downregulation and proinflammatory activation of this remodelling pathway with APOE4 and potentially providing explanation for atypical permeability within the BBB construct. The Occludin and ECM Remodelling axis further reinforces this connection. APOE4 models showed reduced phosphorylation of occludin at threonine (Thr) - a post-translational modification critical for endothelial stability - along elevated markers of permeability such as fibrinogen, fibronectin, and IgG extravasation in perivascular tissue. Occludin regulation is thought to involve the LRP1–PKCη–occludin signalling cascade; however, in APOE4 mice, this pathway may be functionally impaired (Nishitsuji et al., 2011). PKCη also acts as a negative regulator of AKT(51), a kinase central to the PI3K/AKT/mTOR pathway. Regarding the third axis, Insulin/mTOR signalling, several studies included in this review reported insulin resistance, impaired glucose uptake (FDG-PET), and reduced cerebral blood volume (CBV) in APOE4 mice - all of which were attenuated by therapeutic administration of rapamycin, an mTOR inhibitor. These findings suggest a crucial link between metabolic dysfunction and cerebrovascular integrity in APOE4 carriers. Together, these three mechanistic axes (CypA–MMP9, mTOR/insulin, and occludin-related ECM remodelling) provide converging evidence for chronic and progressive BBB structural breakdown in APOE4 carriers. This disruption is characterised by 1) overt direct evidence of BBB permeability (e.g., tracer leakage, blood product accumulation), 2) features which could drive BBB leakage (e.g., tight junction instability, reduced occludin phosphorylation, widened junctions), 3) features which could stem from BBB leakage or worsen it if present (e.g., inflammation and ECM degradation, MMP9, fibronectin), and 4) Impaired vascular metabolism and function (e.g., endothelial glucose uptake, CBF deficits). Limitations This review has several limitations. First, the number of studies eligible for meta-analysis was relatively small, limiting the statistical power and generalisability of pooled findings. Subgroup analyses were further limited by methodological heterogeneity, particularly differences in measurement techniques, biomarkers, and arbitrarily defined age ranges. Additionally, many studies received unclear ratings in key domains of the CAMARADES and SYRCLE quality assessments. Details regarding blinding, sample size calculation, and allocation procedures were often underreported, limiting the ability to assess internal validity. These omissions not only affect the reliability of individual studies but also weaken the overall strength of evidence in the systematic review. Future research should prioritise methodological transparency. Adoption of preclinical reporting standards, such as SYRCLES RoB or ARRIVE guidelines, would improve reproducibility and minimise bias. Finally, consistency in genotype comparators would strengthen the interpretability of APOE -targeted studies. While wildtype mice provide a useful control for transgenic comparisons, APOE3 should be the preferred genetic control when studying humanised APOE isoforms, as it represents the most common and asymptomatic variant. Comparisons between APOE4 and APOE3 are critical to isolate pathogenic effects specific to APOE4 -related dysfunction. Conclusion In conclusion, this systematic review demonstrates that the APOE4 genotype in mice compromises blood-brain barrier integrity through consistent pathogenic alterations in vascular morphology and cerebral blood flow dynamics, alongside clearly defined mechanistic pathways of metabolic and vascular dysfunction. Preclinical studies using humanised APOE models provide valuable insight into these genotype-specific effects and offer a robust platform to inform translational research, including therapeutic development, biomarker discovery, and precision medicine strategies. Declarations Ethics approval and consent to participate Not Applicable Consent for publication Not Applicable Availability of data and materials All data generated or analysed during this study are included in this article and its supplementary information files. The extracted dataset and analysis code that support the findings are available from the corresponding author on reasonable request. The full database search strategies have been deposited on searchRxiv (DOI: 10.1079/searchRxiv.2025.01124). The review protocol was registered with PROSPERO (CRD420251010665): https://www.crd.york.ac.uk/PROSPERO/view/CRD420251010665. Competing interests The authors declare that they have no competing interests. Funding The work of AM is supported by the UK Dementia Research Institute (award UKDRI-4209) through UK DRI Ltd, funded by the UK Medical Research Council (MRC), Alzheimer’s Society UK, Alzheimer’s Research UK, and the British Heart Foundation. AM also holds a UKRI MRC Career Development Award (MR/V032488/1) and a UK DRI Theme Funding Programme Award (DRI-TFP-2024-7). JMW is supported by the UK Dementia Research Institute (award UK DRI-4002) through UK DRI Ltd, funded by the MRC, Alzheimer’s Society, and Alzheimer’s Research UK; by the Row Fogo Charitable Trust [BRO-D.FID3668413] (JMW); and by the NHS Lothian Research and Development Office (MJT). KKL received support from the University of Edinburgh Wellcome Trust Translational Neuroscience 4-year PhD Programme (Grant No. 110002 20132001 TBC 130956 00000000 10002374 000). Authors' contributions KL conceived and designed the study; developed the searches; screened records; extracted and curated data; led analyses; performed primary risk-of-bias/quality assessments; prepared figures/tables; and drafted the manuscript. NF served as second reviewer for screening and full-text review, provided annotations, validated demographic extractions, and conducted risk-of-bias/quality assessments on a subset to verify consistency with KL; the remaining assessments were completed by KL due to resource constraints. KL and NF extracted outcomes independently, with discrepancies resolved by consensus to a single dataset. AM (PI) supervised the study, adjudicated screening disagreements, and provided critical revisions. JW (PI) supervised and provided critical revisions. All authors approved the final manuscript. Acknowledgements We thank Marshall Dozier , College Lead for Library Academic Support, for expert guidance in developing and refining the database search strategies and keyword terms. We also thank Gillian Currie (CAMARADES) for methodological advice to ensure this systematic review met community standards. The CAMARADES consortium is funded by the UK National Centre for the Replacement, Refinement and Reduction of Animals in Research (NC3Rs) to support preclinical systematic reviews. Authors' information (optional) References Huang Y, Mahley RW. Apolipoprotein E: Structure and function in lipid metabolism, neurobiology, and Alzheimer’s diseases. Neurobiol Dis. 2014 Dec 1;72:3–12. Phillips MC. Apolipoprotein E isoforms and lipoprotein metabolism. IUBMB Life. 2014 Sept;66(9):616–23. Blumenfeld J, Yip O, Kim MJ, Huang Y. Cell type-specific roles of APOE4 in Alzheimer disease. Nat Rev Neurosci. 2024 Feb;25(2):91–110. Farrer LA, Cupples LA, Haines JL, Hyman B, Kukull WA, Mayeux R, et al. Effects of Age, Sex, and Ethnicity on the Association Between Apolipoprotein E Genotype and Alzheimer Disease: A Meta-analysis. JAMA. 1997 Oct 22;278(16):1349–56. Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, et al. Gene Dose of Apolipoprotein E Type 4 Allele and the Risk of Alzheimer’s Disease in Late Onset Families. Science. 1993 Aug 13;261(5123):921–3. Foley KE, Wilcock DM. Three major effects of APOEε4 on Aβ immunotherapy induced ARIA. Front Aging Neurosci [Internet]. 2024 May 2 [cited 2025 Feb 11];16. Available from: https://www.frontiersin.org/journals/aging-neuroscience/articles/10.3389/fnagi.2024.1412006/full Hampel H, Elhage A, Cho M, Apostolova LG, Nicoll JAR, Atri A. Amyloid-related imaging abnormalities (ARIA): radiological, biological and clinical characteristics. Brain. 2023 Nov 1;146(11):4414–24. Ward A, Crean S, Mercaldi CJ, Collins JM, Boyd D, Cook MN, et al. Prevalence of apolipoprotein E4 genotype and homozygotes (APOE e4/4) among patients diagnosed with Alzheimer’s disease: a systematic review and meta-analysis. Neuroepidemiology. 2012;38(1):1–17. Lewandowski CT, Weng JM, LaDu MJ. Alzheimer’ss Disease Pathology in ApoE Transgenic Mouse Models: The Who, What, When, Where, Why, and How. Neurobiol Dis. 2020 June;139:104811. Montagne A, Nation DA, Sagare AP, Barisano G, Sweeney MD, Chakhoyan A, et al. APOE4 leads to blood-brain barrier dysfunction predicting cognitive decline. Nature. 2020 May;581(7806):71–6. Montagne A, Nikolakopoulou AM, Huuskonen MT, Sagare AP, Lawson EJ, Lazic D, et al. APOE4 accelerates advanced-stage vascular and neurodegenerative disorder in old Alzheimer’s mice via cyclophilin A independently of amyloid-β. Nat Aging. 2021 June;1(6):506–20. Kapoor A, Nation DA. Platelet-derived growth factor-BB and white matter hyperintensity burden in APOE4 carriers. Cereb Circ - Cogn Behav. 2022 Jan 1;3:100131. Blanchard JW, Bula M, Davila-Velderrain J, Akay LA, Zhu L, Frank A, et al. Reconstruction of the Human Blood-Brain Barrier in vitro reveals a Pathogenic Mechanism of APOE4 in Pericytes. Nat Med. 2020 June;26(6):952–63. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. The BMJ. 2021 Mar 29;372:n71. Bahor Z, Liao J, Currie G, Ayder C, Macleod M, McCann SK, et al. Development and uptake of an online systematic review platform: the early years of the CAMARADES Systematic Review Facility (SyRF). BMJ Open Sci [Internet]. 2021 Mar 31 [cited 2025 June 1];5(1). Available from: http://access.portico.org/stable?au=phzmqrt8kdx Tai LM, Balu D, Avila-Munoz E, Abdullah L, Thomas R, Collins N, et al. EFAD transgenic mice as a human APOE relevant preclinical model of Alzheimerʼns disease [S]. J Lipid Res. 2017 Sept 1;58(9):1733–55. Shenk JC, Liu J, Fischbach K, Xu K, Puchowicz M, Obrenovich ME, et al. The effect of acetyl-L-carnitine and R-α-lipoic acid treatment in ApoE4 mouse as a model of human Alzheimer’s disease. J Neurol Sci. 2009;283(1–2):199–206. Jansen D, Zerbi V, Janssen CIF, van Rooij D, Zinnhardt B, Dederen PJ, et al. Impact of a multi-nutrient diet on cognition, brain metabolism, hemodynamics, and plasticity in apoE4 carrier and apoE knockout mice. Brain Struct Funct. 2014;219(5):1841–68. Zerbi V, Wiesmann M, Emmerzaal TL, Jansen D, Van Beek M, Mutsaers MPC, et al. Resting-state functional connectivity changes in aging apoE4 and apoE-KO mice. J Neurosci Off J Soc Neurosci. 2014;34(42):13963–75. Rohatgi A. WebPlotDigitizer [Internet]. automeris.io; 2025 [cited 2025 Jan 17]. Available from: https://automeris.io Macleod MR, O’Collins T, Howells DW, Donnan GA. Pooling of Animal Experimental Data Reveals Influence of Study Design and Publication Bias. Stroke. 2004 May;35(5):1203–8. Hooijmans CR, Rovers MM, de Vries RB, Leenaars M, Ritskes-Hoitinga M, Langendam MW. SYRCLE’s risk of bias tool for animal studies. BMC Med Res Methodol. 2014 Mar 26;14(1):43. Lin AL, Jahrling JB, Zhang W, DeRosa N, Bakshi V, Romero P, et al. Rapamycin rescues vascular, metabolic and learning deficits in apolipoprotein E4 transgenic mice with pre-symptomatic Alzheimer’s disease. J Cereb Blood Flow Metab. 2017 Jan 1;37(1):217–26. Bell RD, Winkler EA, Singh I, Sagare AP, Deane R, Wu Z, et al. Apolipoprotein E controls cerebrovascular integrity via cyclophilin A. Nature. 2012 May 16;485(7399):512–6. Koizumi K, Hattori Y, Ahn SJ, Buendia I, Ciacciarelli A, Uekawa K, et al. Apoε4 disrupts neurovascular regulation and undermines white matter integrity and cognitive function. Nat Commun. 2018 Sept 19;9(1):3816. Wiesmann M, Zerbi V, Jansen D, Haast R, Lütjohann D, Broersen LM, et al. A Dietary Treatment Improves Cerebral Blood Flow and Brain Connectivity in Aging apoE4 Mice. Neural Plast. 2016;2016:6846721. CAMARADES Systematic Review Wiki [Internet]. [cited 2025 Aug 26]. Meta-Analysis. Available from: https://camarades.de/wiki-content/10-metaanalysis.html Halliday MR, Rege SV, Ma Q, Zhao Z, Miller CA, Winkler EA, et al. Accelerated pericyte degeneration and blood–brain barrier breakdown in apolipoprotein E4 carriers with Alzheimer’s disease. J Cereb Blood Flow Metab. 2016 Jan;36(1):216. Bachmeier C, Shackleton B, Ojo J, Paris D, Mullan M, Crawford F. Apolipoprotein E Isoform-Specific Effects on Lipoprotein Receptor Processing. NeuroMol Med. 2014;16(4):686–96. Ringland C, Schweig JE, Paris D, Shackleton B, Lynch CE, Eisenbaum M, et al. Apolipoprotein E isoforms differentially regulate matrix metallopeptidase 9 function in Alzheimer’s disease. Neurobiol Aging. 2020;95:56–68. Oue H, Yamazaki Y, Qiao W, Yuanxin C, Ren Y, Kurti A, et al. LRP1 in vascular mural cells modulates cerebrovascular integrity and function in the presence of APOE4 . JCI Insight [Internet]. 2023 Apr 10 [cited 2024 July 31];8(7). Available from: https://insight.jci.org/articles/view/163822 Na H, Yang JB, Zhang Z, Gan Q, Tian H, Rajab IM, et al. Peripheral apolipoprotein E proteins and their binding to LRP1 antagonize Alzheimer’s disease pathogenesis in the brain during peripheral chronic inflammation. Neurobiol Aging. 2023;127:54–69. Lin AL, Parikh I, Yanckello LM, White RS, Hartz AMS, Taylor CE, et al. APOE genotype-dependent pharmacogenetic responses to rapamycin for preventing Alzheimer’s disease. Neurobiol Dis. 2020 June 1;139:104834. Thomas R, Morris AWJ, Tai LM. Epidermal growth factor prevents APOE4-induced cognitive and cerebrovascular deficits in female mice. Heliyon [Internet]. 2017;3(6). Available from: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85020180648&doi=10.1016%2fj.heliyon.2017.e00319&partnerID=40&md5=9e788ede5036ea91221ed97b94b08774 Marottoli FM, Katsumata Y, Koster KP, Thomas R, Fardo DW, Tai LM. Peripheral Inflammation, Apolipoprotein E4, and Amyloid-β Interact to Induce Cognitive and Cerebrovascular Dysfunction. ASN Neuro. 2017;9(4):1759091417719201. Alata W, Ye Y, St-Amour I, Vandal M, Calon F. Human apolipoprotein E ɛ4 expression impairs cerebral vascularization and blood-brain barrier function in mice. J Cereb Blood Flow Metab. 2015;35(1):86–94. Nishitsuji K, Hosono T, Nakamura T, Bu G, Michikawa M. Apolipoprotein E regulates the integrity of tight junctions in an isoform-dependent manner in an in vitro blood-brain barrier model. J Biol Chem. 2011;286(20):17536–42. Rhea EM, Hansen K, Pemberton S, Torres ERS, Holden S, Raber J, et al. Effects of apolipoprotein E isoform, sex, and diet on insulin BBB pharmacokinetics in mice. Sci Rep [Internet]. 2021;11(1). Available from: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85115398840&doi=10.1038%2fs41598-021-98061-1&partnerID=40&md5=367fc7cda7da7aeafa7603dd86c6cc43 Jackson RJ, Meltzer JC, Nguyen H, Commins C, Bennett RE, Hudry E, et al. APOE4 derived from astrocytes leads to blood-brain barrier impairment. Brain. 2022;145(10):3582–93. Yamazaki Y, Shinohara M, Yamazaki A, Ren Y, Asmann YW, Kanekiyo T, et al. ApoE (Apolipoprotein E) in Brain Pericytes Regulates Endothelial Function in an Isoform-Dependent Manner by Modulating Basement Membrane Components. Arter Thromb Vasc Biol. 2020;40(1):128–44. Barisano G, Kisler K, Wilkinson B, Nikolakopoulou AM, Sagare AP, Wang Y, et al. A ‘multi-omics’ analysis of blood-brain barrier and synaptic dysfunction in APOE4 mice. J Exp Med. 2022;219(11). Bonnar O, Shaw K, Anderle S, Grijseels DM, Clarke D, Bell L, et al. APOE4 expression confers a mild, persistent reduction in neurovascular function in the visual cortex and hippocampus of awake mice. J Cereb Blood Flow Metab. 2023 Nov;43(11):1826–41. Anderle S, Bonnar O, Henderson J, Shaw K, Chagas AM, McMullan L, et al. APOE4 and sedentary lifestyle synergistically impair neurovascular function in the visual cortex of awake mice. Commun Biol. 2025;8(1):144. Johnson LA, Torres ER, Weber Boutros S, Patel E, Akinyeke T, Alkayed NJ, et al. Apolipoprotein E4 mediates insulin resistance-associated cerebrovascular dysfunction and the post-prandial response. J Cereb Blood Flow Metab. 2019;39(5):770–81. Rhea EM, Torres ERS, Raber J, Banks WA. Insulin BBB pharmacokinetics in young apoE male and female transgenic mice. PLoS ONE [Internet]. 2020;15(1). Available from: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85078745276&doi=10.1371%2fjournal.pone.0228455&partnerID=40&md5=a33e8c83538370a3279425e62da5608b Onos KD, Lin PB, Pandey RS, Persohn SA, Burton CP, Miner EW, et al. Assessment of neurovascular uncoupling: APOE status is a key driver of early metabolic and vascular dysfunction. Alzheimers Dement. 2024;20(7):4951–69. Bhattarai P, Yilmaz E, Cakir EÖ, Korkmaz HY, Lee AJ, Ma Y, et al. APOE-ε4-induced Fibronectin at the blood-brain barrier is a conserved pathological mediator of disrupted astrocyte-endothelia interaction in Alzheimer’s disease [Internet]. bioRxiv; 2025 [cited 2025 July 28]. p. 2025.01.24.634732. Available from: https://www.biorxiv.org/content/10.1101/2025.01.24.634732v1 Laing K. BioRender. 2025 [cited 2025 Nov 2]. Evidence-mapping schematic of studies included in the narrative synthesis. Available from: https://BioRender.com/b0tt4qm Kridel SJ, Chen E, Kotra LP, Howard EW, Mobashery S, Smith JW. Substrate Hydrolysis by Matrix Metalloproteinase-9*. J Biol Chem. 2001 Jan 1;276(23):20572–8. Cui N, Hu M, Khalil RA. Biochemical and Biological Attributes of Matrix Metalloproteinases. Prog Mol Biol Transl Sci. 2017;147:1–73. Rao R. OCCLUDIN PHOSPHORYLATION IN REGULATION OF EPITHELIAL TIGHT JUNCTIONS. Ann N Y Acad Sci. 2009 May;1165:62–8. Yanckello LM, Hoffman JD, Chang YH, Lin P, Nehra G, Chlipala G, et al. Apolipoprotein E genotype-dependent nutrigenetic effects to prebiotic inulin for modulating systemic metabolism and neuroprotection in mice via gut-brain axis. Nutr Neurosci. 2022 Aug;25(8):1669–79. Tables Table 1 . List of studies selected for review, including strain, group size, age, and sex. Studies included in synthesis are in bold. Study Strain (Supplier) Group Size Age (s) Sex Ref Nishitsuji et al., 2011 Human apoE mice, generated using homologous recombination method in embryonic stem cells - Primary cultures of mouse brain capillary endothelial cells (mBECs) were prepared from 3-week-old mice - (37) Bell et al., 2012 TR-APOE mice (C57BL/6 background), generated in-house n = 5 /group 4-9 months - (24) Alata et al., 2015 E2, E3, and E4 targeted replacement mice (C57BL/6 background) Taconic Transgenic Models (Hudson, NY, USA) n = 5-6/group 12 months Male, Female (36) Thomas et al., 2017 EFAD with the 5 Familial Alzheimer’s disease (FAD) mutations (APP K670N/M671L + I716 V + V717I and PS1 M146L + L286 V) with APOE -targeted replacement mice n = 6-8/group 6-8.5 months Male, Female (34) Lin et al., 2017 APOE4 transgenic mice with GFAP promoter (C57BL/6 background) Jackson Laboratory (Bar Harbor, Maine, USA) n = 6/group 1-7 months Female (23) Marottoli et al., 2017 5xFAD +/− / APOE +/+ (EFAD) mice (C57BL/6 background) n = 8/group 4-6 months Male (35) Koizumi et al., 2018 ApoE3-TR and ApoE4-TR mice (C57BL/6 background) n = 5/group 3-4 months Male (25) Johnson et al., 2019 Human E3- and E4-targeted replacement mice Cerebral Blood Volume measure: n = 7-8/group 15 months Female (44) Yamazaki et al., 2020 ApoE3-TR and ApoE4-TR mice (C57BL/6 background) Taconic Transgenic Models (Hudson, NY, USA) n = 4/group 22 months Male, Females (40) Ringland et al., 2020 ApoE3-TR and ApoE4-TR mice Taconic Transgenic Models (Hudson, NY, USA) EFAD (5xFAD +/− APOE +/+ ) n = 4 for each genotype 6 months Female (30) Lin et al., 2020 EFAD with the 5 Familial Alzheimer’s disease (FAD) mutations (APP K670N/M671L + I716 V + V717I and PS1 M146L + L286 V) with APOE -targeted replacement mice (C57BL/6 background) n = 8/group 2 months Male, Female (33) Montagne et al., 2021 ApoE3-TR and ApoE4-TR mice (C57BL/6 background) IHC: n = 5-6/group BBB: n = 6-8/ group CBF: n = 12-14/group 18-24 months Male, Female (11) Rhea et al., 2021 Human E3- and E4-targeted replacement mice n = 10/group 15 months Male, Female (38) Barisano et al., 2022 Human APOE3 and APOE4 KI flox/flox mice (C57BL/6 background) n = 8/group 2–3, 4–6, and 9–12 months Male, Female (41) Jackson et al., 2022 Conditional humanized APOE knock-in (KI) mice n = 6/group 8-8.5 months - (39) Yanckello et al., 2022 Human APOE3(E3FAD mice) and APOE4gene (E4FAD mice) n = 8/group; (male: female= 1:1) 7 months Male, Female (52) Onos et al., 2024 B6J. APOE3 KI (hAPOEε3, available as B6.Cg-Apoeem2(APOE*)Adiuj/J, B6J. APOE4 KI (hAPOEε4, available as B6(SJL)-Apoetm1.1(APOE*4)Adiuj/J Jackson Laboratory (JAX) (JAX#029018, JAX#027894) n = 2-5/group 4-12 months Male, Female (46) Bhattarai et al., 2025 humanized targeted-replacement APOE-ε4 and APOE-ε3 mice N = 6 (50% female) for every experiment 18-22 months of age Male, Female (47) Table 2. PICO elements (Population, Intervention, Comparator, Outcome) across studies. BBB Integrity: grouped IHC-based vascular markers which labelling basement membrane components (Col-IV, Glut1, Laminin). BBB Leakage: measured with integrated density of fibrinogen and MRI-derived Ktrans, reflecting functional barrier permeability. Study Comparator Intervention Components Outcome Domain Outcome Measure Time Points Brain Regions Data: Mean ± SEM, N / group Effect & SE Included in synthesis Bell et al., 2012 APOE3 vs APOE4 BBB Pericyte Markers IHC (% Pericyte Coverage) 2 weeks, 9 months Cortex, Hippocampus Yes Yes Yes CBF CBF 14 C-iodoantipyrine autoradiograms 2 weeks, 9 months Sensorimotor, Ectorhinal Cortex, Hippocampus, Caudate Putamen, Corpus Callosum, Thalamus Yes Yes Yes Alta et al., 2014 APOE3 vs APOE4 BBB BBB Integrity IHC (% Area Occupied) 12 months Cortex Yes Yes Yes Thomas et al., 2017 APOE3 vs APOE4 (E3FAD vs E4FAD) BBB BBB Integrity IHC (% Area Occupied) 22 months Cortex Yes Yes Yes Marotolli et al., 2017 APOE3 vs APOE4 (E3FAD vs E4FAD) BBB BBB Integrity IHC (% Area Occupied) 4-6 months Cortex Yes Yes Yes Koizumi et al., 2018 APOE3 vs APOE4 BBB BBB Integrity IHC (% Area Occupied) 3-4 months Cortex Yes Yes Yes CBF CBF ASL 3-4 months Cortex, Caudate Nucleus Yes Yes Yes Yamazaki et al., 2020 APOE3 vs APOE4 BBB BBB Integrity Relative Intensity Signal 22 months Cortex Yes Yes Yes Montagne et al., 2021 APOE3 vs APOE4 BBB Pericyte Markers IHC (% Pericyte Coverage) 18-24 months Cortex, Hippocampus Yes Yes Yes BBB Leakage IHC (Integrated Density x10^3) 18-24 months Cortex, Hippocampus Yes Yes Yes Ktrans (x10^-3 min^-1) 18-24 months Cortex, Hippocampus Yes Yes Yes CBF CBF ASL (ml 100g-1 min-1) 18-24 months Cortex, Hippocampus Yes Yes Yes Table 3. CAMARADES checklist for study quality. Each study was assessed against the following criteria: (1) publication in a peer-reviewed journal; (2) statement of temperature control; (3) random allocation to treatment or control; (4) blinded induction of the model; (5) blinded assessment of outcome; (6) use of anaesthetic without significant intrinsic neuroprotective activity; (7) use of an appropriate animal model; (8) sample size calculation; (9) compliance with animal welfare regulations; (10) statement of potential conflicts of interest. “Ref” indicates references. An “x” denotes fulfilment of the criterion (score = 1); a blank cell indicates non-fulfilment. An asterisk (*) represents partial fulfilment (score = 0.5). See supplementary material for additional details. Studies in bold italics were included in the meta-analysis. Study 1 2 3 4 5 6 7 8 9 10 Score Nishitsuji et al., 2011 x x x x 4 Bell et al., 2012 x x x x x x x x 8 Alata et al., 2015 x x x * x x x 6.5 Thomas et al., 2017 x x x * x x x 6.5 Lin et al., 2017 x x x x x x x x 8 Marottoli et al., 2017 x x x x x x x 7 Koizumi et al., 2018 x x x x x x x x 8 Johnson et al., 2019 x x x * x x x 6.5 Yamazaki et al., 2020 x x * x x x 5.5 Ringland et al., 2020 x x * x x x 5.5 Lin et al., 2020 x x x x x * x x x x 9.5 Montagne et al., 2021 x x x x x x x x x 9 Yamazaki et al., 2021 x x x x x x x 7 Rhea et al., 2021 x x x x x x x x 8 Barisano et al., 2022 x x x x x x x x x 9 Jackson et al., 2022 x x x x x x x 7 Yanckello et al., 2022 x x x x x x x 7 Bonnar et al., 2023 x x x * x x x x x 8.5 Onos et al., 2024 x x x x x x x x x 9 Anderle et al., 2025 x x x x x x x 7 Bhattarai et al., 2025 * X X X X X 5.5 Additional Declarations No competing interests reported. Supplementary Files SupplementaryTable3Laing.zip SupplementaryTables.docx Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 29 Jan, 2026 Reviews received at journal 24 Jan, 2026 Reviewers agreed at journal 04 Jan, 2026 Reviewers invited by journal 12 Nov, 2025 Editor assigned by journal 05 Nov, 2025 Submission checks completed at journal 05 Nov, 2025 First submitted to journal 02 Nov, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8011437","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Systematic Review","associatedPublications":[],"authors":[{"id":579919222,"identity":"60360bd7-9a53-4d94-affd-63586388b97b","order_by":0,"name":"Krystal Laing","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAw0lEQVRIiWNgGAWjYBACCXYGhgMMDDZgEggSiNDCDFacRqIWIDhMghbJZuaHhwt+nZfjO8D88ANjWxphLdLMbAaHZ/bdNpY8wGYswdiWQ1iLHDODwWHentuJGw4wmDEwtlUQo4X9A1DLufoNB9i/EadFmpnH4DDPjwMJBgd4QLYQ4TDJZp6Cw7wNyYYzD/MUSyScI8L7EsfbN3/m+WMnz3e8feOHD2XJhLWAAWMbkABFUAKRGoDgD/FKR8EoGAWjYAQCAB3sN+7r+3wtAAAAAElFTkSuQmCC","orcid":"","institution":"University of Edinburgh","correspondingAuthor":true,"prefix":"","firstName":"Krystal","middleName":"","lastName":"Laing","suffix":""},{"id":579919223,"identity":"4ebe746d-2607-4545-9d40-1c04c7e775b1","order_by":1,"name":"Nela Fialova","email":"","orcid":"","institution":"University of Edinburgh","correspondingAuthor":false,"prefix":"","firstName":"Nela","middleName":"","lastName":"Fialova","suffix":""},{"id":579919224,"identity":"bc14020f-6c3b-4b63-8adf-035ef35e680a","order_by":2,"name":"Joanna Wardlaw","email":"","orcid":"","institution":"University of Edinburgh","correspondingAuthor":false,"prefix":"","firstName":"Joanna","middleName":"","lastName":"Wardlaw","suffix":""},{"id":579919225,"identity":"32086521-e81f-4c98-846f-b05e62fb610e","order_by":3,"name":"Axel Montagne","email":"","orcid":"","institution":"University of Edinburgh","correspondingAuthor":false,"prefix":"","firstName":"Axel","middleName":"","lastName":"Montagne","suffix":""}],"badges":[],"createdAt":"2025-11-02 14:08:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8011437/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8011437/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101436877,"identity":"8596608f-3eb4-4cb3-9c17-aaef32bf69fe","added_by":"auto","created_at":"2026-01-29 16:26:40","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":182992,"visible":true,"origin":"","legend":"\u003cp\u003eFlow diagram of study selection process following PRISMA 2020 guidelines.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8011437/v1/34c723e60e67ebbf57f10808.png"},{"id":101436955,"identity":"2a8083ee-2b94-4a46-8d86-6a6803ff6943","added_by":"auto","created_at":"2026-01-29 16:26:54","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":208236,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMeta-analysis of cerebral blood flow differences between APOE genotypes by age and brain subregion, using robust variance estimation (RVE). \u003c/strong\u003eStandardised mean differences (SMDs) comparing APOE4 vs. APOE3 were derived from outcome comparisons of brain region and age across three studies. RVE was used to account for within-study clustering, with four distinct clusters (Montagne et al., 2021; Koizumi et al., 2018; and Bell et al.,2012, which was divided into clusters “a” and “b” representing two age groups). The RVE summary effect is shown as the dark blue polygon. Between-study heterogeneity was moderate-to-substantial (τ2 = 2.25).\u003c/p\u003e\n\u003cp\u003eNote: Samples sizes were not adjusted for repeated comparisons within studies due to small group sizes, and degrees of freedom were limited (2.66).\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8011437/v1/bbc4ce2ac9b2a929b9981b3b.png"},{"id":101436846,"identity":"5a169a64-9ba5-4431-a8f6-3c2c2d46bb13","added_by":"auto","created_at":"2026-01-29 16:26:26","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":106769,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eMeta-analysis of vascular morphological differences between APOE genotypes using a random-effects model.\u003c/strong\u003e\u003c/em\u003e\u003cem\u003e Standardised mean differences (SMDs) comparing APOE4 mice vs. APOE3 mice were calculated across five studies assessing vascular morphology using IHC markers (laminin, collagen-IV, CD31). The summary effect is shown as the dark blue polygon. Between-study heterogeneity was moderate (τ2 = 0.64).\u003c/em\u003e\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8011437/v1/6959b6a2db3f8383dee5e42c.png"},{"id":101436838,"identity":"10105ed2-53a0-423d-a0c9-1870bf0993ac","added_by":"auto","created_at":"2026-01-29 16:26:22","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":314803,"visible":true,"origin":"","legend":"\u003cp\u003eEvidence mapping schematic of studies included in the narrative synthesis. This figure illustrates three principal pathways reported in the literature: (i) CypA-NFκB-MMP9 signalling, (ii) insulin/PI3K-mTOR signalling, and (iii) occluding and extracellular matrix (ECM) integrity. Nodes are annotated with studies contributing evidence. Blue = APOE4, green = intermediate mediators, yellow = molecular/protein markers, red = outcomes. \u003cem\u003eAbbreviations: CypA, cyclophilin A; NFκB, nuclear factor kappa B; MMP9, matrix metalloproteinase 9; mTOR, mammalian target of rapamycin; P-gp, P-glycoprotein; FFA, free fatty acids; CBF, cerebral blood flow; CBV, cerebral blood volume; TEER, trans-endothelial electrical resistance; FN1, fibronectin. Created in BioRender. Laing, K. (2025)\u003c/em\u003e(48)\u003cem\u003e.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-8011437/v1/6f7c83803929d2d30aaeebd9.png"},{"id":101436970,"identity":"c5d15e6a-c656-46c2-a421-69ca287429cb","added_by":"auto","created_at":"2026-01-29 16:27:00","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2209606,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8011437/v1/88c73bb2-8baa-4914-aab8-1cd90d912c03.pdf"},{"id":101436848,"identity":"a78238b1-a5be-490a-826f-209c94f84e28","added_by":"auto","created_at":"2026-01-29 16:26:27","extension":"zip","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1604550,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable3Laing.zip","url":"https://assets-eu.researchsquare.com/files/rs-8011437/v1/1edaa86e066c8a2ddbf13afd.zip"},{"id":101436839,"identity":"4447dce4-55f1-4730-b719-c82db8311ad9","added_by":"auto","created_at":"2026-01-29 16:26:23","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":25359,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTables.docx","url":"https://assets-eu.researchsquare.com/files/rs-8011437/v1/0d5fd199fe3422c4d9b312f5.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Impact of Apolipoprotein E4 on Blood-Brain Barrier Integrity in Target Replacement Murine Models: A Systematic Review and Meta-Analysis","fulltext":[{"header":"Background","content":"\u003cp\u003eApolipoprotein E (APOE) is a multifunctional lipid transporter involved in lipoprotein metabolism and cholesterol homeostasis (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). The ε4 variant is the strongest genetic risk factor for late-onset Alzheimer\u0026rsquo;s disease (AD) (\u003cspan additionalcitationids=\"CR4\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e) and has also been linked to other disorders, including vascular dementia, atherosclerosis, and cardiovascular disease, through mechanisms such as amyloid-β (Aβ) accumulation, neurodegeneration, and neuroinflammation (\u003cspan additionalcitationids=\"CR7\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). Heterozygosity for APOE4 increases AD risk 3\u0026ndash;4-fold, while homozygosity raises it by ~\u0026thinsp;15-fold (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). Although only 20\u0026ndash;25% of the general population carry an ɛ4 allele, 40\u0026ndash;65% of AD patients are ɛ4 carriers. As of 2020, nearly 50\u0026nbsp;million people were affected globally by AD or related dementias (\u003cspan additionalcitationids=\"CR11 CR12\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe mechanisms by which ε4 increases dementia risk remain incompletely understood despite extensive study. Recently, several Aβ-targeting therapies have demonstrated robust biomarker outcomes but carry an elevated risk of Amyloid-Related Imaging Abnormalities (ARIA) in APOE4 carriers (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). ARIA manifests as cerebral oedema or haemorrhage on MRI following Aβ immunotherapy and can, in severe cases, cause morbidity or death (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). This underscores the need for a deeper understanding of the molecular and cellular mechanisms driving APOE4-related pathology to inform safer, more precise interventions.\u003c/p\u003e \u003cp\u003eHuman studies of APOE4 pathophysiology are constrained by limited ɛ4 homozygosity (~\u0026thinsp;9.6% in AD patients (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e)) and the invasiveness of outcome measures. Rodent models, however, permit controlled investigation of cellular and molecular mechanisms using transgenic manipulations that mimic human disease. These models enable access to tissue-level endpoints, high reproducibility, and defined experimental conditions. Several humanised APOE mouse lines are available, some incorporating AD-related mutations in amyloid precursor protein (APP), microtubule-associated protein tau (MAPT), or familial AD (FAD) genes (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eBeyond central nervous system alterations, APOE4 also contributes to vascular dysfunction during disease progression. A key mechanism involves blood\u0026ndash;brain barrier (BBB) breakdown (\u003cspan additionalcitationids=\"CR11 CR12\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e), which can be assessed through non-invasive measures such as cerebral blood flow and contrast-enhanced permeability. Additional tools\u0026mdash;including immunohistochemistry, in vivo MRI, and behavioural testing\u0026mdash;provide complementary insight into vascular morphology, structural integrity, and cognitive outcomes.\u003c/p\u003e \u003cp\u003eThis review synthesises current evidence on APOE-driven vascular alterations in mouse models, focusing on genotype-related differences in BBB integrity, vascular morphology, and cerebral blood flow. Given the heterogeneity of study designs, outcomes, and ages assessed, these comparisons are crucial for building a coherent picture of APOE4-related vascular pathology and its contribution to disease progression.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cem\u003e\u003cu\u003eIdentification and Search Strategy\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThis review followed the PRISMA 2020 guidelines (14) and was registered in PROSPERO (CRD420251010665), and the detailed search strategy was separately registered with SearchRxiv (DOI: 10.1079/searchRxiv.2025.01124) to enhance methodological transparency and reproducibility. Literature searches were conducted in Scopus, MEDLINE, Embase, and Web of Science between 25 March and 3 April 2025. The MEDLINE search strategy was: \u003cem\u003e(Apolipoproteins E/ OR Apolipoprotein E3/ OR Apolipoprotein E4/) AND (mice/ OR mice, transgenic/) AND ((brain/ OR blood\u0026ndash;brain barrier/) OR (perfusion/ OR cerebrovascular circulation/))\u003c/em\u003e; full strategies for other databases are provided in \u003cstrong\u003eSupplementary Table 1\u003c/strong\u003e. Additional manuscripts identified by manual search were included. Unpublished data and conference abstracts were excluded.\u003c/p\u003e\n\u003cp id=\"_Toc210759627\"\u003e\u003cem\u003e\u003cu\u003eStudy Selection and Data Extraction\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eScreening and data extraction were performed using the Systematic Review Facility (SyRF) platform (15). Two reviewers (K.L. and N.F.) independently screened titles, abstracts, and full texts, resolving disagreements by consensus with a third reviewer (A.M.).\u003c/p\u003e\n\u003cp\u003eEligible studies:\u003c/p\u003e\n\u003col start=\"1\" type=\"1\"\u003e\n \u003cli\u003eUsed humanised targeted replacement APOE mice (APOE3-TR, APOE4-TR)\u0026nbsp;(16)\u0026nbsp;or APOE-TR;5xFAD (EFAD) lines.\u003c/li\u003e\n \u003cli\u003eProvided comparative analysis between APOE3 and APOE4 homozygotes; APOE4-only studies without APOE3 controls were also accepted\u0026nbsp;(17\u0026ndash;19).\u003c/li\u003e\n \u003cli\u003eAssessed cerebrovascular structure or function (e.g., cerebral blood flow, vascular density, permeability, pericyte coverage).\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eExclusion criteria included non-mice models, human studies, irrelevant outcomes (e.g., amyloid pathology only), non-vascular interventions, or descriptive designs. Additional criteria appear in \u003cstrong\u003eSupplementary Table 2.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eKey study characteristics (strain, age, design) were extracted by one reviewer (K.L.) and validated by another (N.F.) (Table 1). Outcome data were extracted independently and consolidated into a consensus dataset. Graphical data were digitised using WebPlotDigitizer (20). No major discrepancies were identified.\u003c/p\u003e\n\u003cp id=\"_Toc210759628\"\u003e\u003cem\u003e\u003cu\u003eRisk of Bias (RoB) and Quality Assessment\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eStudy quality and risk of bias were evaluated using established preclinical tools: the CAMARADES checklist (21) for overall study quality and the SYRCLE Risk of Bias tool (22) for individual bias domains. Assessments were conducted by one reviewer (K.L.) and cross-checked on a subset of studies by a second reviewer (N.F.) to ensure consistency.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cu\u003eSynthesis Methods\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eMeta-analyses assessed the impact of APOE genotype on cerebral blood flow (CBF) and vascular morphology in humanised APOE knock-in or targeted replacement (APOE-KI/TR) mice using RStudio (version 2024.12.1). Key study characteristics\u0026mdash;experimental design, outcomes, age, and region analysed\u0026mdash;were tabulated, and studies grouped by outcome domain (CBF or vascular morphology). Only those providing sufficient quantitative data (mean, SEM, and sample size) for APOE3 versus APOE4 comparisons were included in meta-analyses; others were summarised narratively. SEMs were converted to standard deviations using:\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\"\u003e\u003c/p\u003e\n\u003cp\u003eMultiple studies compared \u0026nbsp;different ages and brain subregions, resulting in several layers of analysis within a single study(11,23\u0026ndash;26). Studies which contributed multiple subgroup comparisons (e.g., data from different brain regions or age groups) were treated as separate data points and analysed independently. Each subgroup was assigned an alphabetical suffix (e.g., a, b, c) and specific comparisons were detailed in the captions of forest plots. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eGiven expected heterogeneity in design, imaging methods, and antibodies, all quantitative syntheses employed random-effects models. When studies contributed multiple within-study comparisons, a hierarchical random-effects model using robust variance estimation (RVE) accounted for non-independence of effect sizes. Between-group differences (APOE3 vs APOE4) were expressed as standardised mean differences (SMDs) with 95% confidence intervals (CIs). Where multiple measures were reported for the same group, pooled values were calculated following CAMARADES guidelines (27).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAs traditional heterogeneity metrics (Cochran\u0026rsquo;s Q, I\u0026sup2;) are not available with RVE, \u0026tau;\u0026sup2; was reported to represent between-study variance. For analyses with a single effect size per study, standard random-effects models (metafor package, R) were used, providing SMDs, 95% CIs, and heterogeneity indices (\u0026tau;\u0026sup2;, I\u0026sup2;, Q).\u003c/p\u003e\n\u003cp\u003eFormal meta-analysis was not undertaken for outcomes with fewer than three studies. In these cases, results were summarised narratively, and consistency across subgroups and study characteristics was assessed qualitatively.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cem\u003e\u003cu\u003eLiterature Retrieval\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe database search identified 1,493 records, with three additional studies found through manual searching. After removal of five duplicates, 1,488 titles and abstracts were screened. Of these, 1,438 were excluded: 711 were irrelevant; 17 involved the wrong population (e.g., human studies (28)); 109 used inappropriate designs (e.g., amyloid clearance (29,30), gene editing, nanomedicine); 384 employed unsuitable models (e.g., ApoE\u0026minus;/\u0026minus;, APP, or Lrp1-KO mice (31)); 79 used unrelated interventions (e.g., traumatic brain injury, artery occlusion, chronic inflammation (32)); and 138 reported unrelated outcomes (e.g., amyloid burden, behaviour, transcriptomics).\u003c/p\u003e\n\u003cp\u003eFifty studies underwent full-text review, and 18 met inclusion criteria for the systematic review (\u003cstrong\u003eFigure 1\u003c/strong\u003e). Of these, seven were eligible for meta-analysis: six reported vascular morphology (four unique, two overlapping with CBF outcomes), and three examined genotype-dependent differences in cerebral blood flow. The remaining 11 were synthesised narratively. Study characteristics and exclusion details are provided in \u003cstrong\u003eSupplementary Table 3\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cu\u003eStudy Characteristics\u003c/u\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCharacteristics of the 18 included studies are summarised in \u003cstrong\u003eTable 1\u003c/strong\u003e. All used transgenic murine models expressing human APOE alleles, with variation in strain, background, supplier, group size, age, and sex. Nine studies employed C57BL/6 background mice, while others used outbred strains, EFAD lines\u0026nbsp;(11,33\u0026ndash;35), or did not specify background. Mouse ages ranged from 2 to 24 months, encompassing early life (2\u0026ndash;4 months), midlife (6\u0026ndash;12 months), and later life (up to 24 months).\u003c/p\u003e\n\u003cp\u003eMethodologies also varied, including assessments of BBB permeability (e.g., gadolinium-based Ktrans, fibrinogen, fibronectin), vascular morphology, and cerebral blood flow. Detailed PICO (Population, Intervention, Comparator, Outcome) information for each study is provided in \u003cstrong\u003eTable 2\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003e[Table 1, Table 2 here]\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cu\u003eQuality Assessment and Risk of Bias\u003c/u\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStudy quality and reporting were evaluated using the CAMARADES checklist and SYRCLE Risk of Bias tool across all 18 studies. The mean CAMARADES score was 7.2/10 (72%), ranging from 4 to 9.5 (\u003cstrong\u003eTable 3\u003c/strong\u003e). All studies used appropriate transgenic mouse models and adhered to animal welfare regulations; all but one (Bhattarai et al., 2025, \u003cem\u003ebioRxiv\u003c/em\u003e) were peer-reviewed. Owing to the transgenic design, all studies met the criterion for blinded induction. However, several domains showed incomplete reporting, and only four studies explicitly described sample size calculations.\u003c/p\u003e\n\u003cp\u003e[\u003cem\u003eTable 3 here\u003c/em\u003e]\u003c/p\u003e\n\u003cp\u003eSeveral domains exhibited unclear risk of bias, particularly performance bias, selection bias, and detection bias, due to insufficient reporting (\u003cstrong\u003e\u003cem\u003eSupplementary Table 4\u003c/em\u003e\u003c/strong\u003e). Randomised housing was poorly reported in most studies. In contrast, attrition bias, reporting bias, and other sources of bias (e.g., conflict of interest declarations) were most consistently rated as low risk. Only Onos et al., (2024) was rated low risk across all SYRCLE domains.\u0026nbsp;\u003c/p\u003e\n\u003cp id=\"_Toc210759634\"\u003e\u003cstrong\u003e\u003cu\u003eStudy Outcomes\u003c/u\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIncluded studies assessed diverse yet overlapping aspects of blood\u0026ndash;brain barrier breakdown (BBB-b). Five studies reported cerebral blood flow changes (11,23\u0026ndash;25,33), and three examined tight junction integrity (24,36,37). Ten studies identified cerebrovascular morphological differences by APOE genotype, including vessel density, pericyte coverage, vessel diameter, and vascular volume (11,23\u0026ndash;25,34\u0026ndash;36,38\u0026ndash;40).\u003c/p\u003e\n\u003cp\u003eMechanistic investigations varied: one study conducted multi-omic profiling (transcriptomic, lipidomic, proteomic) (41), and two evaluated neurovascular coupling in the visual cortex (42,43). Three studies explored Cyclophilin-A\u0026ndash;NF\u0026kappa;B\u0026ndash;MMP9 signalling (23,24,30), while seven addressed pathways involving insulin resistance, metabolism, and mTOR signalling (23,26,33,38,44\u0026ndash;46). Additional studies examined EGF-related resilience (34), peripheral inflammation (35), and glial contributions to BBB-b (25,39,40,47).\u003c/p\u003e\n\u003cp\u003eGiven the limited number of studies per outcome domain, formal assessments of reporting bias (e.g., funnel plot asymmetry, Egger\u0026rsquo;s test) were not performed.\u003c/p\u003e\n\u003cp id=\"_Toc210759635\"\u003e\u003cstrong\u003e\u003cu\u003eAPOE4 Contributions to BBB Destabilisation and Meta-Analysis\u003c/u\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp id=\"_Toc210759636\"\u003e\u003cem\u003e\u003cu\u003eCerebral Blood Flow\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThree studies assessed APOE-dependent differences in cerebral blood flow (CBF) across distinct age groups and brain regions, using dynamic susceptibility-contrast MRI (DSC-MRI) (11), arterial spin labelling MRI (ASL-MRI) (25), or autoradiography with radiolabelled tracers (24). Despite methodological variation, RVE meta-analysis indicated a consistent reduction in CBF in APOE4 mice relative to APOE3 (SMD = \u0026minus;2.87, 95% CI \u0026minus;5.14 to \u0026minus;0.60, df = 2.66; \u003cstrong\u003eFigure 2\u003c/strong\u003e). Between-study heterogeneity was moderate to substantial (\u0026tau;\u0026sup2; = 2.25), reflecting differences in imaging modality.\u003c/p\u003e\n\u003cp\u003eAll three studies were of moderate-to-high quality (mean CAMARADES score 8.3/10). However, risk of bias may have arisen from incomplete reporting of sample size calculations and uncertainty in performance bias domains (e.g., randomised housing). Overall, 10\u0026ndash;60% of SYRCLE items were rated as unclear risk.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cu\u003eVascular Morphology\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSix studies investigated APOE genotype\u0026ndash;dependent differences in vascular morphology using immunohistochemical markers. Two assessed laminin coverage as cortical percent area in 5- and 8-month-old EFAD mice (34,35). Three evaluated collagen-IV expression as a marker of vessel density\u0026mdash;quantified as hippocampal capillary area in 9- and 12-month-old mice (6,43) or cortical signal intensity in 22-month-old mice (40). One measured neocortical microvessel density via CD31 immunostaining in 9-month-old mice (25).\u003c/p\u003e\n\u003cp\u003eA random-effects meta-analysis (k = 6) showed a negative but non-significant trend toward reduced vascular morphology markers in APOE4 versus APOE3 mice (SMD = \u0026minus;0.59; 95% CI \u0026minus;1.39 to 0.20; p = 0.14; \u003cstrong\u003eFigure 3\u003c/strong\u003e). Heterogeneity was moderate to substantial (\u0026tau;\u0026sup2; = 0.64; I\u0026sup2; = 66.3%; Q = 13.35, p = 0.020).\u003c/p\u003e\n\u003cp\u003eIncluded studies had a mean CAMARADES score of 6.9/10. Risk of bias reflected incomplete reporting of temperature control, randomisation, and sample size calculation. According to SYRCLE, 30\u0026ndash;60% of domains were rated unclear, primarily for selection and performance bias (\u003cstrong\u003eSupplementary Table 4\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cu\u003eNarrative Summary of APOE4-Associated Mechanistic Pathways\u003c/u\u003e\u003c/strong\u003e\u003cstrong\u003e\u003cu\u003e\u0026nbsp;\u003c/u\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRemaining studies that could not be quantitatively synthesised were analysed narratively to provide complementary mechanistic insights (\u003cstrong\u003eTable 4\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 4.\u003c/strong\u003e \u003cem\u003eSummary of key mechanistic axes implicated in APOE4-related blood\u0026ndash;brain barrier (BBB) dysfunction.\u003c/em\u003e\u003c/p\u003e\n\u003cp id=\"_Toc210759639\"\u003e\u003cem\u003e\u003cu\u003eInsulin/mTOR Signalling\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSeveral studies have investigated the role of insulin resistance in brain metabolism and endothelial dysfunction in \u003cem\u003eAPOE4\u003c/em\u003e mice, often focusing on the mammalian target of rapamycin (mTOR) signalling pathway.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eJohnson et al. (2019)(44) reported that high-fat diet (HFD)-induced insulin resistance was associated with impaired cognition, reduced cerebral blood volume (CBV), and decreased glucose uptake, with more pronounced effects in 15-month-old \u0026epsilon;4 than \u0026epsilon;3 mice. Administration of an acute exogenous glucose bolus produced a paradoxical improvement in CBV and cognition for \u003cem\u003eAPOE4\u003c/em\u003e mice only, demonstrating a significant genotype effect (p=0.028).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOnos et al. (2024)(46) further confirmed genotype-linked alterations in glucose metabolism. Longitudinal FDG-PET imaging revealed widespread reductions in glucose uptake in \u003cem\u003eAPOE4\u003c/em\u003e mice across multiple brain regions, with complex age- and sex-specific dynamics. Fluctuating uptake patterns suggested a Type 1 neurovascular uncoupling process - where decreased glucose uptake co-occurs with increased perfusion at 8 and 12 months.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eRhea et al. (2021)(38) examined insulin pharmacokinetics in 15-month-old \u003cem\u003eAPOE\u003c/em\u003e-TR mice, identifying genotype- and sex-specific differences in vascular insulin binding. \u003cem\u003eAPOE4\u003c/em\u003e females exhibited higher amounts of reversible binding of \u003csup\u003e125\u003c/sup\u003eI-Insulin in the frontal cortex, than \u003cem\u003eAPOE3\u003c/em\u003e females, proposed to reflect differences in insulin receptor availability or binding site accessibility on the luminal surface of brain endothelial cells. However, downstream dynamics regarding insulin transport across the BBB remained unclear.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eLin et al. (2020)(33) further linked \u003cem\u003eAPOE4\u003c/em\u003e-associated vascular dysfunction to overactivation of mTOR. In 7-month-old E4FAD mice, mTOR hyperactivity was associated with reduced P-glycoprotein (P-gp) transport at the BBB (p\u0026lt;0.001), impaired CBF, disrupted lipid metabolism, and elevated free fatty acids (FFAs). Sixteen weeks of rapamycin treatment restored BBB function and lipid homeostasis in \u003cem\u003eAPOE4\u003c/em\u003e mice. Interestingly, rapamycin administration in \u003cem\u003eAPOE3\u003c/em\u003e mice slowed glucose-associated metabolism \u0026ndash; reducing glycolytic and TCA cycle intermediates \u0026ndash; whereas \u003cem\u003eAPOE4\u003c/em\u003e metabolism was normalised. This effect was suggested to mirror caloric restriction-like outcomes, where mTOR inhibition in relatively healthy animals reduces glucose uptake.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTogether, these findings support a model in which the \u003cem\u003eAPOE4\u003c/em\u003e genotype is associated with dysregulated brain insulin signalling, reduced brain glucose uptake, and a paradoxical relationship with mTOR activation processes further shaped by diet, age, and sex.\u0026nbsp;\u003c/p\u003e\n\u003cp id=\"_Toc210759640\"\u003e\u003cem\u003e\u003cu\u003eCyclophilin A \u0026ndash; NF\u0026kappa;B \u0026ndash; MMP9\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eFour studies demonstrate that the APOE4 genotype contributes to blood-brain barrier (BBB) breakdown via activation of the Cyclophilin A (CypA) \u0026ndash; Nuclear Factor kappa B (NFkB) \u0026ndash; matrix metalloproteinase 9 (MMP9) pathway.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eBell et al. (2012)(24) reported a 5-6-fold increase in CypA levels within the cerebral microvessels of 6-month-old \u003cem\u003eAPOE4\u003c/em\u003e mice under the GFAP promoter, primarily localised to pericytes. Further experiments demonstrated that this increase drives NFkB and MMP9 expression, resulting in pericyte-mediated ECM remodelling. Pharmacological inhibition of CypA with cyclosporine A, an immunosuppressant, attenuated these effects on BBB integrity.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eLin et al. (2017)(23) corroborated CypA\u0026rsquo;s role in \u003cem\u003eAPOE4\u003c/em\u003e-related vascular dysfunction. Here, treatment with rapamycin in 12-month-old \u003cem\u003eAPOE4\u003c/em\u003e mice under the GFAP promoter normalised CypA and NFkB levels in cortical and microvascular compartments, improved BBB integrity, and restored CBF (ASL-MRI). Although total brain CypA levels were not significantly different between genotypes, \u003cem\u003eAPOE4\u003c/em\u003e control mice showed reduced parenchymal and elevated vascular CypA levels, indicating regional heterogeneity in CypA expression.\u003c/p\u003e\n\u003cp\u003eRingland et al. (2020)(30) found impaired regulation of MMP9 in 6-month-old E4FAD mice, reporting a 56% increase in MMP9 immunoreactivity in cortical endothelial cells compared to E3FAD. At 70 weeks, E4FAD mice continued to show elevated MMP9 levels relative to E3FAD mice (p=0.0533), suggesting chronic dysregulation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eIn vivo\u003c/em\u003e experiments by Jackson et al. (2022) confirmed BBB disruption in \u003cem\u003eAPOE4\u003c/em\u003e mice by 9-months-old. Using fluorescein-labelled 40 kDa dextran in tracer extravasation studies, increased permeability was observed. Quantitative PCR revealed a ~30% increase in MMP9 transcript levels in \u003cem\u003eAPOE4\u003c/em\u003e mice, accompanied by elevated MMP9 protein levels. Transmission electron microscopy showed irregular gaps in tight junction membranes, suggesting focal openings rather than a global loss of structure. Additionally, measurement of astrocytic end-foot coverage around vessels demonstrated a significant reduction.\u003c/p\u003e\n\u003cp\u003eTogether, these studies implicate \u003cem\u003eAPOE4\u003c/em\u003e in activating a proinflammatory and extracellular matrix remodelling cascade involving pericyte-endothelial interactions. This pathway contributes to BBB breakdown, including localised tight junction dysregulation, as early as 6 months of age.\u0026nbsp;\u003c/p\u003e\n\u003cp id=\"_Toc210759641\"\u003e\u003cem\u003e\u003cu\u003eOccludin \u0026amp; ECM\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe \u003cem\u003eAPOE4\u003c/em\u003e genotype has been strongly associated with compromised BBB integrity, driven by both structural and functional disruptions of endothelial and perivascular components. Several studies highlight specific markers of BBB dysregulation, including tight junction protein, fibrinogen, fibronectin and extracellular matrix components.\u003c/p\u003e\n\u003cp\u003eNishitsuji et al. (2011)(37) employed an \u003cem\u003ein vitro\u003c/em\u003e BBB model incorporating primary astrocytes from \u003cem\u003eAPOE3\u003c/em\u003e- or \u003cem\u003eAPOE4\u003c/em\u003e-knock-in mice and found significantly reduced trans endothelial electrical resistance (TEER) in the \u003cem\u003eAPOE4\u003c/em\u003e condition. While findings revealed no change in total tight junction protein levels, phosphorylation of occludin at threonine (Thr) residues \u0026ndash; a modification necessary for occluding assembly into tight junctions \u0026ndash; was significantly reduced. This was accompanied by diminished PKC\u0026eta; activation, suggesting a pathological role for \u003cem\u003eAPOE4\u003c/em\u003e in impaired LRP1\u0026ndash;PKC\u0026eta;\u0026ndash;occludin signalling, ultimately compromising tight junction integrity.\u003c/p\u003e\n\u003cp\u003eYanckello et al. (2022) further examined tight junction protein expression in brain capillaries isolated from 7-month-old E3FAD and E4FAD mice, with or without dietary inulin (a prebiotic fibre). While inulin did not significantly alter occludin protein levels, Claudin-1 expression differed at baseline, with E4FAD control mice exhibiting elevated levels (\u003cem\u003ep\u003c/em\u003e = 0.0012).\u003c/p\u003e\n\u003cp\u003eYamazaki et al. (2020)(40) provided further evidence of \u003cem\u003eAPOE\u003c/em\u003e genotype-dependent BBB permeability using both \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo\u003c/em\u003e approaches. TEER was again significantly lower in BBB models containing \u003cem\u003eAPOE4\u003c/em\u003e-expressing pericytes, despite no significant differences in tight junction protein levels. However, collagen IV expression was significantly reduced in \u003cem\u003eAPOE4\u003c/em\u003e-pericyte models. Immunofluorescence confirmed diminished collagen IV in 8-month-old \u003cem\u003eAPOE4\u003c/em\u003e mice. ELISA assays revealed increased cortical IgG and fibrinogen, supporting a role for extracellular matrix impairment in BBB permeability and an inverse relationship between extravasating markers and basement membrane structural integrity.\u003c/p\u003e\n\u003cp\u003eThomas et al. (2017)(34) identified that 8-month-old female E4FAD mice exhibited ~65% higher cortical fibrinogen levels and increased sodium fluorescein (NaFl) leakage compared o to other groups, suggesting a sex-specific vulnerability of the BBB. Treatment with epidermal growth factor (EGF) attenuated fibrinogen extravasation by about 40%.\u003c/p\u003e\n\u003cp\u003eBarisano et al. (2022) used dynamic contrast-enhanced MRI and histological analysis to investigate BBB integrity in \u003cem\u003eAPOE\u003c/em\u003e-targeted replacement mice. Their findings revealed progressive BBB leakage in \u003cem\u003eAPOE4\u003c/em\u003e\u0026nbsp; \u0026nbsp;mice, evidenced by pericapillary fibrinogen accumulation. Tissue analysis also showed a progressive loss of pericyte coverage (CD13\u003csup\u003e+\u003c/sup\u003e), which strongly inversely correlated with fibrinogen deposition, linking structural loss to BBB leakiness. Transcriptomic analysis of endothelial cells revealed upregulation of genes involved in cell adhesion (e.g., cadherins, contactins, catenins), interpreted as a possible compensatory response to vascular compromise.\u003c/p\u003e\n\u003cp\u003eFinally, Bhattarai et al. (2025)(47) reported significantly elevated fibronectin accumulation in \u003cem\u003eAPOE4\u003c/em\u003e mice - a hallmark of extracellular matrix remodelling - without changes in the CD31 endothelial marker, further demonstrating enhanced susceptibility to BBB permeability in the \u003cem\u003eAPOE4\u003c/em\u003e genotype.\u003c/p\u003e\n\u003cp\u003eAltogether, these studies emphasise the relationship between increased BBB permeability and reduced vascular stability in the \u003cem\u003eAPOE4\u003c/em\u003e genotype.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThese three mechanistic axes provide compelling evidence of pervasive and progressive BBB structural breakdown in \u003cem\u003eAPOE4\u003c/em\u003e contexts (\u003cstrong\u003eFigure 4\u003c/strong\u003e). This disruption is marked by increased permeability - evidenced by tracer extravasation, blood product accumulation, and tight junction widening - and is accompanied by chronic inflammation, possibly driven by sustained extracellular matrix remodelling and perivascular dysfunction.\u003c/p\u003e"},{"header":"Discussion and Future Perspectives","content":"\u003cp\u003eThis systematic review examined associations between the Apolipoprotein E4 (\u003cem\u003eAPOE4\u003c/em\u003e) genotype and BBB integrity in humanised \u003cem\u003eAPOE\u003c/em\u003e-target replacement (TR) and knock-in (KI) mouse models. Recognising that methodological quality and bias assessment are critical for translational validity, this review began by evaluating the quality of included studies before synthesising trends in BBB structure, function, and molecular mechanisms.\u003c/p\u003e\n\u003cp\u003eUsing the CAMARADES checklist and SYRCLE Risk of Bias (RoB) tool(22), we assessed 22 studies that met eligibility criteria for reporting quality. The average CAMARADES score was 7.2/10 (72%), suggesting overall satisfactory experimental design; however, several areas were consistently underreported, particularly random allocation to treatment, blinding of outcome assessment, and transparent reporting of sample size calculation. These items are often assumed or implied within the field, but lack of - or insufficient - reporting of these critical design elements could compromise experimental validity or increase the risk of bias in outcomes. This concern was further underscored by the SYRCLE RoB assessment, which identified high rates of unclear reporting in selection and performance bias domains (e.g., allocation concealment, randomised housing, and caregiver blinding), highlighting the need for more rigorous and standardised reporting practices in preclinical BBB research.\u003c/p\u003e\n\u003cp\u003eTo assess genotype-related differences, meta-analyses were performed examining CBF and vascular morphology in \u003cem\u003eAPOE4\u003c/em\u003e mice vs. \u003cem\u003eAPOE3\u003c/em\u003e mice. Despite heterogeneity in imaging modalities (DSC- and ASL-MRI, and \u003csup\u003e14\u003c/sup\u003eC-autoradiography), studies consistently reported reduced CBF in \u003cem\u003eAPOE4\u003c/em\u003e compared with other \u003cem\u003eAPOE\u003c/em\u003e genotypes across a broad age range (0–24 months).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAnalysis of vascular morphology, evaluating markers of endothelial and basement membrane integrity (laminin, collagen-IV, CD31), revealed a less straightforward trend with middle-aged mice (5-11 months) demonstrated higher vascular expression, compared to older (12-24 months) though the standard mean difference for \u003cem\u003eAPOE4\u003c/em\u003e relative to \u003cem\u003eAPOE3\u003c/em\u003e was not statistically significant. This finding, although not statistically significant, is notable and aligns with broader mechanistic data, namely the strong presence of Cyclophilin A (CypA) – NFkB – MMP9 and Occludin/ECM remodelling axes.\u003c/p\u003e\n\u003cp\u003eStudies within the CypA axis consistently reported elevated levels of CypA, NFkB, and MMP9 – key mediators of inflammatory signalling and ECM remodelling. Although MMP9 can cleave glycoproteins such as laminin, it demonstrates higher binding affinity and substrate specificity to collagens, particularly collagen-IV(49,50). This supports a plausible link between ECM protein downregulation and proinflammatory activation of this remodelling pathway with \u003cem\u003eAPOE4\u003c/em\u003e and potentially providing explanation for atypical permeability within the BBB construct.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe Occludin and ECM Remodelling axis further reinforces this connection. \u003cem\u003eAPOE4\u003c/em\u003e models showed reduced phosphorylation of occludin at threonine (Thr) - a post-translational modification critical for endothelial stability - along elevated markers of permeability such as fibrinogen, fibronectin, and IgG extravasation in perivascular tissue.\u003c/p\u003e\n\u003cp\u003eOccludin regulation is thought to involve the LRP1–PKCη–occludin signalling cascade; however, in \u003cem\u003eAPOE4\u003c/em\u003e mice, this pathway may be functionally impaired (Nishitsuji et al., 2011). PKCη also acts as a negative regulator of AKT(51), a kinase central to the PI3K/AKT/mTOR pathway.\u003c/p\u003e\n\u003cp\u003eRegarding the third axis, Insulin/mTOR signalling, several studies included in this review reported insulin resistance, impaired glucose uptake (FDG-PET), and reduced cerebral blood volume (CBV) in \u003cem\u003eAPOE4\u003c/em\u003e mice - all of which were attenuated by therapeutic administration of rapamycin, an mTOR inhibitor. These findings suggest a crucial link between metabolic dysfunction and cerebrovascular integrity in \u003cem\u003eAPOE4\u003c/em\u003e carriers.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTogether, these three mechanistic axes (CypA–MMP9, mTOR/insulin, and occludin-related ECM remodelling) provide converging evidence for chronic and progressive BBB structural breakdown in \u003cem\u003eAPOE4\u003c/em\u003e carriers. This disruption is characterised by 1) overt direct evidence of BBB permeability (e.g., tracer leakage, blood product accumulation), 2) features which could drive BBB leakage (e.g., tight junction instability, reduced occludin phosphorylation, widened junctions), 3) features which could stem from BBB leakage or worsen it if present (e.g., inflammation and ECM degradation, MMP9, fibronectin), and 4) Impaired vascular metabolism and function (e.g., endothelial glucose uptake, CBF deficits).\u0026nbsp;\u003c/p\u003e\n\u003cp id=\"_Toc210754741\"\u003e\u003cu\u003eLimitations\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eThis review has several limitations. First, the number of studies eligible for meta-analysis was relatively small, limiting the statistical power and generalisability of pooled findings. Subgroup analyses were further limited by methodological heterogeneity, particularly differences in measurement techniques, biomarkers, and arbitrarily defined age ranges.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAdditionally, many studies received unclear ratings in key domains of the CAMARADES and SYRCLE quality assessments. Details regarding blinding, sample size calculation, and allocation procedures were often underreported, limiting the ability to assess internal validity. These omissions not only affect the reliability of individual studies but also weaken the overall strength of evidence in the systematic review.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFuture research should prioritise methodological transparency. Adoption of preclinical reporting standards, such as SYRCLES RoB or ARRIVE guidelines, would improve reproducibility and minimise bias.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFinally, consistency in genotype comparators would strengthen the interpretability of \u003cem\u003eAPOE\u003c/em\u003e-targeted studies. While wildtype mice provide a useful control for transgenic comparisons, \u003cem\u003eAPOE3\u003c/em\u003e should be the preferred genetic control when studying humanised \u003cem\u003eAPOE\u003c/em\u003e isoforms, as it represents the most common and asymptomatic variant. Comparisons between \u003cem\u003eAPOE4\u003c/em\u003e and \u003cem\u003eAPOE3\u003c/em\u003e are critical to isolate pathogenic effects specific to \u003cem\u003eAPOE4\u003c/em\u003e-related dysfunction.\u0026nbsp;\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn conclusion, this systematic review demonstrates that the \u003cem\u003eAPOE4\u003c/em\u003e genotype in mice compromises blood-brain barrier integrity through consistent pathogenic alterations in vascular morphology and cerebral blood flow dynamics, alongside clearly defined mechanistic pathways of metabolic and vascular dysfunction. Preclinical studies using humanised \u003cem\u003eAPOE\u003c/em\u003e models provide valuable insight into these genotype-specific effects and offer a robust platform to inform translational research, including therapeutic development, biomarker discovery, and precision medicine strategies.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cem\u003e\u003cu\u003eEthics approval and consent to participate\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eNot Applicable\u0026nbsp;\u003c/p\u003e\n\u003cp id=\"_Toc210759647\"\u003e\u003cem\u003e\u003cu\u003eConsent for publication\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eNot Applicable\u003c/p\u003e\n\u003cp id=\"_Toc210759648\"\u003e\u003cem\u003e\u003cu\u003eAvailability of data and materials\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAll data generated or analysed during this study are included in this article and its supplementary information files. The extracted dataset and analysis code that support the findings are available from the corresponding author on reasonable request. The full database search strategies have been deposited on searchRxiv (DOI: 10.1079/searchRxiv.2025.01124). The review protocol was registered with PROSPERO (CRD420251010665): https://www.crd.york.ac.uk/PROSPERO/view/CRD420251010665.\u0026nbsp;\u003c/p\u003e\n\u003cp id=\"_Toc210759649\"\u003e\u003cem\u003e\u003cu\u003eCompeting interests\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp id=\"_Toc210759650\"\u003e\u003cem\u003e\u003cu\u003eFunding\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe work of \u003cstrong\u003eAM\u003c/strong\u003e is supported by the UK Dementia Research Institute (award UKDRI-4209) through UK DRI Ltd, funded by the UK Medical Research Council (MRC), Alzheimer\u0026rsquo;s Society UK, Alzheimer\u0026rsquo;s Research UK, and the British Heart Foundation. AM also holds a UKRI MRC Career Development Award (MR/V032488/1) and a UK DRI Theme Funding Programme Award (DRI-TFP-2024-7).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eJMW\u003c/strong\u003e is supported by the UK Dementia Research Institute (award UK DRI-4002) through UK DRI Ltd, funded by the MRC, Alzheimer\u0026rsquo;s Society, and Alzheimer\u0026rsquo;s Research UK; by the Row Fogo Charitable Trust [BRO-D.FID3668413] (JMW); and by the NHS Lothian Research and Development Office (MJT).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eKKL\u003c/strong\u003e received support from the University of Edinburgh Wellcome Trust Translational Neuroscience 4-year PhD Programme (Grant No. 110002 20132001 TBC 130956 00000000 10002374 000).\u003c/p\u003e\n\u003cp id=\"_Toc210759651\"\u003eAuthors\u0026apos; contributions\u003c/p\u003e\n\u003cp\u003eKL conceived and designed the study; developed the searches; screened records; extracted and curated data; led analyses; performed primary risk-of-bias/quality assessments; prepared figures/tables; and drafted the manuscript. NF served as second reviewer for screening and full-text review, provided annotations, validated demographic extractions, and conducted risk-of-bias/quality assessments on a subset to verify consistency with KL; the remaining assessments were completed by KL due to resource constraints. KL and NF extracted outcomes independently, with discrepancies resolved by consensus to a single dataset. AM (PI) supervised the study, adjudicated screening disagreements, and provided critical revisions. JW (PI) supervised and provided critical revisions. All authors approved the final manuscript.\u003c/p\u003e\n\u003cp id=\"_Toc210759652\"\u003eAcknowledgements\u003c/p\u003e\n\u003cp\u003eWe thank \u003cstrong\u003eMarshall Dozier\u003c/strong\u003e, College Lead for Library Academic Support, for expert guidance in developing and refining the database search strategies and keyword terms. We also thank \u003cstrong\u003eGillian Currie\u003c/strong\u003e (CAMARADES) for methodological advice to ensure this systematic review met community standards. The \u003cstrong\u003eCAMARADES consortium\u003c/strong\u003e is funded by the UK National Centre for the Replacement, Refinement and Reduction of Animals in Research (NC3Rs) to support preclinical systematic reviews.\u003c/p\u003e\n\u003cp id=\"_Toc210759653\"\u003eAuthors\u0026apos; information (optional)\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eHuang Y, Mahley RW. Apolipoprotein E: Structure and function in lipid metabolism, neurobiology, and Alzheimer\u0026rsquo;s diseases. Neurobiol Dis. 2014 Dec 1;72:3\u0026ndash;12.\u003c/li\u003e\n \u003cli\u003ePhillips MC. Apolipoprotein E isoforms and lipoprotein metabolism. IUBMB Life. 2014 Sept;66(9):616\u0026ndash;23.\u003c/li\u003e\n \u003cli\u003eBlumenfeld J, Yip O, Kim MJ, Huang Y. Cell type-specific roles of APOE4 in Alzheimer disease. Nat Rev Neurosci. 2024 Feb;25(2):91\u0026ndash;110.\u003c/li\u003e\n \u003cli\u003eFarrer LA, Cupples LA, Haines JL, Hyman B, Kukull WA, Mayeux R, et al. Effects of Age, Sex, and Ethnicity on the Association Between Apolipoprotein E Genotype and Alzheimer Disease: A Meta-analysis. JAMA. 1997 Oct 22;278(16):1349\u0026ndash;56.\u003c/li\u003e\n \u003cli\u003eCorder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, et al. Gene Dose of Apolipoprotein E Type 4 Allele and the Risk of Alzheimer\u0026rsquo;s Disease in Late Onset Families. Science. 1993 Aug 13;261(5123):921\u0026ndash;3.\u003c/li\u003e\n \u003cli\u003eFoley KE, Wilcock DM. Three major effects of APOE\u0026epsilon;4 on A\u0026beta; immunotherapy induced ARIA. Front Aging Neurosci [Internet]. 2024 May 2 [cited 2025 Feb 11];16. Available from: https://www.frontiersin.org/journals/aging-neuroscience/articles/10.3389/fnagi.2024.1412006/full\u003c/li\u003e\n \u003cli\u003eHampel H, Elhage A, Cho M, Apostolova LG, Nicoll JAR, Atri A. Amyloid-related imaging abnormalities (ARIA): radiological, biological and clinical characteristics. Brain. 2023 Nov 1;146(11):4414\u0026ndash;24.\u003c/li\u003e\n \u003cli\u003eWard A, Crean S, Mercaldi CJ, Collins JM, Boyd D, Cook MN, et al. Prevalence of apolipoprotein E4 genotype and homozygotes (APOE e4/4) among patients diagnosed with Alzheimer\u0026rsquo;s disease: a systematic review and meta-analysis. Neuroepidemiology. 2012;38(1):1\u0026ndash;17.\u003c/li\u003e\n \u003cli\u003eLewandowski CT, Weng JM, LaDu MJ. Alzheimer\u0026rsquo;ss Disease Pathology in ApoE Transgenic Mouse Models: The Who, What, When, Where, Why, and How. Neurobiol Dis. 2020 June;139:104811.\u003c/li\u003e\n \u003cli\u003eMontagne A, Nation DA, Sagare AP, Barisano G, Sweeney MD, Chakhoyan A, et al. APOE4 leads to blood-brain barrier dysfunction predicting cognitive decline. Nature. 2020 May;581(7806):71\u0026ndash;6.\u003c/li\u003e\n \u003cli\u003eMontagne A, Nikolakopoulou AM, Huuskonen MT, Sagare AP, Lawson EJ, Lazic D, et al. APOE4 accelerates advanced-stage vascular and neurodegenerative disorder in old Alzheimer\u0026rsquo;s mice via cyclophilin A independently of amyloid-\u0026beta;. Nat Aging. 2021 June;1(6):506\u0026ndash;20.\u003c/li\u003e\n \u003cli\u003eKapoor A, Nation DA. Platelet-derived growth factor-BB and white matter hyperintensity burden in \u003cem\u003eAPOE4\u003c/em\u003e carriers. Cereb Circ - Cogn Behav. 2022 Jan 1;3:100131.\u003c/li\u003e\n \u003cli\u003eBlanchard JW, Bula M, Davila-Velderrain J, Akay LA, Zhu L, Frank A, et al. Reconstruction of the Human Blood-Brain Barrier in vitro reveals a Pathogenic Mechanism of APOE4 in Pericytes. Nat Med. 2020 June;26(6):952\u0026ndash;63.\u003c/li\u003e\n \u003cli\u003ePage MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. The BMJ. 2021 Mar 29;372:n71.\u003c/li\u003e\n \u003cli\u003eBahor Z, Liao J, Currie G, Ayder C, Macleod M, McCann SK, et al. Development and uptake of an online systematic review platform: the early years of the CAMARADES Systematic Review Facility (SyRF). BMJ Open Sci [Internet]. 2021 Mar 31 [cited 2025 June 1];5(1). Available from: http://access.portico.org/stable?au=phzmqrt8kdx\u003c/li\u003e\n \u003cli\u003eTai LM, Balu D, Avila-Munoz E, Abdullah L, Thomas R, Collins N, et al. EFAD transgenic mice as a human APOE relevant preclinical model of Alzheimerʼns disease [S]. J Lipid Res. 2017 Sept 1;58(9):1733\u0026ndash;55.\u003c/li\u003e\n \u003cli\u003eShenk JC, Liu J, Fischbach K, Xu K, Puchowicz M, Obrenovich ME, et al. The effect of acetyl-L-carnitine and R-\u0026alpha;-lipoic acid treatment in ApoE4 mouse as a model of human Alzheimer\u0026rsquo;s disease. J Neurol Sci. 2009;283(1\u0026ndash;2):199\u0026ndash;206.\u003c/li\u003e\n \u003cli\u003eJansen D, Zerbi V, Janssen CIF, van Rooij D, Zinnhardt B, Dederen PJ, et al. Impact of a multi-nutrient diet on cognition, brain metabolism, hemodynamics, and plasticity in apoE4 carrier and apoE knockout mice. Brain Struct Funct. 2014;219(5):1841\u0026ndash;68.\u003c/li\u003e\n \u003cli\u003eZerbi V, Wiesmann M, Emmerzaal TL, Jansen D, Van Beek M, Mutsaers MPC, et al. Resting-state functional connectivity changes in aging apoE4 and apoE-KO mice. J Neurosci Off J Soc Neurosci. 2014;34(42):13963\u0026ndash;75.\u003c/li\u003e\n \u003cli\u003eRohatgi A. WebPlotDigitizer [Internet]. automeris.io; 2025 [cited 2025 Jan 17]. Available from: https://automeris.io\u003c/li\u003e\n \u003cli\u003eMacleod MR, O\u0026rsquo;Collins T, Howells DW, Donnan GA. Pooling of Animal Experimental Data Reveals Influence of Study Design and Publication Bias. Stroke. 2004 May;35(5):1203\u0026ndash;8.\u003c/li\u003e\n \u003cli\u003eHooijmans CR, Rovers MM, de Vries RB, Leenaars M, Ritskes-Hoitinga M, Langendam MW. SYRCLE\u0026rsquo;s risk of bias tool for animal studies. BMC Med Res Methodol. 2014 Mar 26;14(1):43.\u003c/li\u003e\n \u003cli\u003eLin AL, Jahrling JB, Zhang W, DeRosa N, Bakshi V, Romero P, et al. Rapamycin rescues vascular, metabolic and learning deficits in apolipoprotein E4 transgenic mice with pre-symptomatic Alzheimer\u0026rsquo;s disease. J Cereb Blood Flow Metab. 2017 Jan 1;37(1):217\u0026ndash;26.\u003c/li\u003e\n \u003cli\u003eBell RD, Winkler EA, Singh I, Sagare AP, Deane R, Wu Z, et al. Apolipoprotein E controls cerebrovascular integrity via cyclophilin A. Nature. 2012 May 16;485(7399):512\u0026ndash;6.\u003c/li\u003e\n \u003cli\u003eKoizumi K, Hattori Y, Ahn SJ, Buendia I, Ciacciarelli A, Uekawa K, et al. Apo\u0026epsilon;4 disrupts neurovascular regulation and undermines white matter integrity and cognitive function. Nat Commun. 2018 Sept 19;9(1):3816.\u003c/li\u003e\n \u003cli\u003eWiesmann M, Zerbi V, Jansen D, Haast R, L\u0026uuml;tjohann D, Broersen LM, et al. A Dietary Treatment Improves Cerebral Blood Flow and Brain Connectivity in Aging apoE4 Mice. Neural Plast. 2016;2016:6846721.\u003c/li\u003e\n \u003cli\u003eCAMARADES Systematic Review Wiki [Internet]. [cited 2025 Aug 26]. Meta-Analysis. Available from: https://camarades.de/wiki-content/10-metaanalysis.html\u003c/li\u003e\n \u003cli\u003eHalliday MR, Rege SV, Ma Q, Zhao Z, Miller CA, Winkler EA, et al. Accelerated pericyte degeneration and blood\u0026ndash;brain barrier breakdown in apolipoprotein E4 carriers with Alzheimer\u0026rsquo;s disease. J Cereb Blood Flow Metab. 2016 Jan;36(1):216.\u003c/li\u003e\n \u003cli\u003eBachmeier C, Shackleton B, Ojo J, Paris D, Mullan M, Crawford F. Apolipoprotein E Isoform-Specific Effects on Lipoprotein Receptor Processing. NeuroMol Med. 2014;16(4):686\u0026ndash;96.\u003c/li\u003e\n \u003cli\u003eRingland C, Schweig JE, Paris D, Shackleton B, Lynch CE, Eisenbaum M, et al. Apolipoprotein E isoforms differentially regulate matrix metallopeptidase 9 function in Alzheimer\u0026rsquo;s disease. Neurobiol Aging. 2020;95:56\u0026ndash;68.\u003c/li\u003e\n \u003cli\u003eOue H, Yamazaki Y, Qiao W, Yuanxin C, Ren Y, Kurti A, et al. LRP1 in vascular mural cells modulates cerebrovascular integrity and function in the presence of \u003cem\u003eAPOE4\u003c/em\u003e. JCI Insight [Internet]. 2023 Apr 10 [cited 2024 July 31];8(7). Available from: https://insight.jci.org/articles/view/163822\u003c/li\u003e\n \u003cli\u003eNa H, Yang JB, Zhang Z, Gan Q, Tian H, Rajab IM, et al. Peripheral apolipoprotein E proteins and their binding to LRP1 antagonize Alzheimer\u0026rsquo;s disease pathogenesis in the brain during peripheral chronic inflammation. Neurobiol Aging. 2023;127:54\u0026ndash;69.\u003c/li\u003e\n \u003cli\u003eLin AL, Parikh I, Yanckello LM, White RS, Hartz AMS, Taylor CE, et al. APOE genotype-dependent pharmacogenetic responses to rapamycin for preventing Alzheimer\u0026rsquo;s disease. Neurobiol Dis. 2020 June 1;139:104834.\u003c/li\u003e\n \u003cli\u003eThomas R, Morris AWJ, Tai LM. Epidermal growth factor prevents APOE4-induced cognitive and cerebrovascular deficits in female mice. Heliyon [Internet]. 2017;3(6). Available from: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85020180648\u0026amp;doi=10.1016%2fj.heliyon.2017.e00319\u0026amp;partnerID=40\u0026amp;md5=9e788ede5036ea91221ed97b94b08774\u003c/li\u003e\n \u003cli\u003eMarottoli FM, Katsumata Y, Koster KP, Thomas R, Fardo DW, Tai LM. Peripheral Inflammation, Apolipoprotein E4, and Amyloid-\u0026beta; Interact to Induce Cognitive and Cerebrovascular Dysfunction. ASN Neuro. 2017;9(4):1759091417719201.\u003c/li\u003e\n \u003cli\u003eAlata W, Ye Y, St-Amour I, Vandal M, Calon F. Human apolipoprotein E ɛ4 expression impairs cerebral vascularization and blood-brain barrier function in mice. J Cereb Blood Flow Metab. 2015;35(1):86\u0026ndash;94.\u003c/li\u003e\n \u003cli\u003eNishitsuji K, Hosono T, Nakamura T, Bu G, Michikawa M. Apolipoprotein E regulates the integrity of tight junctions in an isoform-dependent manner in an in vitro blood-brain barrier model. J Biol Chem. 2011;286(20):17536\u0026ndash;42.\u003c/li\u003e\n \u003cli\u003eRhea EM, Hansen K, Pemberton S, Torres ERS, Holden S, Raber J, et al. Effects of apolipoprotein E isoform, sex, and diet on insulin BBB pharmacokinetics in mice. Sci Rep [Internet]. 2021;11(1). Available from: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85115398840\u0026amp;doi=10.1038%2fs41598-021-98061-1\u0026amp;partnerID=40\u0026amp;md5=367fc7cda7da7aeafa7603dd86c6cc43\u003c/li\u003e\n \u003cli\u003eJackson RJ, Meltzer JC, Nguyen H, Commins C, Bennett RE, Hudry E, et al. APOE4 derived from astrocytes leads to blood-brain barrier impairment. Brain. 2022;145(10):3582\u0026ndash;93.\u003c/li\u003e\n \u003cli\u003eYamazaki Y, Shinohara M, Yamazaki A, Ren Y, Asmann YW, Kanekiyo T, et al. ApoE (Apolipoprotein E) in Brain Pericytes Regulates Endothelial Function in an Isoform-Dependent Manner by Modulating Basement Membrane Components. Arter Thromb Vasc Biol. 2020;40(1):128\u0026ndash;44.\u003c/li\u003e\n \u003cli\u003eBarisano G, Kisler K, Wilkinson B, Nikolakopoulou AM, Sagare AP, Wang Y, et al. A \u0026lsquo;multi-omics\u0026rsquo; analysis of blood-brain barrier and synaptic dysfunction in APOE4 mice. J Exp Med. 2022;219(11).\u003c/li\u003e\n \u003cli\u003eBonnar O, Shaw K, Anderle S, Grijseels DM, Clarke D, Bell L, et al. APOE4 expression confers a mild, persistent reduction in neurovascular function in the visual cortex and hippocampus of awake mice. J Cereb Blood Flow Metab. 2023 Nov;43(11):1826\u0026ndash;41.\u003c/li\u003e\n \u003cli\u003eAnderle S, Bonnar O, Henderson J, Shaw K, Chagas AM, McMullan L, et al. APOE4 and sedentary lifestyle synergistically impair neurovascular function in the visual cortex of awake mice. Commun Biol. 2025;8(1):144.\u003c/li\u003e\n \u003cli\u003eJohnson LA, Torres ER, Weber Boutros S, Patel E, Akinyeke T, Alkayed NJ, et al. Apolipoprotein E4 mediates insulin resistance-associated cerebrovascular dysfunction and the post-prandial response. J Cereb Blood Flow Metab. 2019;39(5):770\u0026ndash;81.\u003c/li\u003e\n \u003cli\u003eRhea EM, Torres ERS, Raber J, Banks WA. Insulin BBB pharmacokinetics in young apoE male and female transgenic mice. PLoS ONE [Internet]. 2020;15(1). Available from: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85078745276\u0026amp;doi=10.1371%2fjournal.pone.0228455\u0026amp;partnerID=40\u0026amp;md5=a33e8c83538370a3279425e62da5608b\u003c/li\u003e\n \u003cli\u003eOnos KD, Lin PB, Pandey RS, Persohn SA, Burton CP, Miner EW, et al. Assessment of neurovascular uncoupling: APOE status is a key driver of early metabolic and vascular dysfunction. Alzheimers Dement. 2024;20(7):4951\u0026ndash;69.\u003c/li\u003e\n \u003cli\u003eBhattarai P, Yilmaz E, Cakir E\u0026Ouml;, Korkmaz HY, Lee AJ, Ma Y, et al. APOE-\u0026epsilon;4-induced Fibronectin at the blood-brain barrier is a conserved pathological mediator of disrupted astrocyte-endothelia interaction in Alzheimer\u0026rsquo;s disease [Internet]. bioRxiv; 2025 [cited 2025 July 28]. p. 2025.01.24.634732. Available from: https://www.biorxiv.org/content/10.1101/2025.01.24.634732v1\u003c/li\u003e\n \u003cli\u003eLaing K. BioRender. 2025 [cited 2025 Nov 2]. Evidence-mapping schematic of studies included in the narrative synthesis. Available from: https://BioRender.com/b0tt4qm\u003c/li\u003e\n \u003cli\u003eKridel SJ, Chen E, Kotra LP, Howard EW, Mobashery S, Smith JW. Substrate Hydrolysis by Matrix Metalloproteinase-9*. J Biol Chem. 2001 Jan 1;276(23):20572\u0026ndash;8.\u003c/li\u003e\n \u003cli\u003eCui N, Hu M, Khalil RA. Biochemical and Biological Attributes of Matrix Metalloproteinases. Prog Mol Biol Transl Sci. 2017;147:1\u0026ndash;73.\u003c/li\u003e\n \u003cli\u003eRao R. OCCLUDIN PHOSPHORYLATION IN REGULATION OF EPITHELIAL TIGHT JUNCTIONS. Ann N Y Acad Sci. 2009 May;1165:62\u0026ndash;8.\u003c/li\u003e\n \u003cli\u003eYanckello LM, Hoffman JD, Chang YH, Lin P, Nehra G, Chlipala G, et al. Apolipoprotein E genotype-dependent nutrigenetic effects to prebiotic inulin for modulating systemic metabolism and neuroprotection in mice via gut-brain axis. Nutr Neurosci. 2022 Aug;25(8):1669\u0026ndash;79.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003cstrong\u003e. List of studies selected for review, including strain, group size, age, and sex.\u0026nbsp;\u003c/strong\u003eStudies included in synthesis are in bold.\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eStudy\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eStrain (Supplier)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGroup Size\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge (s)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSex\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRef\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003col\u003e\n \u003cli\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eNishitsuji et al., 2011\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003eHuman apoE mice, generated using homologous recombination method in embryonic stem cells\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003ePrimary cultures of mouse brain capillary endothelial cells (mBECs) were prepared from 3-week-old mice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e(37)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003col start=\"2\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBell et al., 2012\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003eTR-APOE mice (C57BL/6 background), generated in-house\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u003cem\u003en\u003c/em\u003e = 5 /group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003e4-9 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e(24)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003col start=\"3\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAlata et al., 2015\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003eE2, E3, and E4 targeted replacement mice (C57BL/6 background)\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eTaconic Transgenic Models (Hudson, NY, USA)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u003cem\u003en\u003c/em\u003e = 5-6/group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003e12 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003eMale, Female\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e(36)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003col start=\"4\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eThomas et al., 2017\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003eEFAD with the 5 Familial Alzheimer\u0026rsquo;s disease (FAD) mutations (APP K670N/M671L + I716 V + V717I and PS1 M146L + L286 V) with \u003cem\u003eAPOE\u003c/em\u003e-targeted replacement mice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u003cem\u003en\u003c/em\u003e = 6-8/group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003e6-8.5 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003eMale, Female\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e(34)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003col start=\"5\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eLin et al., 2017\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003eAPOE4 transgenic mice with GFAP promoter (C57BL/6 background)\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eJackson Laboratory (Bar Harbor, Maine, USA)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u003cem\u003en\u003c/em\u003e = 6/group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003e1-7 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003eFemale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e(23)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003col start=\"6\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMarottoli et al., 2017\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003e5xFAD\u003csup\u003e+/\u0026minus;\u003c/sup\u003e/\u003cem\u003eAPOE\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e (EFAD) mice (C57BL/6 background)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u003cem\u003en\u003c/em\u003e = 8/group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003e4-6 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e(35)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003col start=\"7\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eKoizumi et al., 2018\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003eApoE3-TR and ApoE4-TR mice (C57BL/6 background)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u003cem\u003en\u003c/em\u003e = 5/group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003e3-4 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e(25)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003col start=\"8\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eJohnson et al., 2019\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003eHuman E3- and E4-targeted replacement mice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003eCerebral Blood Volume measure: \u003cem\u003en\u003c/em\u003e = 7-8/group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003e15 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003eFemale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e(44)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003col start=\"9\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eYamazaki et al., 2020\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003eApoE3-TR and ApoE4-TR mice (C57BL/6 background)\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eTaconic Transgenic Models (Hudson, NY, USA)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u003cem\u003en\u003c/em\u003e = 4/group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003e22 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003eMale, Females\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e(40)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003col start=\"10\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eRingland et al., 2020\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003eApoE3-TR and ApoE4-TR mice\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eTaconic Transgenic Models (Hudson, NY, USA)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eEFAD (5xFAD\u003csup\u003e+/\u0026minus;\u003c/sup\u003e \u003cem\u003eAPOE\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e)\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u003cem\u003en\u0026nbsp;\u003c/em\u003e= 4 for each genotype\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003e6 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003eFemale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e(30)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003col start=\"11\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eLin et al., 2020\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003eEFAD with the 5 Familial Alzheimer\u0026rsquo;s disease (FAD) mutations (APP K670N/M671L + I716 V + V717I and PS1 M146L + L286 V) with \u003cem\u003eAPOE\u003c/em\u003e-targeted replacement mice (C57BL/6 background)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u003cem\u003en\u003c/em\u003e = 8/group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003e2 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003eMale, Female\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e(33)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003col start=\"12\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMontagne et al., 2021\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003eApoE3-TR and ApoE4-TR mice (C57BL/6 background)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003eIHC:\u003cem\u003e\u0026nbsp;n\u003c/em\u003e = 5-6/group\u003c/p\u003e\n \u003cp\u003eBBB: n = 6-8/ group\u003c/p\u003e\n \u003cp\u003eCBF:\u003cem\u003e\u0026nbsp;n\u0026nbsp;\u003c/em\u003e= 12-14/group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003e18-24 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003eMale, Female\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e(11)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003col start=\"13\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eRhea et al., 2021\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003eHuman E3- and E4-targeted replacement mice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u003cem\u003en\u003c/em\u003e = 10/group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003e15 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003eMale, Female\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e(38)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003col start=\"14\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eBarisano et al., 2022\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003eHuman \u003cem\u003eAPOE3\u003c/em\u003e and \u003cem\u003eAPOE4\u003c/em\u003e KI\u003csup\u003eflox/flox\u003c/sup\u003e\u003c/p\u003e\n \u003cp\u003emice (C57BL/6 background)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u003cem\u003en\u003c/em\u003e = 8/group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003e2\u0026ndash;3, 4\u0026ndash;6, and 9\u0026ndash;12 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003eMale, Female\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e(41)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003col start=\"15\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eJackson et al., 2022\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003eConditional humanized \u003cem\u003eAPOE\u003c/em\u003e knock-in (KI) mice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u003cem\u003en\u003c/em\u003e = 6/group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003e8-8.5 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e(39)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003col start=\"16\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eYanckello et al., 2022\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003eHuman APOE3(E3FAD mice) and APOE4gene (E4FAD mice)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u003cem\u003en\u003c/em\u003e = 8/group; (male: female= 1:1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003e7 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003eMale, Female\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e(52)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003col start=\"17\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eOnos et al., 2024\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003eB6J.\u003cstrong\u003eAPOE3\u003c/strong\u003e KI (hAPOE\u0026epsilon;3, available as B6.Cg-Apoeem2(APOE*)Adiuj/J, B6J.\u003cstrong\u003eAPOE4\u003c/strong\u003e KI (hAPOE\u0026epsilon;4, available as B6(SJL)-Apoetm1.1(APOE*4)Adiuj/J\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eJackson Laboratory (JAX) (JAX#029018, JAX#027894)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u003cem\u003en\u003c/em\u003e = 2-5/group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003e4-12 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003eMale, Female\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e(46)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003col start=\"18\"\u003e\n \u003cli\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eBhattarai et al., 2025\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31px;\"\u003e\n \u003cp\u003ehumanized targeted-replacement \u003cem\u003eAPOE-\u0026epsilon;4\u003c/em\u003e and \u003cem\u003eAPOE-\u0026epsilon;3\u003c/em\u003e mice\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003eN = 6 (50% female) for every experiment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18px;\"\u003e\n \u003cp\u003e18-22 months of age\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 6px;\"\u003e\n \u003cp\u003eMale, Female\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 4px;\"\u003e\n \u003cp\u003e(47)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp id=\"_Toc208395466\"\u003e\u003cstrong\u003e\u003cem\u003e\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003eTable 2. PICO elements (Population, Intervention, Comparator, Outcome) across studies.\u0026nbsp;\u003c/strong\u003eBBB Integrity: grouped IHC-based vascular markers which labelling basement membrane components (Col-IV, Glut1, Laminin). BBB Leakage: measured with integrated density of fibrinogen and MRI-derived Ktrans, reflecting functional barrier permeability.\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"983\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eStudy\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eComparator\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eIntervention Components\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eOutcome Domain\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eOutcome Measure\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTime Points\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBrain Regions\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eData: Mean \u0026plusmn; SEM, N / group\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eEffect \u0026amp; SE\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eIncluded in synthesis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBell et al., 2012\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eAPOE3 vs APOE4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eBBB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003ePericyte Markers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eIHC (% Pericyte Coverage)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e2 weeks,\u003c/p\u003e\n \u003cp\u003e9 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eCortex, Hippocampus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eCBF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eCBF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e14 C-iodoantipyrine autoradiograms\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e2 weeks,\u003c/p\u003e\n \u003cp\u003e9 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eSensorimotor, Ectorhinal Cortex, Hippocampus, Caudate Putamen, Corpus Callosum, Thalamus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAlta et al., 2014\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eAPOE3 vs APOE4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eBBB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eBBB Integrity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eIHC (% Area Occupied)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e12 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eCortex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eThomas et al., 2017\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eAPOE3 vs APOE4 (E3FAD vs E4FAD)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eBBB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eBBB Integrity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eIHC (% Area Occupied)\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e22 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eCortex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMarotolli et al., 2017\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eAPOE3 vs APOE4\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(E3FAD vs E4FAD)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eBBB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eBBB Integrity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eIHC (% Area Occupied)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e4-6 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eCortex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eKoizumi et al., 2018\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eAPOE3 vs APOE4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eBBB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eBBB Integrity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eIHC (% Area Occupied)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e3-4 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eCortex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eCBF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eCBF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eASL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e3-4 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eCortex, Caudate Nucleus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eYamazaki et al., 2020\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eAPOE3 vs APOE4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eBBB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eBBB Integrity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eRelative Intensity Signal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e22 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eCortex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMontagne et al., 2021\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eAPOE3 vs APOE4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"3\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eBBB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003ePericyte Markers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eIHC (% Pericyte Coverage)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e18-24 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eCortex, Hippocampus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eBBB Leakage\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eIHC (Integrated Density x10^3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e18-24 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eCortex, Hippocampus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eKtrans (x10^-3 min^-1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e18-24 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eCortex, Hippocampus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eCBF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eCBF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eASL (ml 100g-1 min-1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e18-24 months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eCortex, Hippocampus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp id=\"_Toc208395465\"\u003e\u003cstrong\u003e\u003cem\u003e\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003eTable 3. CAMARADES checklist for study quality.\u0026nbsp;\u003c/strong\u003eEach study was assessed against the following criteria: (1) publication in a peer-reviewed journal; (2) statement of temperature control; (3) random allocation to treatment or control; (4) blinded induction of the model; (5) blinded assessment of outcome; (6) use of anaesthetic without significant intrinsic neuroprotective activity; (7) use of an appropriate animal model; (8) sample size calculation; (9) compliance with animal welfare regulations; (10) statement of potential conflicts of interest. \u0026ldquo;Ref\u0026rdquo; indicates references. An \u0026ldquo;x\u0026rdquo; denotes fulfilment of the criterion (score = 1); a blank cell indicates non-fulfilment. An asterisk (*) represents partial fulfilment (score = 0.5). See supplementary material for additional details. Studies in \u003cstrong\u003ebold italics\u003c/strong\u003e were included in the meta-analysis.\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\" class=\"fr-table-selection-hover\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eStudy\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e8\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e9\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e10\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eScore\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNishitsuji et al., 2011\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003eBell et al., 2012\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e8\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003eAlata et al., 2015\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e6.5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003eThomas et al., 2017\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e6.5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLin et al., 2017\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e8\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003eMarottoli et al., 2017\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003eKoizumi et al., 2018\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e8\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eJohnson et al., 2019\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e6.5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003eYamazaki et al., 2020\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e5.5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRingland et al., 2020\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e5.5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLin et al., 2020\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e9.5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003eMontagne et al., 2021\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e9\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eYamazaki et al., 2021\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRhea et al., 2021\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e8\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBarisano et al., 2022\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e9\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eJackson et al., 2022\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eYanckello et al., 2022\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBonnar et al., 2023\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e8.5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eOnos et al., 2024\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e9\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAnderle et al., 2025\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003ex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBhattarai et al., 2025\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003eX\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003eX\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003eX\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003eX\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003eX\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 6px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e5.5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"alzheimers-research-and-therapy","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"azrt","sideBox":"Learn more about [Alzheimer's Research and Therapy](http://alzres.biomedcentral.com/)","snPcode":"13195","submissionUrl":"https://submission.nature.com/new-submission/13195/3","title":"Alzheimer's Research \u0026 Therapy","twitterHandle":"@AlzheimersRes","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Apolipoprotein E4, blood–brain barrier, cerebral blood flow, vascular dysfunction, Alzheimer’s disease, systematic review, meta-analysis","lastPublishedDoi":"10.21203/rs.3.rs-8011437/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8011437/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cu\u003e\u003cem\u003eBackground\u003c/em\u003e\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eThe E4 variant of Apolipoprotein E (APOE) is a primary genetic susceptibility risk factor for late-onset Alzheimer’s disease and has been implicated in cerebrovascular dysfunction. Preclinical mouse models are widely used to study \u003cem\u003eAPOE4\u003c/em\u003e, but cohesive understanding of APOE’s role is still inconsistent and lacking.\u003c/p\u003e\n\u003cp\u003eThe aim of this study was to systematically review and synthesise evidence from preclinical mouse studies assessing \u003cem\u003eAPOE4\u003c/em\u003e related effects on blood-brain barrier (BBB) integrity, vascular morphology and cerebral blood flow (CBF).\u003c/p\u003e\n\u003cp\u003e\u003cu\u003e\u003cem\u003eMain\u003c/em\u003e\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eA systematic search of MEDLINE, Embase, Scopus, and Web of Science was conducted (March-April 2025). Eligible studies included transgenic \u003cem\u003eAPOE\u003c/em\u003e-targeted replacement or knock-in mice reporting vascular outcomes (cerebral blood flow, blood brain barrier permeability, vascular measures). Risk of bias was assessed using SYRCLE and reporting quality with CAMARADES. Random-effects meta-analyses were conducted (where sufficient data was available), otherwise findings were narratively synthesised.\u003c/p\u003e\n\u003cp\u003eEighteen studies met inclusion. Outcome measures varied widely, including diverse approaches to CBF measurement (e.g. arterial spin labelling, autoradiography, DSC-MRI), immunohistochemical measures (e.g. collagen-IV, laminin, CD31), and diverse approaches to measurement of BBB leakage (e.g. fibronectin, fibrinogen, gadolinium-based ktrans). Seven studies contributed to meta-analysis: \u003cem\u003eAPOE4\u003c/em\u003emice showed a consistent reduction in CBF associated with \u003cem\u003eAPOE4\u003c/em\u003e genotype (SMD = -2.87, 95% CI: -5.14 to -0.604, df = 2.66), and a negative non-significant trend towards reduced vascular morphology expression. Narrative synthesis identified three key mechanistic pathways linking APOE4 to vascular dysfunction: (i) insulin resistance and PI3K/AKT-mTOR signalling, (ii) Cyclophilin A–NFκB–MMP9 activation, and (iii) occludin/ECM remodelling. Risk of bias assessment revealed frequent shortcomings in randomisation, blinding, and sample size justification.\u003c/p\u003e\n\u003cp\u003e\u003cu\u003e\u003cem\u003eConclusions\u003c/em\u003e\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003ePreclinical evidence demonstrates that \u003cem\u003eAPOE4\u003c/em\u003e drives alterations in vascular functioning primarily through involvement with pathways related to vascular metabolism, ECM remodelling and BBB leakage. However, heterogeneity in the model (e.g. age, sex, techniques), restricts direct comparability across studies. As such, standardisation or clarification of methodological approaches are necessary for rigorous assessment in the future.\u003c/p\u003e","manuscriptTitle":"Impact of Apolipoprotein E4 on Blood-Brain Barrier Integrity in Target Replacement Murine Models: A Systematic Review and Meta-Analysis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-29 16:25:28","doi":"10.21203/rs.3.rs-8011437/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-01-29T17:04:07+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-24T22:08:44+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"125727974767593724084910471007650290640","date":"2026-01-04T20:25:43+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-11-12T16:33:17+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-11-06T04:52:19+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-11-06T04:51:39+00:00","index":"","fulltext":""},{"type":"submitted","content":"Alzheimer's Research \u0026 Therapy","date":"2025-11-02T13:54:44+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"alzheimers-research-and-therapy","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"azrt","sideBox":"Learn more about [Alzheimer's Research and Therapy](http://alzres.biomedcentral.com/)","snPcode":"13195","submissionUrl":"https://submission.nature.com/new-submission/13195/3","title":"Alzheimer's Research \u0026 Therapy","twitterHandle":"@AlzheimersRes","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"279942eb-cf77-4bc5-8ba7-f90568e4667a","owner":[],"postedDate":"January 29th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-03-14T15:10:26+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-29 16:25:28","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8011437","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8011437","identity":"rs-8011437","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}