https://orcid.org/0000-0002-4286-7265
Figures
Abstract
Background
Impaired brain glucose metabolism is a preclinical feature of neurodegenerative diseases such as Alzheimer’s disease (AD). Infections may promote AD-related pathology. Therefore, we investigated the interplay between infections and APOE4, a strong genetic risk factor for AD.
Methods
We analyzed data on 1,509 participants in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database using multivariate linear regression models. The outcomes were rank-normalized hypometabolic convergence index (HCI), statistical regions of interest (SROI) for AD, and mild cognitive impairment (MCI). Marginal mean estimates for infections, stratified by APOE4 carrier status, were then computed.
Results
Prior infections were associated with greater HCI [β = 0.15, 95% CI: 0.03, 0.27, p = 0.01]. The combined effects of infections and APOE4 carriers on HCI levels were significantly greater than either variable alone. Among APOE4 carriers, the estimated marginal mean was 0.62, rising to 0.77, with infections (p<0.001), indicating an interaction effect. Carriers with multiple infections showed greater hypometabolism (higher HCI), with an estimate of 0.44 (p = 0.01) compared to 0.11 (p = 0.08) for those with a single infection, revealing a dose-response relationship. The estimates for the association of infections with SROI AD and SROI MCI were β = -0.01 (p = 0.02) and β = -0.01 (p = 0.04), respectively.
Citation: Lathika Rajendrakumar A, Arbeev KG, Bagley O, Duan M, Yashin AI, Ukraintseva S, et al. (2025) APOE4 and infectious diseases jointly contribute to brain glucose hypometabolism, a biomarker of Alzheimer’s pathology: New findings from the ADNI. PLoS ONE 20(1): e0316808. https://doi.org/10.1371/journal.pone.0316808
Editor: Yi Su, Banner Alzheimer’s Institute, UNITED STATES OF AMERICA
Received: October 1, 2024; Accepted: December 17, 2024; Published: January 7, 2025
Copyright: © 2025 Lathika Rajendrakumar et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The ADNI consortium determines access to the data used in our analysis and defines the data as a “Limited Data Set” under 45 CFR Part 164.514 (https://adni.loni.usc.edu/wp-content/themes/adni_2023/documents/ADNI_DSP_Policy.pdf). The authors have utilized this data exclusively for secondary analysis. However, all data can be publicly accessed through ADNI (https://adni.loni.usc.edu/data-samples/adni-data/#AccessData) upon request and can be used for replication, following the methods described in the paper. All researchers will have the same level of access as the authors.
Funding: This research was supported by the National Institute on Aging of the National Institutes of Health under Award Numbers R01AG076019 and R01AG062623. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Competing interests: The authors declare no competing interests.
Abbreviations: AD, Alzheimer’s disease; AIC, Akaike Information Criterion; APOE, Apolipoprotein E; Aβ, Amyloid βeta; FDG, 18F-fluorodeoxyglucose; GxE, Gene-environment interaction; GWAS, Genome-wide association studies; HSV, Herpes Simplex Virus; %INCMSE, Percent Increase in Mean Squared error; IQR, Interquartile Range; PET, Positron Emission Tomography; pTau, Phosphorylated Tau; SD, Standard Deviation; SNP, Single Nucleotide Polymorphism; SROI, Statistical Region of Interest; NECTIN2, Nectin Cell Adhesion Molecule 2 (gene); PUD, Peptic Ulcer Disease; UTI, Urinary Tract Infection
Introduction
Alzheimer’s disease (AD) is a slowly developing neurodegenerative disorder that is clinically manifested as dementia [1]. The current figure for the AD burden in older adults in the United States is 6.7 million, and it is poised to rise to 13.8 million by 2060 [2]. The preclinical stage of AD can last many years without obvious signs of dementia [3]. It is crucial to better understand this preclinical stage to develop successful AD prevention [4]. Common preclinical features of AD include toxic protein depositions, neuronal apoptosis, and reduction in hippocampal volume (brain shrinkage), and brain glucose hypometabolism [5, 6]. The brain glucose hypometabolism is observed long before the occurrence of overt symptoms in AD and is partly due to mitochondrial dysfunction [7]. Measuring glucose utilization in the brain using positron emission tomography (PET) and 18F-fluorodeoxyglucose (FDG) allows for convenient examination of hypometabolic patterns in the brain [8]. Brain scans based on FDG PET can effectively detect around 90% of AD-specific metabolic patterns, such as those in the parieto-temporal, frontal, and posterior cingulate regions [9].
A large genetic component drives AD (60–80%), and the entire spectrum of the disease can develop over 15–25 years [10]. Genetic variations in the APOE gene could single-handedly account for a large part of the risk related to AD in old age [11]. On the other hand, addressing modifiable risk factors could reduce or delay up to 40% of dementia risk [12]. Therefore, by focusing on the modifiable risk factors, a substantial part of the AD burden could be alleviated at the population level [13]. Prevention of certain infections can reduce the risk of chronic diseases, including neurological deficits [14–16]. Accumulating evidence suggests that infections could be a significant risk factor for AD that may also facilitate the development of AD pathology at the preclinical stage, though the exact mechanism is unclear and might involve a direct detrimental impact of infection-related factors as well as indirect effects of compromised immunity [17–20].
The connection between infections and AD and related pathology may also be influenced by genetic factors [18, 21, 22]. There are also indications that infections can contribute to brain hypometabolism, one of the earliest features of AD pathology; however, research on this topic is scarce [23]. Here we explored how infectious diseases may influence brain glucose metabolism in presence and absence of APOE4, the strongest genetic risk factor for AD, in participants of the Alzheimer’s Disease Neuroimaging Initiative (ADNI).
Methods
Study design and population
ADNI is a multi-center observational study that began in 2004 under the supervision of Michael W. Weiner. The study recruited individuals within the 55–90 years age range, and enrollment in this cohort occurs in different phases, with previous participants continuing to be in the study and new participants being recruited. The ADNI study only includes subtypes of amnestic MCI. A physician diagnoses MCI and AD based on the medical history, cognition scales such as the Alzheimer’s Disease Assessment Scale–Cognitive (ADAS-COG), the Mini-Mental State Examination (MMSE), and biomarkers.
To compare and gain knowledge about dementia, this database maintains and updates demographic, phenotypic, biomarker, and genetic data gathered from participants with normal cognition, AD, and other forms of cognitive impairment. The availability of such a wide variety of biomarkers provides sufficient information to learn about the evolution and pathology driving AD [24].
More details regarding the study design and objectives can be accessed here (https://adni.loni.usc.edu/study-design/). Broadly, ADNI seeks to integrate information from biomarkers, cognitive measures, and brain scans to improve AD diagnosis and treatment [25]. Brain scans were primarily collected to learn about the structural and metabolic functions of the brain, serving as a standard for differentiating the pathological changes seen in AD from those in normal aging [26].
Predictors: Infections and APOE4
Prior infections were determined by combining information from medical history, baseline symptoms, initial health assessment, and adverse effects datasets. The details of the selected subset of infections included in the final dataset are illustrated in Fig 1 in S1 File. Medical history information was collected during the screening visit using a questionnaire. Non-harmonious disease names were uniformly labeled for analytical purposes. Duplicated participant information having the same infection and diagnosis date, as well as any infections lacking a diagnosis date, were subsequently excluded. Covariates such as age, sex, education, race, marriage status, and APOE4 information were retrieved from the ADNIMERGE file.
The APOE4 carrier status was identified from DNA extracted by Cogenics from a 3 ml aliquot of EDTA blood extracted from participants during their screening visit [27]. Anti-diabetic medications were extracted (list provided in the S2 File) using the Anatomical Therapeutic Chemical (ATC) classification system coding (https://www.who.int/tools/atc-ddd-toolkit/atc-classification). Information regarding smoking and alcohol usage was obtained from the medical history file. Finally, we retained infections that only preceded the HCI measurements.
Outcomes: Brain glucose hypometabolism, AD, and MCI
Multiple PET scanners were used to capture brain images based on a standard protocol [28]. Measures were taken to correct for the related discrepancies [29]. The details regarding the PET scan and related protocols can be viewed elsewhere (https://adni.loni.usc.edu/methods/pet-analysis-method/pet-analysis/). The generated raw PET data are centrally stored at the Laboratory of Neuroimaging (LONI) at the USC Mark and Mary Stevens Neuroimaging and Informatics Institute of the University of California [30].
We retrieved the processed study outcomes from the BAIPETNMRCFDG dataset (https://adni.bitbucket.io/reference/baipetnmrc.html). The main outcome of interest was the hypometabolic convergence index (HCI), developed to reflect AD-specific hypometabolism across regions of the brain by computing voxel-wise z-scores from FDG-PET brain images. Higher HCI values correspond to lower levels of metabolism in the brain [31].
Additionally, we examined the associations for infections with statistical regions of interest (SROI) corresponding to AD and Mild Cognitive Impairment (MCI). SROI associations might provide additional insights into the cerebral metabolic rate for glucose (CMRgl) decline in these brain regions, helping to understand the disease-specific pathology they represent [32]. The Statistical Parametric Mapping (SPM) software was used to generate the HCI and SROI scores [32, 33]. The work of Landau et al. provides further details on the generation and development of regions of interest in the ADNI cohort [34]. A decline in FDG-PET Region of Interest (ROI) values suggests a pathological brain damage and may contribute to the progression of dementia [33].
Statistical analysis
R version 4.3.2 was used for the data linking and statistical analysis [35]. We analyzed the dataset with full covariate and outcome information, without performing any imputations. The ggplot2 package was used to create variable distribution plots [36]. The leptokurtic HCI readings were normalized during the RNomni package [37]. Multivariate linear regression models were conducted for all specified outcomes separately. Age, education, and allele dosages of APOE ε4 were analyzed as continuous variables. Infections, AD, and diabetes medication use were coded as a binary variable (yes or no). Marriage, smoking, and alcohol use were coded as Ever or Never. We explored models with a full set and a reduced set of covariates. The parsimonious model (the best explanatory model) was determined using the Akaike Information Criterion (AIC) in the MuMin package [38]. This package iteratively runs multiple models with varying terms, assigning weights to each model based on its AIC to determine the optimal set of variables. This approach effectively reduces the number of variables in the final model without imposing the constraints typically associated with multiple comparison corrections, such as the Bonferroni test. A two-sided p-value less than 0.05 was considered to support our hypothesis.
A Random Forest-based model was used to rank the significant variables according to their contributions to the best model [39]. For this, we specified the following hyperparameters; 5000 trees, the number of features for splitting corresponding to the square root of the total number of variables, a node size of 1, bootstrapping with replacement, and Mean Squared Error (MSE) as the splitting criterion. The effect modification for infections with HCI by APOE4 carrier status was assessed by visualizing with the rockchalk package [40]. Marginal mean estimates were calculated to show the interaction effects for the infections across categories of APOE4 and sex.
Ethics considerations
The Institutional Review Board of Duke University Health System issued approval for this study (Protocol IDs Pro00109279 and Pro00105389). This publication includes only secondary analyses of existing data available from ADNI, and does not include identifiable human data. Written informed consent for ADNI participants was obtained by the ADNI in accordance with the local legislation and ADNI requirements. ADNI studies follow Good Clinical Practices guidelines, the Declaration of Helsinki, and United States regulations (U.S. 21 CFR Part 50 and Part 56).
Results
Participant characteristics
The final sample included information on 1,509 participants after data linking (Fig 2 in S1 File). As shown in Table 1, the average age among participants was 73.3 years, with an average education duration of 16.0 years (IQR 14.0–18.0). Over 96% of respondents reported being ever married, and 55.8% were males. There was a relatively lower representation of non-white individuals, totaling 116 (7.6%) in the sample. Percentage of individuals with a history of smoking and alcohol use was 27.1% and 3.3%, respectively. Of these, 215 individuals accounting for 14.2% of the total sample size, reported having infections. The median interval between biomarker assessment and infections was 8.4 (IQR: 3.5–28.3). Median HCI was 12.59, and the IQR was 8.4–19.3.
Fig 1 shows the distributions of the original HCI and rank-normalized HCI and also a scatterplot of their relationship. For AD and MCI participants, the mean SROI values were 1.15 and 1.03, respectively. About 3% of participants with diabetes were on medication. As regards the cognitive status of participants in the sample, 333 (22.1%) were cognitively normal (CN), 277 (18.3%) were diagnosed with AD, 334 (22.1%) had Early Mild Cognitive Impairment (EMCI), 458 (30.3%) had Late Mild Cognitive Impairment (LMCI), 106 (7.0%) had Significant Memory Concern (SMC), and 1 (0.0%) did not have a diagnosis. Peptic ulcer disease (PUD) (n = 152), urinary tract infection (UTI) (n = 146), and pneumonia (n = 102) were the most frequent among the selected infections. Fig 3 in S1 File shows the difference in the distribution of HCI values for individuals with infections, AD, and APOE4. The median HCI value among individuals with infections was 13.64, while it was lower (12.48) for those without infections. It was also seen that the HCI had a modest positive correlation with APOE4 (Fig 4 in S1 File).
Association of infections and other predictors with the HCI
Table 1 in S1 File shows the regression estimates for all the predictors in the multivariate linear regression full model for HCI outcome. Marriage status, education, smoking, alcohol, and diabetes medication use were not significant predictors of HCI. Table 2 presents the results of the reduced model, which best describes the model variance. AD status predicted the strongest reduction in brain metabolism [β = 1.04, 95% CI 0.92–1.15, p<0.001], followed by age [β = 0.01, 95% CI 0.01–0.02, p<0.001] and APOE4 carrier status [β = 0.32, 0.25–0.38, p<0.001]. Higher variable relevance is indicated by higher values of %INCMSE and INCNodepurity (Table 2 in S1 File). The regression coefficient for infections was 0.15 [95% CI 0.02–0.27, p = 0.01]. Males and white people were at higher risk of having elevated HCI values. Smoking history was the only non-significant predictor retained in the reduced model. The adjusted R-squared from the reduced model was 26.9%. Males had higher median HCI values.
In the sex-stratified analysis evaluating the effects of infections versus non-infections, males generally demonstrated relatively higher HCI values (Table 3 in S1 File). The difference in normalized marginal means between all groups was statistically significant (p<0.001). The combined effects of infections and APOE4 carrier status on HCI levels are shown in Fig 2. This was significantly greater than the effects of either variable alone. Table 4 in S1 File clarifies these results. Specifically, for individuals without infections and APOE4 carrier status, the estimated marginal mean was 0.03 (p = 0.53). However, this increased significantly to 0.18 (p<0.001) for APOE4 non-carriers in the presence of infections. Notably, among APOE4 carriers, the estimated marginal mean was substantially higher at 0.62, and this value rose to 0.77 with infections (p<0.001), confirming an interaction between the two factors.
This interaction was further demonstrated in the additional analysis (Table 5 in S1 File and Fig 3), indicating that carriers who experienced multiple infections exhibited greater hypometabolism. Among individuals with more than one prior infection (n = 23), the estimate was significantly higher at 0.44 (p = 0.01) compared to those with a single infection, which was 0.11 (p = 0.08), revealing a dose-response relationship.
Association of infections and other predictors with the SROI AD
Table 6 in S1 File provides regression estimates for all the factors investigated for SROI AD. Marriage, race, smoking, and alcohol history were not significant predictors of AD-specific hypometabolism. In the reduced model shown in Table 3, AD was associated with increased region-specific hypometabolism (regression coefficient: -0.08, p<0.001). The use of diabetes medications was associated with decreased brain metabolism (-0.03, p = 0.02). Similar to previous regression, an increase in APOE4 alleles was a strong risk factor for hypometabolism (-0.02, p<0.001). Male gender showed greater hypometabolism (-0.01, p<0.01). Although education was linked to a better metabolic pattern, this relationship was not profound. Age-specific decreases were not as notable as those observed in HCI (-0.003, p<0.001). While statistically significant, the effect estimate for previous infections was lower for AD (-0.01, p = 0.02). These variables collectively predicted 26.8% of the variance in SROI AD.
Association of infections and other predictors with the SROI MCI
The results of the SROI MCI regression (full model) is presented in the Table 7 in S1 File. Generally, the estimates were closer to the SROI AD than HCI. Among the variables that best explained the model (Table 4), AD, APOE4, and diabetes medications had the largest effect estimates. Sex and use of diabetic medications had a marginally greater impact on the MCI region than on the AD region. However, race and education were not identified as significant predictors. Infections were associated with a -0.01 reduction in regional metabolism (p = 0.04). The percentage of variation explained by the model for the SROI MCI was also the highest (28.8%) of the three investigated outcomes.
Discussion
Results of our study suggest that infections and APOE4 can jointly significantly affect brain glucose metabolism, specifically promote hypometabolism, as measured by the increased values of HCI. A history of infections in this ADNI sample corresponds to a greater hypometabolism, specifically a 0.15 unit increase in rank normalized HCI. However, this estimate rose to 0.44 in the presence of multiple infections. Model inclusion of established confounders such as age, sex, race, and education did not diminish these findings. We also adjusted for AD status, which was more prevalent in the group with no infections and associated with reduced brain metabolism. Adjusting for AD status and APOE4 was necessary to reveal the genuine association of previous infections. Additionally, previous infections were significantly associated with regional brain metabolism specific to AD and MCI in our data. The MCI cases are exclusive and not due to AD, as the final diagnosis status was considered in the analysis.
Our findings agree with previous research suggesting that infections may negatively impact brain metabolism [16, 41–43]. Infectious diseases, including those addressed in this study, have been previously linked to AD in other data [22, 44–46]. Our recent paper that used Health and Retirement Study (HRS) data reported associations between AD and various infectious (viral, bacterial, fungal), suggesting that compromised immunity may play a role in AD etiology [20]. The connection between infections and brain hypometabolism may also involve pathological immune responses. Some research provides indirect support to this idea by linking brain hypometabolism to microglia activation [47–49].
Determining the onset time of infection is a major challenge in AD research. Furthermore, the causal inferences are obscured by the fact that individuals with AD often grapple with a variety of infections due to declining immunity, leading to elevated antimicrobial markers [16]. Pathogens have a high affinity to the central nervous system and brain tissue and could affect cognition [50]. Given that brain hypometabolism is an early sign of AD, our findings suggest that infections could potentially trigger this process [51]. However, the progression of hypometabolism may also depend upon the combination of other risk factors [52]. Infections can affect the brain through multiple pathways, both directly and indirectly, particularly when the blood-brain barrier is breached [53, 54]. Infections propagated through the respiratory route can also reach the brain relatively easily [50, 55]. However, upon reaching the brain, different infectious agents employ their preferred mechanisms, such as latent activation and the initiation of inflammation, as seen in the case of the Herpes virus [56]. Pathogen invasion into the brain leads to chronic inflammation, which can compromise the blood-brain barrier [42]. There are distinct differences in inflammatory pathways noted across specific pathogens [19, 57]. Age-related changes could exacerbate these pathological processes even further [58]. Strom and colleagues have also demonstrated that brain hypometabolism correlates with tau pathology and neurodegeneration in crucial dementia-specific regions within the ADNI dataset [59]. These mechanisms could potentially explain a significant portion of the biological processes leading up to hypometabolism.
Studies on the relationship between APOE4 and brain metabolism have produced contrasting findings [59–63]. In their recently published work, Fortea and colleagues found that simply being homozygous for APOE4 is sufficient, in most cases, to guarantee an AD diagnosis [64]. In our analysis, the increase in APOE4 allele was associated with all three outcomes and showed compounding effects with infections and their burden. Even in patients with a single APOE4 variant, which is usually not considered a significant increase in risk compared to homozygous carriers, the presence of infections increases the risk of hypometabolism to nearly the same level as in homozygous APOE4 carriers. One possibility is that the observed effect is due to accelerated neuroinflammation arising from the presence of both risk factors [65]. Risk factors for AD tend to cluster in individuals with APOE risk alleles, including a reduction in brain metabolism [66]. Amyloid-beta and Tau deposition are higher in APOE4 carriers [67]. APOE4 can also accelerate brain degeneration through non-overlapping pathways independent of amyloid deposition and Tau pathology [68–70]. APOE4 alleles both promote and resist infections, depending on the type of infection [71]. Researchers suggest that APOE4 polymorphisms result in increased lipid production [72] and blood-brain barrier loss [73], which could facilitate a conducive environment for pathogens [74]. Supporting evidence from the Northern Manhattan Study showed that the effect-modifying relationship between APOE4 and infectious burden was correlated with decreased cognition [75]. The influence of APOE4 on AD remains incompletely understood, although it is known to engage in intricate interactions with other risk factors for AD, such as age [76, 77]. However, in stark contrast to these findings, a study reported that the effects of APOE4 on cognition are AD-specific. It singles out the cause of cognitive decline as the interaction between APOE4 and amyloid beta in the hippocampus [78].
Our study revealed that diabetes medication was the third-biggest risk factor for AD and MCI-specific brain metabolism, but not for the HCI measure. Previous studies indeed demonstrated that diabetes increases the risk for MCI and AD [79, 80]. Individuals with diabetes and AD often share common biological pathways [81]. Most prominent among these are low-grade chronic inflammation and insulin resistance [82].
Sex differences in mechanisms related to AD warrant in-depth study. Usually, females are susceptible to AD and early brain hypometabolism compared to males [83, 84]. On exposure to prior infections, women are also, particularly at higher risk for reduced hippocampal volume [19]. Males overall had a higher HCI value than females in our sample. It is important to note that males with infections had a slightly higher mean age. But this alone cannot explain the gender difference. Importantly, there was no difference in the increase in marginal means due to infections for both sexes. An earlier study reported that brain hypometabolism increased in men after 70 years of age, while this was not seen in females in a normal brain [85]. However, the applicability of this finding in the AD context needs confirmation. Some participant characteristics in ADNI may differ from the general population due to voluntary recruitment. Variations in the distribution of AD risk factors among genders might also contribute to this finding [86, 87].
Given that AD is not curable, prevention stands as the most viable option at present. Vaccinations may potentially alleviate dementia risk. Influenza vaccines, in particular, are among the candidates demonstrating this preventive potential [4, 79]. However, personal genetics could play a role in determining the efficacy and effectiveness of vaccinations. Recent research has revealed that individuals carrying a polymorphism in the NECTIN2 gene exhibit a decreased susceptibility to AD when compared to non-carriers, when receiving vaccinations for pneumonia and flu [22].
The availability of medical history information and longitudinally standardized FDG PET measurements were important strengths of our study. We were also able to demonstrate the temporality of association, which was rarely described in earlier human studies [88]. There were a couple of study limitations. Of these, the most important is that the medical history is questionnaire-based, suggesting that recall bias may exist and lead to an incorrect exposure classification. Currently, the representation of high-risk groups, such as Afro-American and Hispanic individuals, is limited in the ADNI database, which has constrained the evaluation of the effect modification role of race in relation to infections and brain hypometabolism [89]. In this work, we did not specifically explore the heterogeneity in infections and the brain metabolism relationship. However, previous AD studies indicate that there could be subgroups that may be differentially vulnerable [90–92]. The distribution of the time from the onset of infection and HCI measurement was highly variable in our sample. As individuals with more than one prior infection were limited in our dataset, we could not further investigate whether the timing of infection, along with number of previous infections, would accelerate brain hypometabolism. There may also be a cohort effect, wherein the frequency of infections observed within this group may not accurately fit the current disease landscape. We recommend validating the findings in large cohorts with robust information on prior infections.
Conclusion
This study found that infections and APOE4 jointly promoted brain glucose hypometabolism in older ADNI participants. In individuals with a history of infections who were also carriers of one APOE4 allele, the degree of brain glucose hypometabolism was nearly that seen in APOE4 homozygotes without prior infections. We conclude that prior infections may contribute to AD pathology in synergy with APOE4, thus playing a part in the “multi-hit” mechanism of AD development.
Acknowledgments
Data collection and sharing for this project was funded by the Alzheimer’s Disease Neuroimaging Initiative (ADNI) (National Institutes of Health Grant U01 AG024904) and DOD ADNI (Department of Defense award number W81XWH-12-2-0012). ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: AbbVie, Alzheimer’s Association; Alzheimer’s Drug Discovery Foundation; Araclon Biotech; BioClinica, Inc.; Biogen; Bristol-Myers Squibb Company; CereSpir, Inc.; Eisai Inc.; Elan Pharmaceuticals, Inc.; Eli Lilly and Company; EuroImmun; F. Hoffmann-La Roche Ltd and its affiliated company Genentech, Inc.; Fujirebio; GE Healthcare; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research & Development, LLC.; Johnson & Johnson Pharmaceutical Research & Development LLC.; Lumosity; Lundbeck; Merck & Co., Inc.; Meso Scale Diagnostics, LLC.; NeuroRx Research; Neurotrack Technologies; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier; Takeda Pharmaceutical Company; and Transition Therapeutics. The Canadian Institutes of Health Research is providing funds to support ADNI clinical sites in Canada. Private sector contributions are facilitated by the Foundation for the National Institutes of Health (www.fnih.org). The grantee organization is the Northern California Institute for Research and Education, and the study is coordinated by the Alzheimer’s Disease Cooperative Study at the University of California, San Diego. ADNI data are disseminated by the Laboratory for Neuro Imaging at the University of Southern California.
Data used in preparation of this article were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at: http://adni.loni.usc.edu/wp-content/uploads/how_to_apply/ADNI_Acknowledgement_List.pdf.
References
- 1. Tupler L, Yashkin A, Hoque M, Yashin A, Akushevich I, Stallard E. Differences in cognitive decline due to Alzheimer’s disease, neurodegenerative disorders, and trauma. Innov Aging. 2023;7(Suppl 1):852. PubMed Central PMCID: PMC10737900.
- 2.
2023 Alzheimer’s disease facts and figures. Alzheimer’s Dement. 2023;19:1598–695.
- 3. Dubois B, Hampel H, Feldman HH, Scheltens P, Aisen P, Andrieu S, et al. Preclinical Alzheimer’s disease: Definition, natural history, and diagnostic criteria. Alzheimers Dement. 2016;12(3):323. pmid:27012484; PubMed Central PMCID: PMC6417794.
- 4. Fiandaca MS, Kapogiannis D, Mapstone M, Boxer A, Eitan E, Schwartz JB, et al. Identification of preclinical Alzheimer’s disease by a profile of pathogenic proteins in neurally derived blood exosomes: A case-control study. Alzheimer’s & Dementia. 2015;11(6):600–607. pmid:25130657; PubMed Central PMCID: PMC4329112.
- 5. Blázquez E, Hurtado-Carneiro V, LeBaut-Ayuso Y, Velázquez E, García-García L, Gómez-Oliver F, et al. Significance of Brain Glucose Hypometabolism, Altered Insulin Signal Transduction, and Insulin Resistance in Several Neurological Diseases. Front Endocrinol (Lausanne). 2022;13(873301). pmid:35615716; PubMed Central PMCID: PMC9125423.
- 6. Jack CR, Andrews JS, Beach TG, Buracchio T, Dunn B, Graf A, et al. Revised criteria for diagnosis and staging of Alzheimer’s disease. Nat Med. 2024;30(8):2121–4. pmid:38942991.
- 7. Atamna H, Frey WH. Mechanisms of mitochondrial dysfunction and energy deficiency in Alzheimer’s disease. Mitochondrion. 2007;7(5):297–310. pmid:17625988.
- 8. Herholz K. PET studies in dementia. Ann Nucl Med. 2003;17(2):79–89. pmid:12790355.
- 9. Mosconi L. Brain glucose metabolism in the early and specific diagnosis of Alzheimer’s disease. FDG-PET studies in MCI and AD. Eur J Nucl Med Mol Imaging. 2005;32(4):486–510. pmid:15747152.
- 10. Scheltens P, De Strooper B, Kivipelto M, Holstege H, Chételat G, Teunissen CE, et al. Alzheimer’s disease. Lancet. 2021;397(10284):1590. pmid:33667416; PubMed Central PMCID: PMC8354300.
- 11. Tanzi RE. The Genetics of Alzheimer Disease. Cold Spring Harb Perspect Med. 2012;2(10):a006296. pmid:23028126; PubMed Central PMCID: PMC3475404.
- 12. Livingston G, Huntley J, Sommerlad A, Ames D, Ballard C, Banerjee S, et al. Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet. 2020;396(10248):446. pmid:32738937; PubMed Central PMCID: PMC7392084.
- 13. De La Torre JC. Alzheimer’s disease is incurable but preventable. J Alzheimers Dis. 2010;20(3):70. pmid:20182017.
- 14. Grau AJ, Urbanek C, Palm F. Common infections and the risk of stroke. Nat Rev Neurol. 2010;6:681–94. pmid:21060340.
- 15. Mcmurry HS, Mendenhall E, Rajendrakumar A, Nambiar L, Satyanarayana S, Shivashankar R. Coprevalence of type 2 diabetes mellitus and tuberculosis in low ‐ income and middle ‐ income countries: A systematic review. Diabetes Metab Res Rev. 2019;35(2018). pmid:30144270.
- 16. Vigasova D, Nemergut M, Liskova B, Damborsky J. Multi-pathogen infections and Alzheimer’s disease. Microb Cell Fact. 2021;20(25). pmid:33509204; PubMed Central PMCID: PMC7844946.
- 17. Rippee-Brooks MD, Wu W, Dong J, Pappolla M, Fang X, Bao X. Viral Infections, Are They a Trigger and Risk Factor of Alzheimer’s Disease? Pathogens. 2024;13(3):240. pmid:38535583; PubMed Central PMCID: PMC10974111.
- 18. Paranjpe MD, Belonwu S, Wang JK, Oskotsky T, Gupta A, Taubes A, et al. Sex-Specific Cross Tissue Meta-Analysis Identifies Immune Dysregulation in Women With Alzheimer’s Disease. Front Aging Neurosci. 2021;13(735611). pmid:34658838; PubMed Central PMCID: PMC8515049.
- 19. Popov VA, Ukraintseva S, Duan H, Arbeev KG, Yashin AI. Prior infections are associated with smaller hippocampal volume in older women. Front Dement. 2024;3(2024). pmid:38707479; PubMed Central PMCID: PMC11067727.
- 20. Ukraintseva S, Yashkin AP, Akushevich I, Arbeev K, Duan H, Gorbunova G, et al. Associations of infections and vaccines with Alzheimer’s disease point to a role of compromised immunity rather than specific pathogen in AD. Exp Gerontol. 2024;190. PubMed Central PMCID: PMC11060001. pmid:38548241
- 21. Rajendrakumar L, Ukraintseva S, Arbeev KG, Bagley O, Yashin AI. The association between rs6859 in NECTIN2 gene and Alzheimer’s disease is partly mediated by pTau. Front Aging Neurosc. 2024;16. pmid:39165837; PubMed Central PMCID: PMC11334082.
- 22. Ukraintseva S, Duan M, Simanek AM, Holmes R, Bagley O, Rajendrakumar AL, et al. Vaccination Against Pneumonia May Provide Genotype-Specific Protection Against Alzheimer’s Disease. J Alzheimers Dis. 2023;96(2):499–505. pmid:37807778; PubMed Central PMCID: PMC10657669.
- 23. Hosp JA, Dressing A, Blazhenets G, Bormann T, Rau A, Schwabenland M, et al. Cognitive impairment and altered cerebral glucose metabolism in the subacute stage of COVID-19. Brain. 2021;144(4):1263–76. pmid:33822001; PubMed Central PMCID: PMC8083602.
- 24. Weber CJ, Carrillo MC, Jagust W, Jack CR, Shaw LM, Trojanowski JQ, et al. The Worldwide Alzheimer’s Disease Neuroimaging Initiative: ADNI‐3 updates and global perspectives. Alzheimers Dement (N Y). 2021;7(1):e12226. pmid:35005206; PubMed Central PMCID: PMC8719344.
- 25. Ashford MT, Raman R, Miller G, Donohue MC, Okonkwo OC, Mindt MR, et al. Screening and enrollment of underrepresented ethnocultural and educational populations in the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Alzheimer’s Dement. 2022;1–11. pmid:35213778; PubMed Central PMCID: PMC9402812.
- 26. Petersen RC, Aisen PS, Beckett LA, Donohue MC, Gamst AC, Harvey DJ, et al. Alzheimer’s Disease Neuroimaging Initiative (ADNI) Clinical characterization. Neurology. 2010;74:201–9. pmid:20042704; PubMed Central PMCID: PMC2809036.
- 27. Sudre CH, Cardoso MJ, Frost C, Barnes J, Barkhof F, Fox N, et al. APOE ε4 status is associated with white matter hyperintensities volume accumulation rate independent of AD diagnosis. Neurobiol Aging. 2017;53:67–75. pmid:28235680.
- 28. Langbaum JBS, Chen K, Lee W, Reschke C, Bandy D, Fleisher AS, et al. Categorical and correlational analyses of baseline fluorodeoxyglucose positron emission tomography images from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Neuroimage. 2009;45(4):1116. pmid:19349228; PubMed Central PMCID: PMC2886795.
- 29. Jagust WJ, Bandy D, Chen K, Foster NL, Landau SM, Mathis CA, et al. The Alzheimer’s Disease Neuroimaging Initiative positron emission tomography core. Alzheimer’s & Dementia. 2010;6(3):221–9. pmid:20451870; PubMed Central PMCID: PMC2920531.
- 30. Mueller SG, Weiner MW, Thal LJ, Petersen RC, Jack C, Jagust W, et al. The Alzheimer’s Disease Neuroimaging Initiative. Neuroimaging Clin N Am. 2005;15(4):869–77. pmid:16443497; PubMed Central PMCID: PMC2376747.
- 31. Chen K, Ayutyanont N, Langbaum JBS, Fleisher AS, Reschke C, Lee W, et al. Characterizing Alzheimer’s disease using a hypometabolic convergence index. Neuroimage. 2011;56:52–60. pmid:21276856; PubMed Central PMCID: PMC3066300.
- 32. Chen K, Langbaum JBS, Fleisher AS, Ayutyanont N, Reschke C, Lee W, et al. Twelve-month metabolic declines in probable Alzheimer’s disease and amnestic mild cognitive impairment assessed using an empirically pre-defined statistical region-of-interest: Findings from the Alzheimer’s Disease Neuroimaging Initiative. Neuroimage. 2010;51(2):654–64. pmid:20202480; PubMed Central PMCID: PMC2856742.
- 33. Ewers M, Brendel M, Rizk-Jackson A, Rominger A, Bartenstein P, Schuff N, et al. Reduced FDG-PET brain metabolism and executive function predict clinical progression in elderly healthy subjects. Neuroimage Clin. 2014;4:45–52. pmid:24286024; PubMed Central PMCID: PMC3841292.
- 34. Landau SM, Harvey D, Madison CM, Koeppe RA, Reiman EM, Foster NL, et al. Associations between cognitive, functional, and FDG-PET measures of decline in AD and MCI. Neurobiol Aging. 2011:1207–18. pmid:19660834; PubMed Central PMCID: PMC2891865.
- 35.
R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria [Internet]. 2021 [cited 2023 May 15]. Available from: https://www.R-project.org/
- 36.
Wickham H. ggplot2: Elegant Graphics for Data Analysis Using the Grammar of Graphics [Internet]. Springer-Verlag New York; 2016. Available from: https://ggplot2.tidyverse.org.
- 37. McCaw Z. RNOmni: Rank Normal Transformation Omnibus Test [Internet]. 2020. Available from: https://CRAN.R-project.org/package=RNOmni
- 38. Bartoń K. MuMIn: Multi-model inference [Internet]. Vol. 1, R package version 1.10.0. 2013. Available from: https://cran.r-project.org/web/packages/MuMIn/MuMIn.pdf
- 39. Liaw A, Wiener M. Classification and Regression by randomForest. R News [Internet]. 2002 [cited 2024 May 2];2(3):18–22. Available from: http://www.stat.berkeley.edu/
- 40. Johnson PE. Using rockchalk for Regression Analysis * [Internet]. 2017 [cited 2024 Apr 30]. Available from: http://cran.nexr.com/web/packages/rockchalk/vignettes/rockchalk.pdf
- 41. Ashraf GM, Tarasov V V, Makhmutovа A, Chubarev VN, Avila-Rodriguez M, Bachurin SO, et al. The Possibility of an Infectious Etiology of Alzheimer Disease. Mol Neurobiol. 2019;56:4479–91. pmid:30338482.
- 42. Shinjyo N, Kita K. Infection and Immunometabolism in the Central Nervous System: A Possible Mechanistic Link Between Metabolic Imbalance and Dementia. Front Cell Neurosci. 2021;15(765217). pmid:34795562; PubMed Central PMCID: PMC8592913.
- 43. Butler L, Walker KA. The Role of Chronic Infection in Alzheimer’s Disease: Instigators, Co-conspirators, or Bystanders? Curr Clin Micro Rpt. 2021;8:199–212. pmid:35186664; PubMed Central PMCID: PMC8849576.
- 44. Cairns DM, Itzhaki RF, Kaplan DL. Potential Involvement of Varicella Zoster Virus in Alzheimer’s Disease via Reactivation of Quiescent Herpes Simplex Virus Type 1. J Alzheimers Dis. 2022;88(3):1189–200. pmid:35754275.
- 45. Bukhbinder AS, Ling Y, Hasan O, Jiang X, Kim Y, Phelps KN, et al. Risk of Alzheimer’s Disease Following Influenza Vaccination: A Claims-Based Cohort Study Using Propensity Score Matching. J Alzheimers Dis. 2022;88(3):1061–74. pmid:35723106; PubMed Central PMCID: PMC9484126.
- 46. Noori M, Mahboobi R, Nabavi-Rad A, Jamshidizadeh S, Fakharian F, Yadegar A, et al. Helicobacter pylori infection contributes to the expression of Alzheimer’s disease-associated risk factors and neuroinflammation. Heliyon. 2023;9(9):e19607. pmid:37810022; PubMed Central PMCID: PMC10558876.
- 47. Tondo G, Boccalini C, Caminiti SP, Presotto L, Filippi M, Magnani G, et al. Brain Metabolism and Microglia Activation in Mild Cognitive Impairment: A Combined [18F]FDG and [11C]-(R)-PK11195 PET Study. J Alzheimers Dis. 2021;80(1):433–45. pmid:33579848.
- 48. Kadamani KL, Logan SM, Pamenter ME. Does hypometabolism constrain innate immune defense? Acta Physiol. 2024;240:e14091. pmid:38288574.
- 49. Aldana BI. Microglia-Specific Metabolic Changes in Neurodegeneration. J Mol Biol. 2019;431(9):1830–42. pmid:30878483.
- 50. Guedj E, Million M, Dudouet P, Tissot-Dupont H, Bregeon F, Cammilleri S, et al. 18F-FDG brain PET hypometabolism in post-SARS-CoV-2 infection: substrate for persistent/delayed disorders? Eur J Nucl Med Mol Imaging. 2021;48(2):592–5. pmid:32728799; PubMed Central PMCID: PMC7391029.
- 51. Zilberter Y, Zilberter M. The vicious circle of hypometabolism in neurodegenerative diseases: Ways and mechanisms of metabolic correction. J Neurosci Res. 2017;95:2217–35. pmid:28463438.
- 52. Yang Z, Cummings JL, Kinney JW, Cordes D. Accelerated hypometabolism with disease progression associated with faster cognitive decline among amyloid positive patients. Front Neurosci. 2023;17. pmid:37123373; PubMed Central PMCID: PMC10140339.
- 53. Sweeney MD, Sagare AP, Zlokovic B V. Blood-brain barrier breakdown in Alzheimer disease and other neurodegenerative disorders. Nat Rev Neurol. 2018;14(3):133–50. pmid:29377008; PubMed Central PMCID: PMC5829048.
- 54. Sochocka M, Zwolińska K, Leszek J. The Infectious Etiology of Alzheimer’s Disease. Curr Neuropharmacol. 2017;15(7):996–1009. pmid:28294067; PubMed Central PMCID: PMC5652018.
- 55. Wan D, Du T, Hong W, Chen L, Que H, Lu S, et al. Neurological complications and infection mechanism of SARS-COV-2. Signal Transduct Target Ther. 2021;6(406). pmid:34815399; PubMed Central PMCID: PMC8609271.
- 56. Lövheim H, Gilthorpe J, Adolfsson R, Nilsson LG, Elgh F. Reactivated herpes simplex infection increases the risk of Alzheimer’s disease. Alzheimer’s & Dementia. 2015;11(6):593–9. pmid:25043910.
- 57. Chhatbar C, Prinz M. The roles of microglia in viral encephalitis: from sensome to therapeutic targeting. Cell Mol Immunol. 2021;18:250–8. pmid:33437050; PubMed Central PMCID: PMC7802409.
- 58. Montagne A, Barnes SR, Sweeney MD, Halliday MR, Sagare AP, Zhao Z, et al. Blood-brain barrier breakdown in the aging human hippocampus. Neuron. 2015;85(2):296–302. pmid:25611508; PubMed Central PMCID: PMC4350773.
- 59. Strom A, Iaccarino L, Edwards L, Lesman-Segev OH, Soleimani-Meigooni DN, Pham J, et al. Cortical hypometabolism reflects local atrophy and tau pathology in symptomatic Alzheimer’s disease. Brain. 2022;145(2):713–28. pmid:34373896; PubMed Central PMCID: PMC9014741.
- 60. Corder EH, Jelic V, Bosun H, Lannfelt L, Valind S, Winblad B, et al. No Difference in Cerebral Glucose Metabolism in Patients With Alzheimer Disease and Differing Apolipoprotein E Genotypes. Arch Neurol. 1997;54(3):273–7. pmid:9074396.
- 61. Rabinovici GD, Furst AJ, Alkalay A, Racine CA, O’Neil JP, Janabi M, et al. Increased metabolic vulnerability in early-onset Alzheimer’s disease is not related to amyloid burden. Brain. 2010;133(2):512–28. pmid:20080878; PubMed Central PMCID: PMC2858015.
- 62. Lehmann M, Ghosh PM, Madison C, Karydas A, Coppola G, O’Neil JP, et al. Greater medial temporal hypometabolism and lower cortical amyloid burden in ApoE4-positive AD patients. J Neurol Neurosurg Psychiatry. 2014;85(3):266–73. pmid:23965289; PubMed Central PMCID: PMC3946299.
- 63. Reiman EM, Caselli RJ, Yun LS, Chen K, Bandy D, Minoshima S, et al. Preclinical Evidence of Alzheimer’s Disease in Persons Homozygous for the ε4 Allele for Apolipoprotein E. N Engl J Med. 1996;334:752–8. pmid:8592548.
- 64. Fortea J, Pegueroles J, Alcolea D, Belbin O, Dols-Icardo O, Vaqué-Alcázar L, et al. APOE4 homozygozity represents a distinct genetic form of Alzheimer’s disease. Nat Med. 2024;30:1284–91. pmid:38710950.
- 65. Finch CE, Morgan TE. Systemic Inflammation, Infection, ApoE Alleles, and Alzheimer Disease: A Position Paper. Curr Alzheimer Res. 2007;4(5):185–9. pmid:17430245.
- 66. Duong MT, Nasrallah IM, Wolk DA, Chang CCY, Chang TY. Cholesterol, Atherosclerosis, and APOE in Vascular Contributions to Cognitive Impairment and Dementia (VCID): Potential Mechanisms and Therapy. Front Aging Neurosci. 2021;13(647990). pmid:33841127; PubMed Central PMCID: PMC8026881.
- 67. Fernández-Calle R, Konings SC, Frontiñán-Rubio J, García-Revilla J, Camprubí-Ferrer L, Svensson M, et al. APOE in the bullseye of neurodegenerative diseases: impact of the APOE genotype in Alzheimer’s disease pathology and brain diseases. Mol Neurodegeneration. 2022;17(62). pmid:36153580; PubMed Central PMCID: PMC9509584.
- 68. Farmer BC, Williams HC, Devanney NA, Piron MA, Nation GK, Carter DJ, et al. APOΕ4 lowers energy expenditure in females and impairs glucose oxidation by increasing flux through aerobic glycolysis. Mol Neurodegener. 2021;16(1):62. pmid:34488832; PubMed Central PMCID: PMC8420022.
- 69. Calderón-Garcidueñas L, Hernández-Luna J, Aiello-Mora M, Brito-Aguilar R, Evelson PA, Villarreal-Ríos R, et al. APOE Peripheral and Brain Impact: APOE4 Carriers Accelerate Their Alzheimer Continuum and Have a High Risk of Suicide in PM2.5 Polluted Cities. Biomolecules. 2023;13(6):927. pmid:37371506; PubMed Central PMCID: PMC10296707.
- 70. Sohn HY, Kim SI, Park JY, Park SH, Koh YH, Kim J, et al. ApoE4 attenuates autophagy via FoxO3a repression in the brain. Sci Rep. 2021;11:17604. pmid:34475505; PubMed Central PMCID: PMC8413297.
- 71. Chen F, Ke Q, Wei W, Cui L, Wang Y. Apolipoprotein E and viral infection: Risks and Mechanisms. Mol Ther Nucleic Acids. 2023;33:529–42. pmid:37588688; PubMed Central PMCID: PMC10425688.
- 72. Garcia AR, Finch C, Gatz M, Kraft T, Rodriguez DE, Cummings D, et al. Apoe4 is associated with elevated blood lipids and lower levels of innate immune biomarkers in a tropical amerindian subsistence population. Elife. 2021;10:e68231. pmid:34586066; PubMed Central PMCID: PMC8480980.
- 73. Raulin AC, Doss S V., Trottier ZA, Ikezu TC, Bu G, Liu CC. ApoE in Alzheimer’s disease: pathophysiology and therapeutic strategies. Mol Neurodegeneration. 2022;17(72). pmid:36348357; PubMed Central PMCID: PMC9644639.
- 74. Urosevic N, Martins RN. Infection and Alzheimer’s disease: the APOE epsilon4 connection and lipid metabolism. J Alzheimers Dis. 2008;13(4):35. pmid:18487850.
- 75. Zhao C, Strobino K, Moon YP, Cheung YK, Sacco RL, Stern Y, et al. APOE ε4 modifies the relationship between infectious burden and poor cognition. Neurol Genet. 2020;6(4): e462. pmid:32754642; PubMed Central PMCID: PMC7357411.
- 76. Mosconi L, Herholz K, Prohovnik I, Nacmias B, Cristofaro MTR De, Fayyaz M, et al. Metabolic interaction between ApoE genotype and onset age in Alzheimer’s disease: implications for brain reserve. Jour. 2005;76:15–23. pmid:15607989; PubMed Central PMCID: PMC1739315.
- 77. Khalifa KN. Effect of APOE e4 allele on age-related brain FDG-PET uptake in older individuals at risk of dementia. Alzheimer’s Dement. 2023;19:e067424.
- 78. Fernández A, Vaquero L, Bajo R, Zuluaga P, Weiner MW, Saykin AJ, et al. Apolipoprotein E ɛ4–related effects on cognition are limited to the Alzheimer’s disease spectrum. Geroscience. 2022;44:195–209. pmid:34591236; PubMed Central PMCID: PMC8811053.
- 79. Li W, Risacher SL, Huang E, Saykin AJ. Type 2 diabetes mellitus is associated with brain atrophy and hypometabolism in the ADNI cohort. Neurology. 2016;87(6):595–600. pmid:27385744; PubMed Central PMCID: PMC4977372.
- 80. Raut S, Bhalerao A, Powers M, Gonzalez M, Mancuso S, Cucullo L. Hypometabolism, Alzheimer’s Disease, and Possible Therapeutic Targets: An Overview. Cells. 2023;12(16):2019. pmid:37626828; PubMed Central PMCID: PMC10453773.
- 81. De La Monte SM, Wands JR. Alzheimer’s Disease Is Type 3 Diabetes-Evidence Reviewed. J Diabetes Sci Technol. 2008;2(6):1101–13. pmid:19885299; PubMed Central PMCID: PMC2769828.
- 82. Jayaraj RL, Azimullah S, Beiram R. Diabetes as a risk factor for Alzheimer’s disease in the Middle East and its shared pathological mediators. Saudi J Biol Sci. 2020;27(2):736–50. pmid:32210695; PubMed Central PMCID: PMC6997863.
- 83. Nebel RA, Aggarwal NT, Barnes LL, Gallagher A, Goldstein JM, Kantarci K, et al. Understanding the impact of sex and gender in Alzheimer’s disease: A call to action. Alzheimer’s and Dementia. 2018;14(9):1171–83. pmid:29907423; PubMed Central PMCID: PMC6400070.
- 84. Zhao L, Mao Z, Woody SK, Brinton RD. Sex differences in metabolic aging of the brain: insights into female susceptibility to Alzheimer’s disease. Neurobiol Aging. 2016;42:69–79. pmid:27143423; PubMed Central PMCID: PMC5644989.
- 85. Feng B, Cao J, Yu YP, Yang HY, Jiang YHY, Liu Y, et al. Gender-Related Differences in Regional Cerebral Glucose Metabolism in Normal Aging Brain. Front Aging Neurosci. 2022;14. pmid:35221996; PubMed Central PMCID: PMC8866755.
- 86. Malpetti M, Sala A, Vanoli EG, Gianolli L, Luzi L, Perani D. Unfavourable gender effect of high body mass index on brain metabolism and connectivity. Sci Rep. 2018;8:12584. pmid:30135519; PubMed Central PMCID: PMC6105632.
- 87. Miller KK, Deckersbach T, Rauch SL, Fischman AJ, Grieco KA, Herzog DB, et al. Testosterone administration attenuates regional brain hypometabolism in women with anorexia nervosa. Psychiatry Res Neuroimaging. 2004;132(3):197–207. pmid:15664791.
- 88. Muzambi R, Bhaskaran K, Brayne C, Davidson JA, Smeeth L, Warren-Gash C. Common Bacterial Infections and Risk of Dementia or Cognitive Decline: A Systematic Review. J Alzheimer’s Dis. 2020;76(4):1626. pmid:32651320; PubMed Central PMCID: PMC7504996.
- 89. Lim AC, Barnes LL, Weissberger GH, Lamar M, Nguyen AL, Fenton L, et al. Quantification of race/ethnicity representation in Alzheimer’s disease neuroimaging research in the USA: a systematic review. Commun Med. 2023;3(101). pmid:37491471; PubMed Central PMCID: PMC10368705.
- 90. Veitch DP, Weiner MW, Aisen PS, Beckett LA, DeCarli C, Green RC, et al. Using the Alzheimer’s Disease Neuroimaging Initiative to improve early detection, diagnosis, and treatment of Alzheimer’s disease. Alzheimer’s & Dementia. 2022;18(4):824–57. pmid:34581485; PubMed Central PMCID: PMC9158456.
- 91. Phillips JS, Das SR, McMillan CT, Irwin DJ, Roll EE, Da Re F, et al. Tau PET imaging predicts cognition in atypical variants of Alzheimer’s disease. Hum Brain Mapp. 2018;39(2):691–708. pmid:29105977; PubMed Central PMCID: PMC5764792.
- 92. Sun N, Mormino EC, Chen J, Sabuncu MR, Yeo BTT. Multi-modal latent factor exploration of atrophy, cognitive and tau heterogeneity in Alzheimer’s disease. Neuroimage. 2019;201(116043). pmid:31344486.