Received: 12 February 2020 Revised: 22 May 2020 Accepted: 26 May 2020 Published online: 27 July 2020 DOI: 10.1002/dad2.12052 RESEARCH ARTICLE Obesity impacts brain metabolism and structure independently of amyloid and tau pathology in healthy elderly Jordi Pegueroles1,2 Adriana Pané3 Eduard Vilaplana1,2 Víctor Montal1,2 Alexandre Bejanin1,2 Laura Videla1,2 María Carmona-Iragui1,2 Isabel Barroeta1,2 Ainitze Ibarzabal4 Anna Casajoana5 Daniel Alcolea1,2 Silvia Valldeneu1,2 Miren Altuna1,2 Ana de Hollanda3,6,7 Josep Vidal3,6,8 Emilio Ortega3,6,7 Ricardo Osorio9 Antonio Convit9 Rafael Blesa1,2 Alberto Lleó1,2 Juan Fortea1,2 Amanda Jiménez3,6,7 Alzheimer’s Disease Neuroimaging Initiative∗ 1 Memory Unit, Department of Neurology, Hospital de la Santa Creu i Sant Pau, Biomedical Research Institute Sant Pau, Universitat Autònoma de Barcelona, Barcelona, Spain 2 Centro de Investigación Biomédica en Red de Enfermedades Neurodegenerativas (CIBERNED), Madrid, Spain 3 Obesity Unit, Endocrinology and Diabetes Department, Hospital Clinic Universitari de Barcelona, Barcelona, Spain 4 Obesity Unit, Gastrointestinal Surgery Department, Hospital Clínic de Barcelona, Barcelona, Spain 5 Department of Bariatric Surgery, Bellvitge University Hospital, Barcelona, Spain 6 Institut d’Investigacions Biomèdiques August Pi Sunyer (IDIBAPS), Barcelona, Spain 7 Centro de Investigación Biomédica en Red de la Fisiopatología de la Obesidad y Nutrición (CIBEROBN), Madrid, Spain 8 Centro de Investigación Biomédica en Red de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM), Madrid, Spain 9 Brain, Obesity, and Diabetes Laboratory (BODyLab), New York University School of Medicine, New York, USA Correspondence Juan Fortea, Memory Unit, Department of Abstract Neurology, Hospital of Sant Pau, Biomedi- cal Research Institute Sant Pau, Universitat Aims/hypothesis: Midlife obesity is a risk factor for dementia. We investigated the Autònoma de Barcelona, Sant Antoni Maria impact of obesity on brain structure, metabolism, and cerebrospinal fluid (CSF) core Claret, 167, 08025, Barcelona, Spain. Email: jfortea@santpau.cat Alzheimer’s disease (AD) biomarkers in healthy elderly. Amanda Jiménez, Endocrinology and Diabetes Methods: We selected controls from ADNI2 with CSF AD biomarkers and/or fluo- Department, Hospital Clínic Universitari of Barcelona, Villarroel, 170, 08036, Barcelona, rodeoxyglucose positron emission tomography (FDG-PET) and 3T-MRI. We measured Spain. cortical thickness, FDG uptake, and CSF amyloid beta (Aβ)1-42, p-tau, and t-tau lev- Email: ajimene1@clinic.cat els. We performed regression analyses between these biomarkers and body mass index Jordi Pegueroles and Adriana Pané contributed (BMI). equally to this work. Results: We included 201 individuals (mean age 73.5 years, mean BMI 27.4 kg/m2 ). Juan Fortea and Amanda Jiménez shared the senior authorship. Higher BMI was related to less cortical thickness and higher metabolism in brain areas typically not involved in AD (family-wise error [FWE] <0.05), but not to AD CSF ∗ 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 con- tributed 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. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2020 The Authors. Alzheimer’s & Dementia: Diagnosis, Assessment & Disease Monitoring published by Wiley Periodicals, Inc. on behalf of Alzheimer’s Association Alzheimer’s Dement. 2020;12:e12052. wileyonlinelibrary.com/journal/dad2 1 of 9 https://doi.org/10.1002/dad2.12052 2 of 9 PEGUEROLES ET AL . Funding information Fondo de Investigaciones Sanitario; Insti- biomarkers. It is notable that the impact of obesity on brain metabolism and structure tuto de Salud Carlos III, Grant/Award Numbers: PI14/01126, PI17/01019, was also found in amyloid negative individuals. PI13/01532, PI16/01825, PI18/00335, Conclusions/interpretation: In the cognitively unimpaired elderly, obesity has differ- PI18/00435, PI14/1561, PI17/01896, PI17/00279, FI18/00275; Unión Europea; ential effects on brain metabolism and structure independent of an underlying AD Una manera de hacer Europa; National Insti- pathophysiology. tutes of Health, Grant/Award Numbers: 1R01AG056850 - 01A1, R21AG056974, KEYWORDS R01AG061566; Fundació La Marató de TV3, Grant/Award Numbers: 20141210, 044412; body mass index, cerebrospinal fluid Alzheimer’s disease biomarkers, fluorodeoxyglucose PET, Pla Estratègic de Recerca i Innovació en Salut, magnetic resonance imaging, obesity, pre-clinical Alzheimer’s disease Grant/Award Numbers: SLT006/17/00119, SLT006/17/00125, SLT008/18/00127 1 INTRODUCTION deleterious impact of an excess of body adiposity on brain health and to explore whether obesity contributes to neurodegeneration by Obesity has become a global pandemic with multiple adverse clinical amyloid-dependent or independent mechanisms. consequences.1 Accumulating evidence demonstrates that cognition is Taking advantage of a large multicenter cohort, the Alzheimer’s affected by an excess of body adiposity in both adults and children.2-5 Disease Neurodegenerative Initiative (ADNI), we aimed to investigate Epidemiological studies also indicate that midlife obesity increases the the relationship between BMI and brain structure, brain metabolism, risk of progression to mild cognitive impairment (MCI) and Alzheimer’s and core AD cerebrospinal fluid (CSF) biomarkers in cognitively unim- disease (AD).6,7 On the contrary, higher body mass index (BMI) in late- paired elderly. To examine whether the effects of obesity on brain life might be protective.8 This obesity paradox might be associated with structure and metabolism are independent of an AD process, we also the confounding effect of weight loss in preclinical AD.9 investigated the aforementioned relationships in amyloid negative The exact mechanisms leading to cognitive impairment and neu- individuals. rodegeneration in persons with an excess of body adiposity remain to be fully elucidated.7 Animal models suggest a significant contri- bution of obesity and obesity-related metabolic disturbances to AD 2 METHODS pathophysiology.7,10-13 In contrast, human studies assessing the impact of obesity on amyloid and tau pathology report conflicting findings 2.1 Study participants both in vivo and in post-mortem studies. Thus, higher BMI has been related to higher, but also to lower, AD burden.14-18 On the other hand, Data used in the preparation of this article were obtained from the obesity might contribute to neurodegeneration by mechanisms unre- ADNI database (adni.loni.usc.edu). The ADNI was launched in 2003 lated to AD. Obesity is a state of peripheral low-grade chronic inflam- by the National Institute on Aging (NIA), the National Institute of mation, and it is frequently associated with an abnormal peripheral Biomedical Imaging and Bioengineering (NIBIB), the U.S. Food and sensitivity to insulin effects. In experimental models, obesity-related Drug Administration (FDA), private pharmaceutical companies, and peripheral inflammation has been linked to blood-brain barrier dys- non-profit organizations, as a $60 million, 5-year public-private function, neuroinflammation, and neurodegeneration, whereas central partnership. The primary goal of ADNI has been to test whether insulin resistance has been associated with impaired synaptic plas- serial MRI, positron emission tomography (PET), other biological ticity and memory.19-21 Furthermore, obesity is a strong risk factor markers, and clinical and neuropsychological assessment can be for hypertension, type 2 diabetes, and dyslipidemia, and it is a well- combined to measure the progression of MCI and early AD. The established cerebrovascular risk factor.1 principal investigator of this initiative is Michael W. Weiner, MD, VA Brain atrophy and brain hypometabolism are well-recognized non- Medical Center and University of California, San Francisco. ADNI is specific biomarkers of neurodegeneration. Several studies have shown the result of efforts of many co-investigators from a broad range of an association between high BMI and brain atrophy.22 Fewer studies academic institutions and private corporations, and subjects have been have assessed the relationship between obesity and brain metabolism recruited from >50 sites across the United States and Canada. More in cognitively healthy subjects.23-28 In contrast to the consistent brain information can be found in the acknowledgments section (see also atrophy reported in magnetic resonance imaging (MRI) studies, these http://adni-info.org/). works have reported higher brain metabolism with respect to lean For the present study, we included all cognitively healthy controls controls.23-26,28 from ADNI2 with biometric and biochemical data and either (a) a 3T None of the above-mentioned works, however, integrated biochem- MRI and a fluorodeoxyglucose (FDG) PET scan with a time-lapse inter- ical and neuroimaging data. Multimodal studies might be useful to val between both scans of less than 1 year or (b) CSF measurements better understand the pathophysiological pathways involved in the of amyloid beta (Aβ 1-42), total tau (t-tau), and phosphorylated tau PEGUEROLES ET AL . 3 of 9 (p-tau). Supplementary Figure 1 shows the study flowchart. All human subjects included in the study followed the inclusion criteria from Research in Context ADNI that requires an informed consent. 1. Systematic review: The authors reviewed the litera- Demographic (age, sex, educational level), clinical (presence of type ture using PubMed, meeting abstracts, and presenta- 2 diabetes [T2D], usual medication), neuropsychological (Alzheimer tions. Obesity has been identified as a risk factor for Disease Assessment Scale-Cognitive score [ADAS-Cog], Clinical dementia and is related to brain atrophy in several stud- Dementia Rating Sum of Boxes [CDR-SB], and Mini-Mental State ies. However, few works have explored the relationship Examination [MMSE]), anthropometric (height, weight, and systolic between obesity and brain metabolism. These works have and diastolic blood pressure), and laboratory data (fasting plasma reported higher brain metabolism with respect to lean glucose, cholesterol, and triglycerides) were downloaded from the controls. In addition, the association between obesity ADNI database. BMI was calculated as weight in kilograms divided by and cerebrospinal fluid (CSF) Alzheimer’s disease (AD) height in meters squared. biomarkers is inconclusive. No previous multimodal stud- ies have assessed the relationship between BMI and brain metabolism, structure, and AD biomarkers in cognitively 2.2 MRI analysis healthy elderly. This relevant bibliography is appropri- ately cited. The details of MRI acquisition and pre-processing are available 2. Interpretation: Our findings demonstrate that obesity elsewhere (http://adni-info.org/). We processed the MRIs using the presents a significant but divergent effect on corti- cortical reconstruction pipeline of Freesurfer v5.1. (http://surfer.nmr. cal structure and brain metabolism independent of AD mgh.harvard.edu) as described previously.32,33 Before the statisti- pathology. cal analyses, we checked the estimated surfaces in order to detect 3. Future directions: In this article we propose a framework and correct possible segmentation errors. We applied a smooth- for conducting additional studies such as: (a) the role of ing kernel of 15 mm. To assess atrophy in AD vulnerable areas, inflammation on the brain metabolic alterations found in we extracted the mean cortical thickness values for each subject obesity, and (b) the AD-independent mechanisms leading from a well-validated region of interest, the Dickerson’s finger- to cognitive impairment and dementia in subjects with print (ie, medial temporal cortex, inferior temporal gyrus, temporal obesity. pole, angular gyrus, superior frontal gyrus, superior parietal lobule, supramarginal gyrus, precuneus, and inferior frontal sulcus of both hemispheres).34 2.4 Cerebrospinal fluid (CSF) data 2.3 18-Fluorodeoxyglucose PET scan analysis The details of the CSF analysis have been described in http://adni-info. The details of positron-emission tomography (PET) acquisition and pre- org. Briefly, pristine aliquots were examined by the validated and highly processing are available elsewhere (http://adni-info.org/). Briefly, PET automated Roche Elecsys electrochemiluminescence immunoassays. scans were co-registered and averaged. Subsequently, the image was This methodology minimizes inter-run variability for Aβ1-42, t-tau, and standardized to 1.5 mm cubic voxels and smoothed with an 8 mm full p-tau levels in CSF. The cutoff used for Aβ1-42 was 977 pg/mL.38 width at half maximum (FWHM) kernel to produce an image of a uni- form resolution. Then, each FDG-PET image was spatially normalized to each subject’s Freesurfer anatomical MRI space using a rigid-body 2.5 Statistical methods transformation and intensity-scaled by the pons-vermis region as pre- viously described.35 The resulting images were visually-inspected in Demographic, clinical, anthropometric, and cognitive variables were order to check for35 errors and projected to the middle point of the analyzed by R statistical software (version 3.4.4; http://www.r-project. cortical ribbon.36 Before performing statistical analyses, the resulting org). surfaces were smoothed using a kernel of 10 mm FWHM to obtain To assess the relationship between BMI and both the neuroimag- equivalent relative smoothing kernels for the PET and cortical thick- ing and core AD CSF biomarkers, we performed regression analyses. ness (CTh) maps. Surface-based smoothing introduces less bias than For the neuroimaging metrics, we first executed vertex-wise correla- volume-based methods and substantially improves the reliability and tion analyses between BMI and both CTh and FDG uptake in the whole the intersubject variability.37 In order to assess the brain metabolism in sample including age, sex, and triglycerides, variables significantly cor- AD vulnerable areas, we extracted the mean FDG standardized uptake related with BMI, as covariates. The figures show only those results value ratios (SUVRs) for each subject from a well-validated region of that survived the family-wise error (FWE) correction at P < 0.05 as interest, the Landau signature (ie, left and right angular, temporal, and implemented in Freesurfer. Second, we performed regression analyses posterior cingulate regions).35 between BMI and MRI and FDG-PET AD signatures, mean CTh and 4 of 9 PEGUEROLES ET AL . TA B L E 1 Demographic, clinical, neuropsychological, 3.1 Higher body mass index (BMI) is associated anthropometrical, and biochemical data for the study subsets (CSF with increased cerebral metabolism and cortical and MRI-PET samples) atrophy in areas not typically involved in Alzheimer’s MRI-PET disease (AD) CSF sample sample P N 178 168 Figure 1A and 1B show the association between BMI and FDG uptake Gender = Male (%) 89 (50.0) 85 (50.6) .998 across the cortical mantle. Higher BMI was associated with higher Age, mean (SD) 73.41 (6.25) 73.65 (6.30) .726 cerebral FDG uptake in widespread brain areas including the inferior BMI, mean (SD) 27.36 (4.02) 27.15 (3.92) .627 temporal lobe of the left hemisphere, insula of the right hemisphere and anterior cingulate, medial frontal, and orbitofrontal regions of both SBP, mean (SD) 133.77 (15.90) 134.87 (16.04) .523 hemispheres (FWE < 0.05). Figure 1C and D show the association DBP, mean (SD) 73.61 (9.23) 74.86 (10.08) .231 between BMI and CTh in the same individuals. Higher BMI was asso- Glycemia, mean (SD) 99.79 (18.19) 98.20 (17.89) .413 ciated with lower cortical thickness in areas that included superior Total cholesterol, mean 189.25 (36.69) 190.59 (36.34) .734 medial frontal regions of the right hemisphere and inferior temporal (SD) zones of the left hemisphere (FWE < 0.05). Triglycerides, mean (SD) 133.97 (72.08) 138.55 (79.70) .574 The areas of increased brain metabolism and less cortical thickness Years of Education, mean 16.57 (2.46) 16.52 (2.56) .856 showed little overlap. Of note, these regions also had very little overlap (SD) with the Landau’s and Dickerson’s signature, respectively. Moreover, APOE ε4 = Positive (%) 48 (27.0) 49 (29.2) .737 BMI was not significantly associated with CTh or with FDG in the Dick- MMSE, mean (SD) 29.00 (1.27) 29.03 (1.27) .828 erson’s and Landau’s signatures (P = 0.41 and P = 0.50, respectively) ADAS11, mean (SD) 5.87 (3.13) 5.70 (2.91) .617 (supplementary Figure 2). ADAS13, mean (SD) 9.10 (4.56) 9.00 (4.42) .834 T2D = Positive (%) 31 (17.4) 23 (13.7) .420 CSF Aβ1-42, mean (SD) 1386.95 1397.40 .887 3.2 Higher BMI is not associated with CSF AD (656.68) (661.87) biomarkers CSF t-tau, mean (SD) 235.61 (89.71) 236.15 (92.48) .958 CSF p-tau, mean (SD) 21.48 (9.10) 21.54 (9.10) .955 We found no relationship between BMI and CSF Aβ1-42 (P = 0.15), CSF p-tau (P = 0.43), or CSF total tau (P = 0.35, respectively) levels (Figure 2). Neither did we find any association between BMI and CSF biomarkers when analyzing the amyloid-positive and amyloid-negative SUVRs and the core AD CSF biomarkers including age, sex, and triglyc- groups separately (supplementary Figure 3). erides as covariates. To further assess a potential influence of preclinical AD, we per- formed regression analyses, with and without including the core AD 3.3 The impact of BMI on neuroimaging CSF biomarkers as covariates, and stratified analyses in the amyloid- biomarkers is independent of AD pathophysiology positive and amyloid-negative groups separately. To test the influence of AD biomarkers on neuroimaging outcomes, we repeated the analyses including Aβ1-42 and t-tau as covariates, 3 RESULTS and we performed stratified analyses in amyloid-positive and amyloid- negative individuals. The inclusion of CSF biomarker levels did not We included 201 subjects (50.3% male) in the study with a mean age of affect the results, which remained qualitatively the same (supplemen- 73.5 years (range: 56.2 to 89.1) and a mean BMI of 27.4 kg/m2 (range: tary Figure 4). 20.0 to 39.1). There were no differences in these variables between the The stratified analyses in the amyloid-positive and amyloid-negative subset of patients with MRI and FDG (N = 168) and the subset of sub- participants separately, despite the smaller sample size, showed that jects with CSF (N = 178). Table 1 summarizes the demographic, clin- higher BMI was related to higher FDG uptake in both groups. Brain ical, neuropsychological, anthropometrical, and biochemical data for atrophy in relation to higher BMI was significant only in the amyloid- the two study subsets. There were no differences in any of the variables negative group (Figure 3). between the subsample with CSF biomarkers and the whole sample. Of the whole sample, 29.2% of participants had abnormal levels of Aβ1-42. BMI was weakly correlated with fasting triglycerides levels (r = 0.18, 4 DISCUSSION P = 0.02). There was no correlation between BMI and education, base- line cognitive performance, systolic or diastolic blood pressure, total In this study, we found a differential influence of obesity on brain struc- cholesterol, or fasting glucose levels. ture and brain metabolism. Higher BMI was associated with increased PEGUEROLES ET AL . 5 of 9 F I G U R E 1 Cortical vertex-wise pattern of the relationship between BMI and both FDG (A) and CTh (C) adjusted by age, gender, and triglycerides. Only clusters that survived family-wise error corrected P-value < 0.05 are shown. Red-yellow scale color is used to display positive correlations in relation to FDG uptake and blue-green scale is used for negative correlations in relation with CTh. Scatter plots in B and D show the relationship between BMI and mean FDG uptake and mean CTh in the cluster ROIs in A and C brain metabolic activity, but at the same time with less cortical thick- a negative relationship between BMI and FDG uptake in prefrontal ness. The affected zones did not overlap with the typical AD vulner- areas.27,28 Of note, only 3 of the 21 participants evaluated in the lat- able areas. Furthermore, BMI was not associated with core CSF AD ter study had BMIs in the obesity range.27 All three longitudinal stud- biomarkers and the results were also found in amyloid-negative partic- ies assessed brain metabolism in middle-aged individuals with morbid ipants, suggesting that these changes are independent of an underlying obesity before and after bariatric surgery–induced weight loss.23,24,26 AD pathophysiology. Marques et al. described brain hypermetabolism in 17 women with We first analyzed the relationship between brain metabolism and severe obesity as compared with 16 lean controls, which normalized BMI. Only two cross-sectional and three longitudinal previous studies after weight loss.23 Brain metabolism normalization in this study was have assessed the relationship between brain metabolism and obesity associated with cognitive improvement.23 Tuulari et al. and Rebelos in middle-aged individuals.23,24,26-28 The results of the cross-sectional et al. did not observe differences in brain metabolism between par- studies were conflicting. Wang et al. observed higher brain metabolism ticipants with morbid obesity and controls in fasting conditions, but in the parietal cortices of 20 middle-aged participants with morbid obe- both found higher insulin-stimulated FDG uptake.24,26 The low num- sity as compared with 10 lean controls, whereas Volkow et al. found ber of controls in these two studies (n = 7 and n = 12, respectively) 6 of 9 PEGUEROLES ET AL . FIGURE 2 Scatter plots showing the lack of relationship between BMI and AD CSF biomarkers. From left to right: Aβ1-42, p-tau, and t-tau might have limited the statistical power to detect subtle differences rodegeneration throughout AD pathophysiological mechanisms and during fasting conditions. Nonetheless, in both studies, brain metabolic therefore only those with lower AD burden remain cognitively normal abnormalities normalized after bariatric surgery.24,26 Only one pre- late in life).9,29,30,43 Thus, although a contribution of obesity to AD vious study assessed the relationship between BMI and brain FDG pathophysiological processes cannot be completely ruled out, our uptake in healthy elderly. This study, also in the ADNI cohort, included results reinforce the notion that obesity affects brain metabolism and 222 participants and also showed higher brain metabolic activity in structure by mechanisms independent of AD pathophysiology. relation to higher BMI mostly in women.25 Altogether, our results and The mechanisms mediating the structural and metabolic brain the aforementioned studies suggest that obesity is associated with abnormalities in individuals with obesity are beyond the objectives increased FDG uptake in the brain. of the present work, and deserve further research.47 The finding We also evaluated the relationship between obesity and cortical of higher brain metabolism with increased BMI is relatively unex- thickness in the same sample. In accordance with previous results from pected. We hypothesize that this finding might reflect obesity-induced our group and others, we observed cortical thinning associated with neuroinflammation and astrogliosis. Of interest, in the aforemen- increasing BMI.9,22,39-41 tioned studies with subjects who underwent bariatric surgery, The regions affected by atrophy and higher metabolism in our higher FDG cerebral uptake correlated with markers of systemic study showed little overlap with the typical vulnerable AD regions. inflammation.23,24 In this sense, although brain glucose metabolism is It is important to note that we did not find any association between considered a marker of neuronal activity, FDG-PET signal has recently BMI and the CTh and brain metabolism in two of the most commonly been demonstrated to be also located in astrocytes.44 In addition, used AD signatures.34,35 Furthermore, we did not find any associa- animal studies combining FDG-PET and PET with tracers for activated tion between BMI and CSF amyloid or tau levels, and the inclusion of microglia confirmed a highly co-localized signal of increased glucose CSF biomarkers in the analyses yielded qualitatively the same results. metabolism and neuroinflammation in both wild-type aging mice and Finally, in the analyses stratified by amyloid status, amyloid-negative AD transgenic mice.45,46 It is interesting to note that in wild-type mice participants showed a pattern of changes similar to that observed in an uncoupling between glucose metabolism and neuroinflammation the whole cohort. Altogether, these results suggest that the aforemen- was observed at older ages, that is, neuroinflammation persisted tioned cortical changes are independent of an underlying AD process. but glucose metabolism returned to baseline values. This late-life Other cross-sectional studies in cognitively normal controls uncoupling has been attributed to the progression of age-dependent showed greater amyloid and tau burden associated with lower late-life neurodegeneration.46 In this same line, a triple-tracer study per- BMI.17,18,42 On the contrary, the only two previous longitudinal studies formed in AD transgenic mice showed age-dependent microglial showed greater amyloid deposition late in life in relation with mid-life activation, which positively correlates with amyloid load and brain obesity.15,16 Discrepancies between mid-life and late-life studies metabolism. Nonetheless, in this study, brain hypermetabolism was might be explained by reverse causation (ie, AD-related weight loss observed especially at younger ages, and declined in relation to in preclinical AD), selection and survival biases (ie, higher mortality increasing amyloid burden, thus suggesting that synaptic dysfunction and dementia risk in persons with obesity might determine that only might mask inflammation-related hypermetabolism.45 In the same those specially protected against obesity consequences survived line, in our study, the relationship between obesity and increased and/or maintained normal cognition late in life), or by the existence brain metabolism was attenuated in participants with preclinical AD. of additive and/or competing risk (ie, obesity not only promote neu- Further studies are required to better understand the contribution of PEGUEROLES ET AL . 7 of 9 F I G U R E 3 Cortical vertex-wise pattern of the relationship between BMI and both FDG and CTh depending on their Aβ1-42 status. Only clusters that survived family-wise error corrected P-value < 0.05 are shown. Red-yellow scale color is used to display positive correlations in relation with FDG uptake and blue-green scale is used for negative correlations in relation with CTh peripheral and central nervous system inflammation or other mecha- of insulin sensitivity on brain metabolism was not consistent among nisms to the brain metabolic changes, which are present in obesity. studies.23,24,26,48,49 Fourth, other relevant variables closely related to This study has limitations. First, as it is cross-sectional, a causal body weight, including dietary habits and physical activity, which have relationship between BMI and brain neuroimaging abnormalities been previously related to brain health, are not available in ADNI. cannot be assessed. Second, there is a significant bias in the ADNI In conclusion, obesity presents a significant and divergent effect on cohort, which excluded participants with large vascular burden and cortical structure and brain glucose metabolism in areas not typically comprises mainly Caucasian participants. This selection bias might involved in AD and independent of the AD pathophysiological process. explain the healthier than expected phenotype of cognitively healthy Further studies are needed to explore the association of these changes ADNI participants with obesity. Of note, we did not find the expected with obese-related neuroinflammation or other mechanisms. correlation between BMI and fasting plasmatic glucose and systolic or diastolic blood pressure. This bias precludes the generalization of ACKNOWLEDGMENTS our results to broader obese populations. Third, there is evidence This study was supported by the Fondo de Investigaciones Sanitario that both insulin resistance and variability in fasting glucose lev- (FIS), Instituto de Salud Carlos III (PI14/01126 and PI17/01019 to els can affect FDG uptake among cognitively normal middle-aged Juan Fortea, PI13/01532 and PI16/01825 to Rafael Blesa, PI18/00335 individuals.47,48 Given that there is sparse data available in ADNI to María Carmona-Iragui., PI18/00435 to Daniel Alcolea, PI14/1561 to better characterize the glucometabolic status in our subjects, the and PI17/01896 to Alberto Lleó, PI17/00279 to Amanda Jiménez, degree of increase in FDG uptake reported here needs to be confirmed and FI18/00275 to Víctor Montal) and the CIBERNED program (Pro- among individuals with more detailed evaluation of glucose tolerance gram 1, Alzheimer Disease to Alberto Lleó and SIGNAL study, www. status and appropriate measures of insulin sensitivity. Nonetheless, it signalstudy.es); and partly jointly funded by Fondo Europeo de Desar- should be underscored that no significant correlation between fasting rollo Regional, Unión Europea, Una manera de hacer Europa. This glucose levels and BMI was found in our cohort, and that the impact work was also supported by the National Institutes of Health (NIA 8 of 9 PEGUEROLES ET AL . grants 1R01AG056850 - 01A1; R21AG056974, and R01AG061566 to REFERENCES Juan Fortea), Fundació La Marató de TV3 (20141210 to Juan Fortea, 1. Afshin A, Forouzanfar MH, Reitsma MB, et al. Health effects of over- 044412 to Amanda Jiménez and Rafael Blesa). This work was also sup- weight and obesity in 195 countries over 25 years. N Engl J Med. 2017;377:13-27 ported by the “Pla Estratègic de Recerca i Innovació en Salut” (PERIS) 2. Cournot M, Marquie JC, Ansiau D, et al. Relation between body mass (SLT006/17/00119 to Juan Fortea, and SLT006/17/00125 to Daniel index and cognitive function in healthy middle-aged men and women. Alcolea and SLT008/18/00127 to Amanda Jiménez) and a grant from Neurology. 2006;67:1208-1214. the Fundació Bancaria La Caixa to Rafael Blesa. The work of Adriana 3. Liang J, Matheson BE, Kaye WH, Boutelle KN. Neurocognitive corre- lates of obesity and obesity-related behaviors in children and adoles- Pané is supported by the “ Ajut a la Recerca Josep Font” (Hospital Clinic cents. Int J Obes (Lond). 2014;38:494-506. de Barcelona). In addition, we would like to also thank Dr. Dickerson for 4. Yau PL, Castro MG, Tagani A, Tsui WH, Convit A. Obesity and metabolic allowing the use of the cortical fingerprint in this work. syndrome and functional and structural brain impairments in adoles- Data collection and sharing for this project was funded by the cence. Pediatrics. 2012;130:e856-e864. 5. Yau PL, Kang EH, Javier DC, Convit A. Preliminary evidence of cogni- Alzheimer’s Disease Neuroimaging Initiative (ADNI) (National Insti- tive and brain abnormalities in uncomplicated adolescent obesity. Obe- tutes of Health Grant U01 AG024904) and DOD ADNI (Department sity (Silver Spring). 2014;22:1865-1871. of Defense award number W81XWH-12-2-0012). ADNI is funded by 6. Anstey KJ, Cherbuin N, Budge M, Young J. Body mass index in midlife the National Institute on Aging, the National Institute of Biomedical and late-life as a risk factor for dementia: a meta-analysis of prospec- Imaging and Bioengineering, and through generous contributions from tive studies. Obes Rev. 2011;12:e426-e437. 7. O’Brien PD, Hinder LM, Callaghan BC, Feldman EL. Neurological con- the following: AbbVie, Alzheimer’s Association; Alzheimer’s Drug Dis- sequences of obesity. Lancet Neurol. 2017;16:465-477. covery Foundation; Araclon Biotech; BioClinica, Inc.; Biogen; Bristol- 8. Fitzpatrick AL, Kuller LH, Lopez OL, et al. Midlife and late-life obesity Myers Squibb Company; CereSpir, Inc.; Eisai Inc.; Elan Pharmaceuti- and the risk of dementia. Arch Neurol. 2009;66:336-342. cals, Inc.; Eli Lilly and Company; EuroImmun; F. Hoffmann-La Roche Ltd 9. Pegueroles J, Jimenez A, Vilaplana E, et al. Obesity and Alzheimer’s disease, does the obesity paradox really exist? A magnetic resonance and its affiliated company Genentech, Inc.; Fujirebio; GE Healthcare; imaging study. Oncotarget. 2018;9:34691-34698. IXICO Ltd.; Janssen Alzheimer Immunotherapy Research & Develop- 10. Bhat NR, Thirumangalakudi L. Increased tau phosphorylation and ment, LLC.; Johnson & Johnson Pharmaceutical Research & Develop- impaired brain insulin/IGF signaling in mice fed a high fat/high choles- ment LLC.; Lumosity; Lundbeck; Merck & Co., Inc.; Meso Scale Diagnos- terol diet. J Alzheimers Dis. 2013;36:781-789. 11. Li XH, Lv BL, Xie JZ, Liu J, Zhou XW, Wang JZ. AGEs induce Alzheimer- tics, LLC.; NeuroRx Research; Neurotrack Technologies; Novartis Phar- like tau pathology and memory deficit via RAGE-mediated GSK-3 acti- maceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier; Takeda vation. Neurobiol Aging. 2012;33:1400-1410. Pharmaceutical Company; and Transition Therapeutics. The Canadian 12. Maphis N, Xu G, Kokiko-Cochran ON, et al. Reactive microglia drive tau Institutes of Health Research is providing funds to support ADNI clini- pathology and contribute to the spreading of pathological tau in the brain. Brain. 2015;138:1738-1755. cal sites in Canada. Private sector contributions are facilitated by the 13. Walker JM, Dixit S, Saulsberry AC, May JM, Harrison FE. Reversal Foundation for the National Institutes of Health (www.fnih.org). The of high fat diet-induced obesity improves glucose tolerance, inflam- grantee organization is the Northern California Institute for Research matory response, Κ-amyloid accumulation and cognitive decline in and Education, and the study is coordinated by the Alzheimer’s Disease the APP/PSEN1 mouse model of Alzheimer’s disease. Neurobiol Dis. 2017;100:87-98. Cooperative Study at the University of California, San Diego. ADNI 14. Buchman AS, Schneider JA, Wilson RS, Bienias JL, Bennett DA. Body data are disseminated by the Laboratory for Neuro Imaging at the Uni- mass index in older persons is associated with Alzheimer disease versity of Southern California. pathology. Neurology. 2006;67:1949-1954. 15. Chuang YF, An Y, Bilgel M, et al. Midlife adiposity predicts earlier onset CONFLICTS OF INTEREST of Alzheimer’s dementia, neuropathology and presymptomatic cere- bral amyloid accumulation. Mol Psychiatry. 2016;21:910-915. All authors declare no biomedical financial interests or potential con- 16. Gottesman RF, Schneider AL, Zhou Y, et al. Association between flicts of interest related to this work. midlife vascular risk factors and estimated brain amyloid deposition. JAMA. 2017;317:1443-1450. ORCID 17. Hsu DC, Mormino EC, Schultz AP, et al. Lower late-life body-mass index is associated with higher cortical amyloid burden in clinically Jordi Pegueroles https://orcid.org/0000-0002-3554-2446 normal elderly. J Alzheimers Dis. 2016. Adriana Pané https://orcid.org/0000-0002-7810-1827 18. Vidoni ED, Townley RA, Honea RA, Burns JM. Alzheimer dis- Laura Videla https://orcid.org/0000-0002-9748-8465 ease biomarkers are associated with body mass index. Neurology. María Carmona-Iragui https://orcid.org/0000-0001-6914-2339 2011;77:1913-1920. 19. Mamo JCL, Lam V, Giles C, et al. Antihypertensive agents do not pre- Ainitze Ibarzabal https://orcid.org/0000-0002-1220-6474 vent blood-brain barrier dysfunction and cognitive deficits in dietary- Ana de Hollanda https://orcid.org/0000-0001-9114-8614 induced obese mice. Int J Obes (Lond). 2017;41:926-934. Josep Vidal https://orcid.org/0000-0002-4564-4518 20. Nakandakari SCBR, Munoz VR, Kuga GK, et al. Short-term high-fat Emilio Ortega https://orcid.org/0000-0002-2217-8905 diet modulates several inflammatory, ER stress, and apoptosis mark- ers in the hippocampus of young mice. Brain Behav Immun. 2019;79: Juan Fortea https://orcid.org/0000-0002-1340-638X 284-293. Amanda Jiménez https://orcid.org/0000-0003-4967-1174 PEGUEROLES ET AL . 9 of 9 21. Spinelli M, Fusco S, Mainardi M, et al. Brain insulin resistance impairs a study of fully automated immunoassays in BioFINDER and ADNI hippocampal synaptic plasticity and memory by increasing GluA1 cohorts. Alzheimers Dement. 2018;14:1470-1481. palmitoylation through FoxO3a. Nat Commun. 2017;8:2009. 39. Franz CE, Xian H, Lew D, et al. Body mass trajectories and cortical 22. Willette AA, Kapogiannis D. Does the brain shrink as the waist thickness in middle-aged men: a 42-year longitudinal study starting in expands?. Ageing Res Rev. 2015;20:86-97. young adulthood. Neurobiol Aging. 2019;79:11-21. 23. Marques EL, Halpern A, Correa MM, et al. Changes in neuropsycholog- 40. Shaw ME, Sachdev PS, Abhayaratna W, Anstey KJ, Cherbuin N. Body ical tests and brain metabolism after bariatric surgery. J Clin Endocrinol mass index is associated with cortical thinning with different pat- Metab. 2014;99:E2347-E2352. terns in mid- and late-life. Int J Obes (Lond). 2017; 42(3):455-461. 24. Rebelos E, Immonen H, Bucci M, et al. Brain glucose uptake is associ- doi: https://doi.org/10.1038/ijo.2017.254. ated with endogenous glucose production in obese patients before and 41. Shaw ME, Abhayaratna WP, Anstey KJ, Cherbuin N. Increasing body after bariatric surgery and predicts metabolic outcome at follow-up. mass index at midlife is associated with increased cortical thinning in Diabetes Obes Metab. 2019;21:218-226. Alzheimer’s disease-vulnerable regions. J Alzheimers Dis. 2017;59:113- 25. Sala A, Malpetti M, Ferrulli A, Gianolli L, Luzi L, Perani D. High body 120. mass index, brain metabolism and connectivity: an unfavorable effect 42. Thirunavu V, McCullough A, Su Y, et al. Higher body mass index is asso- in elderly females. Aging (Albany NY). 2019;11. ciated with lower cortical amyloid-Κ burden in cognitively normal indi- 26. Tuulari JJ, Karlsson HK, Hirvonen J, et al. Weight loss after viduals in late-life. J Alzheimers Dis. 2019;69:817-827. bariatric surgery reverses insulin-induced increases in brain glucose 43. Weuve J, Proust-Lima C, Power MC, et al. Guidelines for reporting metabolism of the morbidly obese. Diabetes. 2013;62:2747-2751. methodological challenges and evaluating potential bias in dementia 27. Volkow ND, Wang GJ, Telang F, et al. Inverse association between research. Alzheimers Dement. 2015;11:1098-1109. BMI and prefrontal metabolic activity in healthy adults. Obesity (Silver 44. Zimmer ER, Parent MJ, Souza DG, et al. [18F]FDG PET signal is driven Spring). 2009;17:60-65. by astroglial glutamate transport. Nat Neurosci. 2017;20:393-395. 28. Wang GJ, Volkow ND, Felder C, et al. Enhanced resting activity 45. Brendel M, Probst F, Jaworska A, et al. Glial activation and glucose of the oral somatosensory cortex in obese subjects. Neuroreport. metabolism in a transgenic amyloid mouse model: a triple-tracer PET 2002;13:1151-1155. study. J Nucl Med. 2016;57:954-960. 29. Alhurani RE, Vassilaki M, Aakre JA, et al. Decline in weight and incident 46. Brendel M, Focke C, Blume T, et al. Time courses of cortical glucose mild cognitive impairment. JAMA Neurol. 2016. metabolism and microglial activity across the life span of wild-type 30. Jimenez A, Pegueroles J, Carmona-Iragui M, et al. Weight loss in the mice: a PET study. J Nucl Med. 2017;58:1984-1990. healthy elderly might be a non-cognitive sign of preclinical Alzheimer’s 47. Baker LD, Cross DJ, Minoshima S, Belongia D, Watson GS, Craft S. disease. Oncotarget. 2017;8:104706-104716. Insulin resistance and Alzheimer-like reductions in regional cerebral 31. Muller S, Preische O, Sohrabi HR, et al. Decreased body mass index in glucose metabolism for cognitively normal adults with prediabetes or the preclinical stage of autosomal dominant Alzheimer’s disease. Sci early type 2 diabetes. Arch Neurol. 2011;68:51-57. Rep. 2017;7:1225. 48. Willette AA, Bendlin BB, Starks EJ, et al. Association of insulin resis- 32. Fischl B, Dale AM. Measuring the thickness of the human cerebral tance with cerebral glucose uptake in late middle-aged adults at risk cortex from magnetic resonance images. Proc Natl Acad Sci U S A. for Alzheimer disease. JAMA Neurol. 2015;72:1013-1020. 2000;97:11050-11055. 49. Willette AA, Modanlo N, Kapogiannis D. Insulin resistance predicts 33. Fortea J, Vilaplana E, Alcolea D, et al. Cerebrospinal fluid Κ-amyloid medial temporal hypermetabolism in mild cognitive impairment con- and phospho-tau biomarker interactions affecting brain structure in version to Alzheimer disease. Diabetes. 2015;64:1933-1940. preclinical Alzheimer disease. Ann Neurol. 2014;76:223-230. 34. Dickerson BC, Bakkour A, Salat DH, et al. The cortical signature of Alzheimer’s disease: regionally specific cortical thinning relates to symptom severity in very mild to mild AD dementia and is SUPPORTING INFORMATION detectable in asymptomatic amyloid-positive individuals. Cereb Cortex. Additional supporting information may be found online in the Support- 2009;19:497-510. ing Information section at the end of the article. 35. Landau SM, Harvey D, Madison CM, et al. Associations between cog- nitive, functional, and FDG-PET measures of decline in AD and MCI. Neurobiol Aging. 2011;32:1207-1218. 36. Greve DN, Svarer C, Fisher PM, et al. Cortical surface-based analysis How to cite this article: Pegueroles J, Pané A, Vilaplana E, reduces bias and variance in kinetic modeling of brain PET data. Neu- et al. Obesity impacts brain metabolism and structure roimage. 2014;92:225-236. independently of amyloid and tau pathology in healthy elderly. 37. Greve DN, Salat DH, Bowen SL, et al. Different partial volume correc- Alzheimer’s Dement. 2020;12:1–9. tion methods lead to different conclusions: an 18F-FDG-PET study of agingcorrection methods lead to different conclusions: an (18)F-FDG- https://doi.org/10.1002/dad2.12052 PET study of aging. Neuroimage. 2016;132:334-343. 38. Hansson O, Seibyl J, Stomrud E, et al. CSF biomarkers of Alzheimer’s disease concord with amyloid-Κ PET and predict clinical progression: