# paper.pdf ## Metadata - PDFFormatVersion=1.6 - Language=en - IsLinearized=true - IsAcroFormPresent=false - IsXFAPresent=false - IsCollectionPresent=false - IsSignaturesPresent=false - CreationDate=D:20260611103036+05'30' - Creator=LaTeX with hyperref package - ModDate=D:20260611103212+05'30' - Producer=Acrobat Distiller 26.0 (Windows) - Title=Association of cognitive impairment and APOE ε4 with Centiloids in Hispanic and non‐Hispanic White cohorts - Custom.WPS-ARTICLEDOI=10.1002/alz.71586 - Custom.WPS-JOURNALDOI=10.1002/(ISSN)1552-5279 - Custom.WPS-PROCLEVEL=3 - xmp:createdate=2015-08-18T08:05:36+05:30 - xmp:modifydate=2016-06-21T14:17:29-04:00 - xmp:metadatadate=2016-06-21T14:17:29-04:00 - pdf:producer=Acrobat Distiller 9.0 (Windows) - pdfx:wps-articledoi=10.1002/alz.71586 - pdfx:wps-journaldoi=10.1002/(ISSN)1552-5279 - pdfx:wps-proclevel=3 - jav:journal_article_version=VoR - prism:doi=10.1002/alz.71586 - dc:format=application/pdf - dc:title=Association of cognitive impairment and APOE ε4 with Centiloids in Hispanic and non‐Hispanic White cohorts - xmpmm:documentid=uuid:6433bf7f-a846-3136-a97c-6fa62113bc1d - xmpmm:instanceid=uuid:8be9530d-b918-30a2-b637-6063095b5a50 ## Contents ### Page 1 Received: 16 January 2026 Revised: 16 May 2026 Accepted: 18 May 2026 DOI: 10.1002/alz.71586R E S E A R C H A R T I C L E Association of cognitive impairment and APOE ε4 with Centiloids in Hispanic and non-Hispanic White cohorts Cally Xiao1 Tyler Ard1 Ganna Blazhenets2 Renaud La Joie2 Ioannis Pappas1 Arthur W. Toga1 the Alzheimer’s Disease Neuroimaging Initiative the Health and Aging Brain Study (HABS-HD) Study Team1Laboratory of Neuro Imaging, USC Stevens Neuroimaging and Informatics Institute, Keck School of Medicine of USC, University of Southern California, Los Angeles, California, USA 2Edward and Pearl Fein Memory and Aging Center, Department of Neurology, Weill Institute for Neurosciences, University of California San Francisco, San Francisco, California, USACorrespondenceArthur W. Toga, Laboratory of Neuro Imaging, USC Stevens Neuroimaging and Informatics Institute, Keck School of Medicine of USC, University of Southern California, 2025 Zonal Avenue, Los Angeles, CA 90033, USA. Email:toga@loni.usc.edu Part of the data used in preparation of this article were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. 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 HABS-HD MPIs: Sid E O’Bryant, Kristine Yaffe, Arthur Toga, Robert Rissman, & Leigh Johnson; and the HABS-HD Investigators: Meredith Braskie, Kevin King, James R Hall, Melissa Petersen, Raymond Palmer, Robert Barber, Yonggang Shi, Fan Zhang, Rajesh Nandy, Roderick McColl, David Mason, Bradley Christian, Nicole Phillips, Stephanie Large, Joe Lee, Badri Vardarajan, Monica Rivera Mindt, Amrita Cheema, Lisa Barnes, Mark Mapstone, Annie Cohen, Amy Kind, Ozioma Okonkwo, Raul Vintimilla, Zhengyang Zhou, Michael Donohue, Rema Raman, Matthew Borzage, Michelle Mielke, Beau Ances, Ganesh Babulal, Jorge Llibre-Guerra, Carl Hill and Rocky Vig.Abstract INTRODUCTION: Weak associations of apolipoprotein E (APOE) ε4 with cognitive impairment have been reported in Hispanic cohorts. To determine if APOE ε4 asso- ciates with brain amyloid load differently in Hispanic and non-Hispanic White (NHW) cohorts, we analyzed reported Centiloid values from studies that conducted amyloid positron emission tomography.METHODS: This meta-analysis included 17,017 participants (8.4% Hispanic) from five source studies. Analysis of covariance and logistic regression models adjusting for age, sex, education, and cognitive score were applied.RESULTS: Although cognitive impairment and APOE ε4 were associated with higher Centiloids in both Hispanic and NHW cohorts, cognitively impaired and APOE ε4 posi- tive Hispanic participants had lower Centiloids than NHW participants with the same conditions.DISCUSSION: APOE ε4 may have a weaker influence on brain amyloid load in His- panic cohorts. Further comparative analyses are necessary for Alzheimer’s disease anti-amyloid treatments to be widely applicable to Hispanic populations.K E Y W O R D Samyloid, APOE, biomarker, Centiloid, dementia, ethnicity, Hispanic, MCI, neuroimaging, PETHighlights∙ The Centiloid scale standardizes and allows for the combination of amyloid positron emission tomography (PET) data. 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. © 2026 The Author(s). Alzheimer’s & Dementia published by Wiley Periodicals LLC on behalf of Alzheimer’s Association.Alzheimer’s Dement. 2026;22:e71586. wileyonlinelibrary.com/journal/alz 1 of 13https://doi.org/10.1002/alz.71586 ### Page 2 2 of 13 XIAO ET AL.Funding informationAlzheimer’s Association, Grant/Award Number: SG-20-678486-GAAIN2∙ Higher Centiloids were associated with cognitive impairment and apolipoprotein E (APOE) ε4.∙ Hispanic participants had lower Centiloids than non-Hispanic White (NHW) partici- pants∙ A lower proportion of Hispanic participants had amyloid pathology (Centiloids ≥ 10).∙ APOE ε4 was associated with lower odds of amyloid pathology in the Hispanic cohort.1 BACKGROUNDAdvances in amyloid positron emission tomography (PET) have led to its widespread use in cohort studies and to support diagnoses of Alzheimer’s disease (AD). Subsequently, the Centiloid scale was devel- oped in response to the need for standardization in amyloid PET measurements across tracers, acquisition time, scanners, and imaging centers.1 The Centiloid scale is a linear transformation of standard uptake value ratios (SUVRs), which is a ratio of the global cortical sig- nal to a reference region, typically the whole cerebellum signal. The scale is anchored to 0 for high certainty amyloid negative scans and to 100 for typical scans in patients with AD dementia, and most of its values range from −25 to 150 Centiloids. It has been reported that Centiloids less than 10 exclude amyloid pathology with high certainty, Centiloids greater than 30 correlate to definite amyloid positivity, while Centiloids between 10 and 30 correspond to an intermedi- ate zone of increasing risk for amyloid pathology.2 0The Centiloid scale was first developed for Pittsburgh Compound-B (PiB), and was later adapted for the next generation of fluorine-18 (F-18) tracers such as flutafuranol (FFN),3 florbetaben (FBB),4 flutemetamol (FMM),5 and florbetapir (FBP).6 Centiloid conversion methods and calibration datasets can be found online.7 Using converted Centiloid values, amy- loid PET measurements from multiple study sources can be combined for meta-analyses. One of the major genetic AD risk factors is the apolipoprotein E (APOE) ε4 allele, and APOE ε4 positivity has been associated with increased brain amyloid load, as evidenced by neuropathology8 as well as in vivo by amyloid PET imaging.9–12 However, most prior studies were conducted in non-Hispanic White (NHW) populations, and APOEhas been reported to have varying effects on AD outcomes across racial and ethnic groups.13,14 In particular, Hispanic populations face a higher risk for dementia than NHW populations,15 but the pro- portion of APOE ε4 carriers is lower in Hispanic populations,16 and previous studies have reported weaker associations of the APOE ε4 allele with mild cognitive impairment (MCI) and AD outcomes in His- panic participants compared to NHW participants.13,17–19 In addition, a secondary analysis of a multisite cohort study of Medicare benefi- ciaries who were diagnosed with MCI or dementia reported that the proportion of amyloid positivity and the odds ratio (OR) for amyloid positivity were lower among Hispanic participants than among NHW participants.20 Even though APOE ε4 has been shown to be associated with increased amyloid load in Hispanic participants,21,22 it has been suggested that Hispanic participants with APOE ε4 may have lower amyloid load compared to NHW participants with APOE ε4.23 The aim of this secondary analysis is to compare brain amyloid load measured by amyloid PET between Hispanic and NHW cohorts using multiple study sources for increased statistical power and using reported Centiloid values for combined analyses. We also set out to determine if and how cognitive impairment and presence of the APOEε4 allele correlate with Centiloids in these two cohorts. In this era of emerging anti-amyloid therapies for AD, it is essential to understand the differences in brain amyloid load between ethnicities as modulated by genetic and cognitive statuses in order to widely apply anti-amyloid therapeutics in diverse populations.2 METHODSWe used the Global Alzheimer’s Association Interactive Network (GAAIN) Interrogator data discovery tool24 to find studies that have conducted amyloid PET imaging and reported Centiloid values. GAAIN is a federated data sharing platform for promoting and discovering cohort and clinical studies in AD, related dementias, and aging, and to date has connected over 70 studies and over 700,000 participant records.25,26 We requested full data access from five source studies (Table S1), using the reported Centiloid values as calculated by each study’s imaging team for the subsequent analyses. Centiloid values were transformed from SUVRs using equations specific to each tracer, acquisition method, and processing pipeline, allowing for the harmo- nization of amyloid PET measurements and combination of Centiloid values across studies for meta-analyses.27 Clinical diagnoses of cogni- tively normal (CN), MCI, dementia, and AD etiology if available, were determined following study-specific procedures.28 The Anti-Amyloid Treatment in Asymptomatic Alzheimer’s (A4) study,29 a multisite phase 3 clinical trial that tested the efficacy of solanezumab, recruited participants between 65 to 85 years of age from the United States, Canada, Australia, and Japan, who were CN and who showed elevated brain amyloid from PET scanning.12,30 For this secondary analysis, screening visit data of a total of 4,089 participants with Centiloid values and who self-reported to be Hispanic or NHW were used. This dataset also includes participants who did not show elevated brain amyloid from PET scanning, who were part of either the accompanying Longitudinal Evaluation of Amyloid Risks and Neurode- generation (LEARN) cohort or the screen-fail (SF) cohort.31 Amyloid ### Page 3 XIAO ET AL. 3 of 13RESEARCH IN CONTEXT1. Systematic review: The authors reviewed the literature using traditional sources such as PubMed. Previous stud- ies reported weaker associations of the apolipoprotein E (APOE) ε4 risk allele with cognitive impairment in His- panic cohorts than in non-Hispanic White (NHW) cohorts, and we investigated whether these differences trans- lated into differences in brain amyloid load reported in the standardized amyloid positron emission tomography (PET) Centiloid scale. 2. Interpretation: Our results indicate that while cognitive impairment and APOE ε4 were associated with higher Centiloids in both NHW and Hispanic cohorts, cognitively impaired and APOE ε4 positive Hispanic participants had significantly lower Centiloids than cognitively impaired and APOE ε4 positive NHW participants. 3. Future directions: Further studies that include higher proportions of Hispanic participants who have under- gone amyloid PET imaging are necessary to validate these findings. Our results suggest that the higher dementia risk that Hispanic cohorts face may not be amyloid- dependent. PET imaging for all A4, LEARN, and SF participants was performed using FBP. The Alzheimer’s Disease Neuroimaging Initiative (ADNI) database32 was launched in 2003 as a public-private partnership, led by Principal Investigator Michael W. Weiner, MD. The primary goal of ADNI has been to test whether serial magnetic resonance imaging (MRI), PET, other biological markers, and clinical and neuropsycholog- ical assessments can be combined to measure the progression of MCI and early AD,33 with participating ADNI clinical sites located across the United States and Canada. For this analysis, 1,538 participants with Centiloid values and who self-reported to be Hispanic or NHW were included. The closest clinical visit date was matched to the PET scan date for each participant to determine their cognitive status at the time of scanning. The participants in this analysis underwent amyloid PET imaging with either FBB or FBP.34,35 The Health and Aging Brain Study: Health Disparities (HABS-HD) study36 is a diverse community-dwelling cohort that recruited Mexican American, NHW, and non-Hispanic Black participants in the Dal- las/Fort Worth region of Texas, United States.37 FBB was used for all participants who underwent amyloid PET scanning, and we included records from 1,480 participants who were either NHW or Hispanic. The Imaging Dementia–Evidence for Amyloid Scanning (IDEAS) study38 is a U.S.-wide longitudinal study that assessed the impact of amyloid PET imaging on subsequent changes in clinical management.39 All participants were Medicare beneficiaries, diagnosed with MCI or dementia of uncertain etiology, and at least 65 years of age. For this analysis, 8,995 participants with Centiloid values and self-reported Hispanic or NHW ethnicity were included from IDEAS. Amyloid PET imaging in IDEAS was conducted using either FBB, FBP, or FMM. Edu- cation in the IDEAS dataset was recoded to years from categories as follows: 0 years (no formal education), 6 years (grade school), 10 years (some high school), 12 years (high school including equivalency), 14 years (some college or Associate’s degree), 16 years (Bachelor’s degree), 18 years (Master’s degree), and 20 years (Doctorate). The Standardized Centralized Alzheimer’s and Related Dementias Neuroimaging (SCAN) initiative40 was established to standardize the acquisition and processing of MRI and PET imaging data collected by the Alzheimer’s Disease Research Centers managed by the National Alzheimer’s Coordinating Center (NACC). Demographic data and the closest clinical visit date containing cognitive status from the NACC uniform data set version 3 (UDS v3)41 were matched to the amyloid PET scan date and Centiloid data in SCAN, resulting in 935 partici- pants who self-reported to be Hispanic or NHW. SCAN participants underwent amyloid PET imaging with FBB, FBP, FFN, or PiB. The combined dataset was analyzed in Python (v3.12.4) using the pandas (v2.2.2), scipy (v1.13.1), and statsmodels (v.0.14.2) packages. Analysis of covariance (ANCOVA) models adjusting for age, sex, edu- cation, and cognitive score were applied to quantify the influence of diagnosis, APOE ε4 positivity, and Hispanic ethnicity on Centiloid val- ues. Cognitive score was derived from either the Mini-Mental State Examination (MMSE), or from converting Montreal Cognitive Assess- ment (MoCA) scores to the MMSE scale,42 with 30 as the maximum cognitive normal score, and lower scores indicating worsening cogni- tive impairment. Medical history data such as diabetes, dyslipidemia, hypertension, myocardial infarction, and stroke were collected in at least three source datasets, and a vascular score was calculated by dividing the sum of the conditions by the number of conditions reported for each participant. To determine effect sizes of pairwise comparisons, Cohen’s d was calculated using the difference between estimated marginal means divided by the pooled raw standard devia- tion. Post hoc analyses to correct for multiple comparisons were applied using the Bonferroni method. ANCOVA was then repeated only in par- ticipants with amyloid pathology, defined with a cutoff of 10 Centiloids or higher, excluding definite amyloid negative participants and thus extremely negative Centiloid values that could skew means. This cutoff includes participants who were definitely amyloid positive (Centiloids≥ 30) and those with possible amyloid pathology in the intermediate zone between 10 and 30 Centiloids.2 Binary logistic regression adjust- ing for age, sex, education, and diagnosis was used to calculate ORs and 95% confidence intervals (CIs) of APOE ε4 positivity associating with an outcome of amyloid pathology (Centiloids ≥ 10).3 RESULTSThe combined dataset from five studies included 17,017 participants (53.5% female) who were either Hispanic or NHW, aged over 55, had a cognitive diagnosis status variable, and had Centiloid values derived from amyloid PET scanning (Table 1). Of the combined dataset, 1,427 ### Page 4 4 of 13 XIAO ET AL.TA B L E 1 Descriptive statistics of the combined dataset.Parameter NHW Hispanic Total p-valueNo. (#) of participants (% of total) 15590 (91.6%) 1427 (8.4%) 17017 — # Female (% of total) 8190 (52.5%) 919 (64.4%) 9109 (53.5%) < 0.001Mean age (SD) 73.9 (6.5) 70.3 (8.1) 73.6 (6.7) < 0.001Mean education years (SD) 15.6 (2.9) 12.0 (4.8) 15.3 (3.2) < 0.001Mean Centiloid (SD) 38.0 (46.2) 22.0 (35.7) 36.6 (45.6) < 0.001Adj. mean Centiloid 39.3 24.3 — < 0.001# CN diagnosis (% of total) 5607 (36.0%) 706 (49.5%) 6313 (37.1%) < 0.001# MCI diagnosis (% of total) 6504 (41.7%) 421 (29.5%) 6925 (40.7%) < 0.001# Dementia diagnosis (% of total) 3479 (22.3%) 300 (21.0%) 3779 (22.2%) 0.26 # APOE ε4+ (% with genotype) 2546 (37.4%) 205 (22.5%) 2751 (35.6%) < 0.001# CN and APOE ε4+ (% of CN with genotype) 1877 (34.2%) 137 (20.8%) 2014 (32.8%) < 0.001# MCI and APOE ε4+ (% of MCI with genotype) 418 (45.6%) 50 (24.0%) 468 (41.6%) < 0.001# Dementia and APOE ε4+ (% of dementia with genotype) 251 (62.1%) 18 (40.0%) 269 (59.9%) 0.004# Centiloid ≥ 10 (%) 9557 (61.3%) 653 (45.8%) 10210 (60.0%) < 0.001Mean Centiloid (≥ 10) (SD) 65.4 (38.1) 50.3 (34.6) 64.5 (38.1) < 0.001Adj. mean Centiloid (≥ 10) 62.8 47.1 — < 0.001Vascular score 0.24 0.33 0.25 < 0.001Note: Bold p-values indicate significant differences in proportions or means between ethnicities. Abbreviations: APOE, apolipoprotein E; CN, cognitively normal; MCI, mild cognitive impairment; NHW, non-Hispanic White; SD, standard deviation. participants (8.4%) were Hispanic. On average, Hispanic participants were about four years younger (p < 0.001) and had four years fewer education (p < 0.001) than NHW participants. A significantly higher proportion of Hispanic participants (p < 0.001) were female (64.4%) compared to NHW participants (52.5%). In addition, a significantly higher proportion of Hispanic participants were CN (p < 0.001), and a significantly lower proportion of Hispanic participants were diag- nosed with MCI (p < 0.001). The proportions of participants diagnosed with dementia were not significantly different between ethnicities (p = 0.26). Of those with APOE genotype data, a significantly lower proportion of Hispanic participants had one or two APOE ε4 alleles (p < 0.001). Both the raw mean Centiloid value and the mean Cen- tiloid value adjusted for age, sex, education, and cognitive score were significantly lower for Hispanic participants (p < 0.001). In a two-way ANCOVA model adjusting for age, sex, education, and cognitive score (Table 2), diagnosis (F(2, 16848) = 187.0, p < 0.001) was significantly and positively associated with Centiloid values, while Hispanic ethnicity (F(1, 16848) = 11.1, p < 0.001) and a diagnosis ×Hispanic ethnicity interaction (F(2, 16848) = 12.1, p < 0.001) were sig- nificantly and negatively associated with Centiloid values. With age, sex, education, and cognitive score as covariates, the adjusted mean Centiloid value for CN Hispanic participants was significantly lower (p = 0.003, d = 0.18) than the adjusted mean Centiloid value for CN NHW participants (Figure 1A). The adjusted mean Centiloid value for Hispanic participants diagnosed with MCI was significantly lower (p < 0.001, d = 0.31) than the adjusted mean Centiloid value for NHW participants diagnosed with MCI. The adjusted mean Centiloid value for Hispanic participants diagnosed with dementia was significantlyTA B L E 2 Two-way ANCOVA model correlating diagnosis and ethnicity with Centiloid values (adjusted R2 = 0.122; F(9, 16845) = 261.2, p < 0.001).Parameter Coefficient 95% CI F p-valueDiagnosis 187.0 < 0.001MCI (ref. CN) 14.4 12.8, 16.1 < 0.001Dementia (ref. CN) 20.5 18.1, 22.9 < 0.001Hispanic −6.0 −9.5, -2.5 11.1 < 0.001Diagnosis × Hispanic 12.1 < 0.001MCI (ref. CN) ×Hispanic−8.5 −13.9, -3.1 0.002Dementia (ref. CN)× Hispanic−14.6 −20.7, -8.4 < 0.001Age 0.6 0.5, 0.7 139.8 < 0.001Sex (female) 4.3 3.0, 5.7 40.3 < 0.001Education 0.8 0.5, 1.0 45.2 < 0.001Cognitive Score −1.9 −2.1, -1.7 343.1 < 0.001Note: Bold p-values indicate significant covariates. Abbreviations: ANCOVA, analysis of covariance; CI, confidence interval; CN, cognitive normal; MCI, mild cognitive impairment. lower (p < 0.001, d = 0.41) than the adjusted mean Centiloid value for NHW participants diagnosed with dementia. Significantly higher adjusted mean Centiloid values were seen in MCI versus CN (p < 0.001,d = 0.34) and dementia versus MCI (p < 0.001, d = 0.13) in NHW par- ticipants, but not in Hispanic participants (MCI versus CN, p = 0.17, ### Page 5 XIAO ET AL. 5 of 13(A) (B)F I G U R E 1 Mean Centiloid values of NHW (red) and Hispanic (blue) participants grouped by diagnosis (A), and both diagnosis and APOE ε4 positivity (B). Black lines denote the raw mean Centiloid value of each cohort. Adjusted mean Centiloid values include age, sex, education, and cognitive score as covariates. APOE, apolipoprotein E; CL, Centiloid; CN, cognitively normal; MCI, mild cognitive impairment; NHW, non-Hispanic White.d = 0.19; dementia versus MCI, p > 0.99, d < 0.01). Mean cognitive scores were lower for Hispanic participants in each diagnosis group. AD etiology is confirmed for about 10% of the MCI and demen- tia cases, following the National Institute on Aging and Alzheimer’s Association (NIA-AA) framework.28,43,44 In this smaller subset, the differences in mean Centiloid values adjusted for age, sex, educa- tion, and cognitive score were not significantly different between Hispanic and NHW participants in MCI cases due to AD (p = 0.54,d = 0.23), nor in AD cases (p > 0.99, d = 0.26; Figure S1). In addi- tion, by including a vascular score covariate in the overall ANCOVA model, Hispanic ethnicity was no longer significant on Centiloid val- ues, but the diagnosis × Hispanic ethnicity interaction term was still negatively and significantly associated with Centiloid values (Table S2).APOE genotyping was not systematically conducted in the IDEAS main study, thus subsequent analyses with APOE ε4 positivity did not include IDEAS participants, which excluded over 80% of participants diagnosed with MCI or dementia. However, repeating the previous ANCOVA in participants with APOE genotype data did not change the direction or significance of our main findings (Table S3). To further dissect the mean Centiloid differences between NHW and Hispanic participants, we conducted a three-way ANCOVA adjusting for age, sex, education, and cognitive score (Table 3). Of note, the diagnosis ×APOE ε4 positivity × Hispanic ethnicity interaction (F(2, 7565) = 3.3,p = 0.037) was significantly and negatively associated with Centiloids in MCI cases, but not in dementia cases. With age, sex, education, and cognitive score as covariates (Figure 1B), there were no significant differences in adjusted mean Centiloid values between NHW and His- panic participants without APOE ε4 in any of the diagnosis groups (CNp > 0.99, d = 0.02; MCI p = 0.80, d = 0.17; dementia p = 0.11, d = 0.36). In participants with APOE ε4, significantly lower adjusted mean Centiloid values were seen in Hispanic participants compared to NHW partic- ipants with CN (p = 0.002, d = 0.30) and MCI (p < 0.001, d = 0.71) diagnoses, but not with dementia diagnoses (p > 0.99, d = 0.37). Inter- estingly, in the Hispanic cohort, significantly higher adjusted mean Centiloid values were associated with the presence of APOE ε4 only in the CN (p < 0.001, d = 0.69) and dementia (p = 0.01, d = 0.78) diag- nosis groups, but not in those diagnosed with MCI (p = 0.17, d = 0.43). ### Page 6 6 of 13 XIAO ET AL.TA B L E 3 Three-way ANCOVA model correlating diagnosis, APOE ε4 positivity, and ethnicity with Centiloid values (adjusted R2 = 0.255; F(15, 7565) = 174.1, p < 0.001).Parameter Coefficient 95% CI F p-valueDiagnosis 43.6 < 0.001MCI (ref. CN) 6.7 3.5, 9.9 < 0.001Dementia (ref. CN) 27.6 21.5, 33.8 < 0.001APOE ε4 27.0 25.1, 28.9 783.6 < 0.001Hispanic −0.5 −3.8, 2.9 0.1 0.78 Diagnosis × APOE ε4 12.6 < 0.001MCI (ref. CN) × APOE ε4 12.1 7.3, 16.9 < 0.001Dementia (ref. CN) × APOE ε4 4.9 −2.2, 12.0 0.18 Diagnosis × Hispanic 4.3 0.01MCI (ref. CN) × Hispanic −5.8 −12.6, 1.0 0.09 Dementia (ref. CN) × Hispanic −18.9 −33.2, -4.5 00.01APOE ε4 × Hispanic −11.3 −17.9, -4.6 11.1 < 0.001Diagnosis × APOE ε4 × Hispanic 3.3 0.04MCI (ref. CN) × APOE ε4 × Hispanic −13.5 −26.9, -0.2 0.047Dementia (ref. CN) × APOE ε4 × Hispanic 15.0 −7.3, 37.3 0.18 Age 1.0 0.9, 1.1 261.3 < 0.001Sex (Female) 4.0 2.4, 5.6 25.0 < 0.001Education 0.1 −0.1, 0.4 1.1 0.29 Cognitive Score −1.7 −2.1, -1.3 62.3 < 0.001Note: Bold p-values indicate significant covariates. Abbreviations: ANCOVA, analysis of covariance; APOE, apolipoprotein E; CI, confidence interval; CN, cognitive normal; MCI, mild cognitive impairment. In addition, in Hispanic participants without APOE ε4, adjusted mean Centiloid values were not significantly higher with more impaired diag- noses (MCI vs. CN, p > 0.99, d = 0.04; dementia vs. MCI, p > 0.99,d = 0.23), and in Hispanic participants with APOE ε4, the adjusted mean Centiloid value with dementia diagnoses was significantly higher (p = 0.03, d = 0.74) than the adjusted mean Centiloid value with MCI diagnoses, but adjusted mean Centiloid values between MCI and CN diagnoses were not significantly different (p > 0.99, d = 0.02). Hispanic participants had lower mean cognitive scores than NHW participants in each comparison. Raw mean Centiloid values, adjusted mean Cen- tiloid values, mean cognitive scores, and cohort sizes are noted in Figure 1. The differences in Centiloid values were not driven by any particu- lar source study, as repeating these analyses by leaving out one source study at a time yielded similar results and significance levels (Tables S4-S5). To decipher if the lower Centiloid values seen in Hispanic par- ticipants were due to a lower proportion of participants with amyloid pathology, or due to lower amyloid burden even in those with amyloid pathology, we set a cutoff of 10 or higher Centiloids to exclude definite amyloid negative participants. A significantly lower (p < 0.001) pro- portion of Hispanic participants had amyloid pathology (Centiloids ≥10), and in those with amyloid pathology, both the raw mean Centiloid value and the mean Centiloid value adjusted for age, sex, education, and cognitive score were significantly lower (p < 0.001) in Hispanic participants (Table 1).TA B L E 4 Two-way ANCOVA model correlating diagnosis and ethnicity with Centiloid values ≥ 10 in participants with amyloid pathology (adjusted R2 = 0.138; F(9, 10125) = 181.0, p < 0.001).Parameter Coefficient 95% CI F p-valueDiagnosis 249.9 < 0.001MCI (ref. CN) 19.6 17.8, 21.4 < 0.001Dementia (ref. CN) 23.7 21.2, 26.2 < 0.001Hispanic −11.3 −15.9, -6.8 23.9 < 0.001Diagnosis × Hispanic 0.3 0.71 MCI (ref. CN) ×Hispanic 1.7 −5.1, 8.4 0.63 Dementia (ref. CN)× Hispanic−1.4 −8.2, 5.5 0.70 Age 0.5 0.4, 0.6 80.1 < 0.001Sex (female) 0.3 −1.1, 1.8 0.2 0.64 Education 0.6 0.3, 0.8 19.9 < 0.001Cognitive score −0.9 −1.1, -0.7 90.7 < 0.001Abbreviations: ANCOVA, analysis of covariance; CI, confidence interval; CN, cognitive normal; MCI, mild cognitive impairment. In a two-way ANCOVA model adjusting for age, sex, education, and cognitive score in participants with Centiloid values ≥ 10 (Table 4), diagnosis (F(2, 10125) = 249.9, p < 0.001) was significantly and pos- itively associated with Centiloid values, while Hispanic ethnicity (F(1, ### Page 7 XIAO ET AL. 7 of 13(A) (B)F I G U R E 2 Mean Centiloid values (≥ 10) of NHW (red) and Hispanic (blue) participants grouped by diagnosis (A) and both diagnosis and APOEε4 positivity (B). Black lines denote the raw mean Centiloid value of each cohort. Adjusted mean Centiloid values include age, sex, education, and cognitive score as covariates. APOE, apolipoprotein E; CL, Centiloid; CN, cognitive normal; MCI, mild cognitive impairment; NHW, non-Hispanic White. 10125) = 23.9, p < 0.001) was significantly and negatively associ- ated with Centiloid values. However, the diagnosis × Hispanic ethnicity interaction (F(2, 10125) = 0.3, p = 0.71) was no longer significant. The adjusted mean Centiloid values for Hispanic participants with amyloid pathology were significantly lower than the adjusted mean Centiloid values for NHW participants with amyloid pathology in all diagnosis groups (CN, p < 0.001, d = 0.34; MCI, p = 0.002 d = 0.26; dementia,p < 0.001, d = 0.35; Figure 2A). Significantly higher adjusted mean Cen- tiloid values were associated with more impaired diagnoses in NHW cohorts with amyloid pathology (MCI vs. CN, p < 0.001, d = 0.55; dementia vs. MCI, p < 0.001, d = 0.11), and between MCI and CN Hispanic cohorts with amyloid pathology (p < 0.001, d = 0.69). The adjusted mean Centiloid value for Hispanic participants with amyloid pathology and diagnosed with dementia was not significantly different (p > 0.99, d = 0.03) from the adjusted mean Centiloid value for Hispanic participants with amyloid pathology and diagnosed with MCI. In a three-way ANCOVA model adjusting for age, sex, education, and cognitive score in participants with genotype data and with Cen- tiloid values ≥ 10 (Table 5), diagnosis (F(2, 3915) = 61.8, p < 0.001) and APOE ε4 positivity (F(1, 3915) = 271.3, p < 0.001) were signifi- cantly and positively associated with Centiloid values, while Hispanic ethnicity (F(1, 3915) = 4.5, p = 0.03) was significantly and negatively associated with Centiloid values, but the diagnosis × APOE ε4 positiv- ity × Hispanic ethnicity interaction (F(2, 3915) = 2.0, p = 0.13) was not significant. Further dividing the participants with amyloid pathol- ogy by diagnosis and APOE ε4 positivity (Figure 2B), we found that the adjusted mean Centiloid value for MCI Hispanic participants with amy- loid pathology and APOE ε4 positivity was significantly lower (p = 0.009,d = 0.65) than the adjusted mean Centiloid value for MCI NHW partici- pants with amyloid pathology and APOE ε4 positivity (Figure 2B). APOEε4 positivity was significantly associated with higher adjusted mean Centiloid values in CN NHW participants (p < 0.001, d = 0.63), CN His- panic participants (p = 0.003, d = 0.66), and MCI NHW participants (p < 0.001, d = 0.57), but not for MCI Hispanic participants (p > 0.99,d = 0.16), dementia NHW participants (p > 0.99, d = 0.20), or dementia Hispanic participants (p > 0.99, d = 0.65). In addition, Hispanic par- ### Page 8 8 of 13 XIAO ET AL.TA B L E 5 Three-way ANCOVA model correlating diagnosis, APOE ε4 positivity, and ethnicity with Centiloid values ≥ 10 in participants with amyloid pathology (adjusted R2 = 0.238; F(15, 3915) = 83.0, p < 0.001).Parameter Coefficient 95% CI F p-valueDiagnosis 61.8 < 0.001MCI (ref. CN) 15.3 10.5, 20.1 < 0.001Dementia (ref. CN) 38.8 31.1, 46.3 < 0.001APOE ε4 21.0 18.5, 23.5 271.3 < 0.001Hispanic −5.8 −11.2, -0.5 4.5 0.03Diagnosis × APOE ε4 5.6 00.004MCI (ref. CN) × APOE ε4 0.1 −6.1, 6.3 0.98 Dementia (ref. CN) × APOE ε4 −13.8 −21.9, -5.6 < 0.001Diagnosis × Hispanic 0.7 00.48 MCI (ref. CN) × Hispanic −2.7 −13.9, 8.5 0.63 Dementia (ref. CN) × Hispanic −11.2 −30.3, 7.8 00.25APOE ε4 × Hispanic −4.0 −13.2, 5.2 0.7 0.40 Diagnosis × APOE ε4 × Hispanic 2.0 0.13 MCI (ref. CN) × APOE ε4 × Hispanic −11.0 −29.8, 7.8 0.25 Dementia (ref. CN) × APOE ε4 × Hispanic 20.1 −7.0, 47.2 0.15 Age 1.2 1.0, 1.4 192.4 < 0.001Sex (female) 2.0 −0.1, 4.2 3.4 0.07 Education 0.2 −0.2, 0.6 1.1 0.29 Cognitive score −1.1 −1.6, -0.6 17.0 < 0.001Abbreviations: APOE, apolipoprotein E; ANCOVA, analysis of covariance; CI, confidence interval; CN, cognitive normal; MCI, mild cognitive impairment. ticipants diagnosed with dementia and with amyloid pathology andAPOE ε4 positivity did not have significantly higher (p = 0.06, d = 0.91) adjusted mean Centiloid values than Hispanic participants diagnosed with MCI and with amyloid pathology and APOE ε4 positivity, and His- panic participants diagnosed with MCI and with amyloid pathology andAPOE ε4 positivity did not have significantly higher (p > 0.99, d = 0.05) adjusted mean Centiloid values than CN Hispanic participants with amyloid pathology and APOE ε4 positivity. To verify that these results were not dependent on a specific thresh- old, we also set a stricter Centiloid values ≥ 30 threshold to include only definite amyloid positive participants.2 In those who were amyloid positive, we saw that the adjusted mean Centiloid value for His- panic participants diagnosed with dementia was significantly lower (p = 0.002, d = 0.29) than the adjusted mean Centiloid value for NHW participants diagnosed with dementia (Figure S2A). APOE ε4 and wors- ening diagnoses were associated with higher Centiloid values in NHW participants only (Figure S2B). Finally, using a logistic regression model with Centiloids ≥ 10 as the binary outcome for amyloid pathology, and adjusting for age, sex, education, and diagnosis (Table 6), the OR of NHW participants hav- ing amyloid pathology with APOE ε4 positivity was 4.27 (95% CI: 3.82, 4.77; p < 0.001), which was significantly higher (p = 0.004) than the OR of Hispanic participants having amyloid pathology with APOE ε4 posi- tivity, which was 2.54 (95% CI: 1.82, 3.54; p < 0.001). In addition, the ORs with age and MCI diagnosis associating with Centiloids ≥ 10 were significantly different between NHW and Hispanic models.4 DISCUSSIONThe establishment of the Centiloid scale allowed for the combina- tion and harmonization of amyloid PET data across multiple sites that used different tracers, acquisition times, and scanners. This secondary analysis of 17,017 participants spanned mostly across sites in the United States and Canada, combining data from five large neuroimag- ing cohorts. As Hispanic participants were only 8.4% of the combined dataset and tended to be younger, had fewer years of education, and included more females than NHW participants, we leveraged ANCOVA to compare Centiloid means adjusting for age, sex, education, and cognitive score to help mitigate any group differences. We saw that higher adjusted mean Centiloid values were associ- ated with more impaired cognitive diagnoses and APOE ε4 positivity when analyzing each ethnicity individually, which was in line with previ- ous reports of increased amyloid load being associated with cognitive impairment and APOE ε4 positivity.9–12,21,22 However, when compar- ing amyloid load between ethnicities, Hispanic participants had lower adjusted mean Centiloid values than NHW participants in each diag- nosis group (Figure 1A) and with APOE ε4 positivity by diagnosis group (Figure 1B), which was in accordance with a previous study that reported lower amyloid PET SUVRs in Hispanic participants with APOEε4 than in NHW participants with APOE ε4.23 Interestingly, in Hispanic participants diagnosed with MCI, APOE ε4 positivity was not signifi- cantly associated with higher adjusted mean Centiloid values. These results suggest that cognitive impairment in Hispanic participants may ### Page 9 XIAO ET AL. 9 of 13TA B L E 6 Logistic regression models with Centiloids ≥ 10 as the binary outcome variable.NHW Hispanic Diff. Parameter OR 95% CI p-value OR 95% CI p-value p valueAPOEε4 (ref. no APOEε4) 4.27 3.82, 4.77 < 0.001 2.54 1.82, 3.54 < 0.001 0.004Age 1.05 1.04, 1.06 < 0.001 1.08 1.05, 1.10 < 0.001 0.03Female (ref. male) 1.19 1.07, 1.32 0.002 1.51 1.11, 2.04 0.008 0.15 Education 1.00 0.98, 1.02 0.79 0.99 0.96, 1.02 0.62 0.59 MCI (ref. CN) 1.37 1.18, 1.60 < 0.001 0.82 0.58, 1.15 0.25 0.007Dementia (ref. CN) 3.62 2.74, 4.78 < 0.001 2.10 1.05, 4.20 0.04 0.15Note: Bold p-values indicate significant differences in ORs between ethnicities. Abbreviations: APOE, apolipoprotein E; CI, confidence interval; CN, cognitive normal; MCI, mild cognitive impairment; NHW, non-Hispanic White; OR, odds ratio. be less dependent on amyloid accumulation, and APOE ε4 may have a weaker influence on amyloid load in this population. In cases where AD etiology was confirmed according to the NIA- AA framework, there were no adjusted mean Centiloid differences between Hispanic and NHW participants in MCI due to AD cases or AD cases (Figure S1). However, cohort sizes were much smaller for Hispanic participants. In addition, including a vascular score covariate that was a composite of five comorbidities, which were more prevalent among Hispanic participants, to the ANCOVA model removed the sig- nificance of the Hispanic ethnicity effect on Centiloid values, but not the diagnosis × Hispanic ethnicity interaction term (Table S2). Taken together, these results further suggest that cognitive impairment in Hispanic participants may not be attributed to pure AD pathology. However, Centiloid differences between Hispanic and NHW partici- pants still persisted even after adjusting for vascular risk, indicating that vascular comorbidities do not fully explain the differences. Fur- ther studies that include larger cohorts of participants with confirmed AD etiology and that systematically include additional cardiovascu- lar risk factors are necessary to decipher the mechanism of Centiloid differences. A previous report of the A4 screening data did not find differ- ences in the proportion of elevated amyloid between Hispanic and non-Hispanic participants, which was determined by a combination of SUVR cutoffs and consensus visual reads, although all participants had normal cognition.12 A previous secondary analysis of the IDEAS study, which only included participants diagnosed with MCI and dementia, reported that a lower proportion of Hispanic participants were amy- loid PET positive than NHW participants.20 In our combined analysis using the Centiloid scale, the proportion of Hispanic participants with Centiloid values of 10 or higher was significantly lower than that of NHW participants. The Centiloid scale is anchored at 0 for those who are young and cognitively normal, and although a precise Centiloid cut- off to determine amyloid positivity may range between 10 and 30,1,2 setting a Centiloid cutoff of 10 or higher in this analysis only excluded participants who were definite amyloid negative. In participants who had amyloid pathology with Centiloid values 10 or higher (Figure 2), we found that the adjusted mean Centiloid differences between NHW and Hispanic participants with APOE ε4 and diagnosed with MCI remained significant, and Hispanic participants with APOE ε4 and diagnosed with MCI did not have significantly higher adjusted mean Centiloid values than CN Hispanic participants with APOE ε4. These results suggest thatAPOE ε4 did not further increase amyloid load in Hispanic participants with amyloid pathology. The mean Centiloid differences between His- panic and NHW participants may be due to both a lower proportion of Hispanic participants with amyloid pathology and lower amyloid bur- den in Hispanic participants with amyloid pathology. Finally, the OR ofAPOE ε4 positivity associated with amyloid pathology (Centiloids ≥ 10) was significantly lower in Hispanic participants than in NHW partici- pants, although both ORs were significantly associated with amyloid pathology. This result builds on previous studies that reported weaker associations of APOE ε4 with MCI and AD outcomes in Hispanic par- ticipants compared to NHW participants,13,14,17–19 by showing weaker associations of APOE ε4 with Centiloids and amyloid pathology in Hispanic participants. It should be noted that the source studies used in this secondary analysis reflect the current landscape of available amyloid PET data in diverse populations, although these source studies had different aims and enrollment targets. For example, the IDEAS study enrolled only participants diagnosed with MCI or dementia of uncertain etiology, and A4 enrolled only CN participants who showed signs of elevated amyloid, although our analysis also included non-elevated amyloid par- ticipants from its accompanying LEARN and SF cohorts. ADNI and NACC, the parent study of SCAN, used clinic-based recruitment strate- gies and may have an overrepresented sample of cognitively impaired and APOE ε4 positive participants compared to the population level. HABS-HD is a community-based cohort with balanced proportions of NHW and Hispanic participants and is representative of its location in Texas; however, the proportion of cognitively impaired participants was lower in this study. It would be of interest to compare the correlations of Centiloids with fluid AD biomarkers between Hispanic and NHW participants to further decipher amyloid and AD pathology. A study from the Czech Republic reported high concordance of the phosphorylated (p-)tau181/Aβ42 ratio from cerebrospinal fluid (CSF) with Centiloid values.45 A HABS-HD analysis reported that the Aβ42/40 ratio from plasma was associated with amyloid PET in NHW participants but not ### Page 10 10 of 13 XIAO ET AL.in Hispanic participants.46 Other studies from a smaller but balanced cohort of Hispanic and NHW participants found that Centiloid values positively correlated with plasma p-tau181 and p-tau217 levels in both ethnicities,47,48 but that Hispanic participants with amnestic MCI had significantly lower odds of having pathological plasma p-tau181 levels than NHW participants with amnestic MCI.47 To determine if differences in amyloid load by ethnicity persist over time and result in divergent cognitive decline trajectories, it would be essential to analyze longitudinal data of repeated PET measure- ments and cognitive diagnoses. Even though some source datasets in this secondary analysis were longitudinal studies, there were not many additional visits following amyloid PET measurements, so a cross- sectional design was prioritized for this analysis. In addition, due to the heterogeneity of the Hispanic diaspora, it would be essential to analyze amyloid burden in Hispanic participants by Hispanic origin. Indeed, a previous study reported that the association of APOE ε4 with cognitive decline differed by Hispanic sub-group.49 Other fac- tors that may influence cognitive diagnosis and reserve and which may differ between Hispanic and NHW populations include educa- tion quality,50,51 bilingualism,52 and socioeconomic status and access to healthcare,53 would be essential to include in future studies. Finally, even though we adjusted for age, sex, education, and cognitive score in the ANCOVA models, future studies that include higher proportions of Hispanic participants and are better matched with NHW participants are necessary to confirm these findings. In summary, our results indicate that cognitively impaired andAPOE ε4 positive Hispanic participants had lower amyloid load than cognitively impaired and APOE ε4 positive NHW participants. Taken together with previous studies on amyloid PET, fluid biomarkers, and APOE status, these results suggest that dementia in Hispanic cohorts may be due to non-amyloid or mixed pathologies and may also be influenced by comorbid conditions and other sociodemo- graphic and systemic factors. Further primary and secondary analy- ses that include higher proportions of underrepresented populations are necessary to decipher how these factors may influence anti- amyloid and AD and related dementias treatment strategies in diverse populations.AUTHOR CONTRIBUTIONS Cally Xiao: Conceptualization; formal analysis; methodology; visualiza- tion; writing–original draft. Tyler Ard: Formal analysis; Methodology, writing–review and editing. Ganna Blazhenets: Methodology; writing– review and editing. Renaud La Joie: Methodology; writing–review and editing. Ioannis Pappas: Conceptualization; formal analysis; writing– review and editing. Arthur W. Toga: Conceptualization; supervision; funding acquisition; writing–review and editing.ACKNOWLEDGMENTSData used in preparation of this manuscript were obtained from the Global Alzheimer’s Association Interactive Network (GAAIN), funded by the Alzheimer’s Association (grant number SG-20-678486- GAAIN2) at the Mark and Mary Stevens Neuroimaging and Informatics Institute at Keck School of Medicine of USC. The primary goal of GAAIN is to advance research into the causes, prevention and treat- ment of Alzheimer’s and other neurodegenerative diseases. The A4 Study was a secondary prevention trial in preclinical Alzheimer’s disease, aiming to slow cognitive decline associated with brain amyloid accumulation in clinically normal older individuals. The A4 Study was funded by a public-private-philanthropic partnership, including funding from the National Institutes of Health-National Insti- tute on Aging, Eli Lilly and Company, Alzheimer’s Association, Accel- erating Medicines Partnership, GHR Foundation, an anonymous foun- dation, and additional private donors, with in-kind support from Avid Radiopharmaceuticals, Cogstate, Albert Einstein College of Medicine and the Foundation for Neurologic Diseases. The companion observa- tional Longitudinal Evaluation of Amyloid Risk and Neurodegeneration (LEARN) Study was funded by the Alzheimer’s Association and GHR Foundation. The A4 and LEARN Studies were led by Dr. Reisa Sper- ling at Brigham and Women’s Hospital, Harvard Medical School, and Dr. Paul Aisen at the Alzheimer’s Therapeutic Research Institute (ATRI) at the University of Southern California. The A4 and LEARN Studies were coordinated by ATRI at the University of Southern California, and the data are made available under the auspices of Alzheimer’s Clinical Trial Consortium through the Global Research & Imaging Platform (GRIP). The complete A4 Study Team list is available on: https://www.actcinfo. org/a4-study-team-lists/. We would like to acknowledge the dedica- tion of the study participants and their study partners who made the A4 and LEARN Studies possible. Data collection and sharing for this project were funded by the Alzheimer’s Disease Neuroimaging Initiative (ADNI) (National Insti- tutes 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 Biomed- ical Imaging and Bioengineering, and through generous contribu- tions from the following: AbbVie, Alzheimer’s Association; Alzheimer’s Drug Discovery Foundation; Araclon Biotech; BioClinica, Inc.; Biogen; Bristol-Myers Squibb Company; CereSpir, Inc.; Cogstate; 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 Therapeu- tics. 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 Therapeutic Research Institute at the University of Southern California. ADNI data are disseminated by the Laboratory for Neuro Imaging at the University of Southern California. Research using HABS-HD data reported in this publication was sup- ported by the National Institute on Aging of the National Institutes of Health under Award Numbers R01AG054073 and R01AG058533, ### Page 11 XIAO ET AL. 11 of 13R01AG070862, P41EB015922 and U19AG078109. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The Imaging Dementia–Evidence for Amyloid Scanning (IDEAS) study (NCT #02420756) evaluated the clinical utility of amyloid PET in Medicare beneficiaries with MCI or dementia meeting Appropri- ate Use Criteria for amyloid PET. The IDEAS study was funded by the Alzheimer’s Association, the American College of Radiology, Avid Radiopharmaceuticals Inc (a wholly owned subsidiary of Eli Lilly and Company), General Electric Healthcare, and Life Molecular Imaging (formerly Piramal Imaging). The Centers for Medicare & Medicaid Ser- vices provided coverage for amyloid PET scans under coverage with evidence development. The NACC database is funded by NIA/NIH Grant U24 AG072122. SCAN is a multi-institutional project that was funded as a U24 grant (AG067418) by the National Institute on Aging in May 2020. Data collected by SCAN and shared by NACC are contributed by the NIA-funded ADRCs as follows: Arizona Alzheimer’s Center—P30 AG072980 (PI: Eric Reiman, MD); R01 AG069453 (PI: Eric Reiman (contact), MD); P30 AG019610 (PI: Eric Reiman, MD); and the State of Arizona which provided additional funding supporting our center; Boston University—P30 AG013846 (PI Neil Kowall MD); Cleveland ADRC—P30 AG062428 (James Leverenz, MD); Cleveland Clinic, Las Vegas—P20AG068053; Columbia—P50 AG008702 (PI Scott Small MD); Duke/UNC ADRC— P30 AG072958; Emory University—P30AG066511 (PI Levey Allan, MD, PhD); Indiana University—R01 AG19771 (PI Andrew Saykin, PsyD); P30 AG10133 (PI Andrew Saykin, PsyD); P30 AG072976 (PI Andrew Saykin, PsyD); R01 AG061788 (PI Shannon Risacher, PhD); R01 AG053993 (PI Yu-Chien Wu, MD, PhD); U01 AG057195 (PI Liana Apostolova, MD); U19 AG063911 (PI Bradley Boeve, MD); and the Indiana University Department of Radiology and Imaging Sci- ences; Johns Hopkins—P30 AG066507 (PI Marilyn Albert, Phd.); Mayo Clinic—P50 AG016574 (PI Ronald Petersen MD PhD); Mount Sinai— P30 AG066514 (PI Mary Sano, PhD); R01 AG054110 (PI Trey Hedden, PhD); R01 AG053509 (PI Trey Hedden, PhD); New York University— P30AG066512-01S2 (PI Thomas Wisniewski, MD); R01AG056031 (PI Ricardo Osorio, MD); R01AG056531 (PIs Ricardo Osorio, MD; Girardin Jean-Louis, PhD); Northwestern University—P30 AG013854 (PI Robert Vassar PhD); R01 AG045571 (PI Emily Rogalski, PhD); R56 AG045571, (PI Emily Rogalski, PhD); R01 AG067781, (PI Emily Rogal- ski, PhD); U19 AG073153, (PI Emily Rogalski, PhD); R01 DC008552, (M.-Marsel Mesulam, MD); R01 AG077444, (PIs M.-Marsel Mesulam, MD, Emily Rogalski, PhD); R01 NS075075 (PI Emily Rogalski, PhD); R01 AG056258 (PI Emily Rogalski, PhD); Oregon Health and Science University—P30 AG008017 (PI Jeffrey Kaye MD); R56 AG074321 (PI Jeffrey Kaye, MD); Rush University—P30 AG010161 (PI David Bennett MD); Stanford—P30AG066515; P50 AG047366 (PI Victor Hender- son MD MS); University of Alabama, Birmingham—P20; University of California, Davis—P30 AG10129 (PI Charles DeCarli, MD); P30 AG072972 (PI Charles DeCarli, MD); University of California, Irvine— P50 AG016573 (PI Frank LaFerla PhD); University of California, San Diego—P30AG062429 (PI James Brewer, MD, PhD); University of California, San Francisco—P30 AG062422 (Rabinovici, Gil D., MD); University of Kansas—P30 AG035982 (Russell Swerdlow, MD); Uni- versity of Kentucky—P30 AG028283-15S1 (PIs Linda Van Eldik, PhD and Brian Gold, PhD); University of Michigan ADRC—P30AG053760 (PI Henry Paulson, MD, PhD) P30AG072931 (PI Henry Paulson, MD, PhD) Cure Alzheimer’s Fund 200775 - (PI Henry Paulson, MD, PhD) U19 NS120384 (PI Charles DeCarli, MD, University of Michigan Site PI Henry Paulson, MD, PhD) R01 AG068338 (MPI Bruno Giordani, PhD, Carol Persad, PhD, Yi Murphey, PhD) S10OD026738-01 (PI Douglas Noll, PhD) R01 AG058724 (PI Benjamin Hampstead, PhD) R35 AG072262 (PI Benjamin Hampstead, PhD) W81XWH2110743 (PI Benjamin Hampstead, PhD) R01 AG073235 (PI Nancy Chiar- avalloti, University of Michigan Site PI Benjamin Hampstead, PhD) 1I01RX001534 (PI Benjamin Hampstead, PhD) IRX001381 (PI Ben- jamin Hampstead, PhD); University of New Mexico—P20 AG068077 (Gary Rosenberg, MD); University of Pennsylvania—State of PA project 2019NF4100087335 (PI David Wolk, MD); Rooney Family Research Fund (PI David Wolk, MD); R01 AG055005 (PI David Wolk, MD); University of Pittsburgh—P50 AG005133 (PI Oscar Lopez MD); Uni- versity of Southern California—P50 AG005142 (PI Helena Chui MD); University of Washington—P50 AG005136 (PI Thomas Grabowski MD); University of Wisconsin—P50 AG033514 (PI Sanjay Asthana MD FRCP); Vanderbilt University—P20 AG068082; Wake Forest— P30AG072947 (PI Suzanne Craft, PhD); Washington University, St. Louis—P01 AG03991 (PI John Morris MD); P01 AG026276 (PI John Morris MD); P20 MH071616 (PI Dan Marcus); P30 AG066444 (PI John Morris MD); P30 NS098577 (PI Dan Marcus); R01 AG021910 (PI Randy Buckner); R01 AG043434 (PI Catherine Roe); R01 EB009352 (PI Dan Marcus); UL1 TR000448 (PI Brad Evanoff); U24 RR021382 (PI Bruce Rosen); Avid Radiopharmaceuticals / Eli Lilly; Yale—P50 AG047270 (PI Stephen Strittmatter MD PhD); R01AG052560 (MPI: Christopher van Dyck, MD; Richard Carson, PhD); R01AG062276 (PI: Christopher van Dyck, MD); 1Florida—P30AG066506-03 (PI Glenn Smith, PhD); P50 AG047266 (PI Todd Golde MD PhD). This work was supported by the Global Alzheimer’s Association Interactive Network initiative of the Alzheimer’s Association (SG-20-678486-GAAIN2). The funding source had no role in the design and conduct of the study; collection, management, analysis, or interpretation of the data; prepa- ration, review, or approval of the manuscript; and decision to submit the manuscript for publication.CONFLICT OF INTEREST STATEMENTR.L.J. has served as a paid consultant to GE Healthcare. The other authors declare no conflicts of interest. Author disclosures are avail- able in the Supporting Information.CONSENT STATEMENTAll primary studies were conducted in accordance with the Decla- ration of Helsinki, the International Conference on Harmonization Good Clinical Practice guideline, and local regulatory requirements. All participants provided informed written consent prior to joining the pri- mary studies and no further consent was required for this secondary analysis using anonymized data. ### Page 12 12 of 13 XIAO ET AL.ORCIDCally Xiao https://orcid.org/0000-0001-7172-3175REFERENCES1. Klunk WE, Koeppe RA, Price JC, et al. The Centiloid Project: stan- dardizing quantitative amyloid plaque estimation by PET. Alzheimers Dement. 2015;11:1-15. doi:10.1016/j.jalz.2014.07.003 2. Collij LE, Bollack A, La Joie R, et al. Centiloid recommendations for clin- ical context-of-use from the AMYPAD consortium. Alzheimers Dement. 2024;20:9037-9048. doi:10.1002/alz.14336 3. Rowe CC, Jones G, Doré V, et al. Standardized expression of 18F- NAV4694 and 11C-PiB β-Amyloid PET Results with the centiloid scale.J Nucl Med. 2016;57:1233-1237. doi:10.2967/jnumed.115.171595 4. Rowe CC, Doré V, Jones G, et al. 18F-florbetaben PET beta- amyloid binding expressed in centiloids. Eur J Nucl Med Mol Imaging. 2017;44:2053-2059. doi:10.1007/s00259-017-3749-6 5. Battle MR, Pillay LC, Lowe VJ, et al. Centiloid scaling for quantification of brain amyloid with [18F]flutemetamol using multiple processing methods. EJNMMI Res. 2018;8:107. doi:10.1186/s13550-018-0456-7 6. Navitsky M, Joshi AD, Kennedy I, et al. Standardization of amyloid quantitation with florbetapir standardized uptake value ratios to the Centiloid scale. Alzheimers Dement. 2018;14:1565-1571. doi:10.1016/ j.jalz.2018.06.1353 7. Centiloid Project. The Global Alzheimer’s Association Interactive Network. Accessed May 16, 2026. https://www.gaain.org/centiloid- project/ 8. Schmechel DE, Saunders AM, Strittmatter WJ, et al. Increased amy- loid beta-peptide deposition in cerebral cortex as a consequence of apolipoprotein E genotype in late-onset Alzheimer disease. Proc Natl Acad Sci. 1993;90:9649-9653. doi:10.1073/pnas.90.20.9649 9. Li C, Loewenstein DA, Duara R, et al. The Relationship of Brain Amyloid Load and APOE Status to Regional Cortical Thinning and Cognition in the ADNI Cohort. J Alzheimers Dis. 2017;59:1269-1282. doi:10.3233/ JAD-170286 10. Rowe CC, Ng S, Ackermann U, et al. Imaging β-amyloid burden in aging and dementia. Neurology. 2007;68:1718-1725. doi:10.1212/01. wnl.0000261919.22630.ea 11. Kantarci K, Lowe V, Przybelski SA, et al. APOE modifies the associa- tion between Aβ load and cognition in cognitively normal older adults.Neurology. 2012;78:232-240. doi:10.1212/WNL.0b013e31824365ab 12. Sperling RA, Donohue MC, Raman R, et al. Association of factors with elevated amyloid burden in clinically normal older individuals. JAMA Neurol. 2020;77:735. doi:10.1001/jamaneurol.2020.0387 13. Farrer LA. Effects of age, sex, and ethnicity on the association between apolipoprotein e genotype and Alzheimer Disease: a meta-analysis.JAMA. 1997;278:1349. doi:10.1001/jama.1997.03550160069041 14. Belloy ME, Andrews SJ, Le Guen Y, et al. APOE genotype and Alzheimer disease risk across age, sex, and population ancestry. JAMA Neurol. 2023;80:1284. doi:10.1001/jamaneurol.2023.3599 15. 2025 Alzheimer’s disease facts and figures. Alzheimers Dement2025;21:e70235. doi:10.1002/alz.70235 16. Wang Y-Y, Ge Y-J, Tan C-C, Cao X-P, Tan L, Xu W. The Proportion of APOE4 Carriers Among Non-Demented Individuals: a Pooled Analysis of 389,000 Community-Dwellers. J Alzheimers Dis. 2021;81:1331- 1339. doi:10.3233/JAD-201606 17. Tang M-X. The APOE-∊4 allele and the risk of Alzheimer disease among african americans, whites, and hispanics. JAMA. 1998;279:751. doi:10. 1001/jama.279.10.751 18. Campos M, Edland SD, Peavy GM. Exploratory study of apolipoprotein E ε4 genotype and risk of Alzheimer’s disease in mexican hispanics. J Am Geriatr Soc. 2013;61:1038-1040. doi:10.1111/jgs.12292 19. Xiao C, Pappas I, Aksman LM. Alzheimer’s Disease Neuroimaging Initiative. Comparison of genetic and health risk factors for mild cognitive impairment and Alzheimer’s disease between Hispanic and non-Hispanic white participants. Alzheimers Dement. 2023;19:5086- 5094. doi:10.1002/alz.13110. Toga AW, the Health and Aging Brain Study (HABS-HD) Study Team. 20. Wilkins CH, Windon CC, Dilworth-Anderson P, et al. Racial and ethnic differences in amyloid pet positivity in individuals with mild cogni- tive impairment or dementia: a secondary analysis of the imaging dementia–evidence for amyloid scanning (IDEAS) cohort study. JAMA Neurol. 2022;79:1139. doi:10.1001/jamaneurol.2022.3157 21. O’Bryant SE, Petersen M, Hall J, Johnson L. APOEε4 Genotype Is Related to Brain Amyloid Among Mexican Americans in the HABS- HD Study. Front Neurol. 2022;13:834685. doi:10.3389/fneur.2022. 834685 22. Palta P, Rippon B, Reitz C, et al. Apolipoprotein E genotype and in vivo amyloid burden in middle-aged Hispanics. Neurology. 2020;95:e2086- e2094. doi:10.1212/WNL.0000000000010707 23. Duara R, Loewenstein DA, Lizarraga G, et al. Effect of age, ethnic- ity, sex, cognitive status and APOE genotype on amyloid load and the threshold for amyloid positivity. NeuroImage Clin. 2019;22:101800. doi:10.1016/j.nicl.2019.101800 24. GAAIN Data. The Global Alzheimer’s Association Interactive Network. Accessed May 16, 2026. https://gaaindata.org/ 25. Neu SC, Crawford KL, Toga AW. Sharing data in the global Alzheimer’s association interactive network. NeuroImage. 2016;124:1168-1174. doi:10.1016/j.neuroimage.2015.05.082 26. Toga AW, Neu SC, Bhatt P, Crawford KL, Ashish N. The global Alzheimer’s association interactive network. Alzheimers Dement. 2016;12:49-54. doi:10.1016/j.jalz.2015.06.1896 27. Iaccarino L, Burnham SC, Tunali I, et al. A practical overview of the use of amyloid-PET centiloid values in clinical trials and research.NeuroImage Clin. 2025;46:103765. doi:10.1016/j.nicl.2025.103765 28. McKhann GM, Knopman DS, Chertkow H, et al. The diagnosis of dementia due to Alzheimer’s disease: recommendations from the national institute on aging-Alzheimer’s association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011;7:263-269. doi:10.1016/j.jalz.2011.03.005 29. A4 Study Data. Alzheimer’s clinical trials consortium. Accessed May 16, 2026. https://www.a4studydata.org/ 30. Sperling RA, Donohue MC, Raman R, et al. Trial of solanezumab in preclinical Alzheimer’s disease. N Engl J Med. 2023;389:1096-1107. doi:10.1056/NEJMoa2305032 31. Sperling RA, Donohue MC, Rissman RA, et al. Amyloid and tau predic- tion of cognitive and functional decline in unimpaired older individuals: longitudinal data from the A4 and LEARN studies. J Prev Alzheimers Dis. 2024;11:802-813. doi:10.14283/jpad.2024.122 32. Alzheimer’s Disease Neuroimaging Initiative. ADNI. Accessed May 16, 2026. https://adni.loni.usc.edu/ 33. Weiner MW, Kanoria S, Miller MJ, et al. Overview of Alzheimer’s disease neuroimaging initiative and future clinical trials. Alzheimers Dement. 2025;21:e14321. doi:10.1002/alz.14321 34. Jagust WJ, Landau SM, Koeppe RA, et al. The Alzheimer’s dis- ease neuroimaging initiative 2 PET Core: 2015. Alzheimers Dement. 2015;11:757-771. doi:10.1016/j.jalz.2015.05.001 35. Weber CJ, Carrillo MC, Jagust W, et al. The worldwide Alzheimer’s dis- ease neuroimaging initiative: ADNI-3 updates and global perspectives.Alzheimers Dement (N Y). 2021;7:e12226. doi:10.1002/trc2.12226 36. HABS-HD Study. Institute for translational research, UNT health fort worth. Accessed May 16, 2026. https://www.unthealth.edu/institute- for-translational-research/our-studies/habs-hd.html 37. Petersen ME, Zhou Z, Hall JR, et al. Health and aging brain study– health disparities (HABS-HD) methods and partner characteristics.Alzheimers Dement (N Y). 2025;11:e70140. doi:10.1002/trc2.70140 38. Original IDEAS Study. IDEAS Study. Accessed May 16, 2026. https:// www.ideas-study.org/Original-Study 39. Rabinovici GD, Gatsonis C, Apgar C, et al. Association of amyloid positron emission tomography with subsequent change in clinical man- ### Page 13 XIAO ET AL. 13 of 13agement among medicare beneficiaries with mild cognitive impair- ment or dementia. JAMA. 2019;321:1286. doi:10.1001/jama.2019. 2000 40. Standardized Centralized Alzheimer’s & Related Dementias Neu- roimaging (SCAN). National Alzheimer’s Coordinating Center. Accessed May 16, 2026. https://scan.naccdata.org/ 41. Besser L, Kukull W, Knopman DS, et al. Version 3 of the national Alzheimer’s coordinating center’s uniform data set. Alzheimer Dis Assoc Disord. 2018;32:351-358. doi:10.1097/WAD.00000000000002 79 42. Dubbelman MA, Van De Beek M, Van Gils AM, et al. Comparing and linking the mini-mental state examination and montreal cognitive assessment in the amsterdam dementia cohort. J Int Neuropsychol Soc. 2024;30:867-874. doi:10.1017/S1355617724000341 43. Jack CR, Bennett DA, Blennow K, et al. NIA-AA Research Frame- work: toward a biological definition of Alzheimer’s disease. Alzheimers Dement. 2018;14:535-562. doi:10.1016/j.jalz.2018.02.018 44. Jack CR, Andrews JS, Beach TG, et al. Revised criteria for diagnosis and staging of Alzheimer’s disease: Alzheimer’s association workgroup.Alzheimers Dement. 2024;20:5143-5169. doi:10.1002/alz.13859 45. Cerman J, Škorvagová A, Vyhnálek M, et al. Concordance between amyloid PET and CSF biomarkers in clinical setting: a cross-platform comparison and in-depth analysis of discordant cases. Sci Rep. 2025;15:39963. doi:10.1038/s41598-025-23759-5 46. Petersen ME, Zhang F, Hall J, et al. Characterization of plasma AT(N) biomarkers among a racial and ethnically diverse community-based cohort: an HABS-HD study. Alzheimers Dement (N Y). 2025;11:e70045. doi:10.1002/trc2.70045 47. Asken BM, Wang W, McFarland K, et al. Plasma Alzheimer’s biomark- ers and brain amyloid in Hispanic and non-Hispanic older adults.Alzheimers Dement. 2024;20:437-446. doi:10.1002/alz.13456 48. Asken BM, DeSimone JC, Wang W, et al. Plasma p-tau217 con- cordance with amyloid PET among ethnically diverse older adults.Alzheimers Dement (Amst). 2024;16:e12617. doi:10.1002/dad2.126 17 49. Granot-Hershkovitz E, Tarraf W, Kurniansyah N, et al. APOE alle- les’ association with cognitive function differs across Hispanic/Latino groups and genetic ancestry in the study of Latinos-investigation of neurocognitive aging (HCHS/SOL). Alzheimers Dement. 2021;17:466- 474. doi:10.1002/alz.12205 50. Mantri S, Nwadiogbu C, Fitts W, Dahodwala N. Quality of education impacts late-life cognition. Int J Geriatr Psychiatry. 2019;34:855-862. doi:10.1002/gps.5075 51. Liu C, Murchland AR, VanderWeele TJ, Blacker D. Eliminating racial disparities in dementia risk by equalizing education quality: a sensitiv- ity analysis. Soc Sci Med. 2022;312:115347. doi:10.1016/j.socscimed. 2022.115347 52. Mendez MF, Chavez D, Akhlaghipour G. Bilingualism delays expres- sion of Alzheimer’s clinical syndrome. Dement Geriatr Cogn Disord. 2019;48:281-289. doi:10.1159/000505872 53. Daniel EV, Kleiman MJ, Galvin JE. Exploring reasons for differen- tial vulnerability and Alzheimer’s disease risk in racial and ethnic minorities. J Alzheimers Dis. 2023;91(1):495-506. doi:10.3233/JAD- 220959SUPPORTING INFORMATIONAdditional supporting information can be found online in the Support- ing Information section at the end of this article.How to cite this article: Xiao C, Ard T, Blazhenets G, La Joie R, Pappas I, Toga AW. Association of cognitive impairment andAPOE ε4 with Centiloids in Hispanic and non-Hispanic White cohorts. Alzheimer’s Dement. 2026;22:e71586. https://doi.org/10.1002/alz.71586