Forskningsartikler

B Pozuelo Moyano A von Gunten, C Mueller, R Yaman-Deveci, R Howard, R Stewart 2, A Young, H Costello, S Bulteau, B Calvet, S Bonin-Guillaume, N Hoertel, G H Robert, J Roche, A Lepetit, S Louchart de la Chapelle, L Agüera-Ortiz, J Pla-Vidal, F Bouckaert, O Voshaar, L Fernandes, O Vasiliu, M Dominiak 29, J Priller, M Preisig, F Triolo, M Belvederi Murri, E Aakhus, S Ranjbar, K Swierkosz-Lenart, V Orgeta, P Vandel

Abstract

Background and objectives: Central and general adiposity have been linked to dementia risk, but their relation to blood-based Alzheimer disease (AD) biomarkers is unclear. We examined life course adiposity, measured by waist-to-height ratio (WtHR) and body mass index (BMI), in relation to plasma p-tau217 and to AD dementia verified by biomarker status.

Methods: In this cohort study, we included data on participants from the general population aged 70 years or older from the fourth wave of the Norwegian Trøndelag Health Study (HUNT4; 2017-19). Plasma p-tau217 was collected and standardized clinical cognitive assessments were performed at HUNT4. Plasma p-tau217 concentration at ≥0.63 pg/mL defined positive p-tau217. Positive p-tau217 coupled with a clinical dementia diagnosis defined biomarker-verified AD dementia. WtHR was measured 3 times (HUNT2-4; 1995-2019), and BMI was measured 4 times (HUNT1-4; 1984-2019). We performed linear, logistic, and linear mixed-effects regression adjusting for demographics, APOE ε4, lifestyle, and mental health.

Results: The final study sample comprised 8,797 participants (53.5% women, mean age at HUNT4 77.8 [SD 6.2]). Of these, 2,649 (30.1%) were p-tau217-positive and 659 (7%) had biomarker-verified AD dementia. Midlife WtHR ≥0.60 was associated with 12.8% (95% CI 7.3-19.7) higher late-life p-tau217 concentration and higher risk of positive p-tau217 (relative risk ratios [RRRs] 1.55, 1.21-1.99) and biomarker-verified AD dementia (RRR 1.84, 1.27-2.65) compared with WtHR <0.50. In late life, WtHR ≥0.60 was associated with 15.6% (-19.0 to -12.2) lower p-tau217 concentration and lower risk of positive p-tau217 (RRR 0.49, 0.41-0.59) and biomarker-verified AD dementia (RRR 0.63, 0.46-0.86). BMI showed similar patterns: Midlife obesity was associated with higher p-tau217 concentration and elevated risk of biomarker-verified AD dementia, whereas late-life overweight/obesity was associated with lower p-tau217 and decreased risk of biomarker-verified AD dementia. Among those with positive p-tau217 or biomarker-verified AD dementia, mixed-effects regression showed higher midlife adiposity, reversing by late life.

Discussion: Our results identify midlife central and general adiposity as modifiable risk factors for AD pathology and AD dementia. WtHR may aid early risk stratification and inform interventions targeting central fat reduction.

Ekaterina Zotchev, Bjørn Heine Strand, Anita Sunde, Kay Deckers, Dag Aarsland, Nicholas J Ashton, Henrik Zetterberg, Vegard Fykse Skirbekk, Miguel G Borda, Gill Livingston, Archana Singh-Manoux, Geir Selbaek

Abstract
Background: Adults with Down syndrome (DS) have a high dementia risk, highlighting the need for robust, DS-specific assessment tools. This study evaluated the psychometric properties of the Norwegian version of the CAMDEX-DS-II by examining reliability and validity of the informant interview and the CAMCOG-DS-II cognitive assessment.

Method: In this nationwide study, 108 adults with DS were assessed across 19 hospital units during 2021-2023. Participants underwent a standardised dementia assessment including the CAMDEX-DS-II battery. Reliability was assessed using Cronbach’s alpha, weighted kappa, and intraclass correlation coefficients (ICCs), while validity was evaluated using factor analysis, receiver operating characteristic (ROC) analyses, and external cognitive and functional measures.

Results: The CAMDEX-DS-II informant interview demonstrated good to excellent psychometric properties, with high internal consistency (α ≥ 0.83) in core cognitive-functional sections and strong inter-rater reliability, with most items showing excellent weighted kappa (κ ≥ 0.80). Scores aligned closely with clinician-determined diagnostic classifications. The CAMCOG-DS-II showed very good internal consistency (α = 0.84) and excellent inter-rater reliability (ICCs ≥ 0.90). CAMCOG-DS-II total and domain scores differed significantly across diagnostic groups, with moderate-to-large effect sizes. ROC analyses indicated good overall diagnostic accuracy, with areas under the curve (AUCs) > 0.80, and particularly strong discrimination in individuals with mild ID.

Conclusions: The Norwegian CAMDEX-DS-II provides reliable indicators of dementia-related change in adults with DS. The combined informant interview and cognitive assessment provided evidence based on relations to diagnostic classification and external measures, supporting their clinical utility in the specialist services and contributing to the international evidence base.

Frode Kibsgaard Larsen, Ingrid Tøndel Medbøen, Andre Strydom, Geir Selbæk, Bjørn Heine Strand, Ellen Melbye Langballe