Immune protein levels in cerebrospinal fluid: Associations with memory scores across the AD spectrum. (December 2021)

• Record Type: Journal Article
• Authors: Wesenhagen, Kirsten E. J.; Teunissen, Charlotte E.; Gobom, Johan; Vos, Stephanie J. B.; Blennow, Kaj; Zetterberg, Henrik; Scheltens, Philip; Tijms, Betty M.; Visser, Pieter Jelle
• Is Part Of: Alzheimer's & dementia. Volume 17(2021)Supplement 3
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-12
• Subjects: Alzheimer's disease -- Periodicals; Alzheimer Disease -- Periodicals; Dementia -- Periodicals; Démence; Maladie d'Alzheimer; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.83
• DOI: 10.1002/alz.055451
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: Alzheimer's disease (AD) is characterized by amyloid and tau aggregation. In addition, immune alterations are involved, perhaps with clinical stage‐dependent effects. How cerebrospinal fluid (CSF) levels of immune system‐related proteins associate with memory function in AD remains unknown. We studied this question across the AD clinical spectrum. Method: We selected from ADNI and EMIF‐AD MBD, 425 amyloid‐positive (A+) individuals across clinical stages (preclinical, prodromal and dementia) and 102 amyloid‐negative (A‐) CN controls, with available CSF proteomics and memory scores. Memory test scores of word lists were Z‐transformed based on controls (ADNI), or published norms (EMIF‐AD). Proteins were measured in EMIF‐AD with mass spectrometry (MS) (2537 proteins) and in ADNI by MS (142 proteins) and RBM (83 proteins). Of 165 proteins assessed in both cohorts, we selected 53 proteins associated with 'immune system process' in Gene Ontology (GOv2.1). We Z‐transformed levels relative to controls in each cohort and pooled the datasets. For proteins showing an interaction with diagnosis on memory with p‐value < 0.1, we analysed group‐stratified associations in linear models (predictor: protein levels; outcome: memory scores; covariates: sex, age, years of education, diagnosis; p‐value threshold: 0.05). We annotated memory‐associated proteins for subcategories of the immune system process. Result: We included 102 A‐ controls and 425 A+ individuals (preclinical AD: 105, prodromal AD: 183, and AD dementia: 137, Table 1). Eighteen proteins (34%) showed diagnosis‐dependent memory associations, for which we performed stratified analyses. In the control group, higher levels of 4 proteins (22%) associated with worse memory (Figure 1‐2). In preclinical AD, lower concentrations of 11 proteins associated with worse memory (Figure 1). These proteins were enriched for neutrophil degranulation. Of the 11 proteins, 2 were associated with memory in controls, but in an opposite direction. In prodromal AD, lower concentrations of 4 proteins associated with worse memory. No associations occurred in individuals with AD‐type dementia, probably due to the low memory scores in this group (Figure 2). Conclusion: Our results suggest immune alterations affect memory function very early in AD. Further research is needed to clarify why in controls higher levels of immune proteins associate with lower memory function.