Premium
The potential role of neuroinflammation and synaptic plasticity for neuropsychiatric symptoms
Author(s) -
Blum Hellen,
Lausser Ludwig,
von Einem Björn,
Fissler Patrick,
Kolassa IrisTatjana,
Otto Markus,
Tumani Hayrettin,
Kestler Hans,
Rosenbrock Holger,
von Arnim Christine
Publication year - 2020
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.043310
Subject(s) - neuroinflammation , depression (economics) , dementia , pathogenesis , medicine , logistic regression , synaptic plasticity , biomarker , apolipoprotein e , oncology , disease , psychology , endocrinology , receptor , biology , economics , macroeconomics , biochemistry
Background Neuropsychiatric symptoms (NPS) show a high prevalence within Alzheimer’s dementia. Evidence from animals and humans suggest a role of inflammatory cytokines and second messengers related to synaptic plasticity such as cGMP in Alzheimer´s disease (AD). In addition, studies show associations of serum interleukins (IL) and cGMP with NPS. However, the role of different ILs and cGMP in the pathogenesis of NPS is still not fully clear and studies on CSF are lacking. This study examines the associations between NPS with different established AD markers, pro‐ and anti‐inflammatory ILs and cGMP in the CSF. Method CSF and Serum samples of 120 patients with SCI, MCI or dementia and 19 controls who were previously examined at the memory clinic Ulm were analysed for this study. In clinical routine NPS were assessed by the neuropsychiatric inventory (NPI) total score and the Geriatric Depression Scale‐15 (GDS). Cytokines were determined by MSD V‐Plex assay in serum and CSF of patients, cGMP was determined by ELISA in CSF. Linear and binary logistic regression models were used for investigation. Bootstrapping was used if model assumptions were not fulfilled. Analyses were adjusted for ApoE genotype, age, gender and education. Additionally, an exhaustive screening for predictive marker combinations was performed (NPI>9, yes/no). Result The NPI total score was not associated with any of the evaluated ILs, but there was a positive association with CSF‐Abeta‐42 ( β = .20, p = .038). Higher cGMP was associated with a lower OR for NPI>9 ( OR = 0.69, p = .012). GDS was positively associated with IL‐1RA ( β = .20, p = .029). IL‐6 was negatively associated with CSF‐Abeta‐42 ( β = ‐0.09, p = .035). The predictive models achieved a maximal accuracy of 67.3%. Conclusion This study is one of the first, examining the relationship between CSF inflammatory cytokines, cGMP, AD pathology and NPS in memory clinic outpatients. The results indicate a role of cGMP in NPS. Furthermore, results indicate a potential role of ILs in depression in older adults (IL‐1RA) and in AD. Because of the risk for alpha‐error inflation due to multiple comparisons, further research is needed to replicate these findings.