Premium
Neurofilament light and cognitive performance: Associations with amyloid and vascular pathologies in individuals with mild cognitive impairment
Author(s) -
Bos Isabelle,
Vos Stephanie J.B.,
ten Kate Mara,
Scheltens Philip,
Engelborghs Sebastiaan,
Frisoni Giovanni B,
Blin Olivier,
Richardson Jill,
Bordet Régis,
Verhey Frans R.J.,
Tsolaki Magda,
Popp Julius,
Lovestone Simon,
Streffer Johannes,
Andreasson Ulf,
Blennow Kaj,
Barkhof Frederik,
Zetterberg Henrik,
Visser Pieter Jelle
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.042739
Subject(s) - biomarker , neurodegeneration , medicine , amyloid (mycology) , cognition , oncology , effects of sleep deprivation on cognitive performance , psychology , pathology , disease , psychiatry , biology , biochemistry
Background Neurodegenerative, amyloid and vascular pathologies have all been associated with cognitive performance, but their interrelationship remains uncertain. We investigated the associations between cerebrospinal fluid (CSF) neurofilament light (NfL), a biomarker for neurodegeneration, with four different cognitive measures in individuals with mild cognitive impairment (MCI) with and without biomarker evidence of amyloid and vascular pathologies. Method We included 314 individuals with MCI from the EMIF‐AD Multimodal Biomarker Discovery study based on availability of data on CSF markers, MRI markers and cognitive data. CSF amyloid‐β 42/40 ratio was used to determine amyloid status (A‐/A+). Vascular status (V‐/V+) was based on the presence of at least one abnormal vascular MRI marker (Fazekas≧2, lacunes>0, microbleeds>0). Linear mixed models, with center as random factor, were used to test associations between NfL and four pooled cognitive outcome measures (memory z‐score, language z‐score, executive functioning z‐score and an attention z‐score) and the interaction with amyloid and vascular status. Result On average individuals were 69.2 (SD 7.6) years old and 51% were female. Seventyseven (25%) were classified as A‐V‐, 49 (16%) as A‐V+, 103 as A+V‐ (33%) and 85 (27%) as A+V+. The A+V+ group had significantly higher levels of NfL compared to the A‐V‐ group (p=0.026). There were no other differences between the groups on NfL levels. We found that NfL levels were associated with lower memory (β=‐0.019, p=0.037) and language (β=‐0.022, p=0.019) scores, but not with attention or executive functioning scores. In addition, the associations between NfL levels and memory scores were driven by the individuals with vascular pathology (A‐V+ p=0.007, A+V+ p=0.044) and the association with language scores was driven by the A‐V+ group (p=0.040). Figure 1 shows the relationship between cognitive outcome measures and NfL levels in the AV groups. Conclusion Higher NfL levels were associated with lower memory and language scores, in particular in patients with vascular pathology. NfL was not associated with attention and executive functioning scores and it was not associated with cognitive performance in patients with isolated amyloid pathology. These results could indicate that high NfL levels in prodromal Alzheimer’s disease may be proxy for underlying vascular pathology.