Elsevier

Neurobiology of Aging

Volume 70, October 2018, Pages 233-241
Neurobiology of Aging

Regular article
Neurofilament relates to white matter microstructure in older adults

https://doi.org/10.1016/j.neurobiolaging.2018.06.023Get rights and content

Abstract

Cerebrospinal fluid (CSF) neurofilament light (NFL) is a protein biomarker of axonal injury. To study whether NFL is associated with diffusion tensor imaging (DTI) measurements of white matter (WM) microstructure, Vanderbilt Memory & Aging Project participants with normal cognition (n = 77), early mild cognitive impairment (n = 15), and MCI (n = 55) underwent lumbar puncture to obtain CSF and 3T brain MRI. Voxel-wise analyses cross-sectionally related NFL to DTI metrics, adjusting for demographic and vascular risk factors. Increased NFL correlated with multiple DTI metrics (p-values < 0.05). An NFL × diagnosis interaction (excluding early mild cognitive impairment) on WM microstructure (p-values < 0.05) was detected, with associations strongest among MCI. Multiple NFL × CSF biomarker interactions were detected. Associations between NFL and worse WM metrics were strongest among amyloid-β42–negative, tau-positive, and suspected nonamyloid pathology participants. Findings suggest increased NFL, a biomarker of axonal injury, is correlated with compromised WM microstructure. Results highlight the role of elevated NFL in predicting WM damage in cognitively impaired older adults who are amyloid-negative, tau-positive, or meet suspected nonamyloid pathology criteria.

Introduction

One rapidly emerging biomarker in abnormal cognitive aging and Alzheimer's disease is neurofilament light (NFL), a support protein within large, myelinated axons specifically within the cerebral white matter (WM) (Bergman et al., 2016, Friede and Samorajski, 1970, Skillback et al., 2014, Yates et al., 2009). Increased cerebrospinal fluid (CSF) NFL concentrations are thought to reflect breakdown of large caliber axons (Friede and Samorajski, 1970, Menke et al., 2015), thus compromising WM microstructure with cognitive consequences (Menke et al., 2015, Skillback et al., 2014). Elevated CSF NFL correlates with MRI T1-weighted WM hypointensities, reflecting WM macrostructural damage, particularly among older adults with mild cognitive impairment (MCI) and Alzheimer's disease (Zetterberg et al., 2016). Prior studies have yielded limited and inconsistent associations between CSF NFL and WM microstructure assessed by diffusion imaging among middle aged and older individuals with normal cognition (NC) (Melah et al., 2016, Racine et al., 2017) and more severe traumatic brain injury (Al Nimer et al., 2015). Therefore, it is unknown if CSF NFL relates to more sensitive measures of the WM specific to the microstructure of fiber bundles, among older adults with and without cognitive impairment.

The present study sought to examine the association between CSF NFL obtained in vivo and diffusion tensor imaging (DTI) measures of WM microstructure among older individuals with NC and MCI, a prodromal stage of Alzheimer's disease. Given prior research linking CSF NFL, axonal injury, and WM integrity in murine models (Friede and Samorajski, 1970) and motor neuron disease (Menke et al., 2015), we hypothesized that increased CSF NFL concentrations would correlate with compromised WM microstructure. We also hypothesized effect modification by cognitive diagnosis with associations stronger in MCI, where compromised axonal integrity may underlie clinical signs of cognitive impairment (Nir et al., 2013) and a greater range of WM degeneration is present (Huang et al., 2007). In post hoc analyses, NFL and WM microstructure associations were tested for interactions with established Alzheimer's disease biomarkers, including CSF Aβ42 and t-tau to determine if underlying Alzheimer's disease pathology was driving any significant observations.

Section snippets

Study cohort

The Vanderbilt Memory & Aging Project (Jefferson et al., 2016) is a longitudinal observational study investigating vascular health and brain aging, enriched with older adults with MCI. Cohort inclusion criteria required participants to be at least 60 years of age, speak English, have adequate auditory and visual acuity for testing, and have a reliable study partner. At eligibility, participants underwent a comprehensive assessment and were excluded for a cognitive diagnosis other than NC, early

Participant characteristics

For all participants (n = 147, 72 ± 6 years, 68% males, 93% non-Hispanic white), CSF NFL concentrations ranged 268–3617 pg/mL, CSF Aβ42 concentrations ranged 289–1195 pg/mL, and CSF t-tau concentrations ranged 107–1542 pg/mL. NFL was not correlated with Aβ42 (r = −0.05, p = 0.55) and was correlated with t-tau (r = 0.42, p < 0.001). No collinearity was found between the CSF variables (Variation Inflation Index < 2.9 for all variables in all models). Mean time between brain MRI and lumbar

Discussion

Among community-dwelling older adults without a clinical history of stroke or dementia, higher CSF concentrations of NFL were associated with compromised WM microstructure measured with DTI. Specifically, NFL was negatively associated with FA and positively associated with mean, radial, and axial diffusivity. Diagnostic interactions revealed findings were strongest in the MCI participants. Biomarker interactions including both NC and MCI participants further indicated that findings were

Conclusions

The present study demonstrates a novel association between NFL and WM microstructure, which is modified by cognitive diagnosis (MCI) and biomarkers of Alzheimer's disease (Aβ42) and neurodegeneration (t-tau). Many older adults have comorbid pathologies present in late life (Schneider et al., 2007), and Alzheimer's disease therapies may be less effective when additional pathologies are present (Weekman et al., 2016). Therefore, the modified associations reported here between NFL and DTI in the

Disclosure statement

The authors have no actual or potential conflicts of interest.

Acknowledgments

The authors would like to thank the dedicated Vanderbilt Memory & Aging Project participants, their loved ones, and the devoted staff and trainees who contributed to recruitment, screening, and enrollment of the baseline cohort. The authors would also like to thank the dedicated and skilled laboratory technicians at the Clinical Neurochemistry Laboratory at Sahlgrenska University Hospital, Sweden, who performed all the CSF analyses.

Funding: This research was supported by Alzheimer's Association

References (40)

  • P.N. Sen et al.

    A model for diffusion in white matter in the brain

    Biophysical J.

    (2005)
  • S.M. Smith et al.

    Tract-based spatial statistics: voxelwise analysis of multi-subject diffusion data

    Neuroimage

    (2006)
  • S.M. Smith et al.

    Threshold-free cluster enhancement: addressing problems of smoothing, threshold dependence and localisation in cluster inference

    Neuroimage

    (2009)
  • S.K. Song et al.

    Dysmyelination revealed through MRI as increased radial (but unchanged axial) diffusion of water

    Neuroimage

    (2002)
  • R.A. Sperling et al.

    Toward defining the preclinical stages of Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease

    Alzheimers Dement.

    (2011)
  • D.M. Yates et al.

    Neurofilament subunit (NFL) head domain phosphorylation regulates axonal transport of neurofilaments

    Eur. J. Cell Biol.

    (2009)
  • F. Al Nimer et al.

    Comparative assessment of the prognostic value of biomarkers in traumatic brain injury reveals an independent role for serum levels of neurofilament light

    PLoS One

    (2015)
  • B.B. Bendlin et al.

    CSF t-tau/aβ(42) predicts white matter microstructure in healthy adults at risk for Alzheimer’s disease

    PLoS One

    (2012)
  • J. Bergman et al.

    Neurofilament light in CSF and serum is a sensitive marker for axonal white matter injury in MS

    Neurol. Neuroimmunol. Neuroinflamm.

    (2016)
  • R.B. D'Agostino et al.

    Stroke risk profile: adjustment for antihypertensive medication. The Framingham Study

    Stroke

    (1994)
  • Cited by (0)

    View full text