Reduced myelin contributes to cognitive decline in monogenic small vessel disease.

217 

Abstract Submission 

Jannis Denecke¹, Anna Dewenter¹, Jongho Lee², Nicolai Franzmeier^1,3,4, Lukas Pirpamer⁵, Benno Gesierich^6,1, Marco Duering^6,7,1, Michael Ewers^1,8

¹Institute for Stroke and Dementia Research (ISD), LMU University Hospital, Munich, Germany, ²Laboratory for Imaging Science and Technology, Department of Electrical and Computer Engineering, S, Seoul, Republic of Korea, ³Department of Psychiatry and Neurochemistry, Institute of Neuroscience and Physiology, The Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden, ⁴Munich Cluster for Systems Neurology (SyNergy), Munich, Germany, ⁵Medical Image Analysis Center (MIAC) and Department of Biomedical Engineering, University of Basel, Basel, Switzerland, ⁶Medical Image Analysis Center, University of Basel, Basel, Switzerland, ⁷Department of Biomedical Engineering, University of Basel, Basel, Switzerland, ⁸German Center for Neurodegenerative Disease (DZNE), Munich, Germany

Jannis Denecke  
Institute for Stroke and Dementia Research (ISD), LMU University Hospital
Munich, Germany

Anna Dewenter  
Institute for Stroke and Dementia Research (ISD), LMU University Hospital
Munich, Germany

Jongho Lee  
Laboratory for Imaging Science and Technology, Department of Electrical and Computer Engineering, S
Seoul, Republic of Korea

Nicolai Franzmeier  
Institute for Stroke and Dementia Research (ISD), LMU University Hospital|Department of Psychiatry and Neurochemistry, Institute of Neuroscience and Physiology, The Sahlgrenska Academy, University of Gothenburg|Munich Cluster for Systems Neurology (SyNergy)
Munich, Germany|Gothenburg, Sweden|Munich, Germany

Lukas Pirpamer  
Medical Image Analysis Center (MIAC) and Department of Biomedical Engineering, University of Basel
Basel, Switzerland

Benno Gesierich  
Medical Image Analysis Center, University of Basel|Institute for Stroke and Dementia Research (ISD), LMU University Hospital
Basel, Switzerland|Munich, Germany

Marco Duering  
Medical Image Analysis Center, University of Basel|Department of Biomedical Engineering, University of Basel|Institute for Stroke and Dementia Research (ISD), LMU University Hospital
Basel, Switzerland|Basel, Switzerland|Munich, Germany

Michael Ewers  
Institute for Stroke and Dementia Research (ISD), LMU University Hospital|German Center for Neurodegenerative Disease (DZNE)
Munich, Germany|Munich, Germany

Myelin enwraps axonal connections in the brain and is of critical importance for information transfer between the connected brain regions. Small vessel disease (SVD), a major cause of stroke, is associated with white matter changes such as white matter hyperintensities (WMH). However, myelin alterations in SVD were predominantly characterized in histopathological studies focusing on WMH alterations, with only a few neuroimaging studies having assessed myelin alterations. A major barrier so far in neuroimaging of myelin has been the confounding of the myelin related MR signal by confounding factors such as iron in the case of T2-star weighted images. In order to assess SVD-related myelin alterations in the whole white matter and its association with cognitive decline, we leveraged χ-separation, a newly developed technique, to separate myelin from potentially iron related MRI signal in a pure monogenic caused form of SVD, i.e. CADASIL (which stands for cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy).

We included 65 patients with confirmed CADASIL (mean age 55.1) and 27 cognitively normal controls (NC, age = 71.5) from an in-house study. All Participants were assessed with a 3D-T2-star-weighted multi-echo gradient-echo sequence on 3T MRI scanner, alongside conventional MRI markers (see Baykara et al., 2016). As a measure of myelin, we employed the χ-separation method (Shin et al. 2021). This technique separates the total susceptibility χ into diamagnetic (|χ-negative|, e.g. myelin) and paramagnetic (χ-positive , e.g. iron) sources which both cause a faster spin dephasing, hence magnitude loss, but opposingly influence the phase. For comparison, we assessed DTI-based mean diffusivity (MD), i.e. a standard measure of microstructural white matter changes unspecific to myelin. For each participant, ROI values of |χ-negative|, χ-positive, and MD were extracted from areas of WMH and normal appearing white matter (NAWM). Difference-scores between CADASIL and group averaged NC scores were computed to derive abnormality scores for each ROI. In addition to those ROI values, all measures were obtained from the left anterior thalamic radiation (ATR) and the genu of the corpus callosum (CCg) which are regarded as strategic fiber tracts for information processing speed (Duering et al., 2011). As a measure of processing speed, the power-transformed average of the TMT A & B test scores, normalized by age and education (Tombaugh, 2004) were computed. Linear regression and ridge regression were used to test our hypotheses.

We found significantly reduced |χ-negative| values in the WMH, NAWM, and fiber tracts including the ATR and CCg in CADASIL compared to the controls, adjusted for age, sex, education, and χ-positive values (Figure 1A). The decrease in |χ-negative| difference scores in CADASIL was stronger in WMH compared to NAWM (Figure 1B), suggesting pronounced myelin damage in WMH areas. In contrast, χ-positive values were reduced in WMH areas but not NAWM or tracts (Figure 1, 2^nd row). Consistent with previous findings, MD values were increased in CADASIL WMH areas and negatively correlated to |χ-negative| scores (Fig 1C, standardized β = -0.46, t(60) = -4.3, p < .001). For cognition, a decrease in |χ-negative| values in WMH (Figure 2A, p = .029, partial-R² = .08) and CCg (Figure 2B, p = .016) but not ATR (Figure 2C, p = .49) was associated with slower scores of processing speed, controlled for MD, χ-positive, age, sex, and education.

We found that |χ-negative| values were reduced in CADASIL independently of MR signal highly sensitive to iron, suggesting that myelin is significantly reduced in monogenic SVD. The contribution of χ-negative to cognitive decline was in addition to that by the increase in MD values, suggesting that MR measures with increased specificity to myelin alterations contribute to explain cognitive decline in monogenic SVD.

Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) ¹

Cortical Cyto- and Myeloarchitecture ²

Cerebrovascular Disease

Cognition

Degenerative Disease

Demyelinating

Myelin

Tractography

White Matter

Other - small vessel disease; white matter hyperintensity