8389
Spinal Cord Atrophy As a Marker of Disability Progression: Beyond Trial Data

Thursday, June 2, 2022
Prince George's Exhibit Hall (Gaylord National Resort & Convention Center)
Blake E Dewey, PhD , Neurology, Johns Hopkins University School of Medicine, Baltimore, MD
Nicole Bou Rjeily, MD , Neurology, Johns Hopkins University School of Medicine, Baltimore, MD
Chen Hu, MS , Neurology, Johns Hopkins University School of Medicine, Baltimore, MD
Christy Hulett, BA , Neurology, Johns Hopkins University School of Medicine, Baltimore, MD
Gabriella Dagher, BS , Neurology, Johns Hopkins University School of Medicine, Baltimore, MD
Alexandra Zambriczki Lee, BS , Neurology, Johns Hopkins University School of Medicine, Baltimore, MD
Alyssandra Valenzuela, BS , Neurology, Johns Hopkins University School of Medicine, Baltimore, MD
Erin Brennan, MSN, PMHNP-BC , Neurology, Johns Hopkins University School of Medicine, Baltimore, MD
Anna DuVal, BA, MPH , Neurology, Johns Hopkins University School of Medicine, Baltimore, MD
Peter A Calabresi, MD , Neurology, Johns Hopkins University School of Medicine, Baltimore, MD
Vadim Zipunnikov, PhD , Biostatistics, Johns Hopkins University Bloomberg School of Public Health, Baltimore, MD
Kathryn C Fitzgerald, ScD , Neurology, Johns Hopkins University School of Medicine, Baltimore, MD
Ellen M Mowry, MD, MCR , Dept. of Neurology, Johns Hopkins School of Medicine, Baltimore, MD



Background:

Spinal cord (SC) atrophy has been established as an important marker of disability progression but is infrequently used as a quantitative measure in clinical trials of people with MS, as high-quality, high-resolution MRI of the SC is not routinely acquired. High quality T1-weighted (T1w) brain imaging, however, is a key part of standard of care imaging in MS. It has been recently proposed that T1w brain images with sufficient SC coverage could substitute for dedicated SC imaging in analysis of SC atrophy. Use of these images would allow SC atrophy to be used as a marker in larger populations with longer follow-up than is currently feasible.

Objectives:

To determine if SC cross-sectional area (CSA) measurements in T1w brain images can be used as a marker for disability similarly to dedicated T2-weighted (T2w) SC images.

Methods:

Sagittal 3D T1w MPRAGE or Multi-Echo MPRAGE brain images and T2w SC images were acquired on people with MS as a part of an ongoing, prospective study (progressive: n=47, relapsing: n=103) using two different 3T scanners. The Spinal Cord Toolbox was used to automatically label the vertebral regions. CSA was averaged over the C2-C3 region. To account for different scanners, COMBAT was used to harmonize the data. Spearman rank correlation and corrected correlations with multivariate linear regression models were used for statistical analysis.

Results:

CSA measurements from T1w brain and T2w SC images were highly correlated (r=0.93, p<0.0001). CSA from T2w SC images was significantly correlated with EDSS (r=-0.26, p=0.001), timed 25ft walk (T25W) (r=-0.29, p=0.001), and 9-hole peg test (9HPT) (r=-0.38, p<0.0001). CSA from T1w brain images was similarly correlated (EDSS: r=-0.34, p<0.0001, T25W: r=-0.38, p<0.0001, 9HPT: r=-0.42, p<0.0001). After correction for age and sex, however, T25W was no longer significantly correlated with CSA from either image type. Additionally, CSA from both image types showed a significant difference between relapsing and progressive subjects after adjusting for age and sex (T1w: p=0.0006, T2w: p=0.003).

Conclusions:

SC CSA measurements from T1w brain and T2w SC images show strong correlation to each other as well as to mobility-related disability scores. Correlation was also stronger using T1w brain images. These results demonstrate the ability to utilize routinely acquired T1w brain images to enrich clinical trial datasets targeting progression, which may or may not include dedicated SC imaging.

See more of: Late-breaking Abstracts
See more of: Late-Breaking
<< Previous Abstract | Next Abstract >>