Contributions
Abstract: P494
Type: Poster
Abstract Category: Pathology and pathogenesis of MS - Imaging
Purpose: Accurate and reliable imaging biomarkers of axonal degeneration and demyelination are crucial to develop therapies for multiple sclerosis (MS) through clinical trials. Feasibility of multicentre clinical studies must be a key component of imaging development so that enough progressive MS patients may be recruited for the required statistical power. Recent developments in quantitative cervical spinal cord (CSC) magnetic resonance imaging (MRI) offers encouraging imaging biomarkers to inform decision making in clinical trials of progressive MS patients. In this two-site pilot study, we assessed both intra- and inter-site reproducibility of MRI measurements of CSC area and microstructural integrity at 3 tesla (3T).
Methods: Three healthy volunteers and three progressive MS patients were recruited at each of the two sites. Intra-site short (n=12) and long (n=6) term, as well as inter-site short (effective n=9) and long (n=3) term, reproducibility was evaluated by repeated scans including traveling subjects. The quantitative CSC MRI protocol, harmonized between sites (Siemens scanners), included structural imaging (0.9 mm isotropic T2-weighted (T2w) and 0.8 mm × 0.8 mm × 3 mm phase sensitive inversion recovery (PSIR)), axial 0.9 mm × 0.9 mm × 5 mm cardiac gated diffusion MRI (dMRI) and 0.75 mm × 0.75 mm × 5 mm magnetic transfer (MT) imaging. The dMRI protocol incorporated the latest multiband sequence and image reconstruction development with extensive piloting. The MT on and off images were co-registered before calculating MT ratio (MTR) image. The dMRI images were aligned using an updated 2-dimentional registration scheme before calculating the diffusion tensor maps.
Results: Preliminary data analysis indicated good intra- and inter-site reproducibility, both short and long term, with small variations. For example: standard deviation < 2 mm2 for vertebral-level-wise spinal cord area from C3 to L1 measured from T2w image and reliable grey-matter-to-white-matter contrast-to-noise ratio (> 2) from PSIR image; Kolmogorov-Smirnov test statistics < 0.2 for the difference in the distributions of MTR and DTI maps. Full data analysis is ongoing to quantify reproducibility and inter-site biases.
Conclusions: Reproducible quantitative cervical spinal cord MRI is achievable from multiple sites (single MRI scanner vendor) with careful acquisition protocol harmonization and centralized data analysis.
Disclosure:
Joo-won Kim: nothing to disclose
Dr. Matilde Inglese has received research grants from NIH, NMSS, Novartis Pharmaceuticals Corp., Teva Neuroscience.
Courtney Dula: nothing to disclose
Samantha Lancia: nothing to disclose
Christina Alfonso: nothing to disclose
Peng Sun: nothing to disclose
Sheng-Kwei Song is supported by grants from NIH, NMSS, and DOD.
Robert T Naismith has received fees for consulting from: Alkermes, Acorda, Bayer, Biogen, EMD Serono, Genentech, Genzyme, Mallinckrodt, EMD Serono, Novartis, Pfizer, Teva
Junqian Xu was supported by Radiological Society of North America (RSNA) research scholar grant RSCH1328 and International Progressive Multiple Sclerosis Alliance (IPMSA) infrastructure award PA0097
Abstract: P494
Type: Poster
Abstract Category: Pathology and pathogenesis of MS - Imaging
Purpose: Accurate and reliable imaging biomarkers of axonal degeneration and demyelination are crucial to develop therapies for multiple sclerosis (MS) through clinical trials. Feasibility of multicentre clinical studies must be a key component of imaging development so that enough progressive MS patients may be recruited for the required statistical power. Recent developments in quantitative cervical spinal cord (CSC) magnetic resonance imaging (MRI) offers encouraging imaging biomarkers to inform decision making in clinical trials of progressive MS patients. In this two-site pilot study, we assessed both intra- and inter-site reproducibility of MRI measurements of CSC area and microstructural integrity at 3 tesla (3T).
Methods: Three healthy volunteers and three progressive MS patients were recruited at each of the two sites. Intra-site short (n=12) and long (n=6) term, as well as inter-site short (effective n=9) and long (n=3) term, reproducibility was evaluated by repeated scans including traveling subjects. The quantitative CSC MRI protocol, harmonized between sites (Siemens scanners), included structural imaging (0.9 mm isotropic T2-weighted (T2w) and 0.8 mm × 0.8 mm × 3 mm phase sensitive inversion recovery (PSIR)), axial 0.9 mm × 0.9 mm × 5 mm cardiac gated diffusion MRI (dMRI) and 0.75 mm × 0.75 mm × 5 mm magnetic transfer (MT) imaging. The dMRI protocol incorporated the latest multiband sequence and image reconstruction development with extensive piloting. The MT on and off images were co-registered before calculating MT ratio (MTR) image. The dMRI images were aligned using an updated 2-dimentional registration scheme before calculating the diffusion tensor maps.
Results: Preliminary data analysis indicated good intra- and inter-site reproducibility, both short and long term, with small variations. For example: standard deviation < 2 mm2 for vertebral-level-wise spinal cord area from C3 to L1 measured from T2w image and reliable grey-matter-to-white-matter contrast-to-noise ratio (> 2) from PSIR image; Kolmogorov-Smirnov test statistics < 0.2 for the difference in the distributions of MTR and DTI maps. Full data analysis is ongoing to quantify reproducibility and inter-site biases.
Conclusions: Reproducible quantitative cervical spinal cord MRI is achievable from multiple sites (single MRI scanner vendor) with careful acquisition protocol harmonization and centralized data analysis.
Disclosure:
Joo-won Kim: nothing to disclose
Dr. Matilde Inglese has received research grants from NIH, NMSS, Novartis Pharmaceuticals Corp., Teva Neuroscience.
Courtney Dula: nothing to disclose
Samantha Lancia: nothing to disclose
Christina Alfonso: nothing to disclose
Peng Sun: nothing to disclose
Sheng-Kwei Song is supported by grants from NIH, NMSS, and DOD.
Robert T Naismith has received fees for consulting from: Alkermes, Acorda, Bayer, Biogen, EMD Serono, Genentech, Genzyme, Mallinckrodt, EMD Serono, Novartis, Pfizer, Teva
Junqian Xu was supported by Radiological Society of North America (RSNA) research scholar grant RSCH1328 and International Progressive Multiple Sclerosis Alliance (IPMSA) infrastructure award PA0097