Structural, Functional MRI Measures Improve Prediction of MS Long-Term Worsening

Nerve cells, or neurons, 3d illustration.
Study investigators evaluated the value of integrating structural and functional network MRI measures to predict clinical disability deterioration of multiple sclerosis.

The integration of structural and functional magnetic resonance imaging (MRI) measures significantly improved prediction of long-term clinical worsening in patients with multiple sclerosis (MS), according to study findings published in Neurology: Neuroimmunology & Neuroinflammation.

Currently, there is an active debate regarding the power of conventional MRI for predicting the clinical course of MS. In recent years, there have been efforts to improve the predictive value of MRI in MS by evaluating clinically relevant brain compartments, including gray matter (GM), strategic white matter (WM) tracts, and the spinal cord. Additionally, the mapping of functional and structural brain networks has also been regarded as possibly clinically relevant.

To date, the combined predictive value of functional and structural network techniques on predicting clinical worsening in patients with MS has not been fully elucidated. A team of Italian investigators sought to close this research gap by assessing the integration of functional and structural network MRI measures for the prediction of clinical disability deterioration over a median of 6.4 years in patients with MS.

In the study, researchers obtained baseline 3D T1-weighted and resting-state functional MRI scans from 233 patients with MS from a prospective hospital cohort and 77 healthy controls. The MS cohort included 157 patients with relapsing-remitting (RR) disease (RRMS), 59 with secondary progressive (SP) MS (SPMS), and 17 with primary progressive (PP) MS.

Patients in the study underwent a neurologic evaluation at both baseline and at a median 6.4-year follow-up. The study researchers assessed the Expanded Disability Status Scale (EDSS) score, disease-modifying treatment (DMT) changes, and occurrence of clinical relapses.

At the follow-up period, patients with MS were then classified as having clinically stable or worsened disease, based on changes in disability. Additionally, the study researchers evaluated the incidence of SPMS conversion in patients with RRMS.

In addition to obtaining global brain volumetry, the researchers performed an independent component analysis to identify the primary patterns in functional connectivity (FC) and GM networks.

The median follow-up EDSS score was 4.0 (median EDSS score change, 0.5; P <.0001). Approximately 45% (n=105) of patients with MS had experienced clinical worsening by the median 6.4-year follow-up, while the remaining 128 patients had stable disease.

Additionally, 16% (n=26) of the patients who initially had RRMS experienced disease conversion to SPMS. Factors independently associated with conversion to SPMS included baseline disability, normalized GM volume, and GM atrophy (false discovery rate P range =0.01–0.09; out-of-bag [OOB] accuracy, 0.84).

Normalized GM and brain volumes, increased FC of the left precentral gyrus in the sensorimotor network (SMN), decreased FC between default-mode networks, and GM atrophy in the frontoparietal network were identified as predictors of clinical worsening in a treatment-adjusted random forest model (OOB accuracy, 0.74).

The inclusion of network MRI variables (baseline EDSS score, normalized GM volume, GM sensorimotor network 1, and DMT change) improved prediction of disability worsening (P =.009) and SPMS conversion (OOB-AUC, 0.84; 95% CI, 0.76–0.91; P =.02).

A limitation of this study was the lack of baseline and follow-up cognitive assessments, an important limitation considering that cognitive changes can significantly impact patient functioning. As such, the investigators indicated they could not evaluate worsening of clinically important scores other than the EDSS, such as the Multiple Sclerosis Functional Compositem.

The study researchers suggest that “the added value of other MRI and serologic biomarkers, such as WM network damage assessed by diffusion-weighted MRI, demyelination/remyelination indices derived from magnetization transfer imaging, or neurofilament light chain, might be the topic of future investigations.”

Disclosure: Multiple authors declared affiliations with the pharmaceutical industry. Please refer to the original article for a full list of disclosures.


Rocca MA, Valsasina P, Meani A, et al. Network damage predicts clinical worsening in multiple sclerosis: A 6.4-year study. Neurol Neuroimmunol Neuroinflamm. 2021;8(4):e1006. doi:10.1212/NXI.0000000000001006