PURPOSE

To investigate white matter integrity in the brain of relapsing-remitting multiple sclerosis (RRMS) patients under fingolimod treatment with diffusion MRI (dMRI) metrics longitudinally. Fingolimod is an oral therapy for RRMS, but disease progression may occur despite effective control of inflammation¹. dMRI provides quantitative measures of tissue integrity in Normal Appearing White Matter (NAWM) and may also provide useful predictive biomarkers of injury that do not manifest on a routine exam. In a 2 year longitudinal study of fingolimod-treated MS patients, we used dMRI with high angular resolution diffusion imaging (HARDI) to detect evolution of white matter injury, focusing on corticospinal tracts (CST) in brain. Our results suggest elevated injury in NAWM over the first year after initiation of treatment with stable white matter integrity over the second year.

METHODS

Fifteen RRMS patients were scanned under an IRB-approved protocol on a 3 tesla MRI—either TIM-Trio with a 12-channel head coil or a Prisma with a 16-channal head coil (Siemens AG, Erlangen, Germany). Whole brain dMRI, resting state fMRI (rs-fMRI) and anatomical T1W images of each subject were acquired prior to initiating fingolimod treatment and at 6, 12, 18 and 24 months after initiating fingolimod treatment. dMRI was acquired using a HARDI protocol (2mm isotropic, 71 diffusion-weighting gradients with b=1000sec/mm2 and 8 b=0 volumes). The diffusion tensor images were calculated after motion correction, taking into account systematic differences between the Trio and Prisma in terms of noise floor². The left and right CST were mapped in each subject with probabilistic tractography³, between ROIs placed in M1 (hand knob) and cerebral peduncle, resulting in a track density map. The M1 ROI was defined based upon maximum correlation between the left & right hand knobs⁴ using afni InstaCorr⁵ on rs-fMRI after physiologic noise and motion correction, and the cerebral peduncle ROI was drawn manually on the color FA map. Metrics of tissue integrity (TD, LD, FA and MD) from dMRI were averaged over the track density map on white matter of the whole brain. RANOVA in matlab (http://www.mathworks.com) was performed on mean values of TD, LD, FA and MD at all time points for the both sides of CST. In addition, Tukey-Kramer tests were performed to determine which pairs of time points differ.

RESULTS

RANOVA analysis based on the data from all time points revealed significant differences in TD, MD, FA and LD over 2 years (p<0.0004, table 1). TD was significantly higher at month 12 and 18 than at baseline (p<0.036) and at months 12, 18 and 24 as compared to month 6 (p<0.043) (figure 1, table 2), with no significant changes in TD seen during first 6 months or during the last 12 months of treatment. No significant changes in FA, LD and MD were seen during the first 6 months and the last 6 months of treatment. FA was significantly higher at month 24 than at month12 (p<0.014, table 2). MD was significantly higher at month 12, 18 and 24 than at baseline (p<0.006) and at month 6 (p<0.02, table 2). LD was significantly higher at month 18 and 24 than at base line (p<0.004), at month 6 (p<0.019) and at month12 (p<0.021) (table 2) separately.

DISCUSSION

The results suggest that MS patients experience a continued decline in white matter integrity during the first year of treatment with fingolimod with stabilization during the second year. Without a control group, it is unclear if these trends reflect on the impact of a treatment on disease progression. However, the results suggest that dMRI is sensitive to the evolution of tissue microstructure over time after treatment. Further study may determine if such measurements indeed reflect disease progression.

CONCLUSION

dMRI can detect longitudinal changes to tissue microstructure in MS patients undergoing disease-modifying therapy and may therefore have potential for the evaluation of treatment efficacy.

Acknowledgements

We gratefully acknowledge the support from Novartis and the contributions of Dr. Thorsten Feiweier, Siemens AG, Healthcare, for developing the advanced diffusion MRI pulse sequence used in this work.