Use of diffusion spectrum imaging in preliminary longitudinal evaluation of amyotrophic lateral sclerosis: development of an imaging biomarker

Abstract

Previous diffusion tensor imaging (DTI) studies have shown white matter pathology in amyotrophic lateral sclerosis (ALS), predominantly in the motor pathways. Further these studies have shown that DTI can be used longitudinally to track pathology over time, making white matter pathology a candidate as an outcome measure in future trials. DTI has demonstrated application in group studies, however its derived indices, for example fractional anisotropy, are susceptible to partial volume effects, making its role questionable in examining individual progression. We hypothesize that changes in the white matter are present in ALS beyond the motor tracts, and that the affected pathways and associated pattern of disease progression can be tracked longitudinally using automated diffusion connectometry analysis. Connectometry analysis is based on diffusion spectrum imaging and overcomes the limitations of a conventional tractography approach and DTI. The identified affected white matter tracts can then be assessed in a targeted fashion using High definition fiber tractography (a novel white matter MR imaging technique). Changes in quantitative and qualitative markers over time could then be correlated with clinical progression. We illustrate these principles toward developing an imaging biomarker for demonstrating individual progression, by presenting results for five ALS patients, including with longitudinal data in two. Preliminary analysis demonstrated a number of changes bilaterally and asymmetrically in motoric and extramotoric white matter pathways. Further the limbic system was also affected possibly explaining the cognitive symptoms in ALS. In the two longitudinal subjects, the white matter changes were less extensive at baseline, although there was evidence of disease progression in a frontal pattern with a relatively spared postcentral gyrus, consistent with the known pathology in ALS.

Keywords: amyotrophic lateral sclerosis; diffusion spectrum imaging; extramotoric; longitudinal; motoric; white matter.

FIGURE 1

Length threshold as defined by number of voxels in relation to the “false discovery rate” (FDR; each voxel = 2 mm). A length threshold of 24 mm corresponding to 12 voxels was chosen, therefore leading to an FDR of 0.16.

FIGURE 2

(A) Baseline diffusion connectometry analysis for patient 1, demonstrated the affected areas as being inferior occipito-frontal fasciculus (Inf. Occ. Fr.), cingulum (Cing.), and posterior part of the corpus callosum (CC). (B, posterior view) and (C, lateral view) a year later, extensive changes are now evident bilaterally in the corpus callosum; cinguli and posterior limb of the internal capsule (Int. Cap.) and in parts of inferior occipito-frontal fasciculi. (D) Cortical end-points of the affected fiber tracts at the time of the second scan demonstrate involvement of fibers originating from precentral gyrus (Precent. gyr.) and frontal areas with relative sparing of postcentral gyrus (Postcent. gyr.).

FIGURE 3

(A) Baseline diffusion connectometry analysis for patient 2, demonstrated the affected areas as being parts of the corpus callosum (CC), corticospinal (Cort-sp.) tracts, and inferior occipito-frontal (Inf. Occ. Fr.) fasciculi bilaterally. (B, left lateral view) and (C, posterior view) 6 months later shows extensive bilateral changes now involving corticospinal tracts, frontal fibers coursing toward the brainstem in the posterior limb of the internal capsule (Int. Cap.); cinguli (Cing.); superior longitudinal (Sup. Long.); and inferior occipito-frontal fasciculi. (D) Cortical end points for the second scan display predominant involvement of fibers originating from the precentral gyrus (Precent. gyr.) and frontal areas.

FIGURE 4

Baseline scan (posterior view) in patient 3, demonstrating evidence of extensive motoric and extramotoric changes in the white matter tracts [corpus callosum (CC); internal capsule (IC); Superior longitudinal (Sup. Long.) fasciculus; Cingulum (Cing.)] with changes being more pronounced on the right; also noted was involvement of the left superior cerebellar peduncle (Sup. Cereb.).

FIGURE 5

Baseline scans in patients 4 (A, left lateral view) and 5 (B, posterior view) demonstrating bilateral changes in the motoric and extramotoric pathways, being more pronounced on the right. Bilateral involvement of the cerebellar outflow tracts (Cerebel. outflow) was noted in patient 5. Bilaterally involved tracts in both included corpus callosum (CC); corticospinal tract (Cort-sp.); frontal fibers in the posterior limb of the internal capsule (Int. Cap.) coursing toward the brainstem; cinguli (Cing.) and parts of superior longitudinal (Sup. Long.) and inferior occipito-frontal (Inf. Occ. Fr.) fasciculi.

FIGURE 6

Targeted fiber tracking in patient 1 demonstrated evidence of progressive compromise of the right corticospinal tract (Cort-sp.) between the baseline scan (A) and the subsequent scan (B), a year later, on which discontinuity (discon.) in the caudal segment was noted.