Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Longitudinal changes in caudate nucleus myelination in developmental language disorder
Poster Session D, Thursday, October 1, 4:30 - 6:30 pm, Wangari Maathai
Nilgoun Bahar1,2, Saloni Krishnan3, Kate E. Watkins1; 1University of Oxford, 2University of California, San Francisco (UCSF), 3University College London (UCL)
Introduction: Developmental language disorder (DLD) is a neurodevelopmental condition characterised by persistent difficulties acquiring one’s native language. Previous neuroimaging studies have identified structural differences across distributed language-related networks in DLD, including cortical and subcortical regions implicated in procedural learning. Converging evidence has suggested alterations within frontostriatal circuitry, including the caudate nucleus (CN). Using quantitative multi-parameter mapping (MPM), Krishnan and colleagues (2022) reported lower magnetisation transfer saturation (MTsat) values bilaterally in the dorsal CN in children with DLD relative to typically developing (TD) peers, suggesting differences in myelin-related tissue properties. Higher MTsat values in the left CN were also associated with better language proficiency across participants. These findings motivated the present longitudinal follow-up examining whether previously identified differences persist across development or instead reflect delayed maturation that normalises over time. To our knowledge, this is the first longitudinal quantitative MRI study investigating brain microstructure in DLD. Methods: Whole-brain quantitative MRI scans were collected from 41 participants from the original cross-sectional study. Participants were first assessed between ages 10–15 years and followed up between ages 14–20 years (mean interval = 4y 6m). The sample included 19 participants with DLD and 22 TD controls. MRI data were acquired using the same MPM protocol and processed using the hMRI toolbox to generate MTsat maps. Given prior evidence implicating the CN in DLD, analyses focused on this region. Left and right CN masks derived from the Harvard–Oxford subcortical atlas were used to extract MTsat values at both timepoints. Longitudinal changes were assessed using a repeated-measures ANOVA with time (1, 2) and hemisphere (left, right) as within-subject factors, and group (TD, DLD) as a between-subject factor. As previous findings indicated lower CN MTsat values in DLD, the group comparison was tested directionally using a one-tailed test. Results: Consistent with the hypothesis that participants with DLD would show lower CN MTsat values than TD, the planned one-tailed group comparison revealed a significant main effect of group, F(39) = 3.34, p = .035. There was also a significant main effect of time, F(1,39) = 11.41, p = .001, indicating increased MTsat values over time across groups and hemispheres. In addition, there was a significant main effect of hemisphere, F(1,39) = 24.28, p < .001, indicating higher MTsat values in the left than right CN. The group × time interaction was not significant (p = .653), indicating similar developmental trajectories across both groups. Discussion: These findings extend prior cross-sectional evidence by demonstrating persistently lower CN MTsat values in DLD throughout adolescence. Although both groups showed age-related increases in MTsat consistent with ongoing maturation of the dorsal striatum, participants with DLD exhibited lower MTsat values in the CN across timepoints, with no evidence of developmental “catch-up.” The absence of a group × time interaction suggests that these group differences remain stable over time. The observed left-greater-than-right hemisphere effect additionally suggests lateralised microstructural differences within the dorsal striatum. Overall, our findings indicate that altered CN microstructure may reflect a persistent neurodevelopmental characteristic associated with DLD rather than a transient maturational delay.
Topic Areas: Disorders: Developmental, Language Development/Acquisition