Search Abstracts | Symposia | Slide Sessions | Poster Sessions
White matter microstructure in youth with a history of autism spectrum disorder who have lost the autism diagnosis
Poster Session B, Wednesday, September 30, 4:30 - 6:30 pm, Wangari Maathai
Inge-Marie Eigsti1, Saroj Khanal3, Nabin Koirala1, Deborah A. Fein1, Michael Stevens2; 1University of Connecticut, 2Institute of Living; Hartford, Connecticut, 3Nepal Applied Mathematics and Informatics Institute for Research, Kathmandu, Nepal
Approximately 15% of individuals diagnosed with autism early in development no longer display any symptoms by adolescence; they have “lost the autism diagnosis” (LAD). Alongside well-documented behavioral differences, fMRI studies reveal compensatory functional activation and residual autism-like activation patterns in LAD, but no evidence of normalization toward the activation patterns of neurotypical (NT) peers, suggesting equifinality (e.g., multiple neurodevelopmental pathways yielding similar behavioral outcomes). Research to date has not examined neurostructural differences in LAD. Given the prominence of white matter atypicalities in autism and evidence supporting disconnection models of autism, the present study compared white matter microstructural organization in LAD and neurotypical peers matched on age and IQ. Additionally, we examined the extent to which white matter microstructural variation mediates or moderates behavioral outcomes in LAD with a focus on language abilities, with the goal of identifying neurostructural substrates that may underlie the equifinal developmental trajectories observed in this population. We collected whole-brain multi-shell diffusion MRI data from 27 adolescents and young adults with LAD (n=10) and neurotypicals (NT; n=17) ages 13-35 using a 3T MRI scanner. Diffusion data were preprocessed using FSL's FDT (FMRIB's Diffusion Toolbox), including eddy current correction, motion correction, and brain extraction. White matter microstructural properties were derived using DTIFIT to generate voxelwise maps of fractional anisotropy (FA), mean (MD), axial (AD), and radial diffusivity (RD). Probabilistic tractography was obtained using FSL's BEDPOSTX and PROBTRACKX2 reconstructing major white matter tracts. Tract-based spatial statistics (TBSS) were additionally employed to perform voxelwise analyses and associations between white matter microstructural indices and behavioral measures [ADOS-2 total score; Pragmatic Language Composite] were evaluated using permutation-based inference (randomise, 5000 permutations) with threshold-free cluster enhancement (TFCE) and family-wise error (FWE) correction. There were no LAD/NT group differences; however, higher ADOS-2 total scores (e.g., more autism characteristics) were significantly associated with higher mean diffusivity (MD) across several white matter tracts, including the genu and body of the corpus callosum, left and right anterior corona radiata, right posterior thalamic radiation, and right superior corona radiata (p<0.05, FWE corrected). These microstructural changes were also positively correlated with Pragmatic Language abilities (Composite score). The present findings provide some of the first evidence that loss of autism diagnosis is accompanied by distinct white matter microstructural differences, with elevated mean diffusivity across commissural and projection fiber tracts. The pattern of associations between MD and both ADOS and Pragmatic Language Composite scores suggests that individuals with greater white matter microstructural disorganization in frontal tracts exhibit worse pragmatic language outcomes and more residual autism characteristics despite no longer meeting diagnostic criteria, indicating that white matter organization in this population retains a neurostructural signature of the underlying neurodevelopmental history of autism. These results extend prior functional neuroimaging findings of equifinality by suggesting that behavioral normalization in LAD does not correspond to neurostructural normalization, and that compensatory or atypical white matter organization may represent a neurostructural signature of this phenotype.
Topic Areas: Language Development/Acquisition, Disorders: Developmental