Development of microstructural correlations of white matter in ASD between 6-24 months
Poster No:
450
Submission Type:
Abstract Submission
Authors:
Sooyeon Sung1, Jed Elison1, Brittany Howell2
Institutions:
1University of Minnesota, Minneapolis, MN, 2Virginia Tech, Roanoke, VA
First Author:
Co-Author(s):
Introduction:
Neuroimaging studies suggest that autism spectrum disorder (ASD) is associated with altered white matter microstructure. Evidence from prospective studies demonstrated that infants later diagnosed with ASD had higher fractional anisotropy (FA) values at 6 months, but lower FA values at 24 months compared to infants not diagnosed with ASD. Most studies, however, focus on group differences in the mean values of DTI parameters rather than inter-tract correlations. Some previous studies have shown that the network inefficiency and less coherent white matter microstructure linkage pattern are observed in ASD participants. This study examined the development of white matter correlations in infants diagnosed with ASD later (ASD+) compared to infants not diagnosed with ASD (ASD-) based on FA as quantified using diffusion MRI (dMRI).
Methods:
The data was collected as part of the Infant Behavior Imaging Study (IBIS), an ongoing longitudinal study investigating brain and behavioral development in infants at high and low familial likelihood for developing ASD. The total sample included 324 (276 ASD-, 48 ASD+), 340 (296 ASD-, 44 ASD+) and 311 (261 ASD-, 50 ASD+) participants at 6, 12, and 24 months, respectively. Diagnostic classification of ASD was determined at 24 months using DSM-IV-TR criteria. Diffusion MRI was collected at four US sites using Siemens 3T TIM Trio MR scanners using a 12-channel head coil. The dMRI sequence was acquired as an ep2d_diff pulse sequence (TR=12,800–13,300ms, TE=102ms, b=0 - 1,000 s/mm2, 25 gradient directions, 2mm res). Tracts were traced as previously described. Tracts are shown in Figure 1.
First, the mean FA value of each tract was compared between the ASD- and ASD+ groups at each time point using pair-wise t-test. Second, correlation coefficient between each tract and every other tract was calculated for the ASD- and ASD+ groups and the equality of two correlation matrices were tested by Jennrich's test at each time point. Lastly, each correlation coefficient was transformed to z-score using Fisher's r-to-z transformation and the differences of the pairwise inter-tract correlations between the ASD- and ASD+ groups were tested using z-test at each time point.
First, the mean FA value of each tract was compared between the ASD- and ASD+ groups at each time point using pair-wise t-test. Second, correlation coefficient between each tract and every other tract was calculated for the ASD- and ASD+ groups and the equality of two correlation matrices were tested by Jennrich's test at each time point. Lastly, each correlation coefficient was transformed to z-score using Fisher's r-to-z transformation and the differences of the pairwise inter-tract correlations between the ASD- and ASD+ groups were tested using z-test at each time point.
Results:
The mean FA value of each tract did not significantly differ between the ASD- and ASD+ groups at any time point (the lowest p = 0.18 FDR corrected). The examination of correlation coefficient matrices showed overall significant differences in the inter-tract correlation matrices between the ASD- and ASD+ groups at all time points (ps < .000). (Figure 2) Compared to the ASD- group, the ASD+ group had stronger inter-tract correlations at 6 months but had weaker inter-tract correlations at 24 months. Finally, the examination of discrete entries of the correlation matrices using z-test showed differences between the ASD- and ASD+ groups at 24 months. The inter-tract relatedness of left cingulum to bilateral arcuate and uncinate was decreased in the ASD+ group at 24 months (ps < .05 uncorrected). The inter-tract relatedness of the body of the corpus callosum to bilateral CST, ALIC, and ATR was also decreased in the ASD+ group at 24 months (ps < .05 uncorrected). (Figure 2)
Conclusions:
The results support the importance of investigating inter-tract relatedness given that the differences in white matter development between the ASD- and ASD+ group were observed in the inter-tract correlations rather than in the mean values of the DTI parameter. The decreased in inter-tract correlations in the ASD+ group at 24 months is consistent with previous findings that showed decreased network efficiency in ASD group. The decreased correlations observed in the cingulum, arcuate, and uncinate and the corpus callosum, CST, ALIC, and ATR are particularly interesting because those tracts have been related to the symptom of ASD. In future studies, the relation of the decreased correlations to ASD symptom severity can be examined.
Disorders of the Nervous System:
Neurodevelopmental/ Early Life (eg. ADHD, autism) 1
Lifespan Development:
Early life, Adolescence, Aging 2
Modeling and Analysis Methods:
Diffusion MRI Modeling and Analysis
Keywords:
Autism
MRI
White Matter
WHITE MATTER IMAGING - DTI, HARDI, DSI, ETC
1|2Indicates the priority used for review
Provide references using author date format
2. Lewis, J. D., Evans, A. C., Pruett, J. R., Botteron, K., Zwaigenbaum, L., Estes, A., ... & Dager, S. (2014). Network inefficiencies in autism spectrum disorder at 24 months. Translational psychiatry, 4(5), e388.
3. Dean III, D. C., Travers, B. G., Adluru, N., Tromp, D. P., Destiche, D. J., Samsin, D., ...Lainhart, J.E., & Alexander, A.L. (2016). Investigating the microstructural correlation of white matter in autism spectrum disorder. Brain connectivity, 6(5), 415-433.
4. Zhao, Y., Yang, L., Gong, G., Cao, Q., & Liu, J. (2022). Identify aberrant white matter microstructure in ASD, ADHD and other neurodevelopmental disorders: A meta-analysis of diffusion tensor imaging studies. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 113, 110477.