An fMRI Approach to Investigate Reading Deficits in Medulloblastoma Survivors
Poster No:
1031
Submission Type:
Abstract Submission
Authors:
Josue Luiz Dalboni da Rocha1, Ping Zou Stinnett2, Matthew Scoggins1, Heather Conklin1, Amar Gajjar1, Robert Ogg1, Ranganatha Sitaram1
Institutions:
1St. Jude Children's Research Hospital, Memphis, TN, 2St. Jude Children Research Hospital, Memphis, TN
First Author:
Co-Author(s):
Introduction:
Medulloblastoma (MB), the most common childhood malignant brain tumor, has a 5-year survival rate near 80% with recent progress in treatment. However, survivors often encounter cognitive challenges, including reading deficits (Zou et al., 2016). These deficits may manifest as difficulties in core processes supporting fluent reading-phonological awareness and rapid visual naming (Stappen et al., 2018). This study reports an fMRI study in MB survivors using Rapid Automatized Naming (RAN) and Orthographic Processing of Letters (OPL) tasks to investigate skills supporting reading (Powell et al., 2014).
Methods:
This study received ethical approval from the Institutional Review Board (IRB) and written informed consent was obtained from each participant. 50 MB survivors (mean age at fMRI = 14.4 years; SD = 4.1) and 96 healthy controls (mean age at fMRI = 14.3 years; SD = 5.1) participated in 3 visits for fMRI acquisition, spaced at 12-month intervals.
For RAN, we examined the Blood Oxygenation Level Dependent (BOLD) differences between the conditions of naming letters and naming colors. For OPL, we evaluated BOLD differences between identifying matching pairs of letters versus lines (Powell et al., 2014).
FMRI analysis utilized between-subject comparison based on a region-of-interest (ROI) activation score, in 3 key steps:
1) Within-Subject Analysis: Voxel-wise t-values were extracted for each subject by comparing two task conditions (i.e., for RAN: letter vs color; for OPL: letters vs lines). These t-values at each voxel were henceforth called voxel-wise activation scores.
2) ROI Averaging: Voxel-wise activation scores were averaged within predefined ROIs for each subject, resulting in a single activation score per ROI, henceforth called ROI activation score. The ROIs were 127 brain regions segmented from the Neuromorphometrics atlas (Tourville et al., 2010).
3) Between-Subject Analysis: An intersubject t-test was conducted to compare the ROI activation score. This step determined whether there were significant differences across subjects within each ROI, for the pair of task conditions in analysis.
For RAN, we examined the Blood Oxygenation Level Dependent (BOLD) differences between the conditions of naming letters and naming colors. For OPL, we evaluated BOLD differences between identifying matching pairs of letters versus lines (Powell et al., 2014).
FMRI analysis utilized between-subject comparison based on a region-of-interest (ROI) activation score, in 3 key steps:
1) Within-Subject Analysis: Voxel-wise t-values were extracted for each subject by comparing two task conditions (i.e., for RAN: letter vs color; for OPL: letters vs lines). These t-values at each voxel were henceforth called voxel-wise activation scores.
2) ROI Averaging: Voxel-wise activation scores were averaged within predefined ROIs for each subject, resulting in a single activation score per ROI, henceforth called ROI activation score. The ROIs were 127 brain regions segmented from the Neuromorphometrics atlas (Tourville et al., 2010).
3) Between-Subject Analysis: An intersubject t-test was conducted to compare the ROI activation score. This step determined whether there were significant differences across subjects within each ROI, for the pair of task conditions in analysis.
Results:
Medulloblastoma survivors exhibited an abnormality in BOLD response for letter versus color naming. This abnormality reached statistical significance (p-value < 0.05, Bonferroni corrected for 381 multiple comparisons) only during the third visit, predominantly observed in the right Fusiform Gyrus (p=0.02) and right Occipital Pole (p=0.04), as in Fig 1.
On the other hand, BOLD response abnormalities in medulloblastoma survivors for letter versus lines equal pair identification were statistically significant (p-value < 0.05, Bonferroni corrected for 127 multiple comparisons) in all visits. The top brain regions affected included right inferior occipital gyrus (p=6*10^-11), and bilateral angular gyrus [right (p=8*10^-7) and left (p=5*10^-6)], as in Fig 2.
On the other hand, BOLD response abnormalities in medulloblastoma survivors for letter versus lines equal pair identification were statistically significant (p-value < 0.05, Bonferroni corrected for 127 multiple comparisons) in all visits. The top brain regions affected included right inferior occipital gyrus (p=6*10^-11), and bilateral angular gyrus [right (p=8*10^-7) and left (p=5*10^-6)], as in Fig 2.
·Fig 1. Right Fusiform Gyrus and Right Occipital Pole: ROIs (in red) where abnormal BOLD response for letter versus color naming was observed.
·Fig 2. Left Angular Gyrus (left side), Right Angular Gyrus and Right Occipital Gyrus (right side): ROIs (in red) where abnormal BOLD response for letter versus lines was observed.
Conclusions:
The processing of letters in RAN appears to be sensitive to phonological processing and is often compromised in children with reading disorders. Conversely, the RAN of colors tends to be impaired in children with attention disorders. The delayed onset abnormality in letter versus color naming emphasizes the importance of long-term monitoring and cognitive support for this population. Additionally, the early detection of abnormalities in letters versus lines suggests potential targets for intervention and rehabilitation. Our findings reveal distinct patterns of BOLD response abnormalities in medulloblastoma survivors, which may provide insights into the neural mechanisms underlying their reading deficits. These findings may signify early orthographic processing issues, hinting at the likelihood that younger children, who rely more on orthographic decoding than phonological skills, may manifest these impairments earlier. Understanding these neurofunctional changes may guide the development of tailored strategies to enhance the cognitive outcomes of medulloblastoma survivors.
Disorders of the Nervous System:
Neurodevelopmental/ Early Life (eg. ADHD, autism) 2
Language:
Reading and Writing 1
Keywords:
Development
FUNCTIONAL MRI
Language
MRI
PEDIATRIC
1|2Indicates the priority used for review
Provide references using author date format
Stappen, C.V., et al. (2018), 'Phonological Awareness and Rapid Automatized Naming Are Independent Phonological Competencies With Specific Impacts on Word Reading and Spelling: An Intervention Study', Front. Psychol. 9:320. DOI: 10.3389/fpsyg.2018.00320
Tourville J., et al. (2010), 'Cortical Parcellation Protocol', http://neuromorphometrics.com/ParcellationProtocol_2010-04-05.PDF
Zou, P., et al. (2016), 'Functional MRI in medulloblastoma survivors supports prophylactic reading intervention during tumor treatment', Brain Imaging and Behavior 10, 258–271. https://doi.org/10.1007/s11682-015-9390-8