A 7T fMRI Pilot Study on Colorectal Cancer-Related Cognitive Impairment
Poster No:
915
Submission Type:
Abstract Submission
Authors:
Sharon Chao1, Nathan Nguyen1, Tara Riddle2, Debapriya Dutta2, Ashley Hill2, Robert Yu2, Suparna Mantha2, Kendrith Rowland2, Zhaoyue Shi2
Institutions:
1Carle Illinois College of Medicine, Urbana, IL, 2Carle Foundation Hospital, Urbana, IL
First Author:
Co-Author(s):
Introduction:
Cognitive impairment affecting executive functions such as information processing speed and working memory has been observed in colorectal cancer patients following diagnosis with no significantly added effects of adjuvant therapy.[1,2] Colorectal cancer-related cognitive impairment (CRCI) has already been studied in the literature using cognitive assessments such as objective neuropsychological testing and subjective self-reported questionnaires.[1,2,3] However, little is known regarding the neural mechanism of colorectal CRCI which makes the development of evidence-based therapies currently unattainable. This study aims to utilize 7T magnetic resonance imaging (MRI) to (1) compare the resting state functional connectivities (rsFC) between early-stage colorectal cancer survivors and healthy controls (HCs), and (2) identify specific brain regions and networks that are associated with CRCI by examining relationships between rsFC node strength and cognitive assessment scores.
Methods:
Ten participants included five colorectal cancer survivors (stage I or II, aged 66.2 ± 6.9 years) who had surgery without adjuvant therapy within six months and five age/sex/education matched HCs (aged 64.4 ± 5.4 years). Each underwent cognitive assessments and MRI on the same day. All participants had no prior malignancy or history of neurological disorders. Cognitive assessments included objective neuropsychological tests such as the Trail Making Test (TMT), Hopkins Verbal Learning Test (HVLT), and Controlled Oral Word Association Test (COWAT), as well as subjective self-reported questionnaires including the Functional Assessment of Cancer Therapy-Cognitive Function (FACT-Cog), Beck Depression Inventory (BDI), and State-Trait Anxiety Inventory (STAI). MRI was performed on a Siemens MAGNETOM Terra 7T scanner. MRI acquisition protocol consisted of a T1-weighted MP2RAGE (TR = 4530 ms, TE = 2.32 ms, 0.7 mm3 isotropic) and T2*-weighted BOLD fMRI (TR = 1000 ms, TE = 25 ms, 1.6 mm3 isotropic, 485 volumes).
Standard preprocessing steps included distortion correction, motion correction, removal of physiological noise, temporal filtering (0.01-0.1 Hz), spatial smoothing (5mm FWHM Gaussian kernel), and spatial normalization to the MNI152 template (2 x 2 x 2 mm3). Regions of interest (ROIs) within the central executive network (CEN) and default mode network (DMN) as provided in Atlas55+ were selected for further analysis.[4] The average time series for each ROI was calculated across all voxels within that ROI. rsFC node strength was calculated as the average Pearson's correlation coefficient among the time series of a given ROI and the time series of all other ROIs in the network.
Standard preprocessing steps included distortion correction, motion correction, removal of physiological noise, temporal filtering (0.01-0.1 Hz), spatial smoothing (5mm FWHM Gaussian kernel), and spatial normalization to the MNI152 template (2 x 2 x 2 mm3). Regions of interest (ROIs) within the central executive network (CEN) and default mode network (DMN) as provided in Atlas55+ were selected for further analysis.[4] The average time series for each ROI was calculated across all voxels within that ROI. rsFC node strength was calculated as the average Pearson's correlation coefficient among the time series of a given ROI and the time series of all other ROIs in the network.
Results:
The survivor group had significantly higher perceived cognitive impairment (p<0.05) and lower information processing speed (p<0.05) compared to HCs, as assessed by the FACT-Cog and TMT-A, respectively. There were no significant differences in the HVLT, COWAT, BDI or STAI scores between the two groups. Additionally, the survivor group showed significantly reduced rsFCs in the medial prefrontal cortex (mPFC), anterior cingulate cortex (ACC), and precuneus within the DMN (Fig. 1A), as well as significantly reduced rsFCs in the left middle frontal gyrus (MFG), left inferior parietal gyrus (IPG), and right superior parietal gyrus (SPG) within the CEN compared to HCs (Fig. 1B). Furthermore, among all participants, the rsFC node strengths of the mPFC and ACC were linearly correlated with perceived cognitive impairment scores, and the rsFC node strengths of the SPG and IPG were linearly correlated with TMT-A scaled scores (Fig. 2).
Conclusions:
This ongoing study identified altered brain regions and networks in early-stage colorectal cancer survivors that may be associated with CRCI. Further data collection is currently underway. Findings will help provide biological targets for developing pharmacological or neuro-cognitive interventions for treating colorectal CRCI.
Higher Cognitive Functions:
Executive Function, Cognitive Control and Decision Making 1
Learning and Memory:
Working Memory
Modeling and Analysis Methods:
fMRI Connectivity and Network Modeling
Task-Independent and Resting-State Analysis 2
Keywords:
Cognition
FUNCTIONAL MRI
HIGH FIELD MR
Other - Cancer-Related Cognitive Impairment
1|2Indicates the priority used for review
Provide references using author date format
2. Visovatti, Moira A., et al. "Assessment of cognitive impairment and complaints in individuals with colorectal cancer." Oncology nursing forum. Vol. 43. No. 2. NIH Public Access, 2016.
3. Vardy, J., et al. "Cognitive function and fatigue after diagnosis of colorectal cancer." Annals of oncology 25.12 (2014): 2404-2412.
4. Doucet, Gaelle E., et al. "Atlas55+: brain functional atlas of resting-state networks for late adulthood." Cerebral Cortex 31.3 (2021): 1719-1731.