Cognitive Outcomes in Frontal Glioma Surgery Predicted by Pre-surgical Resting-State fMRI
Poster No:
1660
Submission Type:
Abstract Submission
Authors:
Kazuki Maruo1, MANUELA MORETTO2, Luca Zigiotto3, Stefano Tambalo4, Silvio Sarubbo5, Jorge Jovicich4
Institutions:
1Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Saxony, 2DEPARTMENT OF INFORMATION ENGINEERING, UNIVERSITY OF PADUA, PADUA, PADUA, 3S. Chiara” University-Hospital, Azienda Provinciale per i Servizi Sanitari, Trento, Trentino, 4Center for Mind/Brain Sciences (CIMeC), University of Trento, Trento, Trentino, 5Department of Neurosurgery, “S. Chiara” University-Hospital, Azienda Provinciale per i Servizi Sani, Trento, Trentino
First Author:
Co-Author(s):
Luca Zigiotto
S. Chiara” University-Hospital, Azienda Provinciale per i Servizi Sanitari
Trento, Trentino
S. Chiara” University-Hospital, Azienda Provinciale per i Servizi Sanitari
Trento, Trentino
Silvio Sarubbo
Department of Neurosurgery, “S. Chiara” University-Hospital, Azienda Provinciale per i Servizi Sani
Trento, Trentino
Department of Neurosurgery, “S. Chiara” University-Hospital, Azienda Provinciale per i Servizi Sani
Trento, Trentino
Introduction:
Brain surgery in glioma patients often leads to cognitive deficits, yet predicting their occurrence remains challenging. The impact of post-surgical cognitive deficits on treatment and quality of life is not well-defined. Resting-state fMRI (rs-fMRI) studies have shown distinctive amplitude of low frequency fluctuation (ALFF) differences in glioma patients compared to healthy controls [1-3], with ALFF in the contralateral parietal lobe linked to executive function [5]. This study explores pre-surgical ALFF and dynamic ALFF (dALFF) as predictors of post-surgical cognitive impairments.
Methods:
The study included 23 patients with frontal (12 left/11 right) high-grade gliomas (6F/17M, 52±16 years) who underwent pre-surgical and post-surgical neuropsychological evaluations including attentive and executive assessments. The pre-surgical MRI protocol, acquired on a 1.5T GE Healthcare scanner, included anatomical images and 12 minutes of rs-fMRI EPI scans (TR/TE 2600/45ms; 4mm³ isotropic). After standard rs-fMRI data preprocessing, voxelwise maps of ALFF and dALFF were computed within the 0.01-0.1Hz frequency range using the DPABI toolbox [6]. For dALFF, a sliding time window of 23 TR with a step size of 2 TR was used, yielding 124 windows. dALFF was derived as the voxelwise standard deviation of ALFF across all windows. We restricted the analyses to six networks4, covering the parietal lobe: Control-A and B (CON-A, CON-B), Dorsal Attention-A and B (DAN-A, DAN-B), Salience/Ventral Attention-A and B (SAN-A, SAN-B). To account for subject-specific variability in the BOLD signal, we computed ΔALFF and ΔdALFF as the difference between ipsilateral and contralateral average values within each network.
For each patient, we defined the attentive and executive impairment (Δattention, Δexecutive) by computing the difference between post- and pre-surgical scores. First, we computed Pearson's correlation between the impairments and ΔALFF/ΔdALFF. Second, we conducted two Partial Least Squares (PLS) regression analyses, where Δattention or Δexecutive served as outcome variables, while 8 pre-surgical variables served as predictors. Leave-one-out cross-validation was employed to compute the root mean squared error of prediction from the PLS models. Finally, for the best model, we computed Pearson's correlation between observed and predicted impairments. The workflow of the analyses is presented in Figure 1.
For each patient, we defined the attentive and executive impairment (Δattention, Δexecutive) by computing the difference between post- and pre-surgical scores. First, we computed Pearson's correlation between the impairments and ΔALFF/ΔdALFF. Second, we conducted two Partial Least Squares (PLS) regression analyses, where Δattention or Δexecutive served as outcome variables, while 8 pre-surgical variables served as predictors. Leave-one-out cross-validation was employed to compute the root mean squared error of prediction from the PLS models. Finally, for the best model, we computed Pearson's correlation between observed and predicted impairments. The workflow of the analyses is presented in Figure 1.
·Figure 1
Results:
Postsurgical cognitive deficits were associated with presurgical ipsi-contralateral ALFF differences (Figure 2A). In control networks, ΔALFF correlated significantly with Δexecutive function (CON-A: r = 0.46, p = 0.02; CON-B: r = 0.41, p = 0.04). In attentive networks, ΔALFF correlated with Δattention (DAN-A: r = 0.42, p = 0.04; DAN-B: r = 0.41, p = 0.04). No significant associations were found for ΔALFF in SAN-A and SAN-B.
Postsurgical cognitive deficits could be predicted by presurgical ipsi-contralateral ALFF differences (Figure 2B). The best models for predicting Δexecutive using CON-A and CON-B consisted of the highest loadings attributed to ΔALFF and ΔdALFF in the first component. The correlation between observed and predicted Δexecutive results in r = 0.48, p = 0.02 (CON-A) and r = 0.5, p = 0.01 (CON-B). The optimal models for predicting Δattention using DAN-A and DAN-B comprised the highest loadings attributed to ΔALFF and ΔdALFF in the first component. The correlation between observed and predicted Δattention resulted in r = 0.39, p = 0.06 (DAN-A) and r = 0.49, p = 0.01 (DAN-B).
Postsurgical cognitive deficits could be predicted by presurgical ipsi-contralateral ALFF differences (Figure 2B). The best models for predicting Δexecutive using CON-A and CON-B consisted of the highest loadings attributed to ΔALFF and ΔdALFF in the first component. The correlation between observed and predicted Δexecutive results in r = 0.48, p = 0.02 (CON-A) and r = 0.5, p = 0.01 (CON-B). The optimal models for predicting Δattention using DAN-A and DAN-B comprised the highest loadings attributed to ΔALFF and ΔdALFF in the first component. The correlation between observed and predicted Δattention resulted in r = 0.39, p = 0.06 (DAN-A) and r = 0.49, p = 0.01 (DAN-B).
·Figure 2
Conclusions:
Our results highlight the promising predictive power of pre-surgical ALFF and dALFF for post-surgical executive and attentional impairments. When the parietal lobe ipsilateral to the frontal glioma shows higher ALFF than the contralateral lobe, cognitive functions are likely preserved. This enables pre-surgical identification of high-risk patients, guiding neurosurgical planning and rehabilitation decisions, ultimately improving long-term clinical outcomes.
Higher Cognitive Functions:
Higher Cognitive Functions Other 2
Modeling and Analysis Methods:
Task-Independent and Resting-State Analysis 1
Keywords:
Data analysis
FUNCTIONAL MRI
Machine Learning
Other - Brain Tumor
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
[2] Ge, H. et al. (2022). Synergetic reorganization of the contralateral structure and function in patients with unilateral frontal glioma. Frontiers in neuroscience, 16, 1016693.
[3] Gupta, L. et al. (2018). Advanced and amplified BOLD fluctuations in high-grade gliomas. Journal of magnetic resonance imaging, 47(6), 1616–1625.
[4] Kong, R. et al. (2021). Individual-Specific Areal-Level Parcellations Improve Functional Connectivity Prediction of Behavior. Cerebral cortex, 31(10), 4477–4500.
[5] Liu, Y. et al. (2020). Structural and Functional Reorganization Within Cognitive Control Network Associated With Protection of Executive Function in Patients With Unilateral Frontal Gliomas. Frontiers in oncology, 10, 794.
[6] Yan, C. G. et al. (2016). DPABI: Data Processing & Analysis for (Resting-State) Brain Imaging. Neuroinformatics, 14(3), 339–351.