Purpose

Blood Oxygen Level Dependent functional magnetic resonance imaging (BOLD fMRI) is an indirect measure of neuronal activity. BOLD fMRI detects alterations in deoxyhemoglobin concentration influenced by hemodynamic factors that occur in response to neuronal activity. The coupling between neuronal activity and the hemodynamic changes occurring in the adjacent vasculature is often disrupted in patients with brain tumors, resulting in abnormally decreased ipsilesional BOLD fMRI activation¹. In this study we propose a novel method of correcting for the effects of NVU on task-based fMRI (tbfMRI) activation using the resting state BOLD frequency domain metric ALFF (amplitude of low-frequency fluctuation²). We will apply this method specifically to correct for NVU affecting the primary motor cortex.

Methods

Twelve de novo brain tumor patients who underwent clinical fMRI exams including tbfMRI and resting state fMRI (rsfMRI) on a 3T MRI system were included in this IRB-approved study. Each patient displayed decreased/absent tbfMRI activation in the primary ipsilesional sensorimotor cortex in the absence of corresponding motor deficit or suboptimal task performance, consistent with NVU.¹ Imaging was performed on a 3.0 T Siemens Trio MRI with a 12-channel head matrix coil using a 3D T1 MPRAGE (TR=2300 ms, TI= 900 ms, TE= 3.5 ms, 9° FA, 24-cm FOV, 256x 256x176 matrix, slice thickness 1 mm) for structural imaging and multiple 2D GE-EPI T2* weighted BOLD sequences for both task & resting functional imaging (TR=2000 ms, TE=30 ms, 90° FA, 24-cm FOV, 64x64x33 matrix, 4 mm slice thickness with 1 mm gap between slices, interleaved acquisition). 180 volumes were acquired in 6 minute long rsfMRI scan. A vertical tongue movement task and a bilateral simultaneous sequential finger tapping task (each 3 minutes duration with alternating 30 second blocks of movement and rest) were used for tbfMRI. Instructions for all tasks were visually cued. SPM12 was used for preprocessing of tbfMRI & rsfMRI data (slice timing correction, realignment, normalization to MNI space at 2mm voxel resolution, and spatially smoothing using a 6 mm FWHM Gaussian kernel). Z-score maps for the motor tasks were obtained from the general linear model (GLM) analysis using standard SPM canonical HRF (reflecting motor activation vs. rest). Pre-processed rsfMRI data were analyzed using the REST(version 1.8)³ toolkit. After de-trending for removal of systematic linear trend and low frequency (0.01-0.08 Hz) bandpass filtering ALFF maps were calculated from rsfMRI data. Pre- and post- central gyri were automatically parcellated using an Automated Anatomical Labeling (AAL) template^4,5 for each patient. CL (contralesional) and IL (ipsilesional) ROIs circumscribing the combination of pre- and post- central gyri (CG) were obtained for each slice. Consecutive axial sections were evaluated along the z-axis along the craniocaudal length of the tumors . Identical ROIs were used for analysis of both maps (tbfMRI & ALFF). Mean of ALFF (mALFF) in CL ROI is calculated. Motor activation maps were further analyzed using Amplitude Measured as a Percentage of Local Excitation (AMPLE) thresholding of 50% (i.e., only voxels with Z scores above 50% of a local cluster Z score maximum were considered “active”)⁶. Ipsilesional voxels with sub-threshold activation (between 25% to 50% of a local cluster Z score maximum) in motor task activation map were considered to have falsely reduced activation due to NVU and were subsequently corrected by enhancing their voxel Z-score values by a factor representing the quotient obtained by dividing the contralesional mALFF by the individual ipsilesional voxel ALFF value. For each patient, motor activation maps pre- and post ALFF correction were obtained and their difference was evaluated.

Results

Motor activation maps for two tumor patients as pre- ALFF correction, post- ALFF correction, as well as their difference map, obtained by subtracting the pre-correction map from the post-correction map, are depicted in Figure 1. The number of voxels in the ipsilesional (IL) ROI for each subject pre- and post- ALFF correction was obtained and a group analysis was performed which reveals significantly increased number of voxels displaying ipsilesional motor cortical activation post-correction compared to pre-ALFF correction (p=0.00002 using paired t-test) .

Discussion

In this preliminary study we demonstrate the feasibility of optimization of primary motor cortical activation in the setting of perirolandic tumor-induced NVU through use of a novel resting state BOLD ALFF-based correction algorithm.

Conclusion

This study is the first to demonstrate successful application of a novel resting state ALFF-based correction algorithm for optimization of motor cortical activation in cases of tumor-related neurovascular uncoupling.

Acknowledgements

This work is partially supported by NIH grant R42 CA173976-02 (NCI).