Free Water Elimination Tractometry Reveals Local and Remote White Matter Disruption in Diffuse Gliomas

Free Water Elimination Tractometry Reveals Local and Remote White Matter Disruption in Diffuse Gliomas | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Free Water Elimination Tractometry Reveals Local and Remote White Matter Disruption in Diffuse Gliomas Daniel J. Zhou, Kelly Chang, Marc Jaskir, Kathryn A. Davis, Joel M. Stein, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7982561/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 10 Dec, 2025 Read the published version in Journal of Neuro-Oncology → Version 1 posted 12 You are reading this latest preprint version Abstract Purpose To apply free water elimination (FWE) tractometry to a large real-world clinical imaging dataset to quantify pathology-specific patterns of white matter involvement and peritumoral tissue disruption in diffuse gliomas. Methods The UCSF Preoperative Diffuse Glioma MRI dataset was analyzed using FWE tractometry. Twenty major white matter tracts were reconstructed and each divided into 100 equidistant nodes. Direct tumor involvement was quantified across enhancing tumor, necrotic core, and edema regions. Remote white matter tissue properties were assessed through hemispheric asymmetry analysis of free water-corrected fractional anisotropy (FW-FA), mean diffusivity (FW-MD), and free water fraction (FWF) in non-tumor involved regions at standardized distances from radiological tumor margins. Results 459 patients with unilateral glioma were included (361 glioblastoma, 87 astrocytoma, 11 oligodendroglioma). Glioblastoma demonstrated greater direct white matter involvement in enhancing tumor and necrotic core compared to astrocytoma and oligodendroglioma (q < 0.001, q = 0.01, respectively). Beyond radiological tumor margins, glioblastoma and astrocytoma exhibited decreased FW-FA, while oligodendroglioma showed increased FW-FA (q = 0.008, q = 0.04, respectively). Distance-based analysis revealed that this effect was most prominent in the proximal peritumoral region and diminished with increasing distance from tumor margins. Conclusion Using FWE tractometry on a large clinical repository, we identified distinct pathology-specific patterns of white matter disruption. Glioblastoma showed extensive direct involvement and peritumoral microstructural changes, while oligodendroglioma demonstrated relatively preserved white matter architecture near tumor margins. These patterns reflect expected biological differences and provide a reproducible framework for characterizing extent of white matter involvement, with potential applications in presurgical planning and understanding recurrence patterns. Glioma white matter diffusion MRI tractometry free water elimination Figures Figure 1 Figure 2 Figure 3 Introduction Large-scale clinical imaging repositories have transformed neuro-oncology research by providing statistical power to detect subtle biological differences obscured in smaller cohorts. Adult-type diffuse gliomas are the most common primary malignant brain tumors and remain difficult to treat [ 1 ]. Understanding how different glioma subtypes interact with brain white matter could contribute to surgical planning and predicting recurrence patterns. However, systematic comparisons of white matter involvement patterns across molecularly-defined glioma subtypes have been limited by the need for large cohorts with comprehensive molecular characterization. Histopathological studies have demonstrated that glioma cells disseminate faster along white matter bundles than through cortex, using myelinated fibers as scaffolds for invasion [ 2 – 5 ]. Tissue biopsy studies have revealed that viable tumor cells remain present beyond enhancing regions visible on neuroimaging and intraoperative surgical margins [ 6 , 7 ]. Glioblastoma disrupts structural brain networks and favors recurrence along impacted white matter pathways, while lower grade gliomas demonstrate more variable infiltration patterns [ 4 , 8 – 11 ]. Conventional diffusion tensor imaging (DTI) and tractography studies of glioma are limited by tumor-induced artifacts, including vasogenic edema, mass effect, and variable infiltration, that alter diffusion measures and impede tractography [ 12 , 13 ]. Free water elimination methods separate tissue-specific diffusion from free water contamination, enabling more accurate assessment of white matter tissue properties [ 14 – 16 ]. MRI-based tractometry samples diffusion metrics at standardized points along anatomically validated fiber bundles, offering a reproducible framework for quantifying white matter involvement across institutions [ 17 , 18 ]. The University of California San Francisco Preoperative Diffuse Glioma (UCSF-PDGM) MRI dataset is one of the largest publicly available preoperative imaging repositories with standardized protocols and expert tumor segmentations, combining high-angular resolution diffusion imaging with comprehensive molecular characterization across gliomas [ 19 , 20 ]. Using this dataset, we aim to: (1) quantify direct tumor involvement of white matter tracts across glioma subtypes, and (2) assess white matter tissue property changes along white matter tracts beyond radiologic tumor margins. Materials and Methods Dataset and Participants The UCSF-PDGM dataset is a publicly available collection of preoperative imaging and clinical data from adult patients with diffuse gliomas [ 19 ]. The dataset includes co-registered MRI T1 and T2 imaging, DTI, and segmented tumor components (enhancing tumor, necrotic core, and edema/infiltration regions) along with accompanying clinical and pathologic data. These protocols have been previously described; in brief, diffusion-weighted MRI was collected at 2 mm 3 isotropic resolution, 55 directions, with b = 2,000 s/mm 2 [ 21 ]. For white matter asymmetry analyses, we included patients with unilateral white matter tract involvement. Bilateral or midline tumors were excluded to avoid confounding asymmetry measurements. The dataset provided diagnoses labeled according to the WHO 2021 classification. After correspondence with the dataset authors, cases labeled “astrocytoma IDH -wildtype” were confirmed to represent tumors with incomplete molecular testing and were therefore reclassified as Not Elsewhere Classified (NEC). To ensure diagnostic accuracy for pathology-specific analyses, we excluded NEC cases and focused on patients with definitive histopathologic and molecular diagnoses of glioblastoma, astrocytoma, or oligodendroglioma. Tractography and Free Water Elimination Diffusion images were preprocessed using QSIPrep, which performs motion correction, distortion correction, denoising, and registration to anatomical space [ 22 ]. Tractography reconstruction, tractometry, and free water elimination (FWE) analysis were performed as previously described, using the open-source pyAFQ software version 2.1, which relies on techniques implemented in the DIPY software [ 15 , 17 ]. Tractography was performed in native subject space, and tract identification was performed using pyAFQ's anatomically-defined waypoint region of interest approach. Twenty major white matter tracts were reconstructed bilaterally, including the anterior thalamic radiation, arcuate fasciculus, posterior arcuate fasciculus, cingulate section of the cingulum bundle, corticospinal tract, inferior fronto-occipital fasciculus, inferior longitudinal fasciculus, superior longitudinal fasciculus, uncinate fasciculus, and vertical occipital fasciculus. Each tract was divided into 100 equidistant nodes along its length, standardizing sampling to equivalent anatomical positions across subjects with varying brain sizes and tract lengths. Tract profiles were generated by sampling diffusion metrics along the trajectory of each tract. Because the data had measurements with only one diffusion weighting, we used an implementation of a free-water DTI model that uses spatial continuity constraints to regularize model fit [ 16 , 23 ]. FWE was applied to separate tissue-specific diffusion from free water contamination, yielding FW-corrected fractional anisotropy (FA), mean diffusivity (MD), and free water fraction (FWF) metrics (Fig. 1 ). All reconstructed tracts underwent automated quality control procedures implemented in pyAFQ, including checks for minimum streamline count, tract length, and anatomical plausibility. Representative reconstructions from a subset of cases spanning all pathology types were visually inspected by a neurologist (D.Z.) to confirm anatomical plausibility and verify that tract trajectories aligned with expected neuroanatomical pathways. Analysis of Direct Tumor Involvement Left and right hemisphere homologues of each tract were averaged together for analysis, yielding 10 unique bilateral tract groups (e.g., left and right inferior longitudinal fasciculi analyzed as one tract group). A tract node was classified as tumor-involved if tumor components (enhancing tumor, necrotic core, or edema) occupied more than 5% of that node's spatial extent. This threshold was chosen to provide objective criteria for distinguishing tumor-involved from unaffected tissue while maintaining sensitivity for subtle infiltrative changes, consistent with approaches in probabilistic tractography studies that demonstrate the importance of permissive thresholds in tumor environments [ 24 , 25 ]. Subject-level involvement was binarized as involved or not involved based on whether the tract had at least one node exceeding the 5% threshold. Tract-level involvement was calculated as the proportion of nodes exceeding the 5% threshold within each tract. Analysis of Tumor Involvement Beyond Radiologic Tumor Margins To compare white matter tissue properties in the hemisphere least affected by the tumor, we averaged the free water-corrected (FW-) FA, MD, and FWF across all nodes from the contralateral hemisphere for each patient. Then, to quantify the differences between the ipsilateral and contralateral white matter beyond the tumor margins, we computed a directional percent asymmetry at nodes with < 5% combined tumor involvement (enhancing tumor, necrotic core, and edema/infiltration), pairing nodes by index across hemispheres within the same tract, using the following equation: $$\:Asymmetry\:\left(\%\right)\:=\:\frac{ipsilateral\:-\:contralateral}{(ipsilateral\:+\:contralateral)/2}\:\times\:\:100\%$$ Therefore, positive values indicated higher values in the ipsilateral white matter. For whole-tract asymmetry comparisons, node-level asymmetries were averaged within the tract and then collapsed to a single subject-level mean per metric. Then, to test for peritumoral gradients, we computed the shortest along-tract node distance from each ipsilateral node to the nearest tumor edge, defined by any radiologic tumor component. Subject-level mean asymmetries were then compared within distance segments of five nodes (1–5, 6–10, 11–15, 16–20, and ≥ 21 nodes from the tumor edge), yielding one value per segment, metric, and subject. Statistical Analysis Analyses were conducted in Python (pandas, NumPy, SciPy, statsmodels, matplotlib). All statistical tests were two-sided, comparing imaging measures across pathology types (glioblastoma, astrocytoma, oligodendroglioma). Continuous variables were reported as median (IQR) or mean ± SD, and categorical variables as counts (percent). Normality was assessed with Shapiro-Wilk tests. Three-group comparisons used: (1) Kruskal-Wallis for overall significance; (2) pairwise Mann-Whitney U tests; (3) linear or logistic regression with dummy coding (glioblastoma reference), adjusted for age and sex where appropriate. Multiple comparisons were controlled using Benjamini-Hochberg false discovery rate (FDR), stratified by test family. Statistical significance was denoted as p < 0.05 for individual tests, q < 0.05 for FDR-corrected comparisons within test families. Tumor proportions were calculated as volume normalized to total brain volume, excluding ventricles. FWF-edema correlation used Spearman's coefficient. Direct tract involvement used logistic regression for binary outcomes and ordinary least-squares (OLS) regression for the proportion of involved nodes. Contralateral hemisphere comparisons used linear regression adjusted for age and sex. Whole-tract and distance-based asymmetry analyses (1–5, 6–10, 11–15, 16–20, ≥ 21 nodes from margins) used Kruskal-Wallis and OLS regression without age/sex covariates since asymmetry is computed within subjects. Results Demographic and Clinical Features The UCSF-PDGM dataset contained preoperative MRI scans from 495 unique patients, of whom 459 (93%) were included in the study. Of the excluded patients, 24 (5%) had tumors classified as Not Elsewhere Classified due to incomplete molecular testing, and 12 (2%) had bilateral white matter tract involvement, precluding hemispheric asymmetry analyses. Of the patients included, 361 (79%) were classified as glioblastoma, 87 (19%) as astrocytoma, and 11 (2%) as oligodendroglioma; mean age at time of MRI scan was 57 years (range 19–97), and 184 (40%) were female (Table 1 ). Patients with glioblastoma were significantly older than patients with astrocytoma or oligodendroglioma (both p < 0.001). No significant differences in sex distribution were observed among groups (p = 0.94). Glioblastoma had significantly larger enhancing tumor and necrotic core volumes compared to astrocytomas and oligodendrogliomas (both p < 0.001). Tractography Quality with Free Water Elimination The white matter tracts were successfully reconstructed bilaterally and passed automated quality control procedures implemented in pyAFQ. Visual inspection of tractography with and without FWE revealed improved tract reconstruction in tumor-affected regions, with FWE enabling tract generation through edematous areas where conventional DTI showed sparse or failed reconstructions (representative example in Fig. 2 ). Free water fraction correlated strongly with edema involvement (ρ = 0.55, p < 0.001), suggesting that FWE successfully detected free water in edematous regions and enabling separation of tissue-specific diffusion from free water contamination (Supplementary Fig. S2 E). Direct Tumor Involvement of White Matter Tracts Quantitative analyses of white matter tract involvement revealed significant pathology-specific differences (Table 2 ). Across all white matter tracts, glioblastoma showed greater mean involvement than astrocytoma and oligodendroglioma for enhancing tumor (5.4% vs. 1.3% vs. 0.0%, q < 0.001), necrotic core (1.7% vs. 0.7% vs. 0.0%, q = 0.01), and combined tumor components (17.3% vs. 15.3% vs. 5.8%, p = 0.001). Edema/infiltration involvement was substantial across all pathology types but remained significantly higher in glioblastoma and astrocytoma compared to oligodendroglioma (16.0% vs. 15.1% vs. 5.8%, q = 0.006). Examination of direct tract involvement patterns at the subject level revealed pathology-dependent differences (Supplementary Fig. S2 -S3). For enhancing tumor involvement, glioblastoma showed widespread tract involvement compared to astrocytoma, with the most pronounced differences in the inferior fronto-occipital fasciculus (56% vs. 20% of subjects, q < 0.001), arcuate fasciculus (45% vs. 9%, q < 0.001), and inferior longitudinal fasciculus (40% vs. 8%, q < 0.001). Necrotic core involvement followed similar patterns, with glioblastoma showing significantly higher rates in the arcuate fasciculus (15% vs. 3%, q = 0.01), inferior fronto-occipital fasciculus (21% vs. 8%, q = 0.009), and inferior longitudinal fasciculus (15% vs. 2%, q = 0.01) compared to astrocytoma. Remote White Matter Tissue Properties FW-corrected FA, MD, and FWF values were calculated for each tract node to assess white matter tissue properties in regions not directly inside the tumor (< 5% for all components combined). Distributional analysis confirmed normality assumptions for FA and FWF asymmetry measures but revealed significant deviations from normality for MD asymmetry (Shapiro-Wilk p < 0.001), justifying the application of non-parametric statistical approaches for unadjusted analyses. Analysis of the contralateral hemisphere to the tumor revealed no significant differences in any diffusion metric after adjusting for age and sex (Supplementary Table S1 ), suggesting that baseline hemispheric differences would not confound subsequent hemisphere asymmetry analyses. Whole-tract asymmetry analysis of all nodes without tumor involvement revealed pathology-specific patterns (Supplementary Table S2 ). Oligodendroglioma demonstrated significantly higher FA asymmetry (+ 2.0 ± 2.2%) compared to glioblastoma (-3.0 ± 7.1%, q = 0.008) and astrocytoma (-2.2 ± 6.7%, q = 0.04). Both glioblastoma and astrocytoma showed negative FA asymmetry and positive MD asymmetry, indicating lower anisotropy and higher diffusivity in the ipsilateral hemisphere. In contrast, oligodendroglioma showed positive FA asymmetry and negative MD asymmetry, suggesting relatively preserved white matter microstructure in the tumor-bearing hemisphere. No significant differences were observed for FWF across any pathology comparisons. Distance-based asymmetry analysis revealed pathology-dependent patterns at specific distances from radiologic tumor margins (Fig. 3 ). Oligodendroglioma demonstrated significantly higher FA asymmetry compared to glioblastoma at 1–5 nodes (+ 4.9 ± 9.0% vs. -3.9 ± 9.8%, q = 0.04) and 6–10 nodes (+ 5.6 ± 9.0% vs. -3.8 ± 9.7%, q = 0.03). The magnitude of this preservation effect diminished with increasing distance from tumor margins, with no significant differences observed beyond 10 nodes. No significant differences were observed between astrocytoma and glioblastoma at any distance, or for MD or FWF across any pathology comparisons. Discussion This study demonstrates how a large-scale, standardized clinical imaging repository enables detection of white matter disruption patterns obscured in smaller cohorts. The application of open-source analytical methods to this publicly available dataset exemplifies how well-curated clinical repositories enable reproducible, multi-institutional investigation of tumor biology. By applying FWE tractometry to the UCSF-PDGM dataset, the largest publicly available glioma cohort with standardized diffusion imaging and comprehensive molecular characterization, we quantified both direct tract involvement and remote peritumoral white matter changes across glioblastomas, astrocytomas, and oligodendrogliomas. Glioblastoma showed extensive direct tract involvement and microstructural disruption extending beyond radiologic margins, while oligodendroglioma demonstrated relative preservation of white matter microstructure within 1–10 nodes of tumor edges. These distinct patterns may inform presurgical risk assessment and radiation treatment planning, as peritumoral tract disruption could influence functional outcomes and guide therapeutic targeting beyond visible tumor margins. White matter involvement across pathology types reflect expected differences in tumor biology. Glioblastoma showed the most extensive involvement across contrast-enhancing and necrotic components, reflecting its aggressive properties and extensive invasion along white matter pathways. Astrocytoma showed intermediate involvement patterns, with lower enhancing tumor and necrotic core involvement than glioblastoma but more extensive involvement than oligodendroglioma. Understanding these pathology-specific invasion patterns could aid in treatment planning decisions, including surgical margin determination and radiation field design and inform prognosis based on the degree of white matter involvement at presentation. Analysis of remote white matter revealed pathology-specific peritumoral effects extending beyond visible tumor margins. Oligodendroglioma demonstrated positive FA asymmetry and negative MD asymmetry, suggesting better-preserved ipsilateral white matter microstructure, compared to both glioblastoma and astrocytoma at the whole-tract level. This preservation was most pronounced within the proximal 10% of tract length from radiologic tumor margins, suggesting that oligodendroglioma’s slower growth and less infiltrative behavior result in less disruption of adjacent white matter. In contrast, both glioblastoma and astrocytoma showed negative FA asymmetry and positive MD asymmetry, consistent with microstructural disruption in the tumor-bearing hemisphere even beyond radiologically visible boundaries. While FWE isolated tissue-specific diffusion from free water contamination, the observed microstructural changes could reflect several processes: subclinical tumor infiltration beyond MRI detection limits, white matter deformation from mass effect, alterations in the tissue microenvironment from peritumoral processes, or secondary network effects from tract disruption. Although our cross-sectional imaging data did not distinguish among these possibilities, the distance-dependent gradient, where disruption was most pronounced proximally and attenuated with distance from tumor margins, raises the possibility that these signatures may capture spatial gradients of tumor infiltration. If these patterns do reflect occult infiltrative extent, they could potentially inform estimates of subclinical disease burden relevant to surgical planning and recurrence risk. Validating this hypothesis would require prospective studies correlating preoperative tractometry findings with histopathological analysis of tissue sampled beyond radiographic margins and longitudinal mapping of recurrence patterns. Our finding of no significant contralateral hemisphere differences contrasts with prior DTI studies reporting increased MD and reduced FA in the contralateral hemisphere [ 26 , 27 ]. This discrepancy may reflect differences in patient populations, analysis methodologies, or our application of FWE techniques to better isolate tissue-specific diffusion properties. While the absence of contralateral effects could reflect limited statistical power or true sparing of the contralateral hemisphere, it suggests that the ipsilateral asymmetries are not fully attributable to global bilateral processes, supporting the spatial specificity of our distance-dependent findings. FWE tractometry addressed key limitations of conventional DTI in tumor environments. This correction is particularly important in glioma imaging, where vasogenic edema and infiltration confound standard DTI metrics [ 14 – 16 ]. By isolating tissue microstructure from free water effects, FWE enabled more valid comparisons of white matter integrity across pathology types with different edema burdens. The strong correlation between FWF and edema validates the methodological approach and supports the biological interpretability of our tissue-specific measurements. The study has several limitations that should be considered. The cross-sectional design precludes assessment of temporal changes in white matter properties or correlation with postoperative outcomes such as survival or functional status. The UCSF-PDGM dataset represents a single-institution cohort with specific imaging protocols, which may limit generalizability to other populations or scanners. However, the standardized protocols within this cohort enabled rigorous quantitative comparison without confounding scanner effects, a key advantage over heterogeneous multi-center datasets. Our use of single-shell data for FWE had inherent limitations compared to multi-shell acquisition. The spatial regularization constraints required for single-shell FWE may reduce sensitivity to subtle tissue changes [ 23 , 28 ]. Our exclusion of patients with bilateral or midline tumors, while necessary for proper asymmetry analysis, could introduce selection bias toward more focal lesions limiting generalizability to bilateral tumors. The dataset was curated prior to the 2021 WHO molecular classification. While we excluded cases with incomplete molecular testing to ensure diagnostic accuracy, this reduced the sample sizes for some pathology comparisons. The small oligodendroglioma sample limits statistical power for detecting subtle effects, though the large differences observed compared to glioblastoma may be biologically meaningful. The 5% probability threshold for defining tumor involvement was chosen to provide objective criteria for analysis, though the optimal threshold for distinguishing tumor-involved from unaffected tissue remains to be established. Finally, our imaging-based findings lacked clinical data such as functional outcomes, cognitive assessments, or histopathological validation of white matter involvement patterns, limiting the ability to establish clinical correlations of the observed diffusion changes. This study demonstrates how a large-scale, publicly available clinical imaging dataset can reveal pathology-specific patterns of white matter disruption that could inform tumor biology and clinical decision-making. By applying FWE tractometry to the UCSF-PDGM cohort, we established a reproducible quantitative framework that addresses key limitations of conventional DTI in tumor environments and can be applied across institutions using standardized diffusion protocols. The pathology-specific patterns observed could provide imaging-based correlates of known biological differences with potential implications for surgical and radiation therapy planning. The open-source, reproducible methodology established here exemplifies how well-curated imaging repositories enable methodological innovation and biological discovery in neuro-oncology. Future integration of these tract-based imaging metrics with longitudinal clinical data from cancer registries and prospective surgical cohorts will validate their prognostic utility and advance their translation into clinical decision support tools. This work provides a foundation for multi-institutional efforts to standardize quantitative white matter assessment in glioma and demonstrates the continued value of investment in open neuroscience data infrastructure. Declarations Funding: D.Z.’s work was funded by the National Institute of Neurological Disorders and Stroke (5T32NS091008-07) and the American Epilepsy Society Research and Training Fellowship for Clinicians (1282834). N.S.’s work was funded by the National Institute of Neurological Disorders and Stroke (K99NS138680). AR and KC’s work was funded by National Institutes of Health grants MH121868, MH121867, R25MH112480, R01AG060942, and U19AG066567, and R01EB027585, as well as by National Science Foundation grant 1934292. Open-source software development was also supported by the Chan Zuckerberg Initiative’s Essential Open Source Software for Science program, the Alfred P. Sloan Foundation, and the Gordon & Betty Moore Foundation. Competing Interests: All authors report no competing interest or conflict of interest. Ethics Approval: This study utilized the publicly available, de-identified UCSF Preoperative Diffuse Glioma MRI dataset. All methods were carried out in accordance with relevant guidelines and regulations. The original data collection, including all experimental protocols, was performed in accordance with the Declaration of Helsinki and was approved by the University of California San Francisco Institutional Review Board. As a secondary analysis of publicly available de-identified data, this study was granted exemption from requiring additional institutional review board approval. Consent to Participate: Informed consent to participate in the study was obtained from all participants (or their legal guardian) at the time of original data collection. As this study involved secondary analysis of publicly available de-identified data, additional consent was not required. Consent to Publish: Not applicable. Code and Data Availability: The UCSF-PDGM dataset analyzed in this study is publicly available through The Cancer Imaging Archive (TCIA) at https://doi.org/10.7937/TCIA.2020.C1GQ4842. All tractography and tractometry analyses were performed using pyAFQ version 2.1, an open-source Python package available at https://tractometry.org/pyAFQ/ and free-water elimination software available at https://github.com/nrdg/fwe/. Results of tractometry analysis are available at: https://figshare.com/articles/dataset/Free-water_elimination_tractometry_from_the_The_University_of_California_San_Francisco_Preoperative_Diffuse_Glioma_MRI_dataset/30402736. Code to reproduce the statistical analysis and visualizations in the paper is available at https://github.com/nrdg/fwe_tractometry_glioma. Author Contributions: Study concept/design, material preparation, data collection and analysis were performed by Daniel Zhou, Kelly Chang, and Ariel Rokem. The first draft of the manuscript was written by Daniel Zhou, and all authors helped with data interpretation and commented on previous versions of the manuscript. All authors read and approved the final manuscript. References Ostrom QT, Cioffi G, Gittleman H, et al (2019) CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the United States in 2012-2016. Neuro Oncol 21:v1–v100. https://doi.org/10.1093/neuonc/noz150 Pedersen PH, Edvardsen K, Garcia-Cabrera I, et al (1995) Migratory patterns of lac-z transfected human glioma cells in the rat brain. Int J Cancer 62:767–771. https://doi.org/10.1002/ijc.2910620620 Beliën AT, Paganetti PA, Schwab ME (1999) Membrane-type 1 matrix metalloprotease (MT1-MMP) enables invasive migration of glioma cells in central nervous system white matter. J Cell Biol 144:373–384. https://doi.org/10.1083/jcb.144.2.373 Latini F, Fahlström M, Beháňová A, et al (2021) The link between gliomas infiltration and white matter architecture investigated with electron microscopy and diffusion tensor imaging. Neuroimage Clin 31:102735. https://doi.org/10.1016/j.nicl.2021.102735 Brooks LJ, Clements MP, Burden JJ, et al (2021) The white matter is a pro-differentiative niche for glioblastoma. Nat Commun 12:2184. https://doi.org/10.1038/s41467-021-22225-w Eidel O, Burth S, Neumann J-O, et al (2017) Tumor Infiltration in Enhancing and Non-Enhancing Parts of Glioblastoma: A Correlation with Histopathology. PLOS ONE 12:e0169292. https://doi.org/10.1371/journal.pone.0169292 Pekmezci M, Morshed RA, Chunduru P, et al (2021) Detection of glioma infiltration at the tumor margin using quantitative stimulated Raman scattering histology. Sci Rep 11:12162. https://doi.org/10.1038/s41598-021-91648-8 Mickevicius NJ, Carle AB, Bluemel T, et al (2015) Location of brain tumor intersecting white matter tracts predicts patient prognosis. J Neurooncol 125:393–400. https://doi.org/10.1007/s11060-015-1928-5 Marino S, Menna G, Doglietto F, et al (2025) A white matter-centered approach to investigate recurrence pathways in high-grade gliomas: a single-center retrospective study. J Neurooncol 174:177–190. https://doi.org/10.1007/s11060-025-05050-9 Wei Y, Li C, Cui Z, et al (2023) Structural connectome quantifies tumour invasion and predicts survival in glioblastoma patients. Brain 146:1714–1727. https://doi.org/10.1093/brain/awac360 Rauch P, Gmeiner M, Aichholzer M, et al (2025) Low-grade gliomas do not grow along white matter tracts: evidence from quantitative imaging. Brain Commun 7:fcaf157. https://doi.org/10.1093/braincomms/fcaf157 Kinoshita M, Goto T, Okita Y, et al (2010) Diffusion tensor-based tumor infiltration index cannot discriminate vasogenic edema from tumor-infiltrated edema. J Neurooncol 96:409–415. https://doi.org/10.1007/s11060-009-9979-0 Rokem A, Yeatman JD, Pestilli F, et al (2015) Evaluating the Accuracy of Diffusion MRI Models in White Matter. PLoS One 10:e0123272. https://doi.org/10.1371/journal.pone.0123272 Vidyadharan S, Rao BVVSNP, Yogeeswari P, et al (2024) Accurate low and high grade glioma classification using free water eliminated diffusion tensor metrics and ensemble machine learning. Sci Rep 14:19844. https://doi.org/10.1038/s41598-024-70627-9 Chang K, Burke L, LaPiana N, et al (2024) Free water elimination tractometry for aging brains. bioRxiv 2024.11.10.622861. https://doi.org/10.1101/2024.11.10.622861 Pasternak O, Sochen N, Gur Y, et al (2009) Free water elimination and mapping from diffusion MRI. Magn Reson Med 62:717–730. https://doi.org/10.1002/mrm.22055 Kruper J, Yeatman JD, Richie-Halford A, et al (2021) Evaluating the Reliability of Human Brain White Matter Tractometry. Aperture Neuro 1:1–25. https://doi.org/10.52294/e6198273-b8e3-4b63-babb-6e6b0da10669 Neher P, Hirjak D, Maier-Hein K (2024) Radiomic tractometry reveals tract-specific imaging biomarkers in white matter. Nat Commun 15:303. https://doi.org/10.1038/s41467-023-44591-3 Calabrese E, Villanueva-Meyer JE, Rudie JD, et al (2022) The University of California San Francisco Preoperative Diffuse Glioma MRI Dataset. Radiol Artif Intell 4:e220058. https://doi.org/10.1148/ryai.220058 Abbad Andaloussi M, Maser R, Hertel F, et al (2025) Exploring adult glioma through MRI: A review of publicly available datasets to guide efficient image analysis. Neurooncol Adv 7:vdae197. https://doi.org/10.1093/noajnl/vdae197 Calabrese E, Rudie JD, Rauschecker AM, et al (2021) Feasibility of Simulated Postcontrast MRI of Glioblastomas and Lower-Grade Gliomas by Using Three-dimensional Fully Convolutional Neural Networks. Radiology: Artificial Intelligence 3:e200276. https://doi.org/10.1148/ryai.2021200276 Cieslak M, Cook PA, He X, et al (2021) QSIPrep: an integrative platform for preprocessing and reconstructing diffusion MRI data. Nat Methods 18:775–778. https://doi.org/10.1038/s41592-021-01185-5 Golub M, Neto Henriques R, Gouveia Nunes R (2021) Free-water DTI estimates from single b-value data might seem plausible but must be interpreted with care. Magnetic Resonance in Medicine 85:2537–2551. https://doi.org/10.1002/mrm.28599 Kis D, Szivos L, Rekecki M, et al (2022) Predicting the true extent of glioblastoma based on probabilistic tractography. Front Neurosci 16:886465. https://doi.org/10.3389/fnins.2022.886465 Aabedi AA, Young JS, Chang EF, et al (2022) Involvement of White Matter Language Tracts in Glioma: Clinical Implications, Operative Management, and Functional Recovery After Injury. Front Neurosci 16:. https://doi.org/10.3389/fnins.2022.932478 Kallenberg K, Goldmann T, Menke J, et al (2014) Abnormalities in the normal appearing white matter of the cerebral hemisphere contralateral to a malignant brain tumor detected by diffusion tensor imaging. Folia Neuropathol 52:226–233. https://doi.org/10.5114/fn.2014.45563 Genç B, Aslan K, Özçağlayan A, İncesu L (2023) Microstructural Abnormalities in the Contralateral Normal-appearing White Matter of Glioblastoma Patients Evaluated with Advanced Diffusion Imaging. Magn Reson Med Sci 23:479–486. https://doi.org/10.2463/mrms.mp.2023-0054 Correia MM, Henriques RN, Golub M, et al (2024) The trouble with free-water elimination using single-shell diffusion MRI data: A case study in ageing. Imaging Neuroscience 2:imag–2–00252. https://doi.org/10.1162/imag_a_00252 Tables Table 1. Demographic and Clinical Comparisons Characteristic Glioblastoma (n=361) Astrocytoma (n=87) Oligodendroglioma (n=11) p-value Demographic features Age, median (IQR) 62 (55-71) 36 (31-44.5) 45 (34.5-52) <0.001 Female, n (%) 146 (40%) 34 (39%) 4 (36%) 0.94 Tumor grade Low-Grade (2), n (%) 0 33 (38%) 9 (82%) <0.001 High-Grade (3-4), n (%) 361 (100%) 54 (62%) 2 (18%) <0.001 Tumor proportion of total brain volume Enhancing tumor, median (IQR) a 0.006 (0.002-0.016) 0 (0-0.0005) 0 (0-0) <0.001 Necrotic core, median (IQR) b 0.016 (0.008-0.026) 0 (0-0.003) 0 (0-0) <0.001 Edema/infiltration, median (IQR) 0.044 (0.022-0.071) 0.053 (0.030-0.100) 0.024 (0.015-0.036) 0.001 Total tumor, median (IQR) 0.073 (0.040-0.110) 0.059 (0.030-0.112) 0.025 (0.015-0.036) <0.001 Abbreviations: IQR, interquartile range. a 34% of astrocytomas and 9% of oligodendrogliomas had detectable enhancement. b 47% of astrocytomas and 36% of oligodendrogliomas had detectable necrosis. Table 2. Tumor Involvement of Total White Matter by Pathology Tumor component Glioblastoma, mean ± SD Astrocytoma, mean ± SD Oligodendroglioma, mean ± SD adj p-value fdr q-value Combined 0.173 ± 0.094 0.153 ± 0.097 0.058 ± 0.052 0.001 - Enhancing tumor 0.054 ± 0.046 0.013 ± 0.030 0.000 ± 0.001 <0.001 <0.001 Necrotic core 0.017 ± 0.030 0.007 ± 0.020 0.000 ± 0.000 0.01 0.01 Edema/infiltration 0.160 ± 0.094 0.151 ± 0.095 0.058 ± 0.052 0.005 0.006 Abbreviations: FDR, false discovery rate. SD, standard deviation. Additional Declarations No competing interests reported. Supplementary Files figs1.png figs2.png figs3.png gliomatractometrysupplementary.docx Cite Share Download PDF Status: Published Journal Publication published 10 Dec, 2025 Read the published version in Journal of Neuro-Oncology → Version 1 posted Editorial decision: Revision requested 24 Nov, 2025 Reviews received at journal 23 Nov, 2025 Reviewers agreed at journal 10 Nov, 2025 Reviewers agreed at journal 09 Nov, 2025 Reviewers agreed at journal 09 Nov, 2025 Reviews received at journal 04 Nov, 2025 Reviewers agreed at journal 03 Nov, 2025 Reviewers agreed at journal 03 Nov, 2025 Reviewers invited by journal 31 Oct, 2025 Editor assigned by journal 31 Oct, 2025 Submission checks completed at journal 31 Oct, 2025 First submitted to journal 29 Oct, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7982561","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":543578829,"identity":"2985a869-c972-4933-8bb5-9dd05fcc8b55","order_by":0,"name":"Daniel J. Zhou","email":"data:image/png;base64,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","orcid":"","institution":"University of Pennsylvania","correspondingAuthor":true,"prefix":"","firstName":"Daniel","middleName":"J.","lastName":"Zhou","suffix":""},{"id":543578830,"identity":"70cf74bd-6f7e-4815-b598-e8dfd1679145","order_by":1,"name":"Kelly Chang","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Kelly","middleName":"","lastName":"Chang","suffix":""},{"id":543578831,"identity":"7c551103-8e04-4122-a6cf-ded6f2a6f817","order_by":2,"name":"Marc Jaskir","email":"","orcid":"","institution":"University of Pennsylvania","correspondingAuthor":false,"prefix":"","firstName":"Marc","middleName":"","lastName":"Jaskir","suffix":""},{"id":543578832,"identity":"d1801de0-1fde-45c9-b953-0dcdf071da52","order_by":3,"name":"Kathryn A. Davis","email":"","orcid":"","institution":"University of Pennsylvania","correspondingAuthor":false,"prefix":"","firstName":"Kathryn","middleName":"A.","lastName":"Davis","suffix":""},{"id":543578833,"identity":"630e427a-7fe1-40b7-b453-6b58431b5416","order_by":4,"name":"Joel M. Stein","email":"","orcid":"","institution":"University of Pennsylvania","correspondingAuthor":false,"prefix":"","firstName":"Joel","middleName":"M.","lastName":"Stein","suffix":""},{"id":543578834,"identity":"87e22e55-501c-42d2-8af5-0cf5ce2a45fb","order_by":5,"name":"Nishant Sinha","email":"","orcid":"","institution":"University of Pennsylvania","correspondingAuthor":false,"prefix":"","firstName":"Nishant","middleName":"","lastName":"Sinha","suffix":""},{"id":543578835,"identity":"33cbada8-fe52-457d-aca9-c93a587d4f1b","order_by":6,"name":"Richard E. Phillips","email":"","orcid":"","institution":"University of Pennsylvania","correspondingAuthor":false,"prefix":"","firstName":"Richard","middleName":"E.","lastName":"Phillips","suffix":""},{"id":543578836,"identity":"57d5e93d-8d95-433d-8262-4d5342da5dbc","order_by":7,"name":"Manuel Ferreira Jr.","email":"","orcid":"","institution":"University of Washington Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Manuel","middleName":"","lastName":"Ferreira","suffix":"Jr."},{"id":543578837,"identity":"8723fb80-8f1d-4186-bf10-42d1114ba70b","order_by":8,"name":"Thomas J. Grabowski","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Thomas","middleName":"J.","lastName":"Grabowski","suffix":""},{"id":543578838,"identity":"3c6ea9ce-220a-4d02-8910-35c33115d3f1","order_by":9,"name":"Ariel Rokem","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Ariel","middleName":"","lastName":"Rokem","suffix":""}],"badges":[],"createdAt":"2025-10-29 18:08:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7982561/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7982561/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s11060-025-05370-w","type":"published","date":"2025-12-10T15:59:10+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":95806440,"identity":"50131a20-e641-4537-ba05-6b83ff464349","added_by":"auto","created_at":"2025-11-13 08:47:29","extension":"png","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4095297,"visible":true,"origin":"","legend":"","description":"","filename":"fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/c8653021a5ea1f850f3bc70b.png"},{"id":95806123,"identity":"a8ed8891-c3b7-440e-a702-26b6df6bece3","added_by":"auto","created_at":"2025-11-13 08:47:16","extension":"png","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":578399,"visible":true,"origin":"","legend":"","description":"","filename":"fig2reduced.png","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/32453110e5995dd333894c23.png"},{"id":95806248,"identity":"5ad90d80-3654-40de-84a6-25c718062964","added_by":"auto","created_at":"2025-11-13 08:47:21","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":76963,"visible":true,"origin":"","legend":"","description":"","filename":"gliomatractometrymanuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/5fea3e8f721c74aff99b69e1.docx"},{"id":95806438,"identity":"bc5d96b4-6dfa-432e-b69d-bcb87a6b121f","added_by":"auto","created_at":"2025-11-13 08:47:29","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":175403,"visible":true,"origin":"","legend":"","description":"","filename":"fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/f8a961aea022502746eaf932.png"},{"id":95806546,"identity":"be349b16-5dda-42f8-9a09-92fbd8cb4b08","added_by":"auto","created_at":"2025-11-13 08:47:34","extension":"json","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11783,"visible":true,"origin":"","legend":"","description":"","filename":"2de4cc8127c949eb8eb1a2ebe035ce3f.json","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/03059bb0b2ff587a9d87e62e.json"},{"id":95806558,"identity":"7786e6d3-4a8b-4401-992b-0fcdfe81b3d7","added_by":"auto","created_at":"2025-11-13 08:47:36","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1013741,"visible":true,"origin":"","legend":"","description":"","filename":"figs1.png","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/16d6cc05409c3a197101c320.png"},{"id":95806245,"identity":"94f1a42a-0757-44ed-b3d8-d5a7679a6c2b","added_by":"auto","created_at":"2025-11-13 08:47:21","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":425647,"visible":true,"origin":"","legend":"","description":"","filename":"figs2.png","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/2c3f4a75891c74ef1ea3498f.png"},{"id":95806283,"identity":"6492583a-0c1b-4b18-bbb6-306c8ecd18b2","added_by":"auto","created_at":"2025-11-13 08:47:22","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":450119,"visible":true,"origin":"","legend":"","description":"","filename":"figs3.png","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/2c30dbb0bbae119c14c3a169.png"},{"id":95806333,"identity":"8fea2f47-339d-4308-85ee-d8a77c253549","added_by":"auto","created_at":"2025-11-13 08:47:24","extension":"docx","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1068673,"visible":true,"origin":"","legend":"","description":"","filename":"gliomatractometrysupplementary.docx","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/3cbd5ae442fbd8250ac47421.docx"},{"id":95806420,"identity":"11a9126f-1d44-4828-b96d-e9af68b77730","added_by":"auto","created_at":"2025-11-13 08:47:28","extension":"xml","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":101610,"visible":true,"origin":"","legend":"","description":"","filename":"2de4cc8127c949eb8eb1a2ebe035ce3f1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/5f99eeaa958e016f6acf5563.xml"},{"id":95806303,"identity":"a811c45f-8b03-4d2d-90df-09066eaf6bb6","added_by":"auto","created_at":"2025-11-13 08:47:23","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4095297,"visible":true,"origin":"","legend":"","description":"","filename":"fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/647e9bbff374a7d89ebeb7f5.png"},{"id":95806411,"identity":"6ede106c-64a7-433e-8b13-54cf13c33dd5","added_by":"auto","created_at":"2025-11-13 08:47:27","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":578399,"visible":true,"origin":"","legend":"","description":"","filename":"fig2reduced.png","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/7b3b9de0989d15511376f659.png"},{"id":95806372,"identity":"f561c03e-a838-4fd3-b649-023d3f8f8423","added_by":"auto","created_at":"2025-11-13 08:47:26","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":175403,"visible":true,"origin":"","legend":"","description":"","filename":"fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/81d1d74a68073be6aa6e1121.png"},{"id":95806351,"identity":"d06e58de-6765-4bde-b0e6-d0c742fda516","added_by":"auto","created_at":"2025-11-13 08:47:24","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":633293,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefig1.png","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/421cfa9e6d115fe1488ec925.png"},{"id":95806320,"identity":"23beee31-6ba1-4225-9570-3a60d7c2baa4","added_by":"auto","created_at":"2025-11-13 08:47:23","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":85072,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefig2reduced.png","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/7e4112981a9d0d1da9c243d6.png"},{"id":95806550,"identity":"0abc476e-9274-49c8-9c07-ee7696df18db","added_by":"auto","created_at":"2025-11-13 08:47:34","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":92192,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefig3.png","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/214d6c7b2adbce2939ca00ef.png"},{"id":95806155,"identity":"47efa959-8296-4661-9919-6ff5bbbc0b89","added_by":"auto","created_at":"2025-11-13 08:47:18","extension":"xml","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":98757,"visible":true,"origin":"","legend":"","description":"","filename":"2de4cc8127c949eb8eb1a2ebe035ce3f1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/c97f1ed2857f2c46c8908790.xml"},{"id":95806339,"identity":"ab649095-4de5-4d74-8ffb-cb30cb88dcc2","added_by":"auto","created_at":"2025-11-13 08:47:24","extension":"html","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":110725,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/5ee8f35161ac472bc9d81d41.html"},{"id":95806362,"identity":"db6ff4ee-eb6e-4dce-92d9-d67b90e40429","added_by":"auto","created_at":"2025-11-13 08:47:25","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":3270419,"visible":true,"origin":"","legend":"\u003cp\u003eFree water elimination and tractometry pipeline. (A) Preprocessed diffusion MRI acquisitions. (B) Free water modeling produces free water-corrected (FW) fractional anisotropy (FW-FA), free water fraction (FWF), and mean diffusivity (FW-MD) metrics. (C) The free water model metrics are used to compute free water eliminated diffusion MRI values from the preprocessed diffusion images. (D) The free water eliminated images are used to create tracts and tract profiles. Left and right inferior longitudinal fasciculi are shown with corresponding tract profiles colored by FW-FA values. The black, red, and blue regions correspond to the enhancing tumor, necrotic core, and edema/infiltration regions, respectively. The blue background in the right longitudinal fasciculus represents the nodes considered involved (\u0026gt;5%) in the edema region\u003c/p\u003e","description":"","filename":"fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/8268a11f711dff5490fc589f.png"},{"id":95806370,"identity":"1cb0f914-c39e-457b-babc-4a0929ec16df","added_by":"auto","created_at":"2025-11-13 08:47:26","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":533130,"visible":true,"origin":"","legend":"\u003cp\u003eRepresentative example demonstrating improved tractography reconstruction with free water elimination. Comparison of the right inferior longitudinal fasciculus tractography results without (left) versus with (right) free water elimination for a representative glioblastoma patient (Subject #393, UCSF-PDGM dataset). The top panels show 3D reconstructions of white matter tracts overlaid on tumor segmentations (red: enhancing tumor, light blue: edema). Without free water elimination, only sparse tracts were successfully reconstructed in tumor-affected regions. With free water elimination, multiple major white matter bundles were successfully generated. Bottom panels show corresponding tract profiles of free water-corrected fractional anisotropy (FW-FA) along node position, with colored regions indicating tumor component involvement (red: enhancing tumor, blue: edema). Free water elimination enabled successful tract reconstruction through edematous regions where conventional tractography failed, demonstrating the methodological advantage for analyzing white matter in tumor environments\u003c/p\u003e","description":"","filename":"fig2reduced.png","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/f0b0f7a382401e9e9447d2b9.png"},{"id":95806361,"identity":"263dc53d-e3cc-414d-b93b-5444ef925f10","added_by":"auto","created_at":"2025-11-13 08:47:25","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":175403,"visible":true,"origin":"","legend":"\u003cp\u003eDistance-based white matter asymmetry index. For each tract, subject-level mean asymmetries of free-water corrected diffusion metrics, including fractional anisotropy (FW-FA), mean diffusivity (FW-MD), and free water fraction (FWF) were grouped by distance from the tumor margin (1–5, 6–10, 11–15, 16–20, and 21+ nodes). Unadjusted pairwise comparisons used Mann-Whitney U tests between pathology groups (Astrocytoma, Glioblastoma, Oligodendroglioma). Bars represent group means with error bars as the standard error of the mean. False discovery rate (FDR) correction was applied separately within each diffusion measure, creating 3 independent correction families for pairwise Mann-Whitney U tests. Green asterisks (*) denote distance bins with significant pairwise differences between glioblastoma and oligodendroglioma at q\u0026lt;0.05 after FDR correction\u003c/p\u003e","description":"","filename":"fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/18dc1675a08b5832a1f3befb.png"},{"id":98244782,"identity":"c82b98b4-bcc9-4886-a9c6-4328395c8f45","added_by":"auto","created_at":"2025-12-15 16:15:08","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4782256,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/12d63911-ac8d-40af-960e-482e4d98b11a.pdf"},{"id":95806244,"identity":"bac56749-742d-44d8-9a5d-4f591b195fce","added_by":"auto","created_at":"2025-11-13 08:47:20","extension":"png","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1013741,"visible":true,"origin":"","legend":"","description":"","filename":"figs1.png","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/39a4e7a0368e596e6badc622.png"},{"id":95806414,"identity":"6a9541c2-9cef-4029-9cb5-72173afff2e6","added_by":"auto","created_at":"2025-11-13 08:47:28","extension":"png","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":425647,"visible":true,"origin":"","legend":"","description":"","filename":"figs2.png","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/f4be05d562eb511e27f03f07.png"},{"id":95806153,"identity":"8d170667-eed7-4567-9625-da420898fd95","added_by":"auto","created_at":"2025-11-13 08:47:18","extension":"png","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":450119,"visible":true,"origin":"","legend":"","description":"","filename":"figs3.png","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/e93dbf15c7a104ba2952b91d.png"},{"id":95806365,"identity":"005c155a-f715-4b78-98ec-61c0f5778dc9","added_by":"auto","created_at":"2025-11-13 08:47:26","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":1068673,"visible":true,"origin":"","legend":"","description":"","filename":"gliomatractometrysupplementary.docx","url":"https://assets-eu.researchsquare.com/files/rs-7982561/v1/fda2047b83911c7a6ea78649.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Free Water Elimination Tractometry Reveals Local and Remote White Matter Disruption in Diffuse Gliomas","fulltext":[{"header":"Introduction","content":"\u003cp\u003eLarge-scale clinical imaging repositories have transformed neuro-oncology research by providing statistical power to detect subtle biological differences obscured in smaller cohorts. Adult-type diffuse gliomas are the most common primary malignant brain tumors and remain difficult to treat [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Understanding how different glioma subtypes interact with brain white matter could contribute to surgical planning and predicting recurrence patterns. However, systematic comparisons of white matter involvement patterns across molecularly-defined glioma subtypes have been limited by the need for large cohorts with comprehensive molecular characterization.\u003c/p\u003e\u003cp\u003eHistopathological studies have demonstrated that glioma cells disseminate faster along white matter bundles than through cortex, using myelinated fibers as scaffolds for invasion [\u003cspan additionalcitationids=\"CR3 CR4\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Tissue biopsy studies have revealed that viable tumor cells remain present beyond enhancing regions visible on neuroimaging and intraoperative surgical margins [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Glioblastoma disrupts structural brain networks and favors recurrence along impacted white matter pathways, while lower grade gliomas demonstrate more variable infiltration patterns [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan additionalcitationids=\"CR9 CR10\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eConventional diffusion tensor imaging (DTI) and tractography studies of glioma are limited by tumor-induced artifacts, including vasogenic edema, mass effect, and variable infiltration, that alter diffusion measures and impede tractography [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Free water elimination methods separate tissue-specific diffusion from free water contamination, enabling more accurate assessment of white matter tissue properties [\u003cspan additionalcitationids=\"CR15\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. MRI-based tractometry samples diffusion metrics at standardized points along anatomically validated fiber bundles, offering a reproducible framework for quantifying white matter involvement across institutions [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eThe University of California San Francisco Preoperative Diffuse Glioma (UCSF-PDGM) MRI dataset is one of the largest publicly available preoperative imaging repositories with standardized protocols and expert tumor segmentations, combining high-angular resolution diffusion imaging with comprehensive molecular characterization across gliomas [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. Using this dataset, we aim to: (1) quantify direct tumor involvement of white matter tracts across glioma subtypes, and (2) assess white matter tissue property changes along white matter tracts beyond radiologic tumor margins.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eDataset and Participants\u003c/h2\u003e\u003cp\u003eThe UCSF-PDGM dataset is a publicly available collection of preoperative imaging and clinical data from adult patients with diffuse gliomas [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. The dataset includes co-registered MRI T1 and T2 imaging, DTI, and segmented tumor components (enhancing tumor, necrotic core, and edema/infiltration regions) along with accompanying clinical and pathologic data. These protocols have been previously described; in brief, diffusion-weighted MRI was collected at 2 mm\u003csup\u003e3\u003c/sup\u003e isotropic resolution, 55 directions, with b\u0026thinsp;=\u0026thinsp;2,000 s/mm\u003csup\u003e2\u003c/sup\u003e [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. For white matter asymmetry analyses, we included patients with unilateral white matter tract involvement. Bilateral or midline tumors were excluded to avoid confounding asymmetry measurements.\u003c/p\u003e\u003cp\u003eThe dataset provided diagnoses labeled according to the WHO 2021 classification. After correspondence with the dataset authors, cases labeled \u0026ldquo;astrocytoma \u003cem\u003eIDH\u003c/em\u003e-wildtype\u0026rdquo; were confirmed to represent tumors with incomplete molecular testing and were therefore reclassified as Not Elsewhere Classified (NEC). To ensure diagnostic accuracy for pathology-specific analyses, we excluded NEC cases and focused on patients with definitive histopathologic and molecular diagnoses of glioblastoma, astrocytoma, or oligodendroglioma.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eTractography and Free Water Elimination\u003c/h3\u003e\n\u003cp\u003eDiffusion images were preprocessed using QSIPrep, which performs motion correction, distortion correction, denoising, and registration to anatomical space [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Tractography reconstruction, tractometry, and free water elimination (FWE) analysis were performed as previously described, using the open-source pyAFQ software version 2.1, which relies on techniques implemented in the DIPY software [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. Tractography was performed in native subject space, and tract identification was performed using pyAFQ's anatomically-defined waypoint region of interest approach. Twenty major white matter tracts were reconstructed bilaterally, including the anterior thalamic radiation, arcuate fasciculus, posterior arcuate fasciculus, cingulate section of the cingulum bundle, corticospinal tract, inferior fronto-occipital fasciculus, inferior longitudinal fasciculus, superior longitudinal fasciculus, uncinate fasciculus, and vertical occipital fasciculus. Each tract was divided into 100 equidistant nodes along its length, standardizing sampling to equivalent anatomical positions across subjects with varying brain sizes and tract lengths.\u003c/p\u003e\u003cp\u003eTract profiles were generated by sampling diffusion metrics along the trajectory of each tract. Because the data had measurements with only one diffusion weighting, we used an implementation of a free-water DTI model that uses spatial continuity constraints to regularize model fit [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. FWE was applied to separate tissue-specific diffusion from free water contamination, yielding FW-corrected fractional anisotropy (FA), mean diffusivity (MD), and free water fraction (FWF) metrics (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eAll reconstructed tracts underwent automated quality control procedures implemented in pyAFQ, including checks for minimum streamline count, tract length, and anatomical plausibility. Representative reconstructions from a subset of cases spanning all pathology types were visually inspected by a neurologist (D.Z.) to confirm anatomical plausibility and verify that tract trajectories aligned with expected neuroanatomical pathways.\u003c/p\u003e\n\u003ch3\u003eAnalysis of Direct Tumor Involvement\u003c/h3\u003e\n\u003cp\u003eLeft and right hemisphere homologues of each tract were averaged together for analysis, yielding 10 unique bilateral tract groups (e.g., left and right inferior longitudinal fasciculi analyzed as one tract group). A tract node was classified as tumor-involved if tumor components (enhancing tumor, necrotic core, or edema) occupied more than 5% of that node's spatial extent. This threshold was chosen to provide objective criteria for distinguishing tumor-involved from unaffected tissue while maintaining sensitivity for subtle infiltrative changes, consistent with approaches in probabilistic tractography studies that demonstrate the importance of permissive thresholds in tumor environments [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. Subject-level involvement was binarized as involved or not involved based on whether the tract had at least one node exceeding the 5% threshold. Tract-level involvement was calculated as the proportion of nodes exceeding the 5% threshold within each tract.\u003c/p\u003e\n\u003ch3\u003eAnalysis of Tumor Involvement Beyond Radiologic Tumor Margins\u003c/h3\u003e\n\u003cp\u003eTo compare white matter tissue properties in the hemisphere least affected by the tumor, we averaged the free water-corrected (FW-) FA, MD, and FWF across all nodes from the contralateral hemisphere for each patient. Then, to quantify the differences between the ipsilateral and contralateral white matter beyond the tumor margins, we computed a directional percent asymmetry at nodes with \u0026lt;\u0026thinsp;5% combined tumor involvement (enhancing tumor, necrotic core, and edema/infiltration), pairing nodes by index across hemispheres within the same tract, using the following equation:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:Asymmetry\\:\\left(\\%\\right)\\:=\\:\\frac{ipsilateral\\:-\\:contralateral}{(ipsilateral\\:+\\:contralateral)/2}\\:\\times\\:\\:100\\%$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eTherefore, positive values indicated higher values in the ipsilateral white matter.\u003c/p\u003e\u003cp\u003eFor whole-tract asymmetry comparisons, node-level asymmetries were averaged within the tract and then collapsed to a single subject-level mean per metric. Then, to test for peritumoral gradients, we computed the shortest along-tract node distance from each ipsilateral node to the nearest tumor edge, defined by any radiologic tumor component. Subject-level mean asymmetries were then compared within distance segments of five nodes (1\u0026ndash;5, 6\u0026ndash;10, 11\u0026ndash;15, 16\u0026ndash;20, and \u0026ge;\u0026thinsp;21 nodes from the tumor edge), yielding one value per segment, metric, and subject.\u003c/p\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003eStatistical Analysis\u003c/h2\u003e\u003cp\u003eAnalyses were conducted in Python (pandas, NumPy, SciPy, statsmodels, matplotlib). All statistical tests were two-sided, comparing imaging measures across pathology types (glioblastoma, astrocytoma, oligodendroglioma). Continuous variables were reported as median (IQR) or mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, and categorical variables as counts (percent).\u003c/p\u003e\u003cp\u003eNormality was assessed with Shapiro-Wilk tests. Three-group comparisons used: (1) Kruskal-Wallis for overall significance; (2) pairwise Mann-Whitney U tests; (3) linear or logistic regression with dummy coding (glioblastoma reference), adjusted for age and sex where appropriate. Multiple comparisons were controlled using Benjamini-Hochberg false discovery rate (FDR), stratified by test family. Statistical significance was denoted as p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 for individual tests, q\u0026thinsp;\u0026lt;\u0026thinsp;0.05 for FDR-corrected comparisons within test families.\u003c/p\u003e\u003cp\u003eTumor proportions were calculated as volume normalized to total brain volume, excluding ventricles. FWF-edema correlation used Spearman's coefficient. Direct tract involvement used logistic regression for binary outcomes and ordinary least-squares (OLS) regression for the proportion of involved nodes. Contralateral hemisphere comparisons used linear regression adjusted for age and sex. Whole-tract and distance-based asymmetry analyses (1\u0026ndash;5, 6\u0026ndash;10, 11\u0026ndash;15, 16\u0026ndash;20, \u0026ge;\u0026thinsp;21 nodes from margins) used Kruskal-Wallis and OLS regression without age/sex covariates since asymmetry is computed within subjects.\u003c/p\u003e\u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\n \u003ch2\u003eDemographic and Clinical Features\u003c/h2\u003e\n \u003cp\u003eThe UCSF-PDGM dataset contained preoperative MRI scans from 495 unique patients, of whom 459 (93%) were included in the study. Of the excluded patients, 24 (5%) had tumors classified as Not Elsewhere Classified due to incomplete molecular testing, and 12 (2%) had bilateral white matter tract involvement, precluding hemispheric asymmetry analyses. Of the patients included, 361 (79%) were classified as glioblastoma, 87 (19%) as astrocytoma, and 11 (2%) as oligodendroglioma; mean age at time of MRI scan was 57 years (range 19\u0026ndash;97), and 184 (40%) were female (Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). Patients with glioblastoma were significantly older than patients with astrocytoma or oligodendroglioma (both p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). No significant differences in sex distribution were observed among groups (p\u0026thinsp;=\u0026thinsp;0.94). Glioblastoma had significantly larger enhancing tumor and necrotic core volumes compared to astrocytomas and oligodendrogliomas (both p\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003eTractography Quality with Free Water Elimination\u003c/h3\u003e\n\u003cp\u003eThe white matter tracts were successfully reconstructed bilaterally and passed automated quality control procedures implemented in pyAFQ. Visual inspection of tractography with and without FWE revealed improved tract reconstruction in tumor-affected regions, with FWE enabling tract generation through edematous areas where conventional DTI showed sparse or failed reconstructions (representative example in Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e). Free water fraction correlated strongly with edema involvement (\u0026rho;\u0026thinsp;=\u0026thinsp;0.55, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), suggesting that FWE successfully detected free water in edematous regions and enabling separation of tissue-specific diffusion from free water contamination (Supplementary Fig. \u003cspan class=\"InternalRef\"\u003eS2\u003c/span\u003eE).\u003c/p\u003e\n\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\n \u003ch2\u003eDirect Tumor Involvement of White Matter Tracts\u003c/h2\u003e\n \u003cp\u003eQuantitative analyses of white matter tract involvement revealed significant pathology-specific differences (Table \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e). Across all white matter tracts, glioblastoma showed greater mean involvement than astrocytoma and oligodendroglioma for enhancing tumor (5.4% vs. 1.3% vs. 0.0%, q\u0026thinsp;\u0026lt;\u0026thinsp;0.001), necrotic core (1.7% vs. 0.7% vs. 0.0%, q\u0026thinsp;=\u0026thinsp;0.01), and combined tumor components (17.3% vs. 15.3% vs. 5.8%, p\u0026thinsp;=\u0026thinsp;0.001). Edema/infiltration involvement was substantial across all pathology types but remained significantly higher in glioblastoma and astrocytoma compared to oligodendroglioma (16.0% vs. 15.1% vs. 5.8%, q\u0026thinsp;=\u0026thinsp;0.006).\u003c/p\u003e\n \u003cp\u003eExamination of direct tract involvement patterns at the subject level revealed pathology-dependent differences (Supplementary Fig. \u003cspan class=\"InternalRef\"\u003eS2\u003c/span\u003e-S3). For enhancing tumor involvement, glioblastoma showed widespread tract involvement compared to astrocytoma, with the most pronounced differences in the inferior fronto-occipital fasciculus (56% vs. 20% of subjects, q\u0026thinsp;\u0026lt;\u0026thinsp;0.001), arcuate fasciculus (45% vs. 9%, q\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and inferior longitudinal fasciculus (40% vs. 8%, q\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Necrotic core involvement followed similar patterns, with glioblastoma showing significantly higher rates in the arcuate fasciculus (15% vs. 3%, q\u0026thinsp;=\u0026thinsp;0.01), inferior fronto-occipital fasciculus (21% vs. 8%, q\u0026thinsp;=\u0026thinsp;0.009), and inferior longitudinal fasciculus (15% vs. 2%, q\u0026thinsp;=\u0026thinsp;0.01) compared to astrocytoma.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\n \u003ch2\u003eRemote White Matter Tissue Properties\u003c/h2\u003e\n \u003cp\u003eFW-corrected FA, MD, and FWF values were calculated for each tract node to assess white matter tissue properties in regions not directly inside the tumor (\u0026lt;\u0026thinsp;5% for all components combined). Distributional analysis confirmed normality assumptions for FA and FWF asymmetry measures but revealed significant deviations from normality for MD asymmetry (Shapiro-Wilk p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), justifying the application of non-parametric statistical approaches for unadjusted analyses. Analysis of the contralateral hemisphere to the tumor revealed no significant differences in any diffusion metric after adjusting for age and sex (Supplementary Table \u003cspan class=\"InternalRef\"\u003eS1\u003c/span\u003e), suggesting that baseline hemispheric differences would not confound subsequent hemisphere asymmetry analyses.\u003c/p\u003e\n \u003cp\u003eWhole-tract asymmetry analysis of all nodes without tumor involvement revealed pathology-specific patterns (Supplementary Table \u003cspan class=\"InternalRef\"\u003eS2\u003c/span\u003e). Oligodendroglioma demonstrated significantly higher FA asymmetry (+\u0026thinsp;2.0\u0026thinsp;\u0026plusmn;\u0026thinsp;2.2%) compared to glioblastoma (-3.0\u0026thinsp;\u0026plusmn;\u0026thinsp;7.1%, q\u0026thinsp;=\u0026thinsp;0.008) and astrocytoma (-2.2\u0026thinsp;\u0026plusmn;\u0026thinsp;6.7%, q\u0026thinsp;=\u0026thinsp;0.04). Both glioblastoma and astrocytoma showed negative FA asymmetry and positive MD asymmetry, indicating lower anisotropy and higher diffusivity in the ipsilateral hemisphere. In contrast, oligodendroglioma showed positive FA asymmetry and negative MD asymmetry, suggesting relatively preserved white matter microstructure in the tumor-bearing hemisphere. No significant differences were observed for FWF across any pathology comparisons.\u003c/p\u003e\n \u003cp\u003eDistance-based asymmetry analysis revealed pathology-dependent patterns at specific distances from radiologic tumor margins (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e). Oligodendroglioma demonstrated significantly higher FA asymmetry compared to glioblastoma at 1\u0026ndash;5 nodes (+\u0026thinsp;4.9\u0026thinsp;\u0026plusmn;\u0026thinsp;9.0% vs. -3.9\u0026thinsp;\u0026plusmn;\u0026thinsp;9.8%, q\u0026thinsp;=\u0026thinsp;0.04) and 6\u0026ndash;10 nodes (+\u0026thinsp;5.6\u0026thinsp;\u0026plusmn;\u0026thinsp;9.0% vs. -3.8\u0026thinsp;\u0026plusmn;\u0026thinsp;9.7%, q\u0026thinsp;=\u0026thinsp;0.03). The magnitude of this preservation effect diminished with increasing distance from tumor margins, with no significant differences observed beyond 10 nodes. No significant differences were observed between astrocytoma and glioblastoma at any distance, or for MD or FWF across any pathology comparisons.\u003c/p\u003e\n \u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study demonstrates how a large-scale, standardized clinical imaging repository enables detection of white matter disruption patterns obscured in smaller cohorts. The application of open-source analytical methods to this publicly available dataset exemplifies how well-curated clinical repositories enable reproducible, multi-institutional investigation of tumor biology. By applying FWE tractometry to the UCSF-PDGM dataset, the largest publicly available glioma cohort with standardized diffusion imaging and comprehensive molecular characterization, we quantified both direct tract involvement and remote peritumoral white matter changes across glioblastomas, astrocytomas, and oligodendrogliomas. Glioblastoma showed extensive direct tract involvement and microstructural disruption extending beyond radiologic margins, while oligodendroglioma demonstrated relative preservation of white matter microstructure within 1\u0026ndash;10 nodes of tumor edges. These distinct patterns may inform presurgical risk assessment and radiation treatment planning, as peritumoral tract disruption could influence functional outcomes and guide therapeutic targeting beyond visible tumor margins.\u003c/p\u003e\u003cp\u003eWhite matter involvement across pathology types reflect expected differences in tumor biology. Glioblastoma showed the most extensive involvement across contrast-enhancing and necrotic components, reflecting its aggressive properties and extensive invasion along white matter pathways. Astrocytoma showed intermediate involvement patterns, with lower enhancing tumor and necrotic core involvement than glioblastoma but more extensive involvement than oligodendroglioma. Understanding these pathology-specific invasion patterns could aid in treatment planning decisions, including surgical margin determination and radiation field design and inform prognosis based on the degree of white matter involvement at presentation.\u003c/p\u003e\u003cp\u003eAnalysis of remote white matter revealed pathology-specific peritumoral effects extending beyond visible tumor margins. Oligodendroglioma demonstrated positive FA asymmetry and negative MD asymmetry, suggesting better-preserved ipsilateral white matter microstructure, compared to both glioblastoma and astrocytoma at the whole-tract level. This preservation was most pronounced within the proximal 10% of tract length from radiologic tumor margins, suggesting that oligodendroglioma\u0026rsquo;s slower growth and less infiltrative behavior result in less disruption of adjacent white matter. In contrast, both glioblastoma and astrocytoma showed negative FA asymmetry and positive MD asymmetry, consistent with microstructural disruption in the tumor-bearing hemisphere even beyond radiologically visible boundaries.\u003c/p\u003e\u003cp\u003eWhile FWE isolated tissue-specific diffusion from free water contamination, the observed microstructural changes could reflect several processes: subclinical tumor infiltration beyond MRI detection limits, white matter deformation from mass effect, alterations in the tissue microenvironment from peritumoral processes, or secondary network effects from tract disruption. Although our cross-sectional imaging data did not distinguish among these possibilities, the distance-dependent gradient, where disruption was most pronounced proximally and attenuated with distance from tumor margins, raises the possibility that these signatures may capture spatial gradients of tumor infiltration. If these patterns do reflect occult infiltrative extent, they could potentially inform estimates of subclinical disease burden relevant to surgical planning and recurrence risk. Validating this hypothesis would require prospective studies correlating preoperative tractometry findings with histopathological analysis of tissue sampled beyond radiographic margins and longitudinal mapping of recurrence patterns.\u003c/p\u003e\u003cp\u003eOur finding of no significant contralateral hemisphere differences contrasts with prior DTI studies reporting increased MD and reduced FA in the contralateral hemisphere [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. This discrepancy may reflect differences in patient populations, analysis methodologies, or our application of FWE techniques to better isolate tissue-specific diffusion properties. While the absence of contralateral effects could reflect limited statistical power or true sparing of the contralateral hemisphere, it suggests that the ipsilateral asymmetries are not fully attributable to global bilateral processes, supporting the spatial specificity of our distance-dependent findings.\u003c/p\u003e\u003cp\u003eFWE tractometry addressed key limitations of conventional DTI in tumor environments. This correction is particularly important in glioma imaging, where vasogenic edema and infiltration confound standard DTI metrics [\u003cspan additionalcitationids=\"CR15\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. By isolating tissue microstructure from free water effects, FWE enabled more valid comparisons of white matter integrity across pathology types with different edema burdens. The strong correlation between FWF and edema validates the methodological approach and supports the biological interpretability of our tissue-specific measurements.\u003c/p\u003e\u003cp\u003eThe study has several limitations that should be considered. The cross-sectional design precludes assessment of temporal changes in white matter properties or correlation with postoperative outcomes such as survival or functional status. The UCSF-PDGM dataset represents a single-institution cohort with specific imaging protocols, which may limit generalizability to other populations or scanners. However, the standardized protocols within this cohort enabled rigorous quantitative comparison without confounding scanner effects, a key advantage over heterogeneous multi-center datasets. Our use of single-shell data for FWE had inherent limitations compared to multi-shell acquisition. The spatial regularization constraints required for single-shell FWE may reduce sensitivity to subtle tissue changes [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. Our exclusion of patients with bilateral or midline tumors, while necessary for proper asymmetry analysis, could introduce selection bias toward more focal lesions limiting generalizability to bilateral tumors. The dataset was curated prior to the 2021 WHO molecular classification. While we excluded cases with incomplete molecular testing to ensure diagnostic accuracy, this reduced the sample sizes for some pathology comparisons. The small oligodendroglioma sample limits statistical power for detecting subtle effects, though the large differences observed compared to glioblastoma may be biologically meaningful. The 5% probability threshold for defining tumor involvement was chosen to provide objective criteria for analysis, though the optimal threshold for distinguishing tumor-involved from unaffected tissue remains to be established. Finally, our imaging-based findings lacked clinical data such as functional outcomes, cognitive assessments, or histopathological validation of white matter involvement patterns, limiting the ability to establish clinical correlations of the observed diffusion changes.\u003c/p\u003e\u003cp\u003eThis study demonstrates how a large-scale, publicly available clinical imaging dataset can reveal pathology-specific patterns of white matter disruption that could inform tumor biology and clinical decision-making. By applying FWE tractometry to the UCSF-PDGM cohort, we established a reproducible quantitative framework that addresses key limitations of conventional DTI in tumor environments and can be applied across institutions using standardized diffusion protocols. The pathology-specific patterns observed could provide imaging-based correlates of known biological differences with potential implications for surgical and radiation therapy planning. The open-source, reproducible methodology established here exemplifies how well-curated imaging repositories enable methodological innovation and biological discovery in neuro-oncology. Future integration of these tract-based imaging metrics with longitudinal clinical data from cancer registries and prospective surgical cohorts will validate their prognostic utility and advance their translation into clinical decision support tools. This work provides a foundation for multi-institutional efforts to standardize quantitative white matter assessment in glioma and demonstrates the continued value of investment in open neuroscience data infrastructure.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding: \u003c/strong\u003eD.Z.\u0026rsquo;s work was funded by the National Institute of Neurological Disorders and Stroke (5T32NS091008-07) and the American Epilepsy Society Research and Training Fellowship for Clinicians (1282834). N.S.\u0026rsquo;s work was funded by the National Institute of Neurological Disorders and Stroke (K99NS138680). AR and KC\u0026rsquo;s work was funded by National Institutes of Health grants MH121868, MH121867, R25MH112480, R01AG060942, and U19AG066567, and R01EB027585, as well as by National Science Foundation grant 1934292. Open-source software development was also supported by the Chan Zuckerberg Initiative\u0026rsquo;s Essential Open Source Software for Science program, the Alfred P. Sloan Foundation, and the Gordon \u0026amp; Betty Moore Foundation. \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests: \u003c/strong\u003eAll authors report no competing interest or conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics Approval:\u003c/strong\u003e This study utilized the publicly available, de-identified UCSF Preoperative Diffuse Glioma MRI dataset. All methods were carried out in accordance with relevant guidelines and regulations. The original data collection, including all experimental protocols, was performed in accordance with the Declaration of Helsinki and was approved by the University of California San Francisco Institutional Review Board. As a secondary analysis of publicly available de-identified data, this study was granted exemption from requiring additional institutional review board approval.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to Participate:\u003c/strong\u003e Informed consent to participate in the study was obtained from all participants (or their legal guardian) at the time of original data collection. As this study involved secondary analysis of publicly available de-identified data, additional consent was not required.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to Publish: \u003c/strong\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCode and Data Availability:\u003c/strong\u003e The UCSF-PDGM dataset analyzed in this study is publicly available through The Cancer Imaging Archive (TCIA) at https://doi.org/10.7937/TCIA.2020.C1GQ4842. All tractography and tractometry analyses were performed using pyAFQ version 2.1, an open-source Python package available at https://tractometry.org/pyAFQ/ and free-water elimination software available at https://github.com/nrdg/fwe/. Results of tractometry analysis are available at: https://figshare.com/articles/dataset/Free-water_elimination_tractometry_from_the_The_University_of_California_San_Francisco_Preoperative_Diffuse_Glioma_MRI_dataset/30402736. Code to reproduce the statistical analysis and visualizations in the paper is available at https://github.com/nrdg/fwe_tractometry_glioma. \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions:\u003c/strong\u003e \u003c/p\u003e\n\u003cp\u003eStudy concept/design, material preparation, data collection and analysis were performed by Daniel Zhou, Kelly Chang, and Ariel Rokem. The first draft of the manuscript was written by Daniel Zhou, and all authors helped with data interpretation and commented on previous versions of the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eOstrom QT, Cioffi G, Gittleman H, et al (2019) CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the United States in 2012-2016. Neuro Oncol 21:v1\u0026ndash;v100. https://doi.org/10.1093/neuonc/noz150\u003c/li\u003e\n\u003cli\u003ePedersen PH, Edvardsen K, Garcia-Cabrera I, et al (1995) Migratory patterns of lac-z transfected human glioma cells in the rat brain. Int J Cancer 62:767\u0026ndash;771. https://doi.org/10.1002/ijc.2910620620\u003c/li\u003e\n\u003cli\u003eBeli\u0026euml;n AT, Paganetti PA, Schwab ME (1999) Membrane-type 1 matrix metalloprotease (MT1-MMP) enables invasive migration of glioma cells in central nervous system white matter. J Cell Biol 144:373\u0026ndash;384. https://doi.org/10.1083/jcb.144.2.373\u003c/li\u003e\n\u003cli\u003eLatini F, Fahlstr\u0026ouml;m M, Beh\u0026aacute;ňov\u0026aacute; A, et al (2021) The link between gliomas infiltration and white matter architecture investigated with electron microscopy and diffusion tensor imaging. Neuroimage Clin 31:102735. https://doi.org/10.1016/j.nicl.2021.102735\u003c/li\u003e\n\u003cli\u003eBrooks LJ, Clements MP, Burden JJ, et al (2021) The white matter is a pro-differentiative niche for glioblastoma. Nat Commun 12:2184. https://doi.org/10.1038/s41467-021-22225-w\u003c/li\u003e\n\u003cli\u003eEidel O, Burth S, Neumann J-O, et al (2017) Tumor Infiltration in Enhancing and Non-Enhancing Parts of Glioblastoma: A Correlation with Histopathology. PLOS ONE 12:e0169292. https://doi.org/10.1371/journal.pone.0169292\u003c/li\u003e\n\u003cli\u003ePekmezci M, Morshed RA, Chunduru P, et al (2021) Detection of glioma infiltration at the tumor margin using quantitative stimulated Raman scattering histology. Sci Rep 11:12162. https://doi.org/10.1038/s41598-021-91648-8\u003c/li\u003e\n\u003cli\u003eMickevicius NJ, Carle AB, Bluemel T, et al (2015) Location of brain tumor intersecting white matter tracts predicts patient prognosis. J Neurooncol 125:393\u0026ndash;400. https://doi.org/10.1007/s11060-015-1928-5\u003c/li\u003e\n\u003cli\u003eMarino S, Menna G, Doglietto F, et al (2025) A white matter-centered approach to investigate recurrence pathways in high-grade gliomas: a single-center retrospective study. J Neurooncol 174:177\u0026ndash;190. https://doi.org/10.1007/s11060-025-05050-9\u003c/li\u003e\n\u003cli\u003eWei Y, Li C, Cui Z, et al (2023) Structural connectome quantifies tumour invasion and predicts survival in glioblastoma patients. Brain 146:1714\u0026ndash;1727. https://doi.org/10.1093/brain/awac360\u003c/li\u003e\n\u003cli\u003eRauch P, Gmeiner M, Aichholzer M, et al (2025) Low-grade gliomas do not grow along white matter tracts: evidence from quantitative imaging. Brain Commun 7:fcaf157. https://doi.org/10.1093/braincomms/fcaf157\u003c/li\u003e\n\u003cli\u003eKinoshita M, Goto T, Okita Y, et al (2010) Diffusion tensor-based tumor infiltration index cannot discriminate vasogenic edema from tumor-infiltrated edema. J Neurooncol 96:409\u0026ndash;415. https://doi.org/10.1007/s11060-009-9979-0\u003c/li\u003e\n\u003cli\u003eRokem A, Yeatman JD, Pestilli F, et al (2015) Evaluating the Accuracy of Diffusion MRI Models in White Matter. PLoS One 10:e0123272. https://doi.org/10.1371/journal.pone.0123272\u003c/li\u003e\n\u003cli\u003eVidyadharan S, Rao BVVSNP, Yogeeswari P, et al (2024) Accurate low and high grade glioma classification using free water eliminated diffusion tensor metrics and ensemble machine learning. Sci Rep 14:19844. https://doi.org/10.1038/s41598-024-70627-9\u003c/li\u003e\n\u003cli\u003eChang K, Burke L, LaPiana N, et al (2024) Free water elimination tractometry for aging brains. bioRxiv 2024.11.10.622861. https://doi.org/10.1101/2024.11.10.622861\u003c/li\u003e\n\u003cli\u003ePasternak O, Sochen N, Gur Y, et al (2009) Free water elimination and mapping from diffusion MRI. Magn Reson Med 62:717\u0026ndash;730. https://doi.org/10.1002/mrm.22055\u003c/li\u003e\n\u003cli\u003eKruper J, Yeatman JD, Richie-Halford A, et al (2021) Evaluating the Reliability of Human Brain White Matter Tractometry. Aperture Neuro 1:1\u0026ndash;25. https://doi.org/10.52294/e6198273-b8e3-4b63-babb-6e6b0da10669\u003c/li\u003e\n\u003cli\u003eNeher P, Hirjak D, Maier-Hein K (2024) Radiomic tractometry reveals tract-specific imaging biomarkers in white matter. Nat Commun 15:303. https://doi.org/10.1038/s41467-023-44591-3\u003c/li\u003e\n\u003cli\u003eCalabrese E, Villanueva-Meyer JE, Rudie JD, et al (2022) The University of California San Francisco Preoperative Diffuse Glioma MRI Dataset. Radiol Artif Intell 4:e220058. https://doi.org/10.1148/ryai.220058\u003c/li\u003e\n\u003cli\u003eAbbad Andaloussi M, Maser R, Hertel F, et al (2025) Exploring adult glioma through MRI: A review of publicly available datasets to guide efficient image analysis. Neurooncol Adv 7:vdae197. https://doi.org/10.1093/noajnl/vdae197\u003c/li\u003e\n\u003cli\u003eCalabrese E, Rudie JD, Rauschecker AM, et al (2021) Feasibility of Simulated Postcontrast MRI of Glioblastomas and Lower-Grade Gliomas by Using Three-dimensional Fully Convolutional Neural Networks. Radiology: Artificial Intelligence 3:e200276. https://doi.org/10.1148/ryai.2021200276\u003c/li\u003e\n\u003cli\u003eCieslak M, Cook PA, He X, et al (2021) QSIPrep: an integrative platform for preprocessing and reconstructing diffusion MRI data. Nat Methods 18:775\u0026ndash;778. https://doi.org/10.1038/s41592-021-01185-5\u003c/li\u003e\n\u003cli\u003eGolub M, Neto Henriques R, Gouveia Nunes R (2021) Free-water DTI estimates from single b-value data might seem plausible but must be interpreted with care. Magnetic Resonance in Medicine 85:2537\u0026ndash;2551. https://doi.org/10.1002/mrm.28599\u003c/li\u003e\n\u003cli\u003eKis D, Szivos L, Rekecki M, et al (2022) Predicting the true extent of glioblastoma based on probabilistic tractography. Front Neurosci 16:886465. https://doi.org/10.3389/fnins.2022.886465\u003c/li\u003e\n\u003cli\u003eAabedi AA, Young JS, Chang EF, et al (2022) Involvement of White Matter Language Tracts in Glioma: Clinical Implications, Operative Management, and Functional Recovery After Injury. Front Neurosci 16:. https://doi.org/10.3389/fnins.2022.932478\u003c/li\u003e\n\u003cli\u003eKallenberg K, Goldmann T, Menke J, et al (2014) Abnormalities in the normal appearing white matter of the cerebral hemisphere contralateral to a malignant brain tumor detected by diffusion tensor imaging. Folia Neuropathol 52:226\u0026ndash;233. https://doi.org/10.5114/fn.2014.45563\u003c/li\u003e\n\u003cli\u003eGen\u0026ccedil; B, Aslan K, \u0026Ouml;z\u0026ccedil;ağlayan A, İncesu L (2023) Microstructural Abnormalities in the Contralateral Normal-appearing White Matter of Glioblastoma Patients Evaluated with Advanced Diffusion Imaging. Magn Reson Med Sci 23:479\u0026ndash;486. https://doi.org/10.2463/mrms.mp.2023-0054\u003c/li\u003e\n\u003cli\u003eCorreia MM, Henriques RN, Golub M, et al (2024) The trouble with free-water elimination using single-shell diffusion MRI data: A case study in ageing. Imaging Neuroscience 2:imag\u0026ndash;2\u0026ndash;00252. https://doi.org/10.1162/imag_a_00252\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable 1. Demographic and Clinical Comparisons\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"655\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 25px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 201px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCharacteristic\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGlioblastoma (n=361)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAstrocytoma (n=87)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eOligodendroglioma (n=11)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 61px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ep-value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"bottom\" style=\"width: 226px;\"\u003e\n \u003cp\u003eDemographic features\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 61px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 25px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 201px;\"\u003e\n \u003cp\u003eAge, median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e62 (55-71)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e36 (31-44.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e45 (34.5-52)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 61px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 25px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 201px;\"\u003e\n \u003cp\u003eFemale, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e146 (40%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e34 (39%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e4 (36%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 61px;\"\u003e\n \u003cp\u003e0.94\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"bottom\" style=\"width: 226px;\"\u003e\n \u003cp\u003eTumor grade\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 61px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 25px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 201px;\"\u003e\n \u003cp\u003eLow-Grade (2), n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e33 (38%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e9 (82%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 61px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 25px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 201px;\"\u003e\n \u003cp\u003eHigh-Grade (3-4), n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e361 (100%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e54 (62%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e2 (18%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 61px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"bottom\" style=\"width: 226px;\"\u003e\n \u003cp\u003eTumor proportion of total brain volume\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 121px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 121px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 61px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 25px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 201px;\"\u003e\n \u003cp\u003eEnhancing tumor, median (IQR)\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 121px;\"\u003e\n \u003cp\u003e0.006 (0.002-0.016)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 121px;\"\u003e\n \u003cp\u003e0 (0-0.0005)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 125px;\"\u003e\n \u003cp\u003e0 (0-0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 61px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 25px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 201px;\"\u003e\n \u003cp\u003eNecrotic core, median (IQR)\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 121px;\"\u003e\n \u003cp\u003e0.016 (0.008-0.026)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 121px;\"\u003e\n \u003cp\u003e0 (0-0.003)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 125px;\"\u003e\n \u003cp\u003e0 (0-0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 61px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 25px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 201px;\"\u003e\n \u003cp\u003eEdema/infiltration, median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 121px;\"\u003e\n \u003cp\u003e0.044 (0.022-0.071)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 121px;\"\u003e\n \u003cp\u003e0.053 (0.030-0.100)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 125px;\"\u003e\n \u003cp\u003e0.024 (0.015-0.036)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 61px;\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 25px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 201px;\"\u003e\n \u003cp\u003eTotal tumor, median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 121px;\"\u003e\n \u003cp\u003e0.073 (0.040-0.110)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 121px;\"\u003e\n \u003cp\u003e0.059 (0.030-0.112)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 125px;\"\u003e\n \u003cp\u003e0.025 (0.015-0.036)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 61px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eAbbreviations:\u003c/strong\u003e IQR, interquartile range. \u003csup\u003ea\u003c/sup\u003e 34% of astrocytomas and 9% of oligodendrogliomas had detectable enhancement. \u003csup\u003eb\u003c/sup\u003e 47% of astrocytomas and 36% of oligodendrogliomas had detectable necrosis.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2. Tumor Involvement of Total White Matter by Pathology\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"587\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 129px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTumor component\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGlioblastoma,\u003cbr\u003e\u0026nbsp;mean \u0026plusmn; SD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAstrocytoma,\u003cbr\u003e\u0026nbsp;mean \u0026plusmn; SD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 129px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eOligodendroglioma,\u003cbr\u003e\u0026nbsp;mean \u0026plusmn; SD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 61px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eadj p-value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 61px;\"\u003e\n \u003cp\u003e\u003cstrong\u003efdr q-value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 129px;\"\u003e\n \u003cp\u003eCombined\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 103px;\"\u003e\n \u003cp\u003e0.173 \u0026plusmn; 0.094\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 103px;\"\u003e\n \u003cp\u003e0.153 \u0026plusmn; 0.097\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 129px;\"\u003e\n \u003cp\u003e0.058 \u0026plusmn; 0.052\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 61px;\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 61px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 129px;\"\u003e\n \u003cp\u003eEnhancing tumor\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 103px;\"\u003e\n \u003cp\u003e0.054 \u0026plusmn; 0.046\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 103px;\"\u003e\n \u003cp\u003e0.013 \u0026plusmn; 0.030\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 129px;\"\u003e\n \u003cp\u003e0.000 \u0026plusmn; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 61px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 61px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 129px;\"\u003e\n \u003cp\u003eNecrotic core\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 103px;\"\u003e\n \u003cp\u003e0.017 \u0026plusmn; 0.030\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 103px;\"\u003e\n \u003cp\u003e0.007 \u0026plusmn; 0.020\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 129px;\"\u003e\n \u003cp\u003e0.000 \u0026plusmn; 0.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 61px;\"\u003e\n \u003cp\u003e0.01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 61px;\"\u003e\n \u003cp\u003e0.01\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 129px;\"\u003e\n \u003cp\u003eEdema/infiltration\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 103px;\"\u003e\n \u003cp\u003e0.160 \u0026plusmn; 0.094\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 103px;\"\u003e\n \u003cp\u003e0.151 \u0026plusmn; 0.095\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 129px;\"\u003e\n \u003cp\u003e0.058 \u0026plusmn; 0.052\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 61px;\"\u003e\n \u003cp\u003e0.005\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 61px;\"\u003e\n \u003cp\u003e0.006\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\n\u003cp\u003e\u003cstrong\u003eAbbreviations:\u003c/strong\u003e FDR, false discovery rate. SD, standard deviation.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"journal-of-neuro-oncology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"neon","sideBox":"Learn more about [Journal of Neuro-Oncology](https://www.springer.com/journal/11060)","snPcode":"11060","submissionUrl":"https://submission.nature.com/new-submission/11060/3","title":"Journal of Neuro-Oncology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Glioma, white matter, diffusion MRI, tractometry, free water elimination","lastPublishedDoi":"10.21203/rs.3.rs-7982561/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7982561/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003ePurpose\u003c/h2\u003e\u003cp\u003eTo apply free water elimination (FWE) tractometry to a large real-world clinical imaging dataset to quantify pathology-specific patterns of white matter involvement and peritumoral tissue disruption in diffuse gliomas.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eThe UCSF Preoperative Diffuse Glioma MRI dataset was analyzed using FWE tractometry. Twenty major white matter tracts were reconstructed and each divided into 100 equidistant nodes. Direct tumor involvement was quantified across enhancing tumor, necrotic core, and edema regions. Remote white matter tissue properties were assessed through hemispheric asymmetry analysis of free water-corrected fractional anisotropy (FW-FA), mean diffusivity (FW-MD), and free water fraction (FWF) in non-tumor involved regions at standardized distances from radiological tumor margins.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003e459 patients with unilateral glioma were included (361 glioblastoma, 87 astrocytoma, 11 oligodendroglioma). Glioblastoma demonstrated greater direct white matter involvement in enhancing tumor and necrotic core compared to astrocytoma and oligodendroglioma (q\u0026thinsp;\u0026lt;\u0026thinsp;0.001, q\u0026thinsp;=\u0026thinsp;0.01, respectively). Beyond radiological tumor margins, glioblastoma and astrocytoma exhibited decreased FW-FA, while oligodendroglioma showed increased FW-FA (q\u0026thinsp;=\u0026thinsp;0.008, q\u0026thinsp;=\u0026thinsp;0.04, respectively). Distance-based analysis revealed that this effect was most prominent in the proximal peritumoral region and diminished with increasing distance from tumor margins.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e\u003cp\u003eUsing FWE tractometry on a large clinical repository, we identified distinct pathology-specific patterns of white matter disruption. Glioblastoma showed extensive direct involvement and peritumoral microstructural changes, while oligodendroglioma demonstrated relatively preserved white matter architecture near tumor margins. These patterns reflect expected biological differences and provide a reproducible framework for characterizing extent of white matter involvement, with potential applications in presurgical planning and understanding recurrence patterns.\u003c/p\u003e","manuscriptTitle":"Free Water Elimination Tractometry Reveals Local and Remote White Matter Disruption in Diffuse Gliomas","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-13 07:53:29","doi":"10.21203/rs.3.rs-7982561/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-11-25T03:30:05+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-24T01:44:44+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"301026357063900153154859540084960583385","date":"2025-11-10T18:48:19+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"89882662601009140017281070184246466574","date":"2025-11-09T16:02:35+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"215212015630243095430896375612897112315","date":"2025-11-09T13:59:46+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-05T03:37:47+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"338553929918986459039316337228219415424","date":"2025-11-03T15:22:04+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"318548095117716590914448327426600130576","date":"2025-11-03T13:21:31+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-31T15:09:29+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-31T15:07:38+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-31T14:26:04+00:00","index":"","fulltext":""},{"type":"submitted","content":"Journal of Neuro-Oncology","date":"2025-10-29T18:06:53+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"journal-of-neuro-oncology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"neon","sideBox":"Learn more about [Journal of Neuro-Oncology](https://www.springer.com/journal/11060)","snPcode":"11060","submissionUrl":"https://submission.nature.com/new-submission/11060/3","title":"Journal of Neuro-Oncology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"69f03fc3-e7fd-4b5f-aeb3-0d8668978b68","owner":[],"postedDate":"November 13th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-12-15T16:09:12+00:00","versionOfRecord":{"articleIdentity":"rs-7982561","link":"https://doi.org/10.1007/s11060-025-05370-w","journal":{"identity":"journal-of-neuro-oncology","isVorOnly":false,"title":"Journal of Neuro-Oncology"},"publishedOn":"2025-12-10 15:59:10","publishedOnDateReadable":"December 10th, 2025"},"versionCreatedAt":"2025-11-13 07:53:29","video":"","vorDoi":"10.1007/s11060-025-05370-w","vorDoiUrl":"https://doi.org/10.1007/s11060-025-05370-w","workflowStages":[]},"version":"v1","identity":"rs-7982561","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7982561","identity":"rs-7982561","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}