Lesion-symptom mapping is widely used to identify causal relationships between brain structures and behaviour, and has played a central role in neuropsychologically informed network models of cognition. However, even recent approaches remain constrained by a one-to-one mapping framework, which oversimplifies the complex relationships between network-level damage and cognitive deficits. In addition, the non-orthogonality of cortical and white matter damage makes it difficult to disentangle their distinct contributions. Here, we used graph-based multilayer network analysis to address these limitations and evaluate clinical relevance. Using neuroanatomical and longitudinal neuropsychological data from 252 patients who underwent awake neurosurgery for low-grade glioma, we constructed interactive, three-layer networks for each hemisphere. Layer 1 comprised neuropsychological tasks (NT), layer 2 structural disconnections (SD), and layer 3 cortical damage (CD). Nodes represented tasks, white matter tracts, and cortical parcels, respectively, whereas within-layer edges captured correlations in performance or co-occurring damage patterns. Multilayer community detection identified domain- and hemisphere-specific brain-behaviour motifs linking executive, language, and spatial functions to distinct combinations of cortical and white matter disruption, a pattern confirmed by two spatial embedding approaches. Centrality analyses revealed a continuum of mapping relationships, ranging from one-to-one to one-to-many associations, indicating that tasks such as verbal fluency are better explained by multiple disconnection mechanisms. Additional analyses uncovered many-to-one and many-to-many relationships and highlighted tracts and cortical regions with domain-general relevance. Together, these findings support a neurobiologically grounded, network-oriented account of how structural brain damage gives rise to cognitive deficits, with implications for clinical care.

Contexte et objectif L’imagerie pondérée par transfert de proton amide APTw permet de visualiser la concentration en protéines des gliomes, contribuant ainsi la caractérisation et à l’évaluation pronostique des gliomes1. Cependant, des artefacts peuvent réduire la spécificité de cette technique, incluant la saturation directe de l’eau, le transfert de magnétisation semi-solide (ssMT) et la relaxation T12. Cette étude vise à évaluer si les corrections successives des métriques APTw (spillover, ssMT, T1) améliorent la detection de la mutation IDH et de la co-déléction 1p/19q, deux marqueurs moléculaires clés pour le diagnostic des gliomes. Méthodologie Quarante-six patients présentant un gliome diffus présumé éligible à la chirurgie ont été inclus prospectivement. Les acquisitions ont été réalisées à l’aide d’une séquence 3D snapshot-GRE CEST à 3 Tesla (B1 = 2 µT, T₁sat = 2 s), complétée par une cartographie B₀/B₁ (WASABI) et T₁ (Saturation Recovery, B₁ = 5 µT, 14 points de récupération). Les métriques analysées étaient :• MTRasym (standard, non corrigé)3,• MTRRex (corrigé pour la saturation directe de l’eau et le ssMT)4,• AREX (corrigé pour la saturation, le ssMT et T1)4.Les comparaisons statistiques ont utilisé des tests U de Mann-Whitney (correction de Benjamini-Hochberg) et des tests t de Welch sur les médianes, avec un seuil de significativité fixé à p corrigé < 0,05. Résultats Les métriques corrigées (MTRRex et AREX) séparaient significativement mieux les gliomes IDH-mutés (p ≤ 1,4×10⁻⁶) des gliomes IDH normal que la métrique standard MTRasym (p = 0,008) (Figure 1, A).Au sein des gliomes IDH-mutés, l’AREX distinguait les oligodendrogliomes (codélétés 1p/19q) des astrocytomes (non codélétés, p = 0,010), avec des valeurs plus élevées pour les oligodendrogliomes (Figure 1, B).La séquence de correction (MTRasym → MTRRex → AREX, Figure 2) améliorait la cohérence intra-groupe et la séparabilité des sous-types moléculaires. Conclusion La correction des artefacts liés au contenu liquide des lésions optimise la précision diagnostique de l’imagerie APT. Cette approche offre une méthode robuste, précise et biologiquement fondée pour la stratification moléculaire non invasive des gliomes.

The INDIGO trial demonstrated improved progression-free survival under vorasidenib in IDH-mutated grade 2 glioma patients. However, the pattern of vorasidenib effects remains unknown. We thus performed a retrospective analysis of tumor kinetics in eight patients enrolled in the INDIGO trial. Tumor diameters (computed from FLAIR volumetry) and their slopes were individually estimated, before and under vorasidenib. Three patients showed no change in slope. Three patients exhibited a marked slope decrease, and two demonstrated slope inversion. Tumor diameter slopes allows individual assessment of radiological response, providing clinically relevant insights that cannot be captured by the cohort-level analysis performed in the INDIGO trial.

We identified in two awake surgery cases a postoperative double dissociation between phonological and graphemic output buffer deficits. Using lesion-symptom mapping from ischaemic mini-strokes and preoperative tractography, we demonstrated that the phonological (resp. graphemic) disorder fitted with ventral (resp. dorsal) damage to the AF. Further studies are needed to confirm our hypothesis of a ventro-dorsal functional organization within the AF for phonological versus graphemic processing.

Monitored anesthesia care (MAC) with continuous sedation for awake craniotomy allows rapid recovery but increases risks of hypoventilation and hypoxemia. In April 2019, our center replaced propofol-based sedation with dexmedetomidine-based protocols to improve respiratory safety. We evaluated the impact of this protocol change on respiratory and hemodynamic outcomes. This single-center retrospective before-after observational study analyzed all adult patients undergoing awake craniotomy between April 2017 and July 2021. The propofol group (April 2017-March 2019) received propofol-remifentanil sedation; the dexmedetomidine group (April 2019-July 2021) received dexmedetomidine-based sedation with supplemental low-dose propofol. Primary endpoints were hypoxemia (SpO2 < 92

Introduction Selective amygdalo-hippocampectomy (SAH) is commonly performed in drug-resistant mesial temporal lobe epilepsy (MTLE) with seizure-free outcome in 70-80% of patients, however cognitive risks with respect to surgical approaches remain debated. Research question The preliminary experience of a new surgical route through the orbit are evaluated with the aim to improve the benefits/risks of SAH. Methods Patients were recruited among a cohort of 100 patients investigated in the Epilepsy Unit of the department of Neurosurgery, between 2022 and 2024. A SAH through a transorbital approach with an eyebrow incision was performed in 3 patients (2 males, 22-51 years) after investigation for pure MTLE (right side in 2, non-lesional in 1).Surgical outcome was assessed with respect to seizures according to the International League Against Epilepsy (ILAE) classification. Results No local or general complications occurred postoperatively except a transient diplopia due to ocular oedema. Mean volume of resection was of 9.2 cc, and virtual dissection of tractograms demonstrated that the resection cavity spared the optic, uncinate and inferior longitudinal pathways in all cases.With a 12 to 31 month-follow-up (mean 22 months), seizure-free outcome (ILAE class 1) was obtained in all patients. Neuropsychological evaluation showed increased or stable cognitive and memory performances, with an improvement of quality of life (QOL) in all. Aesthetical results were excellent. Conclusion Transorbital eyebrow approach is safe and effective for SAH, with a positive impact on cognitive functions and QOL. Although preliminary, these findings suggest that it may be an option in patients with pure MTLE.

Abstract Background Returning to work after IDH-mutant glioma surgery can be hindered by cognitive impairment. In this study, we analyzed the predictive value of verbal fluency on return to work (RTW). Methods We retrospectively analyzed 72 surgeries (including 6 reoperations) performed between 2012 and 2025 for diffuse IDH-mutant glioma grades 2 and 3 in 66 patients (mean age 38.5 ± 8.5). All patients were employed at surgery and assessed for RTW status 1 year postoperatively. Multiple semantic fluency (SF) and phonemic fluency (PF) metrics were first compared against 274 matched controls. Clinical and professional variables, along with pre-to-postoperative changes in verbal SF and PF scores, were then analyzed for their predictive value on RTW. Results RTW rate was 77.8% (mean delay 5.2 months ± 2.9). In both PF and SF, patients produced significantly fewer words (P < .001) and more repetitions (P < .05) than controls. Multivariate analysis identified 4 independent predictors of non-RTW: grade 3 glioma (OR = 0.07; 95% CI, 0.01-0.47; P = .006), precarious employment contract vs salaried or self-employed workers (OR ≈ 40, P < .05), decrease in PF total Z-score (OR = 4.28; 95% CI, 1.30-14.05; P = .017), and decrease in SF total Z-score (OR = 4.15; 95% CI, 1.31-13.13; P = .015). Conclusions Four months post-surgery, 1 standard deviation decline in both total PF and SF scores is a predictor of non-return to work. For these patients, reinforced cognitive rehabilitation and personalized workplace accommodations are warranted to facilitate work resumption.

Abstract Purpose Picture naming is a widely used task for perioperative language assessment and brain mapping in awake surgery for brain tumors. Although there is a consensus on this task between centers performing this type of surgery, the tests used vary from one team to another. The MULTIMAP picture naming task, initially developed in 2021 by the Basque Center on Cognition, Brain and Language, aimed to enhance the perioperative assessment of lexical access of both nouns and verbs in awake surgery patients and to facilitate standardized protocols for international multicenter studies. This study presents the French adaptation, digitization, and standardization of extraoperative MULTIMAP. Methods The tool was standardized on a sample of 416 healthy subjects recruited from the French population, whose performances (score and time) were statistically analyzed. Of the 100 items tested, we retained the 80 (40 objects, 40 actions) that showed the highest naming accuracy and were balanced between nouns and verbs for main psycholinguistic variables. Results Gender had no effect on performance, the level of education had a significant effect for the action naming task only, and age for both tasks, allowing to determine percentiles according to these variables. Performances in object and action naming were correlated; however, a significant but minimal difference between the average score and time between tasks was found. Conclusion The digitized French version of MULTIMAP is a promising tool, that awaits further validation in patients with acquired brain lesions, especially in the context of brain tumor awake surgery.

Isocitrate dehydrogenase (IDH) mutation and 1p/19q codeletion are key molecular markers for glioma classification. Amide proton transfer weighted (APTw) and nuclear Overhauser effect-weighted (NOEw) markers showed promise for glioma characterization, by probing protein-related tissue properties. However, their interpretation is confounded by direct water saturation, macromolecules (semi-solid magnetization transfer-ssMT), and T1 relaxation. Here, we aimed to assess the performance of three APTw and NOEw metrics-uncorrected, spillover/ssMT-corrected (FMC), and fully spillover/ssMT- and T1-corrected (FMTC)-for glioma stratification. Fifty patients with suspected gliomas were prospectively enrolled (12 IDH-wild-type, 38 IDH-mutant, of which 21 with 1p/19q codeletion). Acquisitions were performed at 3 T using a 3D gradient echo readout with chemical exchange saturation transfer (B1 = 2 μT for APTw, 0.6 μT for NOEw; T1sat = 2 s), WASABI (WAter Shift And B1) for B0/B1 mapping, and saturation recovery for T1 mapping. Glioma subtypes were compared using metrics extracted from manually segmented masks, using two-tailed Mann-Whitney U tests, and the Benjamini-Hochberg false discovery rate (FDR) correction. Effect sizes were quantified using Cliff's δ with 95% bootstrap confidence intervals and classification performance was assessed by receiver operating characteristic analyses (area under the curve, AUC). IDH-mutant and wild-type gliomas differed significantly for the uncorrected APTw metric (p = 0.005, AUC = 0.79), with stronger discrimination following correction-APTw-FMC (p < 0.001, AUC = 0.94) and APTw-FMTC (p < 0.001, AUC = 0.96), both with large effect sizes. Only APTw-FMTC distinguished 1p/19q codeleted from non-codeleted gliomas before FDR correction (uncorrected p = 0.01, AUC = 0.74). The NOEw metrics did not differ between any molecular subgroups, likely due to limited sensitivity of this contrast at 3 T. These results suggest that correcting for fluid, ssMT, and T1 effects enhances the accuracy of APTw metrics, offering a more robust and biophysically grounded approach to noninvasive glioma diagnosis.

IntroductionMagnetic resonance (MR) imaging is essential for diagnosing central nervous system (CNS) tumors, guiding surgical planning, treatment decisions, and assessing postoperative outcomes and complications. While recent work has advanced automated tumor segmentation and report generation, most efforts have focused on preoperative data, with limited attention to postoperative imaging analysis.MethodsThis study introduces a comprehensive pipeline for standardized postsurgical reporting in CNS tumors. Using the Attention U-Net architecture, segmentation models were trained, independently targeting the preoperative tumor core, non-enhancing tumor core, postoperative contrast-enhancing residual tumor, and resection cavity. In the process, the influence of varying MR sequence combinations was assessed. Additionally, MR sequence classification and tumor type identification for contrast-enhancing lesions were explored using the DenseNet architecture. The models were integrated seamlessly into an automated and standardized reporting pipeline, following the RANO 2.0 guidelines. Training was conducted on multicentric datasets comprising 2000 to 7000 patients, incorporating both private and public data, using a 5-fold cross-validation.ResultsEvaluation included patient-, voxel-, and object-wise metrics, with benchmarking against the latest BraTS challenge results. The segmentation models achieved average voxel-wise Dice scores of 87%, 66%, 70%, and 77% for the tumor core, non-enhancing tumor core, contrast-enhancing residual tumor, and resection cavity, respectively. Classification models reached 99.5% balanced accuracy in MR sequence classification and 80% in tumor type classification.DiscussionThe pipeline presented in this study enables robust, automated segmentation, MR sequence classification, and standardized report generation aligned with RANO 2.0 guidelines, enhancing postoperative evaluation and clinical decision-making. The proposed models and methods were integrated into Raidionics, open-source software platform for CNS tumor analysis, now including a dedicated module for postsurgical analysis.

BACKGROUND AND OBJECTIVES:Currently, there are no standardized clinical mapping protocols for monitoring of executive functions during awake glioma surgery, primarily due to a lack of evidence-based data for cognitive mapping. By aligning procedures and documentation practices across institutions, clinicians can overcome the current fragmentation in the field and iteratively work toward generating reproducible, high-quality Data sets that will better clarify the clinical relevance of white matter pathways involved in executive functions. A previously conducted pilot study led to the development of a standardized monitoring protocol and demonstrated that pooling of data is feasible when surgical teams commit to the study requirements. The primary goal of this multicenter study protocol is to investigate whether using this standardized protocol can identify white matter tracts involved in executive functions. METHODS:In this prospective, clinical observational study, we will continue data collection in 4 neurosurgical departments from the previously conducted pilot study and expand to other hospitals providing neurosurgical care. We aim to include adult patients that will undergo awake primary glioma surgery and undergo monitoring of executive functions with a uniform set of tasks for the following white matter tracts: frontal aslant tract, superior longitudinal fasciculus II and II, arcuate fasciculus, inferior fronto-occipital fasciculus. Data will be collected in a standardized manner for each patient before, during, and after surgery. EXPECTED OUTCOMES:The primary objective of this study was to determine if executive functions can be effectively monitored using a standardized protocol during awake glioma surgery in multiple neurosurgical centers. DISCUSSION:Despite limitations inherent to multicenter and observational studies, this study represents a necessary step toward developing a validated uniform way of collecting intraoperative findings on mapping of executive functions. The generation of high-quality Data sets is highly needed to extend the scientific basis for monitoring of white matter pathways involved in executive functions.

BACKGROUND AND OBJECTIVES: Patients with newly diagnosed lower-grade glioma (World Health Organization grade II and III) are typically of working age. However, work resumption after surgical resection is uncertain, possibly due to loss of capacity from resection of tumor-infiltrated brain regions. Therefore, we explore the association between work resumption and the resection location in addition to other patient, tumor, and treatment characteristics. METHODS: This retrospective cohort consisted of adults undergoing first-time resection for lower-grade glioma between 2011 and 2020 in hospitals in France, Germany, and the Netherlands. Employment was evaluated at baseline and within 1 year after surgery to determine work resumption. The association between work resumption and patient, tumor, and treatment characteristics was analyzed using logistic regression. Resection cavities were segmented from postoperative MRI scans, registered to standard brain space and related to gray nuclei, cortical networks, and white matter tracts using atlas parcellations. To identify brain regions potentially involved with work resumption, the association between work resumption and resection location was analyzed using Bayesian hurdle regression. The identified regions and characteristics were jointly analyzed in their association with work resumption using multiple logistic regression. RESULTS: Of 207 patients, 181 (87%) were employed at baseline. Of these employed patients, 111 (61%) had resumed work at follow-up. Male sex, younger age, and larger extent of resection were independent significant predictors of work resumption. Resection location was not associated with work resumption. CONCLUSION: Almost two-thirds of patients resumed work 1 year after surgery. Work resumption was associated with patient characteristics (male sex and younger age) and extent of resection, but not with resection location.

BACKGROUND AND OBJECTIVES:The posterior petrosal approach is a technically challenging procedure used to treat complex, deep-seated lesions. The petrosal ridge, located between the middle and posterior fossae, can be anatomically compared with the sphenoid ridge, which separates the anterior and middle fossae. Analogous to the orbitotemporal periosteal fold, which connects the orbit to the middle fossa dura, the endolymphatic sac (ELS) acts as a fold connecting the vestibule to the posterior fossa dura. This fold presents an obstacle to further elevating the dura from the posterior surface of the petrous bone. The objective of this study is to provide anatomic details and key landmarks for the preservation and safe peeling of the ELS, thereby expanding the surgical field during posterior petrosectomy and combined petrosal approaches. METHODS:Five formalin-fixed cadaveric heads were injected with colored silicone, and 10 sides were used for cadaveric dissections. RESULTS:At the level of the superomedial margin of the operculum, the periosteal layer is incised to enter a surgical plane between the dura propria and the ELS. The dura is then peeled away from the posterior aspect of the ELS, and the periosteal layer is further incised along the ELS. A dural thickening continuing medially from the operculum, corresponding to the Tubingen line, allows access to the plane between the dural leaflets before reaching the full extent of the ELS. CONCLUSION:Analogous to the orbitotemporal periosteal fold in anterior approaches, the ELS can be safely peeled from the dura of the posterior fossa, thereby enhancing the extradural exposure in these approaches.

BACKGROUND:Acquired brain injury can lead to subtle cognitive disorders that can be challenging to detect albeit impacting patients' long-term functional prognosis. Cog-First has been developed as a tablet-based self-administered cognitive screening tool to assess executive function, memory and attention in approximately 20 minutes, in the acute phase following brain injury. AIM:The aim of this study was to establish reliable normative data for Cog-First to enable meaningful comparisons between patients and a reference population. DESIGN:Cross-sectional study. SETTING:This study was conducted at the PRISME platform of Paris Brain Institute. POPULATION:Four hundred and six healthy French-speaking healthy volunteers were randomly selected from the Paris Brain Institute's database. METHODS:Each participant underwent the Cog-First assessment, which comprises seven subtests, in standardized conditions. Ninety-five participants performed the alternative version one month later to assess the test-retest effect. The effects of gender, age, years of education and test version, as well as their two-way interactions, were evaluated by generalized linear models (GLMs). Formulas from the GLMS were extracted to calculate a corrected score that removes the effects of age, sex, version and years of education. This enables us to derive percentiles in a population of healthy volunteers, allowing the development of the standardization process. RESULTS:The results revealed a significant influence of gender, age, level of education and version on several sub-scores. Based on these results, the standardization process was implemented by calculating the percentiles on the corrected scores in the population of healthy volunteers. Test-retest analyses indicated a learning effect on four out of seven subtests. CONCLUSIONS:The standardization of Cog-First resulted in the development of score formulas adjusted for gender, age, education and version, integrated within the software for automated scoring. CLINICAL REHABILITATION IMPACT:This study establishes reliable norms for Cog-First, enabling meaningful score interpretation and clinical use, thereby facilitating early detection of cognitive impairments and potentially improving patient outcomes. Further research is necessary to determine the tool's applicability and sensitivity in brain-injured patients.

Cavernous malformations (CM) of the brain are vascular abnormalities that carry a risk of bleeding, posing significant neurological and life-threatening challenges, particularly in posterior fossa. The efficacy of radiosurgery for cavernomas still remains a matter of debate, largely due to technical and statistical limitations. In this study, we present a series of posterior fossa cavernomas treated using CyberKnife radiosurgery, employing an innovative approach that integrates both technical and statistical advancements. We conducted a prospective series involving 35 posterior fossa cavernomas in 33 patients treated with low-dose radiosurgery protocols (12 Gy in a single fraction or 18 Gy in 3 fractions). Compared to previously published series, our approach targeted a larger treatment volume, encompassing the entire hemosiderin ring surrounding the cavernoma. Radiosurgery was indicated for cases of hemorrhage or progressive neurological deficits in anatomically challenging, nonsurgical areas. The statistical analysis was designed to address the unknown onset time of cavernoma prior to radiosurgery, enabling a more accurate calculation of the hemorrhage incidence rate before treatment. Follow-up evaluations, including clinical assessments and MRI, were conducted at 3-6-9-12-18-24 months and subsequently on an annual basis. With a mean follow-up duration of 26 months, exceeding the previously described latency period, and a median [IQR] follow-up of 13 months [8.7-30.4] which represents approximately half the latency period, only one patient experienced a recurrence of hemorrhage, occurring 20 months post-treatment and remaining asymptomatic. No patients exhibited radio-induced parenchymal changes or clinical deterioration following radiosurgery. These preliminary results support the strategy of increasing the target volume while reducing the radiation dose for cavernous malformations. We further recommend incorporating sensitivity analyses to evaluate the robustness of results, particularly in the context of uncertainties surrounding the time of onset of cavernomas.

OBJECTIVE The extent of resection (EOR) and postoperative residual tumor (RT) volume are prognostic factors in glioblastoma. Calculations of EOR and RT rely on accurate tumor segmentations. Raidionics is an open-access software that enables automatic segmentation of preoperative and early postoperative glioblastoma using pretrained deep learning models. The aim of this study was to compare the prognostic value of manually versus automatically assessed volumetric measurements in glioblastoma patients. METHODS Adult patients who underwent resection of histopathologically confirmed glioblastoma were included from 12 different hospitals in Europe and North America. Patient characteristics and survival data were collected as part of local tumor registries or were retrieved from patient medical records. The prognostic value of manually and automatically assessed EOR and RT volume was compared using Cox regression models. RESULTS Both manually and automatically assessed RT volumes were a negative prognostic factor for overall survival (manual vs automatic: HR 1.051, 95% CI1.034-1.067 [p < 0.001] vs HR 1.019, 95% CI1.007-1.030 [p = 0.001]). Both manual and automatic EOR models showed that patients with gross-total resection have significantly longer overall survival compared with those with subtotal resection (manual vs automatic: HR 1.580, 95% CI1.291-1.932 [p < 0.001] vs HR 1.395, 95% CI1.160-1.679 [p < 0.001]), but no significant prognostic difference of gross-total compared with near-total (90%-99%) resection was found. According to the Akaike information criterion and the Bayesian information criterion, all multivariable Cox regression models showed similar goodness-of-fit. CONCLUSIONS Automatically and manually measured EOR and RT volumes have comparable prognostic properties. Automatic segmentation with Raidionics can be used in future studies in patients with glioblastoma.

Background:Accurate prognosis of glioblastoma is crucial for better-informed treatment decisions, potentially leading to improved disease management. We investigated whether clinical variables, tumor size, and location, can serve as prognostic factors. Methods:A retrospective, multicenter study enrolled 1318 adult patients with histopathologically confirmed glioblastoma undergoing first-time surgery, with survival censored for 188 patients. Pre-operative brain MRIs were used to compute tumor size and derive advanced radiological features describing tumor location, later refined by expert-based opinion. Post-operative MRIs were used to measure the enhancing residual tumor volume. The prognostic quality of all variables, measurements, and features was assessed as inputs of three survival regression models (CoxPH, Random Survival Forests, DeepSurv) to predict overall survival, under five timepoints of patient treatment: onset presentation, assessment by multidisciplinary board, intervention planning, post-intervention evaluation, and chemoradiotherapy planning. Model evaluation was performed with the C-index, Brier Score over Time, and Integrated Brier Score. Results:Multivariable Cox analysis identified most clinical variables and tumor size as strong predictors of patient survival, with varying hazard ratios across timepoints. DeepSurv was consistently the top performing model under all possible inputs and at all timepoints, yielding mean test C-index scores ranging from 61.71% to 70.29%, and mean Integrated Brier Scores ranging from 8.57% to 7.63%. Conclusion:Clinical variables, tumor size, and location carry prognostic value for the overall survival of patients with glioblastoma. The best predictive performance was observed under a Deep Survival model using all variables at the stage of chemoradiotherapy planning.