TMIC-91. Deciphering spatio-temporal trajectories of tissue adaptation after surgical resection of glioblastoma

Abstract Local relapse along the resection cavity is the most frequent mode of glioblastoma progression. Surgically induced molecular alterations and transformation of residual tumor ecosystems driving local tumor relapse remain elusive. Here, we applied MERFISH and Visium spatial transcriptomics to 203 specimens from 179 patients obtained at first surgery, early re-resection (2–40d) and late recurrence, and trained a spatial graph-attention network (sGAT) that predicted time-after-surgery from microenvironmental neighborhoods (AUC=0.83). Explainable-AI approaches (Integrated-gradient/attention) revealed a stereotyped cascade dominated by a hypoxia/HIF-1α-program with glycolytic switch and lactate-fueled-coupling between residual tumor cells, neurons and infiltrating myeloid cells within 2–10d at peri-cavity-rim. Between days 10–30 the niche transitions to TREM2⁺/P2RY12⁺ microglia, CXCL13⁺ T-cell influx and endothelial proliferation. Multiparametric MRI confirmed a stroke-like-hypoperfusion and blood-brain-barrier breach peaking at days 1–5 and resolving by days 60–90. To validate time-resolved tissue adaptation and causality, we performed spatial metabolomic and transcriptomic profiling on our murine-cortical-resection model. The sGAT classifier trained on human data assigned the exact post-operative status to murine samples (R² = 0.91). MALDI-imaging mass spectrometry confirmed the enhanced lactate consumption in proximity to the resection cavity enriched for neurons and altered microglia. Knockout of the lactate transporter MCT2 in neurons (rAAV:::hSyn:::shRNA-MCT2) demonstrated reduced tumor proliferation and lower-invasive growth (padj<0.05) with reduced tumor-neuron synaptic contacts in a rabies-ΔG-GFP retrograde tracing model using human neocortical slices (padj < 0.01). Similar reduction of tumor-neuron synaptic contacts was observed after depletion of microglia with PLX5622, supporting a co-operative metabolic-inflammatory-axis. Post-resection metabolic adaptation and tumor regrowth link was confirmed in vivo leveraging a LDH inhibitor or ROS scavenger demonstrating a significant decrease in tumor relapse. Our data delineate a narrow, druggable window immediately after surgery in which hypoxic-inflammatory-reprogramming metabolically and synoptically primes glioblastoma for recurrence, nominating peri-cavity lactate-shuttling or microglial activation as actionable adjuvant targets.

Authors

Institutions

Publication Details

Journal
Neuro-Oncology
Published
2025-11-01
DOI
https://doi.org/10.1093/neuonc/noaf201.1846
Primary Topic
Glioma Diagnosis and Treatment
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

TMIC-91. Deciphering spatio-temporal trajectories of tissue adaptation after surgical resection of glioblastoma

Mathieu Larroquette, Roman Sankowski, Tao Peng, Jasim Kada Benotmane et al.
Neuro-Oncology
Glioma Diagnosis and Treatment
article

TMIC-91. Deciphering spatio-temporal trajectories of tissue adaptation after surgical resection of glioblastoma

Mathieu Larroquette, Roman Sankowski, Tao Peng, Jasim Kada Benotmane, Thomas Daubon, Giulia Villa, Oliver Schnell, Michel Salzet, Yahaya A Yabo, Claire Larrieu, M. Arrieta, Dieter Henrik Heiland, Audrey Burban, Franz Ricklefs, Macha Nikolski, Varun Venkataramani, Yanis Zirem, Antonio Pagano Zottola, Mathis Pinglaut, Junyi Zhang, Elena Grabis, Doriane Bomont, Axell Davoust
article en

Abstract

Abstract Local relapse along the resection cavity is the most frequent mode of glioblastoma progression. Surgically induced molecular alterations and transformation of residual tumor ecosystems driving local tumor relapse remain elusive. Here, we applied MERFISH and Visium spatial transcriptomics to 203 specimens from 179 patients obtained at first surgery, early re-resection (2–40d) and late recurrence, and trained a spatial graph-attention network (sGAT) that predicted time-after-surgery from microenvironmental neighborhoods (AUC=0.83). Explainable-AI approaches (Integrated-gradient/attention) revealed a stereotyped cascade dominated by a hypoxia/HIF-1α-program with glycolytic switch and lactate-fueled-coupling between residual tumor cells, neurons and infiltrating myeloid cells within 2–10d at peri-cavity-rim. Between days 10–30 the niche transitions to TREM2⁺/P2RY12⁺ microglia, CXCL13⁺ T-cell influx and endothelial proliferation. Multiparametric MRI confirmed a stroke-like-hypoperfusion and blood-brain-barrier breach peaking at days 1–5 and resolving by days 60–90. To validate time-resolved tissue adaptation and causality, we performed spatial metabolomic and transcriptomic profiling on our murine-cortical-resection model. The sGAT classifier trained on human data assigned the exact post-operative status to murine samples (R² = 0.91). MALDI-imaging mass spectrometry confirmed the enhanced lactate consumption in proximity to the resection cavity enriched for neurons and altered microglia. Knockout of the lactate transporter MCT2 in neurons (rAAV:::hSyn:::shRNA-MCT2) demonstrated reduced tumor proliferation and lower-invasive growth (padj<0.05) with reduced tumor-neuron synaptic contacts in a rabies-ΔG-GFP retrograde tracing model using human neocortical slices (padj < 0.01). Similar reduction of tumor-neuron synaptic contacts was observed after depletion of microglia with PLX5622, supporting a co-operative metabolic-inflammatory-axis. Post-resection metabolic adaptation and tumor regrowth link was confirmed in vivo leveraging a LDH inhibitor or ROS scavenger demonstrating a significant decrease in tumor relapse. Our data delineate a narrow, druggable window immediately after surgery in which hypoxic-inflammatory-reprogramming metabolically and synoptically primes glioblastoma for recurrence, nominating peri-cavity lactate-shuttling or microglial activation as actionable adjuvant targets.

Neuro-OncologyVol. 27(Supplement_5)
Centre National de la Recherche Scientifique (FR), Université de Bordeaux (FR), Universität Hamburg (DE), University of Freiburg (DE), German Cancer Research Center (DE), Friedrich-Alexander-Universität Erlangen-Nürnberg (DE), Heidelberg University (DE), University Medical Center Freiburg (DE), University Hospital Heidelberg (DE), Universitätsklinikum Erlangen (DE), University Medical Center Hamburg-Eppendorf (DE), National Center for Tumor Diseases (DE), Institut de Biochimie et Génétique Cellulaires (FR), Comprehensive Cancer Center Erlangen (DE), Protéomique, Réponse Inflammatoire et Spectrométrie de Masse (FR)
Openalex Percentile: Top 52%
Glioma Diagnosis and Treatment
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.