Glioblastoma is among the deadliest human tumors, and nearly all patients experience recurrence despite maximal treatment with surgery and radio-chemotherapy. A key driver of this poor prognosis is the diffuse infiltration of tumor cells into the surrounding brain, where they interconnect with each other and with non-malignant cells to form resistant networks. These infiltrating cells are usually not amenable to surgical removal and are thought to constitute the reservoir from which recurrence emerges. Yet their biology, and how they interact with their local environment, remains poorly understood largely because conventional tumor resections rarely capture tissue far away from the tumor bulk.
This project takes advantage of a rare resource: anterior temporal lobectomy specimens from patients with temporal glioblastoma, which uniquely preserve the entire continuum from the tumor core to the distant, far infiltration zone. Using spatial transcriptomics (Xenium) and single-cell RNA sequencing (scFlex) across human ATL specimens, we characterize the transcriptional states of infiltrating tumor cells, identify their supportive microenvironmental niches, and map cell-cell communication between malignant and non-malignant cells. By linking these molecular findings to preoperative MRI features, the project aims to build a spatial multi-omics atlas of glioblastoma infiltration with the long-term goal of refining concepts of surgical resection and radiotherapy planning to better target the cellular reservoir responsible for recurrence.

https://www.ukbonn.de/neurochirurgie/forschung/onkologische-forschung/