Scientists map a million cells revealing the predictable rules of glioblastoma

Glioblastoma, the most common and most lethal brain cancer in adults, has long been described as a chaotic disease. Tumours contain a shifting mixture of cancer cell states that seemed to obey no consistent pattern, which is one reason why treatment so often fails and the cancer returns. Now a large collaborative study argues that this apparent randomness is an illusion. By mapping more than one million individual cells across glioblastoma tumours, researchers at the Wellcome Sanger Institute, the European Molecular Biology Laboratory, the German Cancer Research Center, the Cancer Research UK Cambridge Institute and their collaborators have shown that these aggressive tumours follow predictable structural rules as they grow, and that the same sequence of events repeats from one patient to the next.

The work, published in Nature, combined several state-of-the-art technologies to build what the team calls the GBM-Space atlas, a freely accessible resource that captures glioblastoma in unprecedented molecular and spatial detail. The researchers analysed multiple sections of brain tumour samples from 12 patients, integrating single-cell and spatial atlasing, high-resolution imaging and whole-genome sequencing. This combination allowed them to determine not only the molecular identity of each cancer cell but also its precise location within the tumour, the environment surrounding it, its genetic makeup and the developmental direction it was likely heading. Samples were taken from multiple regions of each tumour, giving a fuller picture of how neighbouring cell populations relate to one another within a single tumour section.

What emerged from this data was a recurring architectural pattern. At the outer rim of the tumour, known as the leading edge, where glioblastoma cells begin to invade surrounding healthy brain tissue, the cancer cells resemble brain stem cells. These developmental-like cells act as a reservoir, giving rise to the other cancer cell states found deeper in the tumour. As the tumour expands, these early cells transform into injury response states, mimicking molecular pathways normally involved in brain tissue repair and wound healing. The injury response cells form an inner structural layer of the tumour. In the later stages of tumour development, the core becomes cut off from oxygen, and at that point all cell types transition from injury response states into severely hypoxic, oxygen-deprived states.

To confirm that this roadmap was not an artefact of human tissue analysis alone, the team also developed three-dimensional models of the brain in a dish, known as brain organoids, and used lab-grown mini tumours to mimic tumour development across different time stages. These experimental systems allowed the researchers to categorise cancer cells at different points along their journey within the different environments of a glioblastoma tumour, and the trajectory they observed matched the pattern seen in patient samples. The result challenges a long-held assumption that glioblastoma cell states are too dynamic and disordered to be systematically targeted.

The clinical implications of a predictable trajectory are significant. Because cancer cells move through the same structural pattern during tumour progression, future drugs could be designed to intercept the early developmental-like cells before they transform into harder-to-reach, resistant states. These early cells sit at the leading edge, where they cannot easily be removed by surgery, and they are thought to often drive cancer recurrence. Dr Richard Mair, co-author at the Cancer Research UK Cambridge Institute and Cambridge University Hospitals NHS Foundation Trust, said the research offers a chance to think differently about treatment design. For too long, he noted, clinicians have approached glioblastoma knowing each tumour contains a complex mixture of cancer cells without a clear understanding of how best to target them all. If the different biological states a tumour is likely to adopt can be anticipated, he suggested, treatments could be developed that target multiple aspects of the disease from the very beginning, informed not just by what a tumour looks like today but by what it is likely to become.

The study also revealed that each stage of cancer cell progression requires a specific immune microenvironment within the tumour. The team suggests that finding ways to interrupt or disrupt these environments could open new avenues for treating aggressive brain tumours. Beyond immune targeting, future therapeutics could focus on the gene regulation networks that control cell state transitions, potentially halting cancer cell progression, or could aim directly at the early developmental cells that seed the rest of the tumour.

That regulatory dimension is the subject of a companion paper, also published in Nature, in which the researchers decoded the gene regulators that drive and restrict glioblastoma cell state transitions. Using the same single-cell multi-omic data, the team identified molecular programs that govern how cancer cells move between states, highlighting potential targets that could help avoid treatment resistance. Together, the two papers unify an understanding of glioblastoma architecture with insights into cancer cell plasticity, offering a holistic view of a cancer that has evaded explanation for decades despite being among the most intensively studied tumour types.

Dr Oliver Stegle, co-senior author of both studies at the European Molecular Biology Laboratory and the German Cancer Research Center, said that combining unbiased single-cell and multi-omics techniques with tailored computational tools allowed the team to go back to the drawing board and challenge long-held assumptions about glioblastoma. The work demonstrates, he said, that complex but predictable molecular networks govern cancer cell development, and shows how computational biology can turn massive datasets into meaningful discoveries for patient care. Dr Omer Bayraktar, co-senior author at the Wellcome Sanger Institute and Professor at the UCL Cancer Institute, added that by building the largest multi-modal atlas of these tumours to date, the team has shown that glioblastoma is not as chaotic as previously accepted, but instead follows a set of rules that can be used to rewrite understanding of the disease and guide much-needed new treatments.

The urgency of that search is reflected in the statistics. Around 3,200 people are diagnosed with glioblastoma every year in the United Kingdom. Average survival is 12 to 18 months, and only 4 per cent of those diagnosed survive five years or more. When treatment is possible, it typically involves surgery followed by chemotherapy and radiotherapy. But the tumours are difficult to treat because of their location and because many invade multiple parts of the brain, which can make surgery difficult or impossible. Even when treatment succeeds in destroying most tumour cells, the highly dynamic population of cancer cell states means that some cells survive, leading to drug resistance and re-emergence of the cancer.

All 12 tumour samples analysed in the study came from patients undergoing neurosurgery as part of their medical treatment at Cambridge University Hospitals NHS Foundation Trust, collected under approved ethical protocols with pre-operative informed consent. The researchers emphasise that the atlas and the roadmap it contains are a foundational resource rather than an immediate therapy. The immediate value lies in giving researchers everywhere a common map of glioblastoma’s architecture, a reference against which new hypotheses about tumour biology can be tested and against which candidate drugs can be evaluated. If the predictable trajectory holds up in further studies, the hope is that intercepting tumours early in their developmental sequence, disrupting the immune niches each stage depends on, or blocking the regulatory programs that drive state transitions could finally give patients with this devastating disease treatments designed for what the tumour is becoming, not merely what it already is.

Subject of Research: Single-cell spatial mapping of glioblastoma tumour architecture and cancer cell state progression

Article Title: Uncovering rules for aggressive brain tumors could lead to better treatments

Article References: Uncovering rules for aggressive brain tumors could lead to better treatments. Original publication (Bibliographic details unavailable.)

Image Credits: AI Generated

DOI: Not provided

Keywords: glioblastoma, single-cell sequencing, spatial atlas, tumour heterogeneity, cancer cell plasticity, brain tumour, hypoxia, gene regulation, brain organoids, treatment resistance, Nature, multi-omics