Glioblastoma’s Next Target: Why Precision Drugs Are Still Struggling to Defeat the Brain’s Most Aggressive Cancer
Glioblastoma remains the most lethal primary brain tumour in adults, even after decades of progress in molecular biology and cancer drug development. The disease is defined by rapid growth, invasive behaviour and a remarkable ability to adapt when treatment applies pressure. Surgery, radiotherapy and temozolomide can extend survival, but recurrence is common, and targeted medicines have so far delivered only modest benefits for most patients. A new review in Nature Reviews Clinical Oncology argues that the central challenge is no longer simply finding mutations. Researchers must determine which alterations actually control tumour behaviour, identify vulnerabilities shared by the largest number of malignant cells and deliver therapies effectively across the blood–brain barrier.
The molecular landscape of glioblastoma is extraordinarily complex. Tumours that appear similar under the microscope can be driven by very different combinations of genetic alterations, while distinct regions within the same tumour may carry different mutations and signalling programs. This intratumoural heterogeneity means that a drug aimed at one mutation may destroy only a fraction of the cancer. The remaining cells can expand, repopulate the tumour and acquire additional resistance mechanisms. Glioblastoma cells also display considerable plasticity, changing their state in response to treatment, oxygen levels, immune signals and interactions with surrounding brain tissue. As a result, a therapy that works in one cellular state may become ineffective when the tumour shifts into another.
Receptor tyrosine kinases remain among the most prominent targets under investigation. These cell-surface proteins transmit signals controlling proliferation, survival, migration and metabolism. Abnormal activation of receptors such as epidermal growth factor receptor, platelet-derived growth factor receptors and vascular endothelial growth factor receptors can push glioblastoma cells into continuous growth. However, blocking a single receptor has often failed because signalling networks are highly redundant. A tumour may activate a parallel receptor, increase downstream signalling or use alternative pathways to maintain survival. The review therefore emphasizes the importance of understanding pathway activity rather than relying solely on the presence of a single genetic alteration. Functional profiling could reveal which signalling circuits remain essential after treatment begins.
Downstream intracellular proteins offer another set of potential targets. The PI3K–AKT–mTOR pathway, the RAS–RAF–MEK–ERK cascade and other signalling networks integrate information from growth-factor receptors and the tumour microenvironment. In glioblastoma, these pathways can be activated through mutations, gene amplification, loss of tumour suppressors or non-genetic mechanisms. Inhibiting them is technically difficult because the same pathways are needed by normal cells, creating toxicity concerns, while feedback loops can rapidly restore signalling after treatment. Effective drug combinations may need to block several nodes at once, but combination therapy increases the risk of adverse effects and complicates clinical testing. The most promising strategy may involve identifying a tumour’s dominant or “master” kinase state and matching it with drugs that suppress the pathway on which its cells are functionally dependent.
Cell-cycle dysregulation is another defining feature of glioblastoma. Alterations affecting cyclin-dependent kinases, retinoblastoma protein, p53-related responses and other cell-cycle regulators allow malignant cells to divide despite DNA damage and abnormal growth signals. Drugs targeting cyclin-dependent kinases or related checkpoints could, in principle, force tumour cells into arrest or increase their sensitivity to radiation and chemotherapy. Yet cell-cycle inhibitors must distinguish between rapidly dividing cancer cells and essential normal tissues. Moreover, glioblastoma contains slow-cycling and stem-like populations that may survive therapies designed primarily to eliminate proliferating cells. These resistant reservoirs can later re-enter the cell cycle, contributing to recurrence. The review presents synthetic-lethal approaches as a way to address this problem by targeting weaknesses that become critical only when a particular tumour suppressor or repair pathway is already lost.
Synthetic lethality occurs when disabling either of two genes or pathways alone is tolerated, but disabling both is fatal to the cell. This principle has transformed treatment strategies in some cancers and may be especially valuable in glioblastoma, where genome-maintenance systems are frequently disrupted. Tumours with defects in DNA repair may become unusually dependent on backup repair mechanisms, replication-stress responses or checkpoint proteins. Drugs that inhibit these compensatory systems could selectively damage cancer cells while sparing healthier cells with intact repair capacity. The approach is not without obstacles: glioblastoma cells can vary in their repair defects, and resistance may emerge through restoration of the damaged pathway or activation of alternative mechanisms. Careful molecular selection and repeated tumour monitoring will therefore be essential.
New therapeutic concepts are also expanding beyond conventional kinase inhibition. Researchers are examining strategies aimed at genome integrity, telomere maintenance and epigenetic regulation. Telomeres protect chromosome ends, but cancer cells often reactivate telomerase or use alternative mechanisms to maintain unlimited replicative potential. Blocking these systems could gradually undermine tumour growth, although the delayed action of telomere-directed treatments may limit their usefulness in rapidly progressing disease. Epigenetic drugs seek to alter gene expression without changing the DNA sequence, potentially reversing malignant cell states or restoring sensitivity to other therapies. Because epigenetic regulation is dynamic and closely linked to cellular identity, these agents could be used to target the plasticity that allows glioblastoma cells to escape treatment.
Drug delivery remains one of the most formidable barriers. The blood–brain barrier is formed by tightly connected endothelial cells, specialized transport systems and supporting pericytes and astrocytes. It protects neural tissue from toxins, but also prevents many anticancer drugs from reaching therapeutic concentrations in the tumour. Glioblastoma can locally disrupt the barrier, yet this disruption is uneven, and infiltrating cells may reside in areas where drug exposure is minimal. The review discusses focused ultrasonography as one method for temporarily opening the barrier, potentially allowing drugs to enter selected regions. Convection-enhanced delivery takes a different approach by infusing treatment directly into tumour tissue under pressure. Other strategies include modifying molecules to improve brain penetration, avoiding active efflux pumps that expel drugs from the central nervous system and developing delivery vehicles that transport therapeutic payloads more selectively.
The review also highlights antibody–drug conjugates and theranostics as emerging modalities. Antibody–drug conjugates combine a targeting antibody with a potent cytotoxic compound, allowing the drug to be carried toward cells displaying a chosen surface marker. Their success in glioblastoma will depend on whether that marker is sufficiently abundant across the tumour and accessible from the bloodstream. Theranostic platforms unite diagnosis and treatment, enabling clinicians to identify molecular targets, deliver a therapeutic agent and monitor its distribution or biological effect. These approaches may become more powerful when combined with multiregional sampling. Instead of analysing a single surgical specimen, clinicians could compare samples from different tumour zones to distinguish early, shared “truncal” alterations from later, region-specific changes. Targeting truncal dependencies could reduce the likelihood that untreated subclones survive.
Genomic complexity is further intensified by extrachromosomal DNA, or ecDNA, circular DNA elements that exist outside normal chromosomes. These structures can carry amplified oncogenes, sometimes at very high copy numbers, and can be unevenly distributed among tumour cells. Their inheritance during cell division is less predictable than that of chromosomal DNA, allowing cancer populations to rapidly change the number and arrangement of oncogenic copies under therapeutic pressure. EcDNA may therefore help explain why glioblastomas can become resistant with extraordinary speed. Liquid biopsies, using tumour-derived DNA, RNA or other molecules found in blood or cerebrospinal fluid, could provide a less invasive way to track these changes. Although technical and sensitivity challenges remain, serial monitoring might reveal emerging resistance before it becomes visible on imaging.
Future clinical trials may need to move beyond the traditional model in which one drug is tested against an entire disease category. Pathway-based classification could group patients according to active biological programs rather than tumour appearance alone, while master kinase mapping could identify the dominant signalling dependencies operating in each tumour. Adaptive trials could then modify treatment as molecular data change, testing rational combinations in smaller, genetically defined populations. Such designs may be especially important for glioblastoma because the disease evolves during therapy and differs substantially from one patient to another. The overall message of the review is cautiously optimistic: progress will require more than discovering additional mutations. It will depend on integrating spatial tumour profiling, functional biology, advanced delivery systems, dynamic monitoring and therapies capable of confronting the tumour’s shifting cellular states.
Subject of Research: Targetable molecular alterations and therapeutic strategies in glioblastoma.
Article Title: Exploring the landscape of targetable alterations in patients with glioblastoma.
Article References: Aquilanti, E., Touat, M., French, P. et al. “Exploring the landscape of targetable alterations in patients with glioblastoma.” Nature Reviews Clinical Oncology (2026). https://doi.org/10.1038/s41571-026-01190-7
Image Credits: AI Generated
DOI: 10.1038/s41571-026-01190-7
Keywords: Glioblastoma, targeted therapy, intratumoural heterogeneity, blood–brain barrier, receptor tyrosine kinases, synthetic lethality, extrachromosomal DNA, liquid biopsy, epigenetic therapy, theranostics.
