The review begins with one of the most striking discoveries in modern cancer neuroscience: metastatic tumor cells do not merely tolerate the brain, they actively mimic it. When patient-derived breast and lung cancer cells come into contact with neurons, they upregulate genes associated with synaptic activity and plasticity, including CNR1, EGR2 and ARC. They also increase shuttling of SNAP25, a core component of the SNARE complex that mediates neurotransmitter release, essentially equipping themselves with the molecular machinery neurons use to communicate. This neuronal-like transformation is not limited to breast cancer; single-cell RNA sequencing of patient samples has identified similar neural gene programs in melanoma and non-small cell lung cancer brain metastases, and remarkably, some of these programs resemble those found in glioma cells, suggesting convergent evolutionary strategies for surviving in neural tissue.

The functional consequences of this mimicry are profound. Brain metastatic cells exploit the brain’s neurotransmitter economy for their own growth. In breast cancer cells, exposure to the brain microenvironment induces uptake of gamma-aminobutyric acid through elevated expression of multiple GABA transporters, and the enzyme GABA transaminase then channels this GABA through the GABA shunt into the citric acid cycle, providing metabolic fuel for proliferation. Glutamate, the brain’s principal excitatory neurotransmitter, is similarly co-opted. Triple-negative and basal-like breast tumors express N-methyl-D-aspartate receptor subunits, with phosphorylated GluN2B enriched in brain metastases compared to primary tumors. These receptors allow tumor cells to form pseudo-tripartite synapses with neurons, analogous to how perisynaptic astrocytes interface with synaptic junctions, and NMDA receptor signaling then promotes tumor cell invasion through calcium-dependent kinase pathways.

This metabolic dependency on neurotransmitters creates unexpected therapeutic openings. Because brain metastatic cells rely on GABA processing through GABA transaminase, drugs that inhibit this enzyme block a critical metabolic pathway. Valproic acid and vigabatrin, both established anticonvulsants, have demonstrated efficacy against brain metastasis xenografts by precisely this mechanism. Preclinical studies have also shown that certain antipsychotics, including trifluoperazine, fluphenazine, clozapine and sertindole, suppress proliferation and viability of melanoma and breast cancer brain metastatic cells, though the exact relationship between their cytotoxic effects and the neuronal-like phenotype of these tumors remains to be clarified. Beyond classical neurotransmitters, the review highlights how fasting-induced ghrelin crosses the blood-brain barrier and stimulates neuronal release of neuropeptide Y, which then acts on Y5 receptors on tumor cells to trigger a metabolic switch toward fatty acid oxidation, providing a potential explanation for why underweight lung cancer patients face elevated brain metastasis risk.

Astrocytes, the most abundant glial cells in the central nervous system, emerge from the review as profoundly ambivalent actors. In the earliest stages of metastatic seeding, they may actively resist tumor invasion by secreting plasminogen activator, which generates plasmin to kill incoming cancer cells. Tumor cells counter this defense by expressing plasminogen activator inhibitory serpins. Astrocytes can also induce a state of dormancy in metastatic cells through two distinct mechanisms: deposition of the extracellular matrix glycoprotein laminin-211, which sequesters the transcription factor YAP away from the nucleus and suppresses proliferation, and inhibition of DNA methyltransferase 1, which triggers epigenetic changes that push tumor cells into quiescence. This dormancy phase is considered a rate-limiting step in metastatic development, meaning astrocytes may temporarily contain disease before conditions permit explosive outgrowth.

Once lesions are established, however, the astrocyte-tumor relationship turns decisively pro-tumor. A subpopulation of reactive astrocytes marked by phosphorylated STAT3 co-localizes with brain metastatic cells and secretes a cocktail of immunosuppressive factors including vascular endothelial growth factor A, tissue inhibitor of metalloproteinases-1 and lipocalin-2, alongside extracellular matrix components such as nidogen-2 and neurocan that form physical barriers excluding CD8-positive T-lymphocytes. TIMP-1 in this secretome binds CD63 on T-lymphocytes, reducing expression of cytotoxicity genes through ERK 1/2 signaling and impairing tumor cell killing. This discovery underpins ongoing clinical testing of silibinin, a STAT3 inhibitor, in combination with immune checkpoint blockade, building on an active phase 2 trial of silibinin monotherapy in resected lung and breast cancer brain metastases. Astrocytes also communicate with tumor cells through connexin 43 gap junctions, transferring calcium away from cancer cells to confer chemotherapy resistance, and shuttling the second messenger cGAMP to activate STING signaling in astrocytes, which then release interferon-alpha and tumor necrosis factor that paradoxically enhance tumor cell survival through STAT1 and NF-kappaB pathways.

Tumor-associated macrophages, comprising both resident microglia and bone marrow-derived monocyte macrophages recruited from the periphery, represent another pillar of the brain metastatic microenvironment. Primary breast tumors can prepare the brain for colonization before cancer cells ever arrive, secreting cyclooxygenase-2 and prostaglandin E2 to draw CD11b-positive myeloid cells across the blood-brain barrier and establish what researchers call premetastatic soil. Once inside, these myeloid cells produce S100A8 and S100A9 proteins that recruit additional immune cells and attract tumor cells. Single-cell RNA sequencing has revealed that tumor-associated macrophages exist along a dynamic differentiation continuum, transitioning from APOE-positive states with strong antigen-presentation capacity toward S100A8-positive states resembling myeloid-derived suppressor cells, with elevated CXCL8 and reduced human leukocyte antigen expression. This trajectory suggests tumor cells progressively educate macrophages from a predominantly anti-tumor to a pro-tumor phenotype over time. The transition is not absolute, however. Inhibition of colony-stimulating factor 1 receptor with the agent BLZ945 prevents early microglia-tumor interactions and reduces breast tumor growth in mice, implying microglia play tumor-supporting roles even in the earliest colonization phases.

Neutrophils, long overlooked in brain metastasis biology, are now recognized as abundant and functionally diverse infiltrating cells. The neutrophil-to-lymphocyte ratio, measurable from routine blood tests, predicts worse outcomes in patients receiving stereotactic radiosurgery and identifies lung cancer patients at elevated risk of brain metastasis. Within tumors themselves, neutrophils are drawn to perivascular niches by chemokines including CXCL8, CXCL1, CXCL2 and CXCL5, where they release neutrophil extracellular traps that facilitate tumor cell migration and invasion. A newly described pathway reveals how tumor cells recruit immunosuppressive neutrophils: phosphorylation of the chromatin regulator EZH2 by the proto-oncogene c-Src redirects EZH2 to bind RNA polymerase II, upregulating the transcription factor c-Jun and driving granulocyte colony-stimulating factor-dependent recruitment of arginase-1 and PD-L1 expressing neutrophils that suppress CD8-positive T-lymphocyte proliferation. Neutrophil phenotypes also vary dramatically by tumor genotype; in TP53-mutant lung cancer brain metastases, neutrophils undergo metabolic reprogramming toward fatty acid oxidation that generates reactive oxygen species and reinforces immunosuppression, while in kataegic breast cancer lesions with focal hypermutation, neutrophils instead display pro-inflammatory features.

T-lymphocytes infiltrate brain metastases at levels far exceeding those seen in gliomas, and this difference likely explains why immune checkpoint inhibitors such as nivolumab, pembrolizumab and ipilimumab have achieved intracranial response rates in melanoma, renal cell carcinoma and lung cancer that were once considered unattainable. Melanoma brain metastases are the most immune-infiltrated, showing high densities of CD8-positive cells and spatial correlation between PD-1 expression on lymphocytes and PD-L1 on tumor cells, a pattern consistent with an active but exhausted immune microenvironment. Breast cancer brain metastases are relatively less infiltrated, though triple-negative tumors harbor more cytotoxic T-lymphocytes than HER2-positive ones. Intriguingly, recent work shows that the immune cell populations and T-cell receptor sequences found within brain metastases are largely mirrored in the cerebrospinal fluid, opening the possibility that a simple lumbar puncture could serve as a minimally invasive method to identify patients likely to benefit from immunotherapy or to guide T-cell receptor-directed treatments. Natural killer cells, though a minor fraction of the infiltrate, have emerged as unexpected allies; they produce chemokines such as CXCL9 and CXCL16 that facilitate CD8-positive T-cell trafficking, and their absence compromises the efficacy of combined PD-1 and CTLA-4 blockade in melanoma brain metastasis models.

The review concludes that no single cell type in the brain metastatic microenvironment is exclusively friend or foe, and that functional plasticity driven by tumor-intrinsic and extrinsic pressures creates profound heterogeneity both within individual lesions and across different primary tumor types. The authors call for accelerated integration of single-cell RNA sequencing, spatial transcriptomics and proteomic analyses on patient-derived tissues, combined with longitudinal sampling to track how the microenvironment evolves during treatment. They envision a future in which tumor-specific vulnerabilities at the microenvironmental level are identified and exploited in a personalised manner, transforming brain metastases from a uniformly devastating diagnosis into a tractable and individually targeted clinical challenge.

Recent groundbreaking research from Ludwig Lausanne, led by Johanna Joyce and former postdoctoral fellow Ángel Álvarez-Prado, marks a pivotal advance in understanding and potentially overcoming GBM’s stubborn defenses. Published in the current issue of Cell Reports, this study zeroes in on ADAR1, a protein that acts like a molecular “off-switch” for the innate antiviral defense system within mammalian cells. By disabling ADAR1, the researchers demonstrate a simultaneous dual disruption of tumor growth dynamics and the tumor’s immunosuppressive microenvironment, offering a novel therapeutic pathway that could radically transform GBM treatment.

At the cellular level, ADAR1 plays a vital role in preventing unwarranted activation of antiviral pathways by chemically modifying endogenous double-stranded RNA (dsRNA) species. Cells inherently produce dsRNA molecules, but they are typically “edited” by ADAR1 to avoid being mistaken for foreign, virus-derived RNA. This editing suppresses the internal antiviral alarm that would otherwise provoke the production of type I interferons and spark a potent immune response. The delicate balance maintained by ADAR1 safeguards against autoimmune pathology but, paradoxically, also shields certain cancer cells from immune detection and destruction.

In cancers such as GBM, a subset of tumor cells often expresses interferon-stimulated genes (ISGs), rendering them potentially susceptible to disruptions in this antiviral equilibrium. Joyce’s lab investigated whether this dependency on ADAR1 could be therapeutically exploited. Using a combination of genetically engineered mouse models mimicking the heterogeneity of human GBM as well as patient-derived tumor cell cultures, the team systematically deleted ADAR1 and observed profound effects. Loss of ADAR1 not only arrested the proliferation of diverse tumor cell populations but also reprogrammed the tumor microenvironment (TME) from its characteristic immunosuppressive state to one that actively recruits and mobilizes immune effector cells.

Mechanistically, ADAR1 deletion unleashed an endogenous antiviral signaling cascade typically muted in tumor cells. This cascade induces intracellular pathways that halt protein synthesis, effectively locking cancer cells in a non-proliferative state. This cellular stress response was striking in tumor cells but absent in normal neural cell cultures, suggesting a therapeutic window that might spare healthy brain tissue. The specificity of this effect opens new avenues for targeted treatments that could avoid the severe collateral damage often seen with conventional therapies.

Crucially, the immunological landscape within the GBM microenvironment underwent a dramatic shift upon ADAR1 loss. The team documented increased infiltration and activity of cytotoxic CD8+ T cells, pro-inflammatory macrophages, and natural killer (NK) cells—key players in anti-tumor immunity. Concurrently, populations of immunosuppressive cells, which usually shield the tumor from immune attack, were depleted. This dual mode of action—direct tumor cell arrest combined with immune activation—embodies a one-two punch that stands to overcome the two fundamental barriers that have long stymied GBM therapy.

Álvarez-Prado, who now leads his own research group at the Luxembourg Institute of Health, highlighted the translational potential of these findings. He noted that targeting ADAR1 could revolutionize GBM treatment by offering a strategy effective across genetically diverse tumors, sparing normal brain cells while simultaneously unleashing the immune system against the cancer. This broad applicability is particularly significant given the notorious intra- and inter-tumoral heterogeneity of GBM, which has been a critical obstacle to uniformly successful treatments.

Looking ahead, the Joyce laboratory intends to focus efforts on the development of small molecule inhibitors of ADAR1 that can efficiently cross the blood-brain barrier, a notorious challenge in neuro-oncology drug design. Preclinical studies using these inhibitors in models that closely recapitulate human disease will be essential for validating this approach and refining dosage and administration regimens. Such studies could pave the way for clinical trials, potentially heralding a new era in GBM therapeutics.

This work builds on a growing body of literature that underscores the role of ADAR1 in cancer immune evasion. Previous research in melanoma demonstrated improved immunotherapy responses following ADAR1 deletion, and the current study extends these insights into the realm of brain cancer. By elucidating the mechanisms by which ADAR1 safeguards tumors from innate immune signaling and revealing the therapeutic vulnerabilities that arise from its loss, this research advances the frontiers of cancer immunology and precision medicine.

The implications of activating the body’s innate virus-fighting machinery against GBM represent a paradigm shift. Rather than relying solely on external drugs or immunotherapies, this strategy harnesses intrinsic cellular antiviral pathways previously suppressed within tumors. Enhancing endogenous immune detection and reprogramming suppressive microenvironments may break the therapeutic stalemate that has persisted for decades in brain cancer treatment.

In summary, this pioneering study ushers in hope against a cancer type that has long evaded effective control. By targeting ADAR1, a molecular switch that balances antiviral immunity within cells, researchers have established a promising avenue for both halting tumor progression and engaging the immune system’s destructive potential. This dual approach might finally shift the landscape of glioblastoma from one of inevitable decline to one of meaningful survival and improved quality of life.

Subject of Research: Glioblastoma multiforme (GBM), ADAR1 protein, tumor microenvironment, cancer immunotherapy
Article Title: ADAR1 Inhibition Reprograms Glioblastoma Microenvironment and Halts Tumor Proliferation
News Publication Date: September 5, 2025
Web References: https://www.ludwigcancerresearch.org/scientist/johanna-joyce/; https://www.cell.com/cell-reports/fulltext/S2211-1247(25)00922-2
Image Credits: Ludwig Cancer Research
Keywords: Glioblastoma, ADAR1, tumor microenvironment, immunotherapy, interferon-stimulated genes, glioblastoma treatment, cancer immunology, brain cancer, innate immunity

Glioblastoma multiforme (GBM) represents a dire medical challenge, notorious for its rapid proliferation, resistance to treatment, and dismal patient prognosis. Despite intensive research, effective targeted therapies remain elusive. The recent findings pinpoint the mitochondrial trifunctional protein subunit alpha (HADHA) as a pivotal metabolic regulator intricately linked to oncogenic signaling pathways involved in tumor survival and expansion. Such a dualistic function in both metabolism and epigenetic control is particularly compelling, bridging two formerly considered disparate realms of cancer biology.

HADHA’s canonical role involves the beta-oxidation of long-chain fatty acids within mitochondria, a critical component of cellular energy homeostasis. However, the novel insight from Wang and colleagues establishes a hitherto unknown function of HADHA in modulating JAK/STAT3 signaling—a pathway notoriously implicated in the proliferation, immune evasion, and stemness of glioblastoma cells. This dual functional capacity implies that metabolic enzymes may exert far-reaching influences beyond canonical bioenergetics, functioning as epigenetic modulators that sculpt oncogenic transcriptional programs.

The study employs a sophisticated integration of metabolomic profiling, chromatin immunoprecipitation sequencing (ChIP-seq), and proteomic analyses to delineate how HADHA influences STAT3 phosphorylation and nuclear translocation. The data reveal that suppression of HADHA disrupts fatty acid oxidation flux, leading to alterations in the cellular acetyl-CoA pool. These metabolic changes then cascade to affect histone acetylation patterns, thereby epigenetically reprogramming STAT3 target gene expression. This mechanistic pathway suggests a feedback loop wherein mitochondrial metabolism directly informs chromatin architecture, fine-tuning gene expression landscapes critical for glioblastoma malignancy.

Importantly, the authors demonstrate that silencing HADHA expression in glioblastoma cell lines markedly diminishes tumor cell viability and invasiveness in vitro, effects that are rescued by enforced activation of STAT3 signaling. Such functional assays affirm the indispensable role of HADHA-mediated metabolic regulation in sustaining JAK/STAT3-driven oncogenic phenotypes. This crosstalk underscores an integrative axis that could be exploited pharmacologically; inhibiting HADHA might concurrently disrupt energy metabolism and epigenetic oncogene expression, delivering a one-two punch to tumor progression.

Further reinforcing clinical relevance, analysis of patient-derived glioblastoma specimens reveals a positive correlation between HADHA expression and STAT3 activation status, as well as poorer overall survival rates. These findings point toward HADHA not only as a mechanistic node but also as a prognostic biomarker for aggressive disease. The ability to stratify patients based on HADHA-STAT3 axis activity could refine precision oncology approaches and inform therapeutic decision-making.

Beyond glioblastoma, this study propels a paradigm shift concerning metabolic enzymes as epigenetic regulators. It embodies the concept that metabolism and gene regulation exist not as isolated processes but as deeply entwined networks that cooperate to drive tumor biology. By illuminating this previously unappreciated metabolic-epigenetic axis, the research opens fertile ground for investigating analogous pathways in other malignancies characterized by metabolic dysregulation and aberrant JAK/STAT signaling.

Equally significant is the methodological rigor with which the team interrogated the regulatory axis. Using CRISPR/Cas9-based genetic editing, targeted metabolite supplementation, and advanced microscopy techniques to visualize STAT3 localization changes, the research offers a multi-dimensional perspective. Such comprehensive approaches ensure that findings are not artifacts of in vitro models but robust phenomena with in vivo translational potential.

The therapeutic implications of these discoveries are vast. Traditional strategies targeting JAK/STAT pathways often encounter obstacles such as compensatory signaling and systemic toxicities. By targeting HADHA, an upstream metabolic regulator, there is potential to circumvent such resistance mechanisms while simultaneously impairing tumor energetics and epigenetic maintenance. Drug development efforts could focus on small molecules or peptides that specifically inhibit HADHA’s enzymatic function or disrupt its interaction with STAT3 co-factors, thus providing finely tuned interventions.

Moreover, combining HADHA inhibition with existing modalities such as temozolomide chemotherapy or immune checkpoint blockade may potentiate anti-tumor efficacy. Given the immunosuppressive microenvironment in glioblastoma, the ability to modulate metabolic-epigenetic pathways influencing immune evasion could reinvigorate host anti-tumor responses. Translational research aimed at evaluating such combinatorial regimens may pave the path for clinical trials.

Wang et al.’s elucidation of the HADHA-JAK/STAT3 axis underscores the growing appreciation for metabolic enzymes as versatile regulators that extend beyond their canonical functions. This blurring of boundaries between metabolism and epigenetics is emblematic of a new frontier in cancer biology, one that promises innovative therapeutic targets grounded in a nuanced understanding of tumor cell physiology.

In sum, the study’s revelations mark a transformative advance in glioblastoma research, offering a mechanistic blueprint for future interventions. The metabolic-epigenetic interplay mediated by HADHA and its impact on JAK/STAT3 signaling could redefine strategies aimed at combating this intractable cancer. As researchers continue to unravel the complexities of tumor biology, insights like these illuminate paths toward more effective, durable, and personalized treatments for patients facing the formidable challenge of glioblastoma.

Subject of Research: HADHA-mediated regulation of JAK/STAT3 signaling in glioblastoma through a metabolic-epigenetic axis

Article Title: HADHA-mediated regulation of JAK/STAT3 signaling in glioblastoma: a metabolic-epigenetic axis

Article References:
Wang, K., Xiao, Y., Wan, J. et al. HADHA-mediated regulation of JAK/STAT3 signaling in glioblastoma: a metabolic-epigenetic axis. Cell Death Discov. 11, 361 (2025). https://doi.org/10.1038/s41420-025-02660-0

Image Credits: AI Generated

DOI: https://doi.org/10.1038/s41420-025-02660-0