ACCESS
Review Article
Volume 3, Article ID: 2026.0025
Yunzhi Zou
zouyz1@sysucc.org.cn
Rong Xiang
xiangrong@sysucc.org.cn
Jixiang Zhao
zhaojx@sysucc.org.cn
Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-sen University Cancer Center, Guangzhou 510060, China
* Author to whom correspondence should be addressed
Received: 31 Oct 2025 Accepted: 21 May 2026 Available Online: 21 May 2026
This article is part of the Special Issue Spatial Omics in Cancer: Decoding Heterogeneity, Microenvironment Crosstalk, and Therapeutic Implications
The human brain exhibits marked regional specialization in cellular composition, gene expression, and tissue architecture. Growing evidence suggests that these spatial differences influence both the anatomical distribution of brain tumors and their subsequent progression. Across tumor types such as gliomas, meningiomas, and brain metastases, specific molecular subtypes frequently show characteristic anatomical distributions, suggesting a functional coupling between tumor genetics and regional brain microenvironments. In this mini review, we summarize recent advances in understanding how regional brain biology shapes the molecular, cellular, and microenvironmental landscapes of brain tumors. This study examines evidence linking anatomical location to tumor genotype, cellular states, and microenvironmental interactions. The study also highlights the emerging concept of “region-aware oncology”, in which tumor behavior is interpreted within the anatomical, cellular, and molecular context of the surrounding brain tissue. A deeper understanding of this spatial coupling may improve tumor classification and support the development of region-informed diagnostic and therapeutic strategies.
The human brain is a highly heterogeneous organ composed of numerous anatomically and functionally distinct regions that differ in developmental origin, cellular composition, and regulatory architecture. This regional heterogeneity forms the molecular and cellular basis of diverse neural functions. Recent advances in multi-omics and spatial transcriptomics have begun to define the region-specific molecular architecture of the normal brain, revealing that each brain region has distinct gene-expression profiles, co-expression networks, and signaling environments [1-3]. Mounting evidence suggests that this intrinsic regional diversity may influence brain tumor initiation, evolution, and therapeutic response. Gliomas, meningiomas, and brain metastases exhibit clear spatial preferences: specific mutations and transcriptional programs are enriched in particular anatomical locations. For example, IDH-mutant gliomas frequently arise in the frontal and temporal lobes [4], whereas H3K27M-mutant diffuse midline gliomas commonly arise in midline structures such as the thalamus and brainstem [5]. Similarly, NF2-driven meningiomas predominate in the convexity, while TRAF7–, KLF4–, AKT1–, and SMO-altered variants localize mainly to the skull base [6-8]. These spatial–genetic patterns suggest that tumor genotypes and anatomical environments may be functionally coupled, reflecting possible co-evolution between oncogenic programs and regional brain ecology [9,10]. However, despite increasing recognition of these relationships, a systematic understanding of region-specific tumor biology remains limited. Most studies have focused on individual tumor types or specific brain regions, whereas integrative analyses incorporating developmental lineage, epigenetic regulation, and microenvironmental context remain scarce. Understanding how the regional molecular baseline of the human brain constrains or permits specific oncogenic pathways could provide new insights into tumor heterogeneity, regional vulnerability, and therapeutic vulnerabilities. This review summarizes current progress in understanding regional molecular heterogeneity in the normal brain and discuss how these intrinsic differences may shape the genetic, transcriptomic, and microenvironmental landscapes of gliomas, meningiomas, and brain metastases. This study further proposes a “region-aware” framework for oncology, defined by the integration of genetic, spatial, and anatomical information to improve understanding of tumor biology and guide the development of more precise therapeutic strategies. Together, these observations support the concept that brain tumors arise within regionally defined molecular ecosystems, in which anatomical context and oncogenic mutations interact to shape tumor behavior.
Relevant English-language studies were identified through searches of PubMed and Web of Science. Search terms included “brain tumor,” “glioma,” “meningioma,” “brain metastasis,” “spatial heterogeneity,” and “brain region.” Priority was given to recent studies investigating regional molecular characteristics and interactions between brain tumors and their microenvironment.
3.1. Regional Cellular Composition and Molecular Heterogeneity in the Normal Brain Understanding the intrinsic molecular organization of the normal brain is essential for interpreting the regional heterogeneity observed in brain tumors. The adult human brain exhibits a high degree of regional specialization at both the molecular and cellular levels. This diversity is reflected not only in cell-type composition and relative abundance but also in region-specific gene-expression patterns, co-expression networks and regulatory programs. In a landmark study, Siletti et al. generated a single-nucleus atlas comprising approximately three million cells from nearly 100 anatomically defined regions across the forebrain, midbrain, and hindbrain, revealing systematic variation in both neuronal and glial populations. The cerebral cortex exhibits a characteristic laminar architecture composed of upper- and deep-layer excitatory neuronal populations, including intratelencephalic, near-projecting, and corticothalamic subtypes, as well as inhibitory neurons derived from distinct developmental lineages. By contrast, the midbrain, brainstem, and hypothalamus contain highly heterogeneous neuronal populations that frequently co-express multiple neurotransmitters and neuropeptides, suggesting more complex and multimodal functional organization than that observed in the cerebral cortex. Regional specialization is also evident among glial cells. Astrocytes and oligodendrocyte progenitor cells (OPCs) in telencephalic regions differ markedly from their counterparts in non-telencephalic areas in both molecular signatures and relative abundance. These differences suggest that even cells derived from shared lineages may follow distinct differentiation trajectories and perform region-specific functions depending on their anatomical milieu [1]. 3.2. Cross-Regional Single-Cell Atlases Reveal Region-Specific Cellular Ecosystems Extending these efforts, Chen et al. integrated 70 single-cell and single-nucleus transcriptomic datasets comprising approximately 11.3 million cells to construct a Brain Cell Atlas encompassing 14 major brain regions and 30 subregions. This integrative analysis characterized region-specific cellular landscapes and identified the gene-regulatory networks associated with each region. Within this atlas, the cerebral cortex is predominantly composed of upper- and deep-layer intratelencephalic neuronal populations, whereas the hippocampus contains distinct neuronal populations associated with the CA1–CA4 subfields and dentate gyrus. By contrast, the thalamus and amygdala are distinguished by excitatory nuclear neurons and specialized regulatory glial subsets. Regional specialization is also apparent at the levels of cell–cell communication and transcriptional control, rather than only in cell-type composition. Microglia, for example, exhibit distinct transcriptional states and ligand–receptor interaction profiles between the prefrontal cortex and hippocampus, consistent with region-specific mechanisms of immune surveillance and metabolic support. Collectively, these integrative atlases indicate that brain regional brain specialization is shaped not only by the distribution of developmental lineages but also by continuous functional adaptation to local microenvironmental cues [2]. 3.3. Gene Co-Expression Networks and Functional Modularization Across Brain Regions From a systems-level perspective, analysis of GTEx RNA-seq data from 13 normal brain regions enabled construction of a genome-wide gene co-expression network (GCN), revealing distinct co-expression modules and associated functional enrichment patterns. Genes characteristic of the cerebral cortex is primarily enriched in pathways related to synaptic transmission and neuronal signaling, whereas genes characteristic of the basal ganglia is associated with motor control and neurotransmitter metabolism. Cerebellar modules are enriched in genes involved in ion homeostasis and neurodevelopment, whereas the spinal cord exhibits a distinct co-expression pattern, forming a separate cluster in dimensionality-reduction analyses. The number of eQTLs and the degree of gene connectivity differ substantially across regions, suggesting hierarchical variation in the complexity and robustness of regional gene-regulatory networks. These differences align closely with functional compartmentalization. The prefrontal and parietal cortices, which are involved in cognition and perception, are enriched in genes related to synaptic signaling and plasticity, whereas the basal ganglia, hypothalamus, and brainstem—regions associated with motor and autonomic regulation—predominantly express genes involved in energy metabolism and neurotransmitter pathways [3]. Taken together, molecular distinctions among brain regions appear to be shaped by developmental origins and organizational hierarchies. According to Siletti et al., telencephalon-derived structures such as the cortex, hippocampus, and striatum exhibit lineage-specific repertoires of excitatory neuron, whereas the midbrain and hindbrain contain a higher proportion of mixed-neurotransmitter cell types. This pattern may contribute to the pronounced cellular heterogeneity observed in non-cortical brain regions [1]. Chen et al. further highlighted locally driven variations in cell–cell communication—such as divergent microglial signaling networks between the hippocampus and prefrontal cortex [2]. Moreover, key genes within region-specific co-expression modules show higher mutation frequencies in various neurological disorders and brain tumors, suggesting that these baseline regional molecular differences might represent a foundational layer predisposing certain areas to disease vulnerability [3]. Overall, these studies demonstrate that the human brain is organized into region-specific molecular ecosystems characterized by distinct cellular compositions, regulatory networks, and intercellular signaling patterns. These intrinsic differences may establish regionally permissive microenvironments that favor the emergence and progression of specific brain tumor types.
4.1. Regional Features of Gliomas Among primary brain tumors, gliomas represent one of the clearest examples of spatially structured molecular heterogeneity. Extensive evidence indicates that glioma anatomical location is closely associated with molecular profiles, cellular states, and clinical outcomes. Distinct neurodevelopmental backgrounds, lineage compositions, and epigenetic landscapes across brain regions shape tumor mutational spectra and biological behavior. Clear spatial preferences have been observed across brain regions, ranging from the frontal and temporal lobes to the thalamus, brainstem, spinal cord, and cerebellum. For instance, Qi et al. reported that IDH-mutant gliomas predominantly occur in the frontal or temporal lobes [4], while Vuong et al. demonstrated that H3K27M-mutant diffuse midline gliomas (DMG) are concentrated in the brainstem, thalamus, and spinal cord [5]. In contrast, Dandapatha et al. found that low-grade BRAF-fusion tumors are mainly located in the cerebellum and posterior fossa. Collectively, these findings support the paradigm that “anatomical location influences tumor biology”—a central principle underlying the spatial heterogeneity of gliomas [9] (Table 1). Molecular landscape of gliomas across main brain regions. 4.1.1. Supratentorial Regions: Molecular Distribution Across Frontal, Temporal, Parietal, and Occipital Lobes Within the supratentorial compartment, gliomas arising in different cortical lobes exhibit distinct molecular characteristics and clinical behaviors. Frontal lobe: The frontal lobe represents the most frequent site of adult gliomas. Qi et al. reported that IDH1/2-mutant tumors are significantly enriched in this region and typically exhibit well-demarcated margins, homogeneous MRI signal characteristics, and minimal contrast enhancement, features consistent with lower invasiveness and a more favorable prognosis [4]. Integrated radiomic analyses further indicate that frontal tumors often display MGMT promoter methylation and 1p/19q codeletion, predicting better chemotherapy response [11]. Functionally, Qi et al. reported that patients with left frontal IDH-mutant tumors exhibit enhanced contralesional frontoparietal network activity, suggesting an interaction between molecular subtype and neural plasticity [12]. Temporal lobe: Temporal gliomas frequently exhibit EGFR amplification and enrichment of RTK II subtypes, which are associated with increased invasiveness and angiogenic activity [13]. Spatial transcriptomics has revealed that in temporal-lobe GBM, perivascular and neuron-interaction layers are especially prominent, suggesting that the local microenvironment might play a key role in shaping tumor multilayered architecture [14]. Parietal lobe: Glioblastomas in the parietal region tend to have the poorest survival. Multimodal analyses indicate enrichment of PTEN mutations and FGFR3-TACC3 fusions, along with pronounced radiographic texture heterogeneity and aggressive imaging features, reflecting greater spatial heterogeneity and more unfavorable biological behavior [11]. Occipital lobe: Occipital gliomas are relatively rare. Cini et al. noted that their mutational spectra are not distinctive and that many lesions likely represent secondary infiltration or dissemination from parietal or temporal tumors, implying a lower intrinsic tendency for primary tumorigenesis in this area [15]. 4.1.2. Insular and Lateralization Effects: Structural Connectivity and Hemispheric Differences Yang et al. demonstrated pronounced hemispheric asymmetry in insular gliomas [16]. In the left hemisphere, MGMT methylation levels positively correlated with damage to the IFOF tract, whereas the opposite pattern was observed on the right. This finding suggests that gliomas may remodel white matter connectivity in a hemisphere-specific manner. Systematic reviews further noted higher frequencies of TP53 mutations and MGMT methylation in left-sided gliomas, often associated with poorer prognosis, implying that molecular features could be intricately linked to functional hemispheric lateralization [15]. 4.1.3. Midline Regions: Spatial Stratification of the Thalamus, Brainstem, and Spinal Cord Thalamus and Pulvinar subregion: Pediatric thalamic gliomas account for approximately 5% of CNS tumors. Chiba et al. found that lesions in the Pulvinar (posterior thalamic nucleus) display higher malignancy, enrichment of H3K27M mutations, and early ventricular dissemination [17]. The aggressiveness of Pulvinar tumors suggests that biological stratification might exist even within thalamic subregions. Diffuse midline glioma (DMG): An integrated analysis of 804 cases revealed region-specific distinctions: brainstem DMGs occur at younger ages, show the highest Ki-67 index, and have the poorest prognosis; thalamic DMGs fare somewhat better; spinal DMGs occur mainly in adults and often harbor TERT-promoter mutations; whereas brainstem DMGs are enriched for HIST1H3B/C and ACVR1 mutations. Thus, despite a shared H3K27M background, the anatomical location and accompanying mutational landscape may determine tumor evolution and survival outcomes [5]. These findings suggest that even within the midline compartment, anatomical subregions correspond to biologically distinct tumor states. Single-cell multi-omics analyses further revealed that H3K27M DMGs share an OPC-like stem-cell state, but differentiation trajectories and immune microenvironments vary by site: pontine lesions contain more hypoxic, stem-like populations; thalamic lesions exhibit more mature glial lineages; and adult-type DMGs show enrichment of MES-like states, suggesting that location, age, and immune milieu jointly shape the tumor’s spatial and cellular architecture [18]. 4.1.4. Infratentorial Regions: Distinct Subtype Signatures of the Cerebellum and Posterior Fossa Dandapatha et al. reported that 91% of pilocytic astrocytomas (PAs) carry BRAF fusions and almost all are located in the cerebellum or posterior fossa. These pediatric, developmentally driven tumors are associated with an excellent prognosis and represent a MAPK-activated benign lineage [9]. In contrast, Oki et al. described that cerebellar GBMs, though rare, constitute a biologically distinct subgroup—lacking IDH, H3F3A, and TERT mutations but showing a strong tendency for distant dissemination and ventricular invasion. This suggests that the infratentorial region might constitute an independent tumor ecosystem [19]. Overall, these findings indicate that glioma molecular subtypes are not randomly distributed across the brain but instead show clear anatomical predilections. Recognition of these spatial patterns may help refine diagnostic stratification and guide region-adapted therapeutic strategies. 4.2. Coupling Between Anatomical Location and Molecular Subtypes in Meningiomas Meningiomas provide another clear example of tight coupling between tumor genotype and anatomical location. As the most common primary intracranial tumor type in adults, meningiomas exhibit clear location-specific molecular and clinical differences, with biological behavior closely correlated with their anatomical origin. Tumors driven by NF2 mutations and 22q loss predominantly arise on the cerebral convexity or supratentorial compartments and tend to recur more frequently, whereas non-NF2 variants, including TRAF7, KLF4, AKT1, SMO, and POLR2A) are enriched at the skull base and posterior fossa, typically presenting as benign, stable WHO grade I tumors [6]. This position-genotype coupling suggests that the cell of origin, developmental lineage, and the local microenvironment collectively shape meningioma biology (Table 2). Regional–molecular coupling in meningiomas. 4.2.1. Supratentorial Meningiomas: NF2-Driven, Convexity-Dominant, and Aggressive Whole-exome sequencing of 300 cases revealed NF2 mutations in about 36% of patients, primarily located on the cerebral convexity and supratentorial tentorium. These tumors frequently exhibit 22q deletion, genomic instability, and higher histologic grade, indicative of increased invasiveness and recurrence potential. In contrast, TRAF7/KLF4/AKT1/SMO mutations are largely restricted to skull-base tumors, showing mutual exclusivity with NF2-driven lesions [7]. The biology of supratentorial meningiomas thus appears to depend on the NF2–YAP signaling axis, which could represent a key driver pathway for high-risk subtypes [6]. 4.2.2. Skull-Base Meningiomas: The TRAF7–KLF4–AKT1–SMO Benign Lineage Arising from the sphenoid ridge, olfactory groove, petrous apex, and sellar region, skull-base meningiomas represent approximately one-quarter of all cases. These “non-NF2” tumors are typically driven by TRAF7 mutations (~24%), which often co-occur with KLF4 K409Q or AKT1 E17K alterations. Nearly all are located in the medial skull base or sphenoid region, characterized by chromosomal stability and benign histology. SMO-mutant meningiomas cluster in the olfactory groove and anterior cranial fossa, suggesting activation of the Hedgehog pathway [6]. Collectively, this group appears to represent a developmentally stable, low-invasive, and anatomically confined lineage, primarily driven by PI3K–AKT–mTOR signaling rather than NF2–YAP pathway activation [8]. 4.2.3. Posterior-Fossa Meningiomas: Anatomical Sub-Regions and Chromosomal Instability Comprehensive genomic profiling of 132 posterior-fossa meningiomas identified three molecular subtypes: Group A: NF2-intact tumors without Merlin-pathway abnormalities, typically located in lateral infratentorial or petrous areas. Group B: NF2-mutant or 22q-deleted tumors without high-risk CNAs, showing broad distribution. Group C: Genomically unstable tumors with multiple high-risk CNAs (1p/14q co-deletion, 9p loss, TERT/CDKN2A mutations), 81% of which occur in midline structures (tentorial notch, clivus, foramen magnum). Group C meningiomas are associated with the poorest prognosis, demonstrating significantly shorter progression-free survival even following gross-total resection [10]. These results suggest that proximity to midline structures might correlate with greater genomic instability and malignant potential, reflecting a developmental lineage-linked risk gradient. 4.2.4. Cerebellopontine-Angle (CPA) Meningiomas: Prognostic Implications of POLR2A Mutation In a cohort of 70 WHO grade I cerebellopontine angle (CPA)meningiomas, POLR2A mutations were found in 17% of cases—all in women and predominantly of the meningothelial subtype. The mutation significantly increased recurrence risk, particularly in patients undergoing subtotal resection. These tumors were most frequently located at the petroclival junction and posterior petrous ridge [20]. Thus, POLR2A might serve as a skull-base–specific molecular biomarker, potentially influencing transcriptional regulation and cellular differentiation in region-defined contexts. Collectively, these studies highlight a strong position–genotype coupling in meningiomas, suggesting that developmental lineage and local microenvironment jointly shape tumor biology. 4.3. Regional Adaptation of Brain Metastases Brain metastasis (BrM) are the most common intracranial malignancies in adults and show a cortical predilection, frequently involving the frontal and parietal lobes as well as the cerebellar hemispheres, whereas deep structures such as the thalamus, basal ganglia, and brainstem are less commonly affected. At the systems level, BrM formation is not merely determined by vascular perfusion patterns but may reflect a coevolution between metastatic cells and regional brain microenvironments. Differences in vascular architecture, glial composition, and immune–stromal context across brain regions likely influence metastatic susceptibility. In the study by Gonzalez et al., 15 parenchymal brain metastases from diverse primaries (breast, lung, melanoma, ovarian, colorectal, renal, etc.) were analyzed, and nearly all were located in the cerebral cortex, indicating a preference for vascular-rich cortical territories [21]. Single-cell transcriptomics and mass cytometry revealed that cortical metastases are characterized by complex vascular networks enriched for arterio-venous-transition endothelial clusters (APLNR+, EFNB2+, NR2F2+), abundant pericytes and MSC-like cells expressing ECM-remodeling genes (COL1A1, TINAGL1, TGFB3), and scarcity of astrocytes, implying local breakdown of the glial barrier during cortical invasion. Cross-tissue transcriptome-wide association analysis (TWAS) further identified CASP8 as a key driver gene in breast cancer brain metastasis, with specific overexpression in the cerebellar hemispheres and frontal cortex [22]. This finding suggests that breast-cancer cells may preferentially colonize the cerebellar–frontal axis, where distinct metabolic and immune properties, together with vascular permeability and glial fiber density, may create a favorable niche for metastatic colonization. Together, these observations indicate that metastatic tumor cells adapt to region-specific vascular and glial niches, supporting the idea that the brain microenvironment plays a decisive role in shaping metastatic colonization patterns.Table 1: Brain Region
Representative Subtype/Mutation
Molecular/Genetic Features
Microenvironmental & Functional Traits
Clinical/Prognostic Implications
Frontal lobe
IDH1/2-mutant glioma
MGMT methylation, 1p/19q codeletion
Lower invasiveness and weak contrast enhancement, preserved network plasticity
Favorable prognosis, good chemotherapeutic response
Temporal lobe
RTK II subtype GBM
EGFR amplification
Perivascular & neuron-interaction layers enriched
Highly invasive, angiogenic
Parietal lobe
GBM with PTEN mutation, FGFR3-TACC3 fusion
High texture heterogeneity
Aggressive phenotype
Poor survival
Occipital lobe
Secondary infiltration from parietal/temporal
No distinctive mutations
Low intrinsic tumorigenesis
Rare primary lesions
Insula/Left vs. Right Hemisphere
IDH-mutant glioma (lateralized)
Left: ↑TP53 mutation, ↑MGMT methylation
White-matter remodeling (IFOF asymmetry)
Left-sided → poorer outcome
Thalamus/Pulvinar
Pediatric H3K27M-mutant glioma
H3 K27M mutation, ventricular spread
High malignancy in Pulvinar subregion
Poor prognosis
Brainstem (DMG)
H3 K27M-mutant DMG
HIST1H3B/C, ACVR1
Hypoxic, OPC-like stem states
Worst prognosis
Spinal cord (DMG)
Adult DMG
TERT-promoter mutation
MES-like enrichment
Intermediate prognosis
Cerebellum/Posterior fossa
BRAF-fusion PA; rare cerebellar GBM
BRAF fusion (MAPK activation); IDH/H3F3A-negative GBM
Developmental benign lineage (PA); invasive GBM niche
PA: excellent prognosis; cerebellar GBM: high dissemination
Table 2: Anatomical Region
Dominant Molecular Subtype(s)
Key Genetic Alterations
Biological/Pathway Features
Clinical/Prognostic Traits
Cerebral Convexity/Supratentorial
NF2-driven meningioma
NF2 mutation, 22q loss
NF2–YAP signaling activation; genomic instability
High-grade (II–III), aggressive, recurrent
Skull Base (Sphenoid ridge, Olfactory groove, Sellar region)
Non-NF2 lineage: TRAF7, KLF4, AKT1, SMO
TRAF7 ± KLF4 (K409Q), AKT1 (E17K), SMO mutation
PI3K–AKT–mTOR and Hedgehog pathway activation
WHO grade I, benign, chromosomally stable, low recurrence
Posterior Fossa (Tentorial notch, Clivus, Foramen magnum)
Group A–C stratification
NF2-intact (A), NF2/22q loss (B), 1p/14q loss, 9p loss, TERT/CDKN2A (C)
Gradual increase in genomic instability from lateral → midline
Group C (midline): worst prognosis, rapid recurrence
Cerebellopontine Angle (CPA)
POLR2A-mutant subtype
POLR2A mutation (female-predominant)
Altered transcriptional regulation, meningothelial histology
Grade I but higher recurrence risk
Posterior Skull Base (Petrous apex, Petroclival junction)
TRAF7–KLF4–AKT1 lineage overlap
TRAF7, AKT1 E17K
PI3K–AKT signaling
Benign, stable
Overall Pattern
NF2 vs. non-NF2 dichotomy
NF2–YAP vs. PI3K–AKT–mTOR pathways
Location-dependent developmental lineage
“Position–Genotype Coupling” defines biology
Growing evidence from these studies indicates that region-specific molecular baselines exert a profound influence on brain tumor biology. Across gliomas (Figure 1), meningiomas, and brain metastases, each tumor type shows characteristic anatomical predilections and molecular features that track with its site of origin, underscoring the importance of the local host microenvironment in tumor initiation and subsequent evolution. From a translational standpoint, the spatial molecular ecology of brain tumors is likely to be a major determinant of clinical course and treatment sensitivity. Integrating anatomical localization with single-cell, spatial transcriptomic, and imaging-based omics data could enable a more nuanced, region-oriented classification framework that better supports precision neurosurgery and region-adapted adjuvant therapies. Looking ahead, it will be important to elucidate how developmental patterning cues, intercellular signaling circuits, and metabolic microenvironments jointly confer regional vulnerability or resistance to oncogenesis, thereby clarifying the intrinsic links between brain regionality and tumor behavior. Despite these advances, our understanding of region-specific mechanisms in brain tumors remains limited. Although the regional molecular features of gliomas have been partially characterized, integrative studies combining developmental lineage, epigenetic regulation, and microenvironmental signaling remain limited. In meningiomas, the coupling between location and genotype has been well established, yet the developmental and biomechanical underpinnings of this relationship remain poorly characterized. In contrast, research on regional heterogeneity in brain metastases remains relatively underdeveloped. Future studies should focus on integrating spatial omics with neuroimaging and radiomic analyses to better characterize region-specific tumor ecosystems. In addition, integrating anatomical location with molecular features may facilitate the development of region-informed risk stratification models, while clinical studies stratified by both tumor genotype and anatomical site may further clarify therapeutic responses across distinct brain regions. Such efforts may advance the emerging concept of “region-aware oncology” and support more precise diagnostic and therapeutic strategies for brain tumors.
In summary, growing evidence indicates that brain tumors are shaped not only by intrinsic genetic alterations but also by the region-specific molecular and cellular context of the host brain. Across gliomas, meningiomas, and brain metastases, distinct anatomical locations are associated with characteristic molecular features, tumor behaviors, and microenvironmental interactions, supporting the concept of region-aware oncology. A deeper understanding of this spatial coupling may improve brain tumor classification and provide a foundation for more precise diagnostic and therapeutic strategies.
APLNR, apelin receptor; ACVR1, activin A receptor type 1; AKT1, AKT serine/threonine kinase 1; BrM, brain metastasis; CNS, central nervous system; COL1A1, collagen type I alpha 1 chain; CPA, cerebellopontine angle; DMG, diffuse midline glioma; ECM, extracellular matrix; EGFR, epidermal growth factor receptor; EFNB2, ephrin B2; eQTL, expression quantitative trait locus; FGFR3–TACC3, fibroblast growth factor receptor 3–transforming acidic coiled-coil containing protein 3 fusion; GBM, glioblastoma; GCN, gene co-expression network; GTEx, Genotype-Tissue Expression Project; H3 K27M, lysine-to-methionine substitution at position 27 of histone H3; H3F3A, H3 histone family member 3A; HIST1H3B/C, histone cluster 1 H3 family member B/C; IDH, isocitrate dehydrogenase; IFOF, inferior fronto-occipital fasciculus; KLF4, Krüppel-like factor 4; MES, mesenchymal; MGMT, O6-methylguanine-DNA methyltransferase; MRI, magnetic resonance imaging; MSC, mesenchymal stromal cell; NF2, neurofibromatosis type 2; NR2F2, nuclear receptor subfamily 2 group F member 2; OPC, oligodendrocyte progenitor cell; PA, pilocytic astrocytoma; PFS, progression-free survival; PI3K, phosphoinositide 3-kinase; POLR2A, RNA polymerase II subunit A; RTK, receptor tyrosine kinase; SMO, Smoothened frizzled class receptor; TERT, telomerase reverse transcriptase; TGFB3, transforming growth factor beta 3; TINAGL1, tubulointerstitial nephritis antigen-like 1; TWAS, transcriptome-wide association study; WHO, World Health Organization; YAP, Yes-associated protein.
Conceptualization, literature review, data curation, writing—original draft preparation: Y.Z; Supervision, conceptualization, writing—review & editing: R.X; Methodology, visualization, validation, writing—review & editing: J.Z. All authors have read and agreed to the published version of the manuscript.
The authors declare no conflicts of interest.
The study did not receive any external funding and was conducted using only institutional resources.
We sincerely thank Prof. Gao Zhang from The University of Hong Kong for his valuable guidance and support throughout the preparation of this review.
During manuscript preparation, the authors used ChatGPT only for language polishing and improving readability. All scientific content, interpretation of the literature, and final conclusions were reviewed and approved by the authors.
[1] Siletti, K.; Hodge, R.; Mossi Albiach, A.; Lee, K.W.; Ding, S.L.; Hu, L.; Lonnerberg, P.; Bakken, T.; Casper, T.; Clark, M.; et al. Transcriptomic Diversity of Cell Types across the Adult Human Brain. Science 2023, 382, eadd7046. [CrossRef] [PubMed]
[2] Chen, X.; Huang, Y.; Huang, L.; Huang, Z.; Hao, Z.Z.; Xu, L.; Xu, N.; Li, Z.; Mou, Y.; Ye, M.; et al. A Brain Cell Atlas Integrating Single-Cell Transcriptomes across Human Brain Regions. Nat. Med. 2024, 30, 2679–2691. [CrossRef]
[3] Hang, Y.; Aburidi, M.; Husain, B.; Hickman, A.R.; Poehlman, W.L.; Feltus, F.A. Exploration into Biomarker Potential of Region-Specific Brain Gene Co-Expression Networks. Sci. Rep. 2020, 10, 17089. [CrossRef] [PubMed]
[4] Qi, S.; Yu, L.; Li, H.; Ou, Y.; Qiu, X.; Ding, Y.; Han, H.; Zhang, X. Isocitrate Dehydrogenase Mutation Is Associated with Tumor Location and Magnetic Resonance Imaging Characteristics in Astrocytic Neoplasms. Oncol. Lett. 2014, 7, 1895–1902. [CrossRef]
[5] Vuong, H.G.; Le, H.T.; Jea, A.; McNall-Knapp, R.; Dunn, I.F. Risk Stratification of H3 K27M-Mutant Diffuse Midline Gliomas Based on Anatomical Locations: An Integrated Systematic Review of Individual Participant Data. J. Neurosurg. Pediatr. 2022, 30, 99–106. [CrossRef]
[6] Szulzewsky, F.; Thirimanne, H.N.; Holland, E.C. Meningioma: Current Updates on Genetics, Classification, and Mouse Modeling. Ups. J. Med. Sci. 2024, 129, e10579. [CrossRef]
[7] Clark, V.E.; Erson-Omay, E.Z.; Serin, A.; Yin, J.; Cotney, J.; Ozduman, K.; Avsar, T.; Li, J.; Murray, P.B.; Henegariu, O.; et al. Genomic Analysis of Non-NF2 Meningiomas Reveals Mutations in TRAF7, KLF4, AKT1, and SMO. Science 2013, 339, 1077–1080. [CrossRef] [PubMed]
[8] Lee, Y.S.; Lee, Y.S. Molecular Characteristics of Meningiomas. J. Pathol. Transl. Med. 2020, 54, 45–63. [CrossRef]
[9] Dandapath, I.; Singh, J.; Das, S.; Jangir, H.; Roy, C.; Dan Charan, B.; Manjunath, N.; Garg, A.; Suri, A.; Kedia, S.; et al. Comprehensive Transcriptomic Profiling of Fusions and Abnormal Variant Transcripts in Pilocytic Astrocytoma Using NanoString nCounter Technology. Pathol. Res. Pract. 2025, 271, 156024. [CrossRef]
[10] Hirano, Y.; Miyawaki, S.; Sakai, Y.; Teranishi, Y.; Umekawa, M.; Hirata, T.; Ogawa, S.; Komura, D.; Katoh, H.; Ikemura, M.; et al. Integrated Molecular and Detailed Anatomical Profiling Identifies a Prognostically Adverse Subtype of Posterior Fossa Meningiomas: High-Risk Copy Number Alterations Are Associated with Midline Predilection and Predict Poor Prognosis. Acta Neuropathol. Commun. 2025, 13, 168. [CrossRef]
[11] Kundal, K.; Rao, K.V.; Dhanda, S.K.; Kumar, N.; Kumar, R. Tumor Location, Genomic Alterations, and Radiomic Features as Predictors of Survival in Glioblastoma: A Multi-Modal Analysis. Neuroradiology 2025, 67, 2713–2725. [CrossRef]
[12] Qi, C.; Wang, R.; Meng, L.; Li, S.; Li, Y. Functional Reorganization of Contralesional Networks Varies According to Isocitrate Dehydrogenase 1 Mutation Status in Patients with Left Frontal Lobe Glioma. Neuroradiology 2022, 64, 1819–1828. [CrossRef]
[13] Foltyn-Dumitru, M.; Alzaid, H.; Rastogi, A.; Neuberger, U.; Sahm, F.; Kessler, T.; Wick, W.; Bendszus, M.; Vollmuth, P.; Schell, M. Unraveling Glioblastoma Diversity: Insights into Methylation Subtypes and Spatial Relationships. Neurooncol. Adv. 2024, 6, vdae112. [CrossRef]
[14] Greenwald, A.C.; Darnell, N.G.; Hoefflin, R.; Simkin, D.; Mount, C.W.; Gonzalez Castro, L.N.; Harnik, Y.; Dumont, S.; Hirsch, D.; Nomura, M.; et al. Integrative Spatial Analysis Reveals a Multi-Layered Organization of Glioblastoma. Cell 2024, 187, 2485–2501.e2426. [CrossRef] [PubMed]
[15] Cini, N.T.; Pennisi, M.; Genc, S.; Spandidos, D.A.; Falzone, L.; Mitsias, P.D.; Tsatsakis, A.; Taghizadehghalehjoughi, A. Glioma Lateralization: Focus on the Anatomical Localization and the Distribution of Molecular Alterations (Review). Oncol. Rep. 2024, 52, 139. [CrossRef] [PubMed]
[16] Yang, Z.C.; Yin, C.D.; Yeh, F.C.; Xue, B.W.; Song, X.Y.; Li, G.; Sun, S.J.; Deng, Z.H.; Hou, Z.G.; Xie, J. Exploring MGMT Methylation-Driven Structural Connectivity Changes in Insular Gliomas: A Tractography and Graph Theoretical Analysis. J. Neurooncol. 2024, 166, 155–165. [CrossRef]
[17] Chiba, K.; Aihara, Y.; Masui, K.; Abe, K.; Komori, T.; Kawamata, T. Pulvinar Locus Is Highly Relevant to Patients’ Outcomes in Surgically Resected Thalamic Gliomas in Children. World Neurosurg. 2020, 134, e530–e539. [CrossRef]
[18] Liu, I.; Jiang, L.; Samuelsson, E.R.; Marco Salas, S.; Beck, A.; Hack, O.A.; Jeong, D.; Shaw, M.L.; Englinger, B.; LaBelle, J.; et al. The Landscape of Tumor Cell States and Spatial Organization in H3-K27M Mutant Diffuse Midline Glioma across Age and Location. Nat. Genet. 2022, 54, 1881–1894. [CrossRef]
[19] Oki, S.; Yamaguchi, S.; Ishi, Y.; Motegi, H.; Gekka, M.; Tanei, Z.I.; Kanno-Okada, H.; Takakuwa, E.; Tanaka, S.; Mori, T.; et al. Recurrence Patterns and Clinical Outcomes in Adult Cerebellar Glioblastoma. Acta Neurochir. (Wien) 2025, 167, 217. [CrossRef]
[20] Okano, A.; Miyawaki, S.; Teranishi, Y.; Hongo, H.; Dofuku, S.; Ohara, K.; Sakai, Y.; Shin, M.; Nakatomi, H.; Saito, N. POLR2A Mutation Is a Poor Prognostic Marker of Cerebellopontine Angle Meningioma. Neurosurgery 2024, 95, 275–283. [CrossRef] [PubMed]
[21] Gonzalez, H.; Mei, W.; Robles, I.; Hagerling, C.; Allen, B.M.; Hauge Okholm, T.L.; Nanjaraj, A.; Verbeek, T.; Kalavacherla, S.; van Gogh, M.; et al. Cellular Architecture of Human Brain Metastases. Cell 2022, 185, 729–745.e720. [CrossRef] [PubMed]
[22] Liu, J.; Guan, X.; Gao, S.; Quan, L.; Dou, M.; Yue, J.; Shi, M.; Yuan, P. Novel Insight of Critical Genes Involved in Breast Cancer Brain Metastasis: Evidence from a Cross-Tissue Transcriptome Association Study and Validation through External Clinical Cohorts. BMC Cancer 2025, 25, 707. [CrossRef]
Disclaimer/Publisher’s Note: The views expressed in this article are those of the author(s) and do not necessarily reflect the views of the publisher or editors. The publisher and editors assume no responsibility for any injury or damage resulting from the use of information contained herein.
This is not the final version of the article. Changes may occur when the manuscript is published in its final format.
©2026 Copyright by the Authors.
Licensed as an open-access article distributed under the terms and conditions of the CC BY 4.0 license