Regional Vulnerability to Brain Metastases Reflects Structural, Metabolic and Genetic Organization of the Healthy Brain
- 1Department of Neurological Surgery, NYU Langone Health
Abstract
Background: Brain metastases (BMs), the most common intracranial tumor, do not distribute uniformly throughout the brain, but instead preferentially localize to specific regions. Building on known associations with perfusion and the gray-white matter junction, we hypothesized that regional vulnerability to BMs also reflects the cellular and molecular organization of the healthy brain. Methods: In a discovery cohort of 1,406 patients, we generated a voxelwise map of BM distribution and compared it with normative atlases of brain microstructure, vasculature, metabolism, cell-type composition, and gene expression. Associations were tested for replication in an independent multicenter validation cohort of 1,209 patients. Results: In the discovery cohort, BM frequency was higher near the gray-white matter junction and in regions with greater cerebral perfusion, glucose metabolism, and mitochondrial density. At the cellular level, BM frequency was inversely related to astrocyte and microglial transcriptional signatures. Using normative gene expression maps derived from postmortem healthy brain tissue, we identified 17 genes spatially associated with BM vulnerability and specific to metastases relative to other intracranial lesion types. These associations replicated in the validation cohort. The immune checkpoint gene LAG3 ranked among the ten genes most positively associated with metastatic vulnerability across all genes tested (r = 0.40-0.45, p < 0.001). LAG3 expression was also elevated in peritumoral brain metastasis tissue relative to control brain tissue in an independent spatial transcriptomic dataset. Conclusions: These findings suggest that BM vulnerability reflects healthy brain organization and identify LAG-3, a clinically targetable immune checkpoint, as a candidate link between regional brain immune biology and metastatic colonization.
Related
- Gastric cancer peritoneal metastases as an immune compartment: immune evasion, systemic immunotherapy and regional treatment strategies
- Metabolic Age as a Potential Marker of Cardiometabolic Vulnerability in Institutionalised Older Adults: A Cross-Sectional Study
- Dopamine Metabolism as a Chemically Constrained Network: Structural Regulation, Redox Liability and Dopaminergic Vulnerability in Parkinson’s Disease
- Aberrant Accumulation of Cyclin D1 Rewires Glutamine Metabolism and Generates a Metabolic Vulnerability in Squamous Cell Carcinoma
- DEGS1-mTORC1-driven lipid metabolic reprogramming fuels rhabdomyosarcoma progression and unveils a therapeutic vulnerability
BibTeX
@misc{turner2026regional,
title = {{Regional Vulnerability to Brain Metastases Reflects Structural, Metabolic and Genetic Organization of the Healthy Brain}},
author = {J. I. Turner and A. Arias and R. MacFadyen and A. Y. Fu and R. Y. Huang and J. Patel and N. Lamba and H. Elhalawani and A. A. Aizer and A. Y. Schwieger and F. Wolpert and M. Galovic and M. Weller and K. Bernstein and M. E. Pacold and E. K. Oermann and F. L. Schaper and D. Kondziolka},
howpublished = {medRxiv},
year = {2026},
month = nov,
doi = {10.64898/2026.11.06.26367469},
url = {https://www.medrxiv.org/content/10.64898/2026.11.06.26367469v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Regional Vulnerability to Brain Metastases Reflects Structural, Metabolic and Genetic Organization of the Healthy Brain
AU - J. I. Turner
AU - A. Arias
AU - R. MacFadyen
AU - A. Y. Fu
AU - R. Y. Huang
AU - J. Patel
AU - N. Lamba
AU - H. Elhalawani
AU - A. A. Aizer
AU - A. Y. Schwieger
AU - F. Wolpert
AU - M. Galovic
AU - M. Weller
AU - K. Bernstein
AU - M. E. Pacold
AU - E. K. Oermann
AU - F. L. Schaper
AU - D. Kondziolka
PY - 2026
DA - 2026/11/06
PB - medRxiv
DO - 10.64898/2026.11.06.26367469
UR - https://www.medrxiv.org/content/10.64898/2026.11.06.26367469v1
AB - Background: Brain metastases (BMs), the most common intracranial tumor, do not distribute uniformly throughout the brain, but instead preferentially localize to specific regions. Building on known associations with perfusion and the gray-white matter junction, we hypothesized that regional vulnerability to BMs also reflects the cellular and molecular organization of the healthy brain. Methods: In a discovery cohort of 1,406 patients, we generated a voxelwise map of BM distribution and compared it with normative atlases of brain microstructure, vasculature, metabolism, cell-type composition, and gene expression. Associations were tested for replication in an independent multicenter validation cohort of 1,209 patients. Results: In the discovery cohort, BM frequency was higher near the gray-white matter junction and in regions with greater cerebral perfusion, glucose metabolism, and mitochondrial density. At the cellular level, BM frequency was inversely related to astrocyte and microglial transcriptional signatures. Using normative gene expression maps derived from postmortem healthy brain tissue, we identified 17 genes spatially associated with BM vulnerability and specific to metastases relative to other intracranial lesion types. These associations replicated in the validation cohort. The immune checkpoint gene LAG3 ranked among the ten genes most positively associated with metastatic vulnerability across all genes tested (r = 0.40-0.45, p < 0.001). LAG3 expression was also elevated in peritumoral brain metastasis tissue relative to control brain tissue in an independent spatial transcriptomic dataset. Conclusions: These findings suggest that BM vulnerability reflects healthy brain organization and identify LAG-3, a clinically targetable immune checkpoint, as a candidate link between regional brain immune biology and metastatic colonization.
ER -