Neural–Tumor and Microenvironmental Interactions in Brain Metastases from Lung Cancer: From Blood–Brain Barrier Traversal to Adaptation within the Brain

Abstract

Brain metastases from lung cancer do not merely reflect passive tumor growth after entry into the central nervous system. Rather, their development is a dynamic process involving transvascular entry into the brain, remodeling of the intracranial microenvironment, metabolic adaptation, and changes in neural function. Advances in single-cell and spatial omics, electrophysiology, and functional models have increasingly clarified the interactions between lung cancer cells and host cells in the brain. Current evidence indicates that brain metastasis is associated with the selection of specific malignant cell states and is shaped by the brain vascular endothelium and blood–brain barrier integrity. Once within the brain, astrocytes, microglia, oligodendrocytes, and adaptive immune cells collectively shape the metastatic niche through signaling, metabolic exchange, and spatial organization. Local nutrient availability and metabolites such as lactate, cholesterol, and other lipids further influence tumor cell survival and growth. Neural adaptation differs markedly across lung cancer subtypes. In small-cell lung cancer (SCLC), more direct evidence supports intrinsic electrical excitability and functional neuron–tumor interactions, whereas evidence in non-small-cell lung cancer (NSCLC) remains centered on neural-like transcriptional states and neurotransmitter-related signaling; whether comparable functional neural connections are a general feature of NSCLC remains unclear. Overall, lung cancer brain metastasis is best viewed as a multilayered adaptive process involving tumor cell states, the brain vasculature, glial and immune microenvironments, metabolism, and neural signaling. Future studies should define the temporal and causal relationships among these mechanisms and establish their generalizability and translational relevance across lung cancer subtypes and patient populations.