# Local metastatic expansion versus secondary intra-organ dissemination: two causes of neurological death explained by fundamentally different metastatic colonization patterns

## TL;DR

This study explains how brain metastases cause neurological death through two distinct mechanisms: local tumor growth and widespread secondary spread.

## Contribution

The study provides the first direct evidence of contralateral recolonization by secondary metastasis-initiating cells and highlights the role of histopathological growth patterns in predicting neurological failure.

## Key findings

- Neurological death in the TUBO model is caused by local tumor expansion compressing vital brain structures.
- The E0771-LG model causes mortality through widespread secondary dissemination and recolonization.
- Secondary metastasis-initiating cells bypass classical vascular routes to reach distant brain regions.

## Abstract

Neurological failure contributes to 15–50% of deaths in patients with brain metastases, yet the underlying mechanisms remain poorly understood. Clinical causes range from local compression to meningeal metastasis. In this context, a link between infiltrative histopathological growth patterns (HGPs) and meningeal metastasis was recently described and prompted this reverse translation study.

We conducted a retrospective postmortem histological assessment and a prospective MRI-based proof-of-concept study to explore neurological decline mechanisms in two experimental brain metastasis models with different HGPs: (i) the non-infiltrative TUBO model, characterized by well-defined tumor borders and a multilayered astrocytic capsule; and (ii) the infiltrative E0771-LG model, exhibiting diffuse infiltration and widespread astrogliosis.

In the TUBO model, neurological death resulted from local metastatic expansion compressing vital structures, while the E0771-LG model caused mortality mainly through widespread secondary dissemination. We provide the first direct evidence of contralateral recolonization by secondary metastasis-initiating cells (secMICs), and highlight the high efficiency of secondary spread. Additionally, we show that secMICs exploit distinct anatomical structures to reach distant brain regions, bypassing classical vascular dissemination routes. Notably, the HGP and its associated features are intrinsic to tumor cells and are established early during metastatic colonization.

This study identifies the HGP as a potential surrogate for predicting the underlying cause of organ failure in brain metastases. Additionally, it highlights the significant role of secondary dissemination and recolonization in brain metastasis, processes that have been largely overlooked in clinical practice. These findings address a critical knowledge gap and may inform future treatment strategies.

The online version contains supplementary material available at 10.1186/s12943-026-02574-0.