Article
Is the invasive growth pattern of malignant gliomas determined by the physical structure of the brain? : results of a computer simulation
Wird das invasive Wachstumsmuster maligner Gliome durch die physikalische Struktur des Gehirns verursacht? : Ergebnisse einer Computersimulation
Search Medline for
Authors
Published: | May 4, 2005 |
---|
Outline
Text
Objective
Malignant gliomas are characterized by their highly infiltrative growth pattern which ultimately results in distant and multifocal tumour recurrence despite aggressive loco-regional therapy. Clinical, radiological and pathological evidence suggests that spreading of the tumour cells might be guided by fiber tracts within the white matter of the brain. In order to better understand the role of fiber tract anatomy during glioma invasion, we introduce computer simulations with a new mathematical model.
Methods
A simple lattice-gas cellular automaton model is used to simulate glioma growth and invasion. We consider healthy immobile brain tissue and potentially mobile malignant cells. Malignant cells may move or rest and only resting malignant cells proliferate (‘Go or grow’ hypothesis). Movement and resting behaviour are controlled by local cell densities and density gradients of brain and cancer cells. Chemotaxis and any cellular interactions independent of cell densities are not explicitly included in the model. The brain tissue is destroyed by cancer cells and it is assumed that the brain cannot regenerate. The physical structure of the brain, in particular its white matter tracts, are modelled as a prepattern represented by a vector field which introduces unidirectional persistence of malignant cell movement.
Results
We only observed invading tumour cells and growing tumour islets in front of the expanding tumour bulk (i.e. the growth pattern typically observed in clinical practice), when the prepattern was included in the simulations. No density gradients of brain or tumour cells are required to observe this behaviour. Without the influence of the prepattern (i.e. when the physical structure of the brain surrounding the tumour is neglected), only isolated glioma cells are found in the vicinity of the main tumour bulk.
Conclusions
Our results suggest that the infiltrative growth of malignant gliomas could be, in part, determined by the physical structure of the surrounding brain rather than by intrinsic properties of the tumour cells. Targeting the substrate of invasion instead of the tumour cells might prove a superior anti-invasion treatment strategy for malignant gliomas.