An adaptive semi-implicit finite element solver for brain cancer progression modeling.

Area of Science:

• Computational Biology
• Oncology
• Biophysics

Background:

• Glioblastoma is a highly aggressive Grade IV glioma with poor patient survival rates.
• Understanding glioblastoma progression requires accurate mechanistic modeling.
• In silico approaches offer valuable tools for quantifying primary brain tumor dynamics.

Purpose of the Study:

• To present a high-performance computing framework for simulating glioblastoma progression.
• To implement a continuum-based finite element model for scalable cancer simulations.
• To investigate the impact of various biological factors on glioblastoma evolution.

Main Methods:

• Developed a finite element framework utilizing open-source libraries and high-performance computing.
• Adopted the established proliferation-invasion-hypoxia-necrosis-angiogenesis model.
• Implemented arbitrary order discretization schemes and adaptive remeshing for accurate simulations.
• Conducted model sensitivity analysis on key biological parameters.

Main Results:

• The framework produced accurate and efficient 2D and 3D simulations of glioblastoma.
• Sensitivity analysis revealed the impact of vascular density, cell invasiveness, and angiogenesis.
• Individualized simulations using MRI data demonstrated the model's ability to capture complex disease dynamics.

Conclusions:

• The proposed framework enables patient-specific simulations of glioblastoma prognosis.
• This approach can bridge clinical imaging data with computational modeling for enhanced cancer understanding.
• The model offers a valuable tool for investigating brain cancer progression and informing treatment strategies.