Premium
Development of an in silico stochastic 4D model of tumor growth with angiogenesis
Author(s) -
Forster Jake C.,
Douglass Michael J.J.,
HarrissPhillips Wendy M.,
Bezak Eva
Publication year - 2017
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1002/mp.12130
Subject(s) - angiogenesis , head and neck , stem cell , stochastic modelling , radiation therapy , hypoxia (environmental) , dosimetry , in silico , pathology , biology , biomedical engineering , cancer research , medicine , nuclear medicine , chemistry , surgery , mathematics , microbiology and biotechnology , statistics , organic chemistry , oxygen , biochemistry , gene
Purpose A stochastic computer model of tumour growth with spatial and temporal components that includes tumour angiogenesis was developed. In the current work it was used to simulate head and neck tumour growth. The model also provides the foundation for a 4D cellular radiotherapy simulation tool. Methods The model, developed in Matlab, contains cell positions randomised in 3D space without overlap. Blood vessels are represented by strings of blood vessel units which branch outwards to achieve the desired tumour relative vascular volume. Hypoxic cells have an increased cell cycle time and become quiescent at oxygen tensions less than 1 mmHg. Necrotic cells are resorbed. A hierarchy of stem cells, transit cells and differentiated cells is considered along with differentiated cell loss. Model parameters include the relative vascular volume (2–10%), blood oxygenation (20–100 mmHg), distance from vessels to the onset of necrosis (80–300 μ m) and probability for stem cells to undergo symmetric division (2%). Simulations were performed to observe the effects of hypoxia on tumour growth rate for head and neck cancers. Simulations were run on a supercomputer with eligible parts running in parallel on 12 cores. Results Using biologically plausible model parameters for head and neck cancers, the tumour volume doubling time varied from 45 ± 5 days ( n = 3) for well oxygenated tumours to 87 ± 5 days ( n = 3) for severely hypoxic tumours. Conclusions The main achievements of the current model were randomised cell positions and the connected vasculature structure between the cells. These developments will also be beneficial when irradiating the simulated tumours using Monte Carlo track structure methods.