A patient-specific computational model of hypoxia-modulated radiation resistance in glioblastoma using 18F-FMISO-PET

Abstract

Glioblastoma multiforme (GBM) is a highly invasive primary brain tumour that has poor prognosis despite aggressive treatment. A hallmark of these tumours is diffuse invasion into the surrounding brain, necessitating a multi-modal treatment approach, including surgery, radiation and chemotherapy. We have previously demonstrated the ability of our model to predict radiographic response immediately following radiation therapy in individual GBM patients using a simplified geometry of the brain and theoretical radiation dose. Using only two pre-treatment magnetic resonance imaging scans, we calculate net rates of proliferation and invasion as well as radiation sensitivity for a patient's disease. Here, we present the application of our clinically targeted modelling approach to a single glioblastoma patient as a demonstration of our method. We apply our model in the full three-dimensional architecture of the brain to quantify the effects of regional resistance to radiation owing to hypoxia in vivo determined by [(18)F]-fluoromisonidazole positron emission tomography (FMISO-PET) and the patient-specific three-dimensional radiation treatment plan. Incorporation of hypoxia into our model with FMISO-PET increases the model-data agreement by an order of magnitude. This improvement was robust to our definition of hypoxia or the degree of radiation resistance quantified with the FMISO-PET image and our computational model, respectively. This work demonstrates a useful application of patient-specific modelling in personalized medicine and how mathematical modelling has the potential to unify multi-modality imaging and radiation treatment planning.

Keywords: glioblastoma; hypoxia; mathematical modelling; patient-specific; radiation resistance.

Figure 1.

Orthogonal views of the patient's diagnostic T1-weighted gadolinium enhanced (T1Gd), T2-weighted MRI and FMISO-PET obtained prior to RT, with the composite RT dose based on MRI-defined margins. The yellow region of the FMISO-PET image indicates hypoxia as defined by tumour-to-blood values greater than or equal to 1.2 [24]. (Online version in colour.)

Figure 2.

The BrainWeb phantom provides a voxel-wise probability map used to define the invasion rate of the tumour in model simulations [20]. Each voxel is composed of grey matter, white matter and/or CSF in relative proportions such that the sum of all tissues in each voxel is unity. The voxels in the phantom are cubic with dimensions 1 × 1 × 1 mm = 1 mm³.

Figure 3.

Left column: tumour size versus time. The dashed line is the model-predicted tumour size on T2-weighted MRI, solid line is T1-weighted gadolinium enhanced (T1Gd) MRI. Black circles are tumour sizes calculated volumetrically with 1 mm error bars based on interobserver measurement uncertainty, and the grey rectangle represents when radiation therapy (RT) was delivered. Middle column: zoom-in of tumour size versus time during RT. Right column: three-dimensional renderings of RT dose, FMISO-PET and model-predicted tumour following RT. Top row: patient-specific simulation of RT without the oxygen enhancement ratio (OER) to model uniform sensitivity to RT. Bottom row: simulation with hypoxia-mediated radiation resistance in regions of FMISO-PET T/B activity greater than 1.2.

Figure 4.

Spatial metric between the model-predicted T1Gd surface (light/cyan contour) and the observed tumour boundary (dark/red contour) on the second pre-RT MRI, indicating a median (±standard deviation) of 2.2 ± 2.2 mm using the observed tumour region (dark/red) as ‘true’. Similar accuracy was observed for the post-RT MRI (table 4). Negative distances indicate an under-estimation of the model-predicted tumour front, whereas positive distances indicate an over-estimation, with zero distance indicating intersection of the surfaces. (Online version in colour.)