Predicting the radiation control probability of heterogeneous tumour ensembles: data analysis and parameter...

Area of Science:

• Radiation Oncology
• Mathematical Biology
• Biophysics

Background:

• Tumor control probability (TCP) models are crucial for radiotherapy planning.
• Existing homogeneous models do not fully capture the complexity of heterogeneous tumors.
• Accurate estimation of radiobiological parameters is essential for predictive modeling.

Purpose of the Study:

• To develop a closed-form formula for TCP of heterogeneous tumors.
• To analytically average a homogeneous TCP formula over key biological and physical parameters.
• To enable straightforward fitting to clinical data for parameter estimation.

Main Methods:

• Analytical averaging of a double-exponential TCP formula over distributions of radiosensitivity, density, repopulation rate, tumor volume, and dose.
• Fitting the derived formula to published clinical TCP data grouped by dose, volume, and treatment duration.
• Determination of fitted parameter values, confidence intervals, and goodness-of-fit statistics.

Main Results:

• The formula provides non-rejectable fits to clinical data for TCP grouped by dose and volume.
• Fits are good when radiosensitivity parameters are near laboratory estimates, though volume dependence parameters may be high.
• Alternative good fits are achieved with a volume parameter of one, using radiosensitivity values around one-third of laboratory estimates.
• Non-rejectable fits to data grouped by dose and treatment duration are possible with parameters near laboratory estimates.

Conclusions:

• The derived closed-form TCP formula effectively models heterogeneous tumors.
• The formula facilitates the estimation of radiobiological parameters from clinical data.
• Parameter estimations show flexibility, allowing for different combinations of radiosensitivity and volume dependence to fit data effectively.