Program Information
A Model of Tumor Control Probability Based On Tumor Shape and Spatial Distribution of Oxygen
D Goldbaum*, R Hamilton , Univ Arizona, Tucson, AZ
Presentations
MO-B-BRD-10 (Monday, March 9, 2015) 10:00 AM - 12:00 PM Room: Ballroom D
Purpose: The purpose of our investigation is to improve our qualitative understanding of the relationship between tumor control probability (TCP) and tumor volume (V), and to present a model that will predict TCP using measurements from the patient to be treated.
Methods: We present qualitative models with phenomenological parameters that can be set by comparison to past patient data. Each model uses a bimodal expression for the cell-survival fraction, where each clonogen is considered to be either normoxic or hypoxic, and where the radiosensitivity of the hypoxic clonogens is characterized by an average hypoxia reduction factor. In each model we provide an argument for how the hypoxia reduction factor should scale with respect to V.
Also, we point out that in a clinical setting one can often characterize tumor shape by measuring the tumor’s diameter along each of its “principal axesâ€. We use these parameters to analyze the tumor as if it were an ellipsoid. Then for more in-depth algebraic analysis, we use the simplifying assumption that the tumor is spherical.
Results: We generated a model where the scaling relationship between TCP and V is similar to those found by fitting clinical data. This suggests that one can qualitatively understand this scaling through the effect of hypoxia, and that one might be able to use tumor measurements from a single patient to predict the TCP vs. dose relationship for that patient.
Conclusion: We generated a phenomenological model, based on reasonable qualitative principles, where the scaling of TCP with respect to V resembles that observed in clinical data. We will use TCP vs. V data to determine values for the model’s phenomenological parameters. Then we will have a sound basis on which to evaluate its clinical value. We will also analyze the model with respect to tumor hypoxia data.
Contact Email: