The Use of Computational Patient Models to Assess the Risk of Developing Radiation-Induced Cancers From Radiation Therapy of the Primary Cancer

Abstract

Even with highly conformal treatment delivery techniques for radiation therapy, one cannot avoid the irradiation of healthy tissues surrounding the tumor. In treatment planning, absorbed dose constraints are defined for organs at risk visible in the patient's computed tomography scan in order to minimize side effects. However, scattered or secondary absorbed dose is also deposited in areas outside of the imaged volume. Although these doses are typically quite low (well below 1% of the therapeutic dose), there is the potential risk for patients to develop a radiation-induced second malignancy later in life. For a systematic analysis of scattered or secondary doses and their carcinogenic effects, accurate patient-specific organ dosimetry is necessary. Computational patient models are needed because low absorbed doses to organs away from the primarily irradiated target area cannot be measured easily. Monte Carlo simulations can incorporate the geometry and tissue properties of computational phantoms to simulate dose deposition events weighted with radiation weighting factors. Epidemiological models can then be applied to relate organ equivalent absorbed doses to cancer risks. This paper describes the use of computational phantoms in conjunction with Monte Carlo dose calculations and epidemiological risk models. We demonstrate how these simulations are applied with the purpose of estimating second cancer risks in radiation therapy and of improving existing epidemiological risk models.