Robust optimization of intensity-modulated radiation therapy accounting for biological uncertainties

Background and Objectives: As a treatment for cancer, IMRT, or intensity-modulated radiation therapy, is a form of radiation treatment that uses beams of radiation that can be adjusted to fit the tumor. This allows for more precise targeting of the tumor and can reduce the amount of normal tissue that is damaged by treatment. Optimization models help to increase the accuracy of the treatment plan by applying constraints on the minimum prescribed dose for tumors and the tolerable amount of radiation for healthy organs. Despite many advances in conducting IMRT treatment plans enhanced by optimization and mathematical programming, IMRT is subject to uncertainties that can have significant effects on the accuracy and efficacy of the treatment plan. These uncertainties arise from a number of factors, including variations in patient anatomy and the inherent biological variability of tumors and normal tissues. Robust optimization has emerged as a powerful tool for addressing uncertainties and improving the accuracy of IMRT treatment planning. Methods In this work, two radiobiological optimization models, including the tumor control probability (TCP) and the normal tissue complication probability (NTCP), have been constructed for a fixed value and an expected value scenario for robust model, on TROT dataset. Models have been implemented using MATLAB, and probability density functions are used to express the uncertain parameters of models. Results The findings imply that distribution function plays a significant role in the final dosage distribution and desired TCP-NTCP results, while robust treatment plans are less sensitive to changes in model parameters and improve the TCP of the tumor. Conclusion Applying these models to generate treatment plans based on each individual's biological parameters would be a promising method for personalized treatment plans aiming to achieve the most effective treatment possible while minimizing the risk of side effects and other complications.