Magnetic resonance imaging-based treatment planning for prostate cancer: Use of population average tissue densities within the irradiated volume to improve plan accuracy

Purpose: The purpose of this study was to investigate the feasibility of using population average tissue densities within the irradiated volume to improve the dosimetric accuracy of magnetic resonance imaging-based treatment plans for prostate cancer. Methods and materials: Computed tomography images and radiation therapy treatment plans from 20 patients with prostate cancer were reviewed retrospectively. Patient anatomy was segmented into fat, nonfat soft tissue, and bone. Population average tissue densities within the irradiated volume were obtained. Two bulk density override plans were generated using the tissue densities reported in International Commission on Radiation Units & Measurements Report 46 and those obtained in this study, respectively. Both plans were compared to the clinically approved computed tomography-based plan to assess dosimetric accuracy. Results: The population average tissue densities within the irradiated volume obtained in this study were found to be different from those reported in International Commission on Radiation Units & Measurements Report 46. Use of the population average tissue densities within the irradiated volume reduced dosimetric errors for all dose metrics, for example, V100 (percentage of prostate volume receiving 100% of the prescription dose; 0.32% vs 1.73%), D95 (dose covering 95% of the target volume; 0.32% vs 0.92%), D50 (dose covering 50% of the target volume; 0.30% vs 0.89%), and maximum dose to bladder (0.37% vs 0.78%), rectum (0.35% vs 0.95%), and penile bulb (0.23% vs 0.49%). All improvements were statistically significant. Conclusions: Use of population average tissue densities within the irradiated volume by the density override technique can improve the dosimetric accuracy of magnetic resonance imaging-based treatment plans for prostate cancer.