Fine-tuning the normal tissue objective in eclipse for lung stereotactic body radiation therapy

The purpose of this study was to characterize the effects of the normal tissue objective (NTO) on lung stereotactic body radiation therapy (SBRT) dose distributions. The NTO is a spatially varying constraint used in Eclipse to limit dose to normal tissues by steepening the dose gradient. However, the multitude of potential NTO setting combinations challenges optimal NTO tuning. In the present study, a broad range of NTO settings are investigated for lung SBRT treatment planning with volumetric modulated arc therapy(VMAT). Ten prior lung SBRT cases were replanned using NTO priorities of 1, 50, 100, 200, 500, and 999 in combination with fall-off values of 0.01, 0.05, 0.10, 0.15, 0.20, 0.30, 0.50, 1.00, and 5.00 mm^-1 and the automatic NTO. NTO distances to planning target volume (PTV), start dose, and end dose were 1 mm, 100%, and 10%, respectively, for all 600 plans. Prescription dose covered 95% of the PTV. The following metrics were recorded: conformity index (CI), ratio of the 50% prescription isodose volume to PTV (R_50%), maximum dose 2 cm away from PTV (D_2cm), lung volume of ≥20 Gy (V_20Gy), maximum PTV dose (PTV_max), and monitor units (MUs). Differences between prior plans and NTO plans were evaluated using the Wilcoxon signed-rank test. Different combinations of NTO settings resulted in wide-ranging plan quality metrics: CI (1.00 to 1.54), R_50% (3.95 to 7.57), D_2cm (33.4% to 67.9%), V_20Gy (1.66% to 2.75%), MU (1.81 cGy^-1 to 4.69 cGy^-1), and PTV_max (118% to 175%). Although no settings were optimal for all metrics, a fall-off of 0.15 mm^-1 and a priority of 500 best satisfied institutional criteria. Compared with prior plans, NTO plans resulted in significantly lower R_50% (4.00 vs 4.35, p = 0.002), lower V_20Gy (1.22% vs 1.32%, p = 0.006), and higher PTV_max (138% vs 122%, p = 0.002). All of the prior and well-tuned NTO plans met Radiation Therapy Oncology Group (RTOG) 0813 guidelines. Lung SBRT dose distributions were characterized across a range of NTO settings. NTO plans with well-tuned settings compared favorably with prior plans.