Vienna, Austria

ESTRO 2023

Session Item

May 15
16:30 - 17:30
Business Suite 3-4
Treatment planning: Photons
Sara Pilskog, Norway
Poster Discussion
Evaluation and validation of density override in VMAT treatment planning for lung SBRT
Chin Loon Ong, The Netherlands


Evaluation and validation of density override in VMAT treatment planning for lung SBRT

Chin Loon Ong1, Eric Franken1

1HagaZiekenhuis, Radiotherapy, Den Haag, The Netherlands

Show Affiliations
Purpose or Objective

Treatment planning in heterogeneous media has always been challenging. Although the accuracy of dose calculation algorithms has significantly improved, dose deposition in near-air density areas remain problematic. In some patients, the lung density is close to air properties (0.01-0.05g/cm3). In absence of tissue mass, the treatment planning system is forced to locally deliver much higher output in order to provide sufficient target coverage. The purpose of this study is to evaluate and validate the use of density override (DO) during volumetric modulated arc therapy (VMAT) planning for lung stereotactic body radiotherapy (SBRT).

Material and Methods

10 SBRT patients with low lung density are included in this study (table 1). For all patients, an internal target volume (ITV) was delineated, using all breathing phases. The margin area between ITV and PTV (ITV-PTV, 5mm) was overridden with a lung density of 0.15 (~-850HU) and 0.26g/cm3 (~-760HU, default mass density value for lung in RaySearch, RS), and 2 new plans were generated using the same constraints, respectively. Each plan was optimized on the average CT in RS such that 100% of the prescribed dose covers 95% of the PTV. Several DVH parameters and the monitor unit (MU) efficiency were compared between the 3 plans. Later, the DO is undone and robust evaluation was performed (standard functionality in RS) in 8 different scenarios with combination of 2mm shifts in all 3 translational directions. A plan is considered robust (pass) if 98% and 100% of the ITV were covered by the 95% and 90% of the prescription dose, respectively. The Dmax of the PTV and OAR of the worst case scenario were also compared.


The DVH analysis of all 3 plans for the 10 patients is shown in table 2. Plans optimized with DO show better PTV coverage, lower PTV Dmax and doses to all other organs and also generated less MU. Plan quality improves when ITV-PTV is overridden with higher lung density. The results of robust evaluation (with DO undone) are also shown in table 2. Plans with no DO and DO of 0.15 always meet the ITV coverage requirements while 2 plans with DO of 0.26 fail. For patient 1, 2 scenarios fail to meet the ITV requirement of V90% and for patient 3, 4 scenarios fail at both ITV V95% and V90%. Both patients have very small target volume and low average HU in ITV-PTV area. In the worst case scenario, the Dmax of the PTV, of a ring at 2cm distance and the most critical organ are always lower in plans optimized with DO.


The use of DO between ITV-PTV area improves VMAT lung SBRT plan quality in patient with very low lung density. Robust evaluation can be used (without DO) to validate the robustness of the plan. In general, the default lung density value of 0.26 will results in robust plan with superior plan quality. But in cases with very small ITV and low average HU in the ITV-PTV margin area, the use of lower lung density of 0.15 is recommended. The use of DO reduces OAR doses and the risk of over dosage caused by setup uncertainties.