Related Experiment Video
Updated: Mar 27, 2026
Radiation Planning Assistant - A Streamlined, Fully Automated Radiotherapy Treatment Planning System
Published on: April 11, 2018
Robust VMAT planning strategy to ensure target coverage and reduce NTCP in head and neck cancer
Marcel A van Schie1, Anna L Petoukhova1, Linda Rossi2
1Department of Medical Physics, Haaglanden Medical Center, Leidschendam, the Netherlands.
Purpose:
Fixed PTV margins of 3 mm as applied in current head and neck cancer (HNC) VMAT treatment planning may be suboptimal for individual patients. In robust planning, the PTV is omitted and the plan is directly optimized on the CTV. In this study we developed a robust VMAT planning strategy to ensure robust CTV coverage for all patients, while keeping toxicity levels as low as possible.
Methods:
We included 20 HNC patients that received a 35 fraction treatment with a primary dose of 70 Gy and elective dose of 54.25 Gy. We applied robust planning in two steps. First, the robust coverage of the clinical PTV-based plans was assessed using the CTV V95% ≥ 98% as coverage criteria in the voxel-wise minimum (vw-min) dose distribution, which was constructed from 6 orthogonal and 8 diagonal treatment delivery scenarios with a 3 mm patient shift. Second, based on the robust evaluation, in the clinical plans that showed robust CTV coverage, OAR doses were reduced while preserving robust CTV coverage. In the non-robust clinical plans (CTV V95% < 98% in the vw-min dose distribution), robust optimization was performed with 6 orthogonal treatment delivery scenarios with a 3 mm patient shift plus a nominal scenario, to achieve robust CTV coverage, while OAR dose was only reduced where possible. Additional treatment planning time was recorded to assess clinical feasibility. Normal tissue complication probabilities (NTCP) were calculated to compare expected toxicities between PTV-based and robust optimized plan.
Results:
While 9/20 clinical plans initially showed insufficient CTV coverage in the vw-min dose distribution, after robust optimization all plans had robust CTV coverage. Robust planning took on average 20 min. For all patients, robust optimized plans had equal or reduced NTCPs compared to the PTV-based plans. Median reductions of OAR mean dose in the oral cavity and parotid glands, as well as all NTCPs, were statistically significant. Median NTCP reductions were largest for dysphagia grade 2+ (-1.4%).
Conclusions:
We introduced a robust VMAT planning strategy for HNC which ensured robust CTV coverage and statistically significant NTCP reductions.
