Year: 2015
Type: conference-abstract
Abstract: Purpose/Objective: Treatment planning for Stereotactic Body Radiation Therapy (SBRT) of liver tumours is often challenging due to large respiratory motion and a close vicinity of OARs such as great vessels and duodenum.In such cases a section of the PTV is edited to spare nearby OARs.We evaluated a PTV-less probabilistic planning technique which directly incorporates respiratory motion and other geometric uncertainties of the GTV in the dose optimization process.An improved balance between OAR exposure and the confidence of proper target dosage is expected, as well as a more efficient planning procedure since less human interaction is required.Materials and Methods: Four liver SBRT plans (3x20 Gy prescription) for which clinical planning had been problematic due to OAR constraints were re-planned using an in-house developed research plug-in for Pinnacle (version 9.100).The plug-in combines the tumour trajectory extracted from the 4D planning CT (~1 cm amplitude for this case) with the Gaussian distribution of random errors (0.25 cm) into a dose blurring kernel, and incorporates a Gaussian distribution of systematic errors as GTV offsets (0.34 cm).Our probabilistic objective assumes shift invariance and aims for a set confidence (e.g.90%) of GTV minimum dose.Clinical and probabilistic plans were compared using in-house software that accurately simulates the effects of motion and uncertainties on an optimized dose distribution by explicitly sampling three daily errors for each systematic error (10000 were simulated), and 100 positions along the breathing trajectory for each daily error.OARs were evaluated via a traditional DVH.Results: In three out of four cases the probabilistic plan showed a clear clinical benefit compared to the conventional plan.In two cases the clinical PTV coverage was lowered to spare the great vessels and duodenum, here probabilistic planning significantly increases GTV coverage, while maintaining a low enough dose on the OAR (e.g.Fig. 1).In a third case a higher dose to the great vessels was tolerated in the clinical plan in favour of getting proper PTV coverage.Using probabilistic planning we were able to reduce the great vessel dose while still reaching 90% target dose confidence.In the fourth case all clinical constraints were met both in the clinical and probabilistic plan.Conclusions: Probabilistic planning can make a valuable contribution to treatment planning in those cases in which it is difficult to meet all clinical criteria.
Source: Radiotherapy and Oncology
Authors: M. Miften, Y. Vinogradskiy, B. Kavanagh, Q. Diot
Institution: The Netherlands Cancer Institute
Language: English
Subfield: Radiation
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $3,770
OpenAlex ID: W2584500796