ESTRO 2023 - Abstract Book

S86

Saturday 13 May

ESTRO 2023

1 Canberra Health Services, Radiation Oncology Department, The Canberra Hospital, Canberra, Australia; 2 Canberra Health Services, Medical Physics and Radiation Engineering Department, The Canberra Hospital, Canberra, Australia; 3 Australian National University, ANU College of Health and Medicine, Canberra, Australia; 4 Australian National University, Division of Genome Sciences & Cancer, John Curtin School of Medical Research, Canberra, Australia; 5 Australian National University, Irradiation Immunity Interaction Lab, Division of Genome Sciences & Cancer, John Curtin School of Medical Research, Canberra, Australia Purpose or Objective Spine stereotactic body radiotherapy (SBRT) is characterised by steep radiation dose gradients to a planning target volume (PTV) that can overlap the thecal sac (TS). Inherent spinal cord (SCo) motion within the spinal canal (SCa) may result in higher dose delivery to SCo, compared to the planned treatment, increasing the risk of radiation myelopathy when appropriate planning organ at risk volume (PRV) is not considered. Anatomical variations outside SCa may also affect accurate treatment delivery. We aim to study the effect of body weight changes on TS dose in volumetric modulated arc therapy (VMAT)-based spine SBRT plans. Materials and Methods Datasets of 5 different patients were used to create clinically acceptable single vertebral level SBRT plans at thoracic and lumbar regions. TS was contoured as a surrogate for SCo PRV. Maximum TS dose (TSmax) was optimised to be under 17Gy in 2 fractions. Mild and moderate weight gain and loss duplicate datasets were created using the original planning scan by overriding densities at the patient external contour to adipose tissue and air, respectively (Fig 1a). Anisotropic body contour changes of +/- 0.6cm anteriorly, 0.3cm laterally and 0.1cm posteriorly were applied for mild weight gain and loss, whereas changes of +/- 1.0cm anteriorly, 0.6cm laterally and 0.2cm posteriorly for moderate weight variations. Volume changes, calculated from 3cm above to 3cm below the respective vertebral body (Fig 1b), were used as surrogates for changes in body weight. Original optimised plans were recalculated on the duplicate datasets and TSmax of each plan was analysed. Analyses considered the effect of PTV shapes (based on consensus guidelines), mediums (lung/soft tissue density), and beam models (6MV flattened and 6MV/10MV flattening filter free beams) on TSmax with body weight changes. Dose calculations were performed using Eclipse Acuros XB algorithm version 15.6.

Results TS doses from 165 spine SBRT plans were analysed. Weight gain and loss had a significant effect on TSmax, with a negative linear relationship between weight changes and TSmax (Fig 2a and b). Mild and moderate weight gain significantly decreased TSmax from ~16.9Gy to ~16.7Gy and ~16.6Gy respectively, whereas mild and moderate weight loss significantly increased it to ~17.1Gy and ~17.3Gy across the modelled datasets. Effect of weight changes on TSmax was modelled using a linear equation. PTV shapes, tissue mediums, and beam models did not influence the observed effect of weight changes on TSmax (Fig 2c, d and e).