ESTRO 2024 - Abstract Book

S4654

Physics - Optimisation, algorithms and applications for ion beam treatment planning

ESTR0 2024

1771

Digital Poster

Evaluating image-guided dose painting for skull-based chordoma patients treated with proton therapy

Caterina Brighi 1 , Giovanni Parrella 2 , Letizia Morelli 2 , Silvia Molinelli 3 , Giuseppe Magro 3 , Giulia Riva 3 , Alberto Iannalfi 3 , Mario Ciocca 3 , Sara Imparato 3 , David E. J. Waddington 1 , Paul J. Keall 1 , Chiara Paganelli 2 , Ester Orlandi 3 , Guido Baroni 2 1 University of Sydney, Image X Institute, Faculty of Medicine and Health, Sydney, Australia. 2 Politecnico di Milano, Department of Electronics, Information and Bioengineering, Milano, Italy. 3 National Center for Oncological Hadrontherapy, CNAO, Pavia, Italy

Purpose/Objective:

Skull-based chordoma (SBC) patients have limited prognosis with local relapse occurring a few years after completion of radiation treatment. Local recurrence occurs due to the ability of tumour cells to develop mechanisms of resistance to radiotherapy. Patient-specific MRI data could be used to tailor radiotherapy plans to the individual tumour radiobiological characteristics, thereby improving the probability of local control. This study presents a workflow to develop MRI-guided dose painting (DP) radiotherapy plans in patients with SBC undergoing proton therapy, and quantifies the estimated therapeutic advantages of DP compared to conventional radiation treatment.

Material/Methods:

Retrospective data from five SBC patients treated with proton therapy with a fixed-beam pencil beam scanning delivery system were used in this study. Diffusion-weighted MRI data were used to derive maps of cellularity from an apparent diffusion coefficient-based model. 1,2 Per-voxel cellularity data within the gross tumour volume (GTV) were converted into dose prescriptions ranging between 74-81 Gy (RBE) using a linear function. 3 A high-risk clinical target volume (CTV HR ) was defined as 3-5 mm expansion of the GTV modified according to the anatomy and the surgical pathway, and a low-risk CTV (CTV LR ) consisted in 5 mm expansion of the CTV HR . The high- and low-risk planning target volumes (PTV HR and PTV LR ) encompassed a 2 mm isotropic expansion of the CTV HR and CTV LR , respectively. As per in the conventional sequential boost (SB) plans, the dose prescribed to the non-GTV CTV HR and CTV LR was 74 Gy (RBE) and 54 Gy (RBE), respectively. 4 DP plans were calculated in RayStation planning system with an inverse planning approach. 5 DP plans were developed with two beams, to simulate an approach similar to what was used in the SB plans, 4 and four beams, to achieve a better dose distribution in the target. The same constraints to organs at risk (OARs) as per SB plans were used for DP plans optimization, with the added constraint of not exceeding a maximum dose of 81 Gy (RBE) in the PTV LR . Dose metrics and conformity to dose prescriptions (evaluated with a quality factor, QF, ranging 0-100% quantifying the percentage of voxels with a planned dose as close as possible to the prescribed dose 5 ) in the targets, clinical goals to OARs and GTV, and dose delivered to areas of recurrence overlapping with CTV LR were compared between the DP and the conventional SB plans. An example of DP plans and SB plan, with relative dose prescriptions and QF maps is shown in Figure 1 . A Poissonian linear-quadratic model was used to estimate tumour control probability (TCP) in the GTV, by using patient-specific per-voxel cellularity data and ion-specific radiosensitivity parameters derived from survival experiments on chordoma cell lines irradiated with proton beams. 6 TCP in the GTV obtained with DP plans was compared with TCP calculated for SB plans.