ESTRO 2024 - Abstract Book

S4885

Physics - Quality assurance and auditing

ESTRO 2024

1 Aarhus University Hospital, Danish Centre for Particle Therapy, Aarhus, Denmark. 2 Aarhus University, Department of Clinical Medicine, Aarhus, Denmark. 3 University of Southern Denmark, Department of Oncology, Vejle Hospital, Vejle, Denmark. 4 Aalborg University Hospital, Department of Oncology, Aalborg, Denmark

Purpose/Objective:

Patients with high-risk prostate cancer are often treated with radiotherapy to the prostate and pelvic lymph nodes and may benefit from proton therapy due to the large treatment volume. However, this tumour site is complex as the target includes both soft tissue structures (the prostate and seminal vesicles) exhibiting inter-fractional anatomical variations as well as the pelvic lymph nodes that are stationary relative to the pelvic bones.

The aim of this study was to evaluate target dose delivery robustness of proton therapy for high-risk prostate cancer within a pilot study before entering into a randomised clinical trial.

Material/Methods:

The ten first patients included in the pilot phase of the randomised clinical trial, PROstate PROTON Trial 1 (NCT05350475) were included in this study. For each patient, weekly repeated computed tomography scans (rCTs) with associated recalculated treatment plans were used. Treatment planning complied with protocol guidelines with 78 Gy to the primary clinical target volume (CTVp; prostate and involved seminal vesicles) and 56 Gy to the elective target (CTVe; the pelvic lymph nodes and the remaining seminal vesicles) in 39 fractions using a simultaneously integrated boost technique. To accommodate both intra- and inter-fractional motion, internal margins of 4 mm in the cranio-caudal direction and 2 mm in other directions were applied to CTVp to form an internal target volume (ITV) to which dose was prescribed during treatment plan optimisation. The ITV was cropped to exclude the rectum. A four-field configuration using pencil beam scanning was used for all patients, with two lateral oblique beams and two posterior oblique beams, to ensure minimal radiation to the primary organs at risk; rectum, bladder and bowel. Further, robust optimisation with multi-field optimisation (MFO) was used with 14 robustness scenarios combining 0 mm and 5 mm isotropic isocenter shift with +/-3.5% range uncertainty. For the purpose of this study, current treatment plans at any given time were recalculated on a total of 69 rCTs for the ten patients. As in our on-line match strategy, the rCTs were registered to the primary CT via a six dimensional match to bony anatomy followed by a match in three translational dimensions to three gold markers (Gold Anchor) implanted in the prostate. Recalculations included two uncertainty scenarios with +/-3.5% range uncertainty. Three of the ten patients had a re-plan during their course of treatment based on a clinical evaluation – e.g. due to excessive amounts of air or stool in the rectum on the original planning CT compared to every-day cone-beam CTs, or to move the isocenter in the treatment plan for practical positioning issues. The volume of the target structures that received 95% of the prescribed dose should be greater than or equal to 98% (V95% ≥ 98%). Dose/volume measures corresponding to this constraint for both CTVp and CTVe were evaluated for each recalculated treatment. Considering the nominal plans and the uncertainty scenarios separately, the data were pooled across patients and all available scans (69 in total).

Results: