S1915
Physics - Dose prediction/calculation, optimisation and applications for photon and electron planning
ESTRO 2026
Purpose/Objective: Short-course radiotherapy (SCRT) for locally advanced rectal cancer (LARC) with either photons or protons is used in the ongoing randomized trial PRORECT (NCT04525989). Radiation-induced bowel toxicity is one of the most prevalent side effects of rectal cancer therapy. However, robust dose-constraints for acute bowel toxicity in SCRT setting are lacking. In PRORECT, dose-constraint for RTOG-defined bowel bag (1) is V18Gy ≤ 450 cm3(2). Recently, the first NTCP model predictive for acute bowel toxicity after SCRT was presented (3). Based on these promising results, as well as experiences from other international centres using the stricter V18 ≤ 250cc constraint (4), we aimed to (a) retrospectively evaluate compliance with this constraint and (b) to create new treatment plans to minimize dose to the bowel bag. Material/Methods: 72 patients treated in the PRORECT study were included. Of these, 36 had received proton therapy (PT) and 36 had received photon therapy (xRT). V18 Gy ≤ 250 cm3 was exceeded in seven of the xRT patients. New xRT plans were created for these seven patients in an attempt to fulfill V18 Gy ≤ 250 cm3. A model based on all 36 xRT patients, taking into account overlap volumes and approximate closest distance between CTV and bowel bag, was created to predict V18 Gy. This model was applied to the 36 PT patients. New xRT plans were created for the seven PT patients with the highest predicted V18 Gy Results: Of the new plans created, three of the seven selected xRT patients and six of the seven selected PT patients received V18 Gy ≤ 250 cm3. The model predicted that all PT patients would satisfy V18 ≤ 250 cm3 except one. Taking into account this and the newly created plans, this translates to 86% of all considered patients satisfying the V18 Gy ≤ 250 cm3 dose guidance to bowel bag. Conclusion: Based on this cohort, it is feasible to reduce the dose- constraint from V18 Gy ≤ 450 cm3 to V18 Gy ≤ 250 cm3 for the bowel bag in clinical practice. References: 1.Gay HA et al. Pelvic normal tissue contouring guidelines for radiation therapy: a Radiation Therapy Oncology Group consensus panel atlas. Int J Radiat Oncol Biol Phys. 2012;83(3):e353-62.2. Devisetty K et al. A multiinstitutional acute gastrointestinal toxicity analysis of anal cancer patients treated with concurrent intensity-modulated radiation therapy (IMRT) and chemotherapy. Radiother Oncol. 2009;93(2):298-301.3, NTCP Models for acute bowel toxicity and chemotherapy compliance based on LARC patients treated with 5x5 Gy and chemotherapy in the RAPIDO trial. Poster E25-479 Tanaka et al. ESTRO 20254. UK national rectal cancer intensity modulated
radiotherapy (IMRT guidance) Keywords: Rectal cancer, bowel bag dose
Digital Poster 3101 The Dark Side of the MLC: A study regarding the umbra interleaf and inter-bank Leakage Íris Almeida 1,2 , Filipe Dias 1,2 , Diana L. Silva 1,2 , José Mesquita 1,2 , Anabela G. Dias 1,2 , Catarina Souto 1,2 , Joana Lencart 1,2 1 Medical Physics, IPO-Porto, Porto, Portugal. 2 Medical Physics, Radiobiology and Radiation Protection Group, IPO-Porto Research Center, Porto, Portugal Purpose/Objective: The purpose of this study was to evaluate the accuracy of the Analytical Anisotropic Algorithm (AAA), and the Acuros XB (AXB) algorithm (Varian Eclipse 16.1.2), when replicating the Multileaf Collimator (MLC) interleaf and inter-bank leakage effects for small photon fields in the umbrae region. Material/Methods: Measurements were performed on a Varian TrueBeam Linear Accelerator equipped with a Millennium 120- leaf MLC. Inplane profiles were acquired for small static MLC-defined fields 1x1, 2x2, and 3x3 cm ² , while the jaws shaped a 10x10 cm ² field. All measurements were performed in a water phantom, for 6 MV at a source-to-surface distance (SSD) of 95 cm, and a depth of 5 cm, using a PTW microDiamond detector. Opposite leaf pairs were closed, when applicable, on the main axis.Plans reproducing the measurement setup were calculated in Eclipse for both AAA and AXB, using a high resolution of 0.1 mm. The relative dose profiles were exported for each field size and calculation algorithm. A Python-basedscript tool was developed, to simultaneously plot measured, AAA and AXB profiles for each field size. The profiles were compared both visually and quantitatively, focusing on the umbrae region. Point- wise differences between measured and calculated profiles were computed in this region.The same analysis was then repeated with the MLC central junction positioned below the collimator jaws, in order to evaluate potential variations in the interleaf leakage modelling. The average point to point difference and corresponding relative dose mean differences were
determined and compared with the two MLC configurations (original and shifted), for AAA, and AXB algorithms. Results: For the 3x3 cm ² field, the mean difference
between AAA calculated and measured profile was - 2.12%, while AXB showed a 0.34% mean difference relative to the measured inplane profile. Graph 1 presents the comparison between the measured, AAA,
Made with FlippingBook - Share PDF online