ESTRO 2025 - Abstract Book

S116

Invited Speaker

ESTRO 2025

Given the advances in OART, we can move beyond the conventional margin approach and consider patient specific, dose-adaptive treatment planning. By appropriate image guidance, intra-fraction motion can be monitored and the accumulated dose can be estimated over the course of treatment. in an online adaptive treatment, technically it is feasible to adjust the treatment based on the accumulated dose. This approach allows for a drastic reduction of PTV margins, since any underdosage during the course of treatment can be compensated for in a subsequent fraction.

4839

Speaker Abstracts Reirradiation treatment planning accounting for previously delivered dose Ane Appelt Leeds Institute of Medical Research, University of Leeds, Leeds, United Kingdom. Leeds Cancer Centre, Department of Medical Physics, Leeds Teaching Hospitals NHS Trust, Leeds, United Kingdom Abstract: Reirradiation is one of the most complex clinical scenarios in radiotherapy, requiring careful integration of and accounting for previously delivered dose at all stages of treatment planning, evaluation, and decision-making. However, this process is complicated by numerous factors; including dose mapping across time and evolving anatomy (often with months or years between treatments), the need for rescaling dose distributions to account for fractionation effects, and the poorly understood dynamics of tissue recovery between treatment courses. Reirradiation treatment planning is particularly challenging, as previous dose must be accounted for as part of plan optimisation, and cumulative dose distributions should preferably be evaluated iteratively. Optimisation and iterative evaluation must to happen in a common space and on a common scale. In other words, it must be done on the same anatomical representation of the patient (same image set) and using an equieffective dose measure (such as EQD2Gy). No commercial treatment planning systems currently support this; posing a significant barrier to robust and safe reirradiation planning and delivery. A recent priority setting exercise by the transatlantic Reirradiation Collaborative Group (ReCOG) identified the need for software tools that integrate cumulative equieffective dose into plan optimisation as one of the most pressing gaps in clinical practice. In the absence of dedicated tools, clinic rely on workarounds; such as extracting previous point doses for organs-at risk (OARs) and manual calculation of indicative optimisation constraints, mapping previous field borders onto current imaging, or converting isodose lines to structures for avoidance purposes. These methods are manual and labor-intensive and often account for prior spatial dose distributions only in a limited manner. Moreover, they do not support cumulative volumetric planning objectives, such as mean dose or Vxx. Recent work has explored ways to address these limitations. Studies have investigated automation and scripting for dose mapping and isodose structure creation (Zhang et al., Phys Imaging Radiat Oncol, 2024) as well as plan optimisation using ‘bias’ or ‘background dose’ plans (García-Alvarez et al., Radiother Oncol, 2025), both leveraging adaptations of existing clinical software solutions. Our group in Leeds has collaborated with a major radiotherapy vendor to develop plan optimisation that directly incorporates previous dose distribution(s), using optimisation cost functions that evaluate cumulative EQD2Gy across multiple treatments (Murray et al., Radiother Oncol, 2023; Ödén et al., Med Phys, 2024). This lecture will review these approaches and discuss ongoing advancements, including the integration of uncertainty and robustness into reirradiation planning (Thompson et al., Cancers, 2023).