ESTRO 2025 Congress report
By the reirradiation focus group (report #2/5)

1. Reirradiation treatment planning accounting for previously delivered dose
   Physics Symposium
   Presenter: Ane Appelt

Effective planning of reirradiation treatment requires accurate consideration of the previously delivered dose. A comprehensive evaluation of cumulative dose is essential and must incorporate various factors, including anatomical changes, fractionation effects, and tissue recovery between treatment courses.

Over time, anatomy can significantly change between the initial and subsequent radiation treatments. In such cases, deformable image registration becomes a crucial tool to align both dose distributions onto a common anatomical reference and enable cumulative dose assessment.

Since biological effects are influenced by the dose per fraction, it is necessary to convert all doses from multiple treatments into a standardised scale such as the equivalent dose in 2Gy fractions (EQD2). This enables meaningful summation of the biological dose.

Tissue recovery is another important consideration. When months or years pass between treatments, normal tissues undergo biological processes such as repopulation, compensation, or regrowth. However, data on tissue recovery are extremely limited, with only sparse evidence from animal models and even less from human studies. The central nervous system and spinal cord are among the few tissues in which some recovery has been documented. In contrast, some tissues may become more radiosensitive over time.

Several strategies, ranging from simple to complex, are available for dose prescription in reirradiation. The most commonly used in clinical practice include previous OAR maximum point dose assessment (a safe and conservative method) and planning avoidance based on prior isodose volumes (a manual, labour-intensive approach that cannot be used to optimise volumetric constraints).

A more advanced technique involves optimisation of the new plan by treating the prior dose as a background dose. This allows for cumulative dose assessment through the use of objective functions in EQD2 space, which may permit the use of a higher prescription dose to the new PTV.

Tools are emerging that can refine this process further. These include EQD2-based planning systems that incorporate α/β ratios that are specific to regions of interest and model tissue recovery within dedicated optimisation algorithms. These advancements are intended to enable combined dose evaluation and improve the safety and efficacy of reirradiation in clinical practice.

In conclusion, the integration of previous dose data into current treatment planning is fundamental for safe and effective reirradiation, and the development of EQD2-driven, biologically informed planning tools will be critical for future clinical implementation.

2. 𝘙𝘦𝘪𝘳𝘳𝘢𝘥𝘪𝘢𝘵𝘪𝘰𝘯 r𝘦𝘷𝘰𝘭𝘶𝘵𝘪𝘰𝘯
   Joint ESTRO-American Association of Physicists in Medicine 𝘗𝘩𝘺𝘴𝘪𝘤𝘴 𝘚𝘺𝘮𝘱𝘰𝘴𝘪𝘶𝘮
   𝗖𝗵𝗮𝗶𝗿𝘀: Daniela Thorwart (Germany), Charles Mayo (USA)
   Speakers: Charles Mayo, Søren M. Bentzen, Marija Popovic, Lone Hoffmann

In this symposium, the focus was on the technical and practical aspects of reirradiation, highlighting the systematic approach required in clinical workflow.

The physicist's role in the clinical reirradiation workflow As the first speaker, Chuck Mayo's presentation provided an initial perspective on the practical implementation of reirradiation. He noted the inherent challenges that physicists face in daily practice during the treatment of patients who require reirradiation, particularly the difficulty of making decisions when definitive answers or established guidelines may not be available. A central theme that can be derived from his remarks is the significant variability in how reirradiation is approached across different treatment centres. He highlighted the necessity for systematic data capture within the clinical workflow. This is particularly important because different centres, or even individual cases within the same centre, may have variations in their protocols, making it difficult to aggregate data and to learn from outcomes without a structured approach to the recording of treatment details.

Reirradiation tolerance: biology often trumps physics A major theme was the critical role of biology in defining r𝘦𝘪𝘳𝘳𝘢𝘥𝘪𝘢𝘵𝘪𝘰𝘯 limits. The session emphasised that dose-limiting toxicity must be evaluated beyond dosimetric parameters, by taking into account biological factors such as endpoint latency, severity, tissue recovery potential, and health-related quality of life. Early toxicity endpoints often indicate that tissues have high proliferative capacity and strong recovery potential. Late toxicity endpoints, in contrast, are shaped by radiation-induced atrophy, vascular injury, and structural damage, and are inherently dose-limited. Radiation pathogenesis must guide planning decisions; dose-limiting toxicities should be understood mechanistically so that planning objectives can be tailored accordingly. Importantly, the dose-volume-response relationship cannot be determined without detailed dosimetry and clinical outcome data, and this highlights the need for good documentation and data sharing.

Optimisation of clinical reirradiation workflows In a presentation from a high-volume centre, staff shared insights from over 3000 reirradiation cases that emphasised the physicist’s pivotal role in ensuring safe, reproducible treatment workflows. Examples of spine, liver, brain, and head-and-neck reirradiation illustrate challenges in image registration, anatomical changes, and the balancing of tumour control with protection of normal tissue. The team’s approach integrated precise cumulative dose estimation, radiobiological corrections and structured peer review, while data on evaluation times confirmed efficient clinical integration.

ESTRO physics reirradiation working group consensus Recognising the need for standardisation, this working group has developed 38 consensus statements that cover cumulative dose evaluation practices. The recommendations highlight the use of complete digital imaging and communications in medicine data, the use of rigid image registration (RIR) or deformable image registration (DIR) depending on anatomical changes, and mandatory biological dose rescaling (e.g., EQD2/biologically equivalent dose) before any dose summation. Special attention is given to adaptive radiotherapy, paediatric patients, and particle therapy. Practical guidance emphasises the balancing of clinical impact with resource demands, and the importance of clear communication of uncertainties. These guidelines offer a practical and structured foundation upon which to reduce inter-institutional variability and enhance patient safety.

Prospective evidence: the CURE lung trial The CURE trial in Denmark is intended to generate prospective, high-quality evidence via the enrolment of 500 patients over five years to evaluate high-dose thoracic reirradiation. Key features include mandatory access to previous 3D dose plans, use of deformable registration for dose accumulation, and adherence to strict national dose constraints based on prior trials and literature. A national radiotherapy quality assurance group developed a robust pre-trial programme that involved the testing of image registration, dose summation accuracy, and inter-centre variability in planning. Consensus was reached that OAR sparing should be prioritised over target coverage when necessary. The trial also incorporates comprehensive data collection, patient-reported outcomes (including decision regret), and long-term follow-up.

3. Economically sustainable radiotherapy
   Proffered Papers session
   Chairs: M. Carmen Rubio Rodríguez, Spain, and Timothy Hanna, Canada

Title: National study of dose accumulation for reirradiation: radiobiological dose scaling increases inter-centre variation

Presenter: Isak Wahlstedt, Denmark (presentation number: E25-1051)

In a compelling session, Isak Wellstead presented results from a national Danish study that had investigated the variability in dose accumulation for reirradiation across multiple radiotherapy centres. The study findings underscored that radiobiological dose scaling, rather than image registration technique, was the main driver of inter-centre variation, with significant implications for clinical decision-making.

Eight anonymised reirradiation cases that involved various anatomical sites (brain, head-and-neck, lung, liver, breast, rectum, anus, and lymphoma) were distributed to eight Danish radiotherapy centres. Each centre was tasked with registering previous dose distributions onto current planning CTs using either RIR or DIR. Among the DIR commercial methods used, five centres used Velocity, two used MIM, and one used RaySearch. Centres then performed dose accumulation in both physical dose and EQD2 scales, incorporating local α/β values and applying recovery correction factors if applicable.

The analysis focused on OAR doses in each case. For example, in the brain case, the optic chiasm was the most dose-limiting structure; in the head-and-neck case, the mandible was challenging due to partial surgical resection. Other dose-limiting OARs included the aorta, heart, chest wall, kidney, bowel bag, and bladder, depending on the anatomical site.

Key findings revealed that:

• The choice of image registration technique (RIR vs. DIR) had a limited impact on inter-centre variability;
• The application of EQD2 scaling, especially when combined with recovery corrections, significantly increased variation in near-maximum OAR doses; and
• In some cases, such as the rectum, dose variation increased from 1.2Gy (physical) to 19.2Gy (EQD2 + recovery) across centres.

This variability in radiobiological modelling led to notable differences in accumulated dose estimates, which may ultimately have affected clinical decisions, including treatment intent (curative vs. palliative) and dose prescription.

In the discussion, Dr Wellstead emphasised the need for national consensus and standardised approaches to dose scaling. He pointed to ongoing prospective efforts, such as the CURE lung trial, as essential platforms from which to develop shared protocols and reduce inter-centre discrepancies in reirradiation planning.

This presentation highlighted how methodological choices—especially in biological dose modelling—can influence the consistency and safety of reirradiation treatments.

 

 

 

 

 

Dr Gian Marco Petrianni
Operative Research Unit of Radiation Oncology
Fondazione Policlinico Universitario Campus Bio-Medico
Rome, Italy
Member of the ESTRO reirradiation focus group
ESTRO 2025 Social media ambassador for reirradiation
g.petrianni@policlinicocampus.it
LinkedIn: Gian Marco Petrianni
X: Gian Marco Petrianni

 

Dr Bartek Tomasik
Department of Oncology and Radiotherapy, Faculty of Medicine
Medical University of Gdańsk
Gdańsk, Poland
Member of the ESTRO reirradiation focus group
ESTRO 2025 Social media ambassador for reirradiation
bartlomiej.tomasik@gumed.edu.pl
LinkedIn: Bartłomiej Tomasik
X: https://x.com/B_Tomasik, @B_Tomasik