ESTRO 2023 - Abstract Book

S139

Saturday 13 May

ESTRO 2023

cost benefits are anticipated the level of evidence required may be less but are radiobiological modelling or dosimetric planning studies sufficient? Where adaption is aimed at reducing toxicity we need to consider how best to measure this – through clinician reporting of radiation side effects or patient reporting of the impact of any side effects on their quality of life; and whether there is a need to formally demonstrate non-inferiority of disease control in a randomised controlled trial. Where radiation is combined with systemic therapy or radiosensitising agents to enhance efficacy, care must be given to assessing not just radiation related side effects but the impact of the combination therapy. Methodological evaluation of response or predictive biomarkers to guide treatment requires biomarker-stratified or biomarker-strategy designs. This talk will expand on some of the methodological issues related to evaluating personalised radiotherapy. [It is likely to raise more questions than answers]. SP-0192 Adaptive radiotherapy in daily practice - What evidence do we still need to be able to implement? P. Balermpas 1 1 Zurich University Hospital, Department of Radiation Oncology, Zurich, Switzerland Abstract Text Introduction Adaptive Radiotherapy (ART) is already widely used in most large centers in daily clinical routine. However, high-quality data on this topic are scarce and only very few prospective trials have been conducted so far. Methods The literature and the available evidence have been carefully reviewed, presented and discussed here Results The broad implementation of image-guided radiotherapy (IGRT) over the last decades led to the advent of adaptive treatments for several indications, with head and neck cancer, prostate cancer and tumors of the upper abdomen being the most prominent examples. Especially recent developments, both in terms of improved imaging and software, allow for more frequent, quicker and even online plan adaptations. For example, hybrid platforms combining a linear accelerator with MRI-imaging, revolutionized "on-board" imaging with improved quality and soft-tissue contrast and machine-learning methods simplified delineation, re-planning and plan quality assurance. Nevertheless, even a generally accepted definition of the term "adaptive radiotherapy" is lacking ,as well as established guidelines of the methods to follow and purposes to pursue. So, the term "adaptive radiotherapy" summarizes very different approaches like either preservation of the original plan despite anatomic changes or maximizing organ sparing, up to more experimental regimens such as spatio-temporal fractionation and different dose-escalation or de-escalation strategies based on response assessment. Every one of these approaches requires different methods and until the procedures and the reporting has been standardized both conducting clinical trials and reproducing clinical results remains a great challenge. Moreover, only very recently investigators started conducting prospective trials evaluating possible benefits of ART. Up to this day, only few results have been published mainly in the indications of head and neck, prostate cancer and upper abdominal tumors, most of them showing clear dosimetric benefits. However, it remains largely unknown if these dosimetric benefits can be translated to improvement in any clinically relevant endpoint. A recent randomized phase iii trial in prostate cancer (MIRAGE) could demonstrate decreased toxicity of MR-guided adaptive stereotactic radiotherapy compared to standard CT-guided approach. Yet, as different platforms and imaging and different target volume margins were used in the two arms, it remains questionable if the differences observed can be attributed to the ART-approach. In contrast to that, the only randomized trial in head and neck cancer examining the possible benefits of ART in reducing xerostomia (ARTIX) failed to meet its primary endpoint. Finally, some promising results from biologically and/or response-based ART for head and neck cancer originate from small phase I-II trials with low patient numbers and short follow up. Conclusion Although different forms of adaptive radiotherapy are already implemented in most large radiotherapy centers, there are many questions to be answered before this modality becomes standard of care. First of all, there is an urgent and unmet need for international guidelines on performing and reporting ART, depending on the techniques used and the goals pursued. This will allow for more and high-quality prospective randomized trials demonstrating not only obvious dosimetric benefits, but also improvement in clinical endpoints. Abstract Text The potential of Artificial Intelligence (AI) in radiation oncology seems tremendous with a vast array of applications, ranging from refinement in diagnostic and disease delineation and staging to optimisation of the clinical workflow, routine integration of multi-parametric and multi-dimensional imaging, treatment response prediction, treatment guidance etc. Several types of gains are expected out of these: shorten patients’ workflows while improving their prognosis, integrate in the routine high- precision adaptive radiation therapy approaches, reduce treatment-associated costs, etc. Importantly, AI applied to cancer care should ultimately lead to treatment harmonization, giving rise to procedures that are less operator-dependent and consequently, progressively, to the reduction of care disparities across centers. However, such promises cannot be taken for granted. The conventional development path requires methodologically-suited quantitative and qualitative evaluations, including important validation steps in independent cohorts, that are key for envisioning further large-scale deployment. For now, AI-based tools suffer from a lack of adhesion of the community overall, intrinsically linked to an insufficient diffusion of novel AI technologies. Confusion remains between Technology Readiness Level (TRL), the actual availability of a tool and its clinical validation. As such, every novel AI-based techniques can, and should be evaluated compared to the techniques of reference. This evaluation will be of key importance for several aspects: convincing care providers and institutions of the value of a given technology with explainable metrics (improved patients’ quality of life and clinical outcomes, decreased need of dedicated resources, reduced time needed per patients, improved accuracy for the definition of patient staging and SP-0193 How to integrate AI/IO in the clinical research setting E. Deutsch 1 1 Gustave Roussy, Radiation Oncology, Paris, France