Evidence-based decision making in image-guided brachytherapy

Motivation The treatment of locally advanced cervical carcinoma has undergone tremendous improvement through the implementation of 3D-image guided brachytherapy during the last decade, which has led to improved local control while reducing treatment-related morbidity. Complex application techniques in combination with magnetic-resonance-imaging-guided treatment planning nowadays allow the treatment of a variety of patients with a wide range of disease-stages. In order to make the knowledge on advantages of certain treatment strategies gained by specialised centers applicable for a maximum number of patients, the Evi-Guide project aims at developing an evidence-based decision support system for image-guided brachytherapy. Additional software development within this project will lead to improved individual online dose-reporting and dose-response modelling for pelvic organs at risk, which are essential components of personalized evidence-based treatment planning. Methods Based on the findings of the multicentric EMBRACE study for image-guided brachytherapy with over 1400 patients from specialized centers, which has been led by our institution, software tools for the evaluation of different existing treatment concepts will be developed by use of combinations of dosimetric and clinical outcome parameters. For this purpose, fundamental clinical parameters, tumour dimensions, dose-volume parameters, application specific parameters and the latest dose- response models based on the EMBRACE study will be used. This will allow the user to derive information about the expected dosimetric and clinical outcome, in terms of probabilities for tumour control expected side effects for urinary bladder, rectum and bowel, of a selected treatment strategy, based on corresponding observations from a large patient cohort. The underlying dose-response models will be improved within the Evi-Guide project by development of software tools for i) optimization of radiobiological standard model parameters for the linear-quadratic model for pelvic radiotherapy, based on the EMBRACE cohort, and ii) reducing uncertainties of the reported dose by implementation of a systematic magnetic-resonance-imaging- based online dose verification procedure. The results of Evi-Guide could be applied worldwide by numerous centers for evidence-based optimization of image guided brachytherapy strategies. The software modules created within this project could serve as a model for decision-support systems for other indications and therapy options beyond brachytherapy. Future inclusion of knowledge emerging from the prospective EMBRACE II study is also considered. The increase of precision in dose reporting and optimization of radiobiological models will lead to an improved understanding of normal tissue complication probabilities, which will also be applicable to other applications in clinical radiation oncology.

Evidence-based decision making in image-guided brachytherapy - "Evi-Guide" The treatment of locally advanced cervical carcinoma has undergone tremendous improvement through the implementation of 3D-image guided brachytherapy during the last decade, which has led to improved local control while reducing treatment-related morbidity. Complex application techniques in combination with magnetic-resonance-imaging-guided treatment planning nowadays allow the treatment of a variety of patients with a wide range of disease-stages. In order to make the knowledge on advantages of certain treatment strategies gained by specialised centers applicable for a maximum number of patients, the "Evi-Guide" project aimed at developing an evidence-based decision support system for image-guided brachytherapy. Additional software development within this project lead to improved individual online dose-reporting and dose-response modelling for pelvic organs at risk, which are essential components of personalized evidence-based treatment planning. Based on the findings of the multicentric EMBRACE study for image-guided brachytherapy with over 1400 patients from specialized centers, which has been led by our institution, software tools for the evaluation of different existing treatment concepts were developed by use of combinations of dosimetric and clinical outcome parameters. For this purpose, fundamental clinical parameters, tumour dimensions, dose-volume parameters, application specific parameters and the latest dose-response models based on the EMBRACE study were used. This will allow users to derive information about the expected dosimetric and clinical outcome, in terms of probabilities for tumour control expected side effects for urinary bladder, rectum and bowel, of a selected treatment strategy, based on corresponding observations from a large patient cohort. The potential to improve underlying dose-response models was investigated within the "Evi-Guide" project by development of software tools for i) optimization of radiobiological standard model parameters for the linear-quadratic model for pelvic radiotherapy, based on the EMBRACE cohort, and ii) reducing uncertainties of the reported dose by implementation of a systematic magnetic-resonance-imaging-based online dose verification procedure. The results of "Evi-Guide" could be applied worldwide by numerous centers for evidence-based optimization of image guided brachytherapy strategies. The software modules created within this project could serve as a model for decision-support systems for other indications and therapy options beyond brachytherapy. Future inclusion of knowledge emerging from the prospective EMBRACE II study is also possible. The increase of precision in dose reporting and optimization of radiobiological models will lead to an improved understanding of normal tissue complication probabilities, which will also be applicable to other applications in clinical radiation oncology.