PRIMER: Development of Daily Online Magnetic Resonance Imaging for Magnetic Resonance Image Guided Radiotherapy

Overview

In radiotherapy high-tech scans with x-rays (CT scans) are taken before and during treatment to locate the tumour and ensure the radiation is hitting the target. These x-rays expose patients to additional radiation and the quality of these scans is often poor which makes it difficult to distinguish tumour from normal tissue and there may be uncertainty in the tumour position due to movement or shrinkage. To allow for these uncertainties a large margin around the tumour is also treated, but this means that large volumes of normal tissue also receive significant doses of radiation, which can result in early and late toxicity. MRI (magnetic resonance imaging) is better than CT scanning at being able to tell the difference between tumour and normal tissues and does not expose patients to additional radiation. A new machine called an MR Linac (or magnetic resonance imaging-guided linear accelerator) integrates high quality MRI with a state-of-the-art radiotherapy machine and the Institute of Cancer Research (ICR)/The Royal Marsden Hospital (RMH) are currently installation a prototype, which will be one of the first in the world. This revolutionary technology has the potential to change the way radiotherapy is delivered. We hope the improved precision and accuracy in hitting the target will mean reductions in margins around tumours and that this will lead to higher cure rates with significantly fewer side effects. Studies are required to simulate treatment on the MR Linac before it can be used in routine clinical practice and to conduct these studies, we need to obtain MRI scans on volunteers and patients who are currently undergoing treatment. This study will involve imaging with MRI in healthy volunteers as well as in patient volunteers before and during their standard course of radiotherapy to allow us to develop MRI sequences derived on the MR Linac for MR Linac-based research focusing on clinical application and establishment into a MR-CT and MR only workflow, treatment adaptation and quality assurance.

This is a study protocol for the acquisition of MR Linac derived images with a view to MRIgRT for patients undergoing radiotherapy with radical intent at sites predicted to benefit most from MRIgRT (namely, for rectum, cervix, lung, prostate, bladder, breast, brain, oesophagus, head and neck, hepato-pancreato-biliary (HPB) and paediatric abdomino-thoracic cancers). These will be used to build a database of MR images for MR Linac-based research focusing on clinical application and establishment into a MR-CT and MR only workflow, treatment adaptation and quality assurance. The primary aims are to 1) develop MR Linac image sequences suitable for seeing tumour/target and normal tissue at the time of radiotherapy and 2) determine variations in image registrations and tissue contouring using the MR Linac images. Four stages of imaging development sub-studies will be included in this protocol for optimisation of MR imaging on the MR Linac. Consideration will be given to image quality, scan times, and IT requirements for reconstruction, storage and image transfer. These imaging sub-studies will be: A) Non-patient Volunteer Imaging Studies of Normal Tissue (n = 18 - 54 per centre). The four anatomical regions selected to be studied are Head and neck, Thorax (chest wall/breast and lung/oesophagus), Abdomen and Pelvis (male and female). Recruitment will continue until imaging from a minimum of three consecutive patients using the same 'exam card' are deemed to be acceptable for normal tissue visualisation for each anatomical site. B) Patient Volunteer Imaging Studies of Normal Tissue (n = 39 - 72 per centre). We will further develop the protocols to visualise site specific tumours in 13 target areas (brain, oropharynx/larynx/hypopharynx, oesophagus, lung, chest wall \_+/- regional lymph nodes), paediatric abdomen-thoracic cancers, hepatobiliary tract, bladder, gynaecology, prostate, rectum, oligomets bone and oliogmets soft tissue). Patients will be recruited for imaging for a minimum of one and maximum of 12 imaging sessions whilst undergoing radiotherapy planning and/or radiotherapy treatment. Repeat imaging will allow for the determination of protocol robustness. The aim will be to produce images appropriate for tumour and normal tissue visualisation with a view to directing radiotherapy. Recruitment will continue until imaging from a minimum of three consecutive patients using the same 'exam card' are deemed to be acceptable for tumour/target visualisation at each site. C) Pathway development studies (n = 39 - 208 per centre). At each centre, it is estimated that 16 imaging sessions per tumour site are required for inter and intra registration observer variability. To establish library of images to be used to develop adaptive radiotherapy workflows, up to 50 imaging sessions may be required depending on the tumour site. Images for normal tissue and tumour visualisation will also be used to refine exam cards from Stage B. This stage requires a minimum of three patient volunteers to be recruited to each tumour site which may (or may not) include images acquired in the previous Stage B of the study, and a maximum of 16 patient volunteers to reach the desired number of imaging sessions. Each volunteer should undergo a minimum of two and not more than twelve imaging sessions (max 1/day, 3/week). The methodology of image acquisition will follow methodology for Stage B as above. The primary aim will be to establish that both online and offline image-matching using MR Linac with a view to guiding radiotherapy treatments can be undertaken with minimal intra and inter user variability. Development of treatment workflow to help implement radiotherapy treatment on the MR Linac can also be carried out. Refined exam cards from Stage B but fine tuning as necessary can be used to acquire sets of images and to use this information to plan treatment by simulating an on-line treatment whilst not delivering radiation. D) On-going image development and continuous quality improvement of images This stage of the study can run in parallel, subsequent to or independently of stages B and C. The purpose will be to recruit patients receiving radiotherapy, non-radiotherapy cancer patients or non-patient volunteers in order to optimise MRI guided radiotherapy delivery. Examples of this include, investigating suitability of radiotherapy immobilisation devices for patient positioning and treatment set-up development and/or development of new/novel imaging sequences, optimisation of existing sequences and undertaking continuing image quality improvement for the tumour sites listed in this protocol. Adult patient with cancers not conventionally treated using radiotherapy could be recruited in order to investigate whether radiotherapy could be used as a treatment modality in the future. Patients with unusual or rare cancer (for example sarcoma, kidney and lymphoma) that do receive radiotherapy could be recruited to investigate the benefit of using the MR Linac for cancers of this type. The total number of volunteers is, as yet, unknown; however the recruitment progress will be monitored by the trial management group (TMG). The TMG will ensure that this protocol is being adhered to, that all volunteers are being imaged safely and that the information gained from the imaging sessions is being used appropriately. The investigator may recruit between one and five volunteers (patient, non- radiotherapy patient or non patient). On going evaluation of technique MRI sequence and/or equipment etc should be carried out to determine number of required volunteers with a maximum of 5 repeated imaging sessions per volunteer. The TMG can restrict the maximum number of volunteers and repeated imaging sessions, if appropriate. A maximum of 150 volunteers will be recruited to this section of the study