Stereotactic body radiotherapy (SBRT) is the preferred treatment for inoperable early-stage non-small cell lung cancer (NSCLC) and for operable cases when patients decline surgery, offering local control (LC) rates of 80-97%. SBRT is also used in oligometastatic cancer. However, its safety and efficacy for ultra-central lung tumours, near the proximal bronchial tree (PBT) and/or esophagus, remain debated. SBRT targeting these tumours has been associated with higher risk of severe toxicity, including pulmonary hemorrhage and pneumonitis. The term "ultra-central lung tumours" refers to tumours abutting the PBT, trachea, or esophagus.

MR-guided online adaptive radiotherapy has been implemented recently using MR-linac technology, where MR-imaging is combined with linac technology in hybrid systems. These systems offer MR imaging for online adaptive radiotherapy within the treatment room and with the patient in treatment position. Despite results regarding technical and clinical feasibility being promising, the high costs of MR-linac systems and the demands for staffing resources of MR-linacs are limiting a broader implementation of this technology. The MAGELLAN trial is currently investigating MR-linac based SBRT in ultracentral lung tumours to reduce severe toxicity, with outcomes pending.

At the Department of Radiation Oncology, University Hospital Zurich, MRI-simulations for RT planning using a dedicated scanner have been performed since 2023 for different tumour entities. We have developed and implemented an MRI-based RT online adaptive re-planning workflow followed by RT delivered at a CBCT-guided, conventional linac. We now aim to expand this MR-informed workflow to online adaptive RT for ultracentral lung tumours. With this study, we will investigate the feasibility of MR-informed, online adaptive SBRT for ultracentral lung tumours using a dedicated MR simulator and treatment delivery using conventional CBCT-linac technology.

Patients will be registered, screened, and checked for their eligibility. Upon signing the written informed consent patients recognised eligible will be enrolled in the study. Patients in treatment position will undergo MRI-simulation using two dedicated MR-scanners: A 0.55 T machine that will also be used for daily adaptations during treatment, and a 1.5 T machine for the standard-of-care planning simulation. Additionally, a standard-of-care planning CT with contrast-enhanced 3DCT and native 4DCT will be performed. Target and OARs will be identified and delineated on the MRI scan, while planning CT-sequences will be co-registered to gain additional information on (breathing) motion. An SBRT treatment plan will be generated. For GTVs with movement > 5 mm in either direction over the breathing phases, the treatment will be carried out in expiration-breath-hold-technique. On the 1st day of treatment, a new MRI scan in treatment position will be acquired on the dedicated MR simulator, and co-registered with the original one. The original target and OAR contours will be propagated to the new image set using rigid and deformable image registration. The contours will be adapted by a radiation oncologist to changes detected within the MRI of the day. The original SBRT treatment plan will be adapted by a medical physicist to the changed and adapted anatomy of targets and OARs by plan re-optimisation. Upon the plan approval by the treating radiation oncologist, patient-specific quality assurance (PSQA) will be performed. Should the target volumes and OARs prove unchanged compared to the simulation, the radiation oncologist may omit contour adaptation and the workflow may proceed in a shortened fashion; the original plan will then be used for dose re-calculation on the images of the day. Patients will be transferred from the MR scanner with a dedicated couch-to-couch-shuttle-system and will be set up on the treatment table of the linac. Their position will be verified by matching the adapted contours of the day to the linac CBCT. For highly mobile tumours that are irradiated in breath-hold the breath-hold position will be verified with repetitive CBCTs as is standard of care. It will be followed by treatment delivery. The procedure will be repeated before delivery of each SBRT fraction. On the first two days an additional MR scan will be performed directly before the CBCT imaging to verify the anatomy of the day. This will be done by matching to the tumour and visual checking of the contours. In case of motion this will be repeated the following days. During simulation and once weekly during treatment an additional (second) MR-scan, as well as additional functional imaging will be acquired on the dedicated MR scanners. Patients with different fractionation schemes can be included in the study. The standard prescription is 50 Gy in 10 fractions (4-5 fractions/week) @80% IDL to the planning target volume (PTV) for primary tumours of the lung, or 45 - 49.5 Gy in 10 - 11 fractions (4-5 fractions/week) @80% IDL to the planning target volume (PTV) for secondary tumours/metastases of the lung. However, other SBRT fractionations can be included as well.

Inclusion Criteria:

• Written informed consent according to Swiss law and ICH/GCP regulations signed and dated by the participant and the investigator before any trial specific procedures (Informed Consent Form);
• Ultracentral (PTV-overlap with bronchial tree, trachea/carina and/or esophagus) primary or secondary lung lesion(s) from a histologically confirmed malignancy, confirmed by imaging and amenable for SBRT;
• Indication for SBRT based on a multidisciplinary tumour board recommendation and / or clinical practice guideline;
• Age≥18 years old;
• Patients who are willing and able to comply with scheduled visits, treatment, and other trial procedures.

Exclusion Criteria:

• Large body size that would not fit the MRI-simulator bore;
• Contraindications for MRI;
• Previous radiation therapy directly overlapping with planned SBRT (type I re-irradiation);
• Women who are pregnant or breast feeding;
• Intention to become pregnant during the trial;
• Lack of safe contraception (Female participants who underwent hysterectomy and/or bilateral oophorectomy or post-menopausal for longer than 2 years are not considered as being of child bearing potential);
• Known or suspected non-compliance, drug or alcohol abuse;
• Inability to follow the procedures of the trial, e.g. due to language problems of the participant;
• Enrolment of the investigator, his/her family members, employees and other dependent persons.