Developing a prospective rapid-learning methodology to evaluate the survival impact of changing radiotherapy practice to include a new heart dose limit for patients with lung cancer in a UK specialist cancer centre (RAPID-RT): a protocol.

introductionThe RAPID-RT study is part of a large-scale research programme investigating the use of routinely collected real-world patient data to rapidly and prospectively evaluate and optimise the impact of changes in radiotherapy practice on clinical outcome, an approach often referred to as 'rapid learning'. As a proof of concept, a prospective, observational clinical study using realworld data is embedded within the programme. This study implements a new dose limit to a defined region of the heart in patients with stage I-III lung cancer treated with curative-intent radiotherapy at The Christie NHS Foundation Trust. The RAPID-RT study includes both methodological and clinical objectives. Its primary aim is to assess the feasibility and clinical acceptability of using rapid learning with real-world data to evaluate outcomes following modifications to standard-of-care radiotherapy protocols. This work has the potential to establish rapid learning as a robust, evidence-based approach for the continuous optimisation of radiotherapy workflows. METHODS AND ANALYSIS: RAPID-RT is a series of prospective single-arm observational studies with historic controls that uses only real-world data. A clinical decision was made to implement a dose limit to a specific region of the heart in all patients with stage I-III non-small cell lung cancer treated with curative-intent radiotherapy. The research focuses on using real-world data, the information collected as a part of patients' routine care, to evaluate the impact of this change on overall survival and treatment-related toxicities. The study employs broad inclusion criteria, and data are extracted directly from the electronic patient record. Patients are provided with clear patient information materials and consent for data use via an informed opt-out process. Outcomes for patients treated before and after the introduction of the dose limit are compared using a Bayesian analytical framework to allow sequential updating of results as patients are recruited to the study. Evidence of clinical impact will guide the clinical team in determining whether refinements to the heart dose limit are necessary. These changes will, in turn, be evaluated in subsequent rapid-learning cycles. The RAPID-RT study aims to complete at least two iterative learning cycles to support the continuous optimisation of radiotherapy protocols. ETHICS AND DISSEMINATION: The study has received ethical approval (REC reference 22/NW/0390) from the North West Haydock Research Ethics Committee, is sponsored by The Christie NHS Foundation Trust and is funded by the UK National Institute for Health and Care Research. The programme management group is supported by an independent programme steering committee, an independent statistical review panel, a clinical management team and patient advisory group. Findings from the RAPID-RT study will be shared widely through conferences, focus groups and a stakeholder event, including a public 'People's Forum' to co-create guiding principles for trusted rapid learning in radiotherapy. In parallel, interviews with participants, professionals and regulators will inform consensus and the development of practical, ethical and legal guidelines to support the adoption of rapid learning across NHS radiotherapy centres. TRIAL REGISTERATION NUMBER: ISRCTN17129364.