REFINE-Lung: A randomised open-label phase III trial of REduced Frequency pembrolizumab ImmuNothErapy for first-line treatment of patients with advanced non-small cell lung cancer (NSCLC) utilising a novel multi-arm frequency-response optimisation design.

RESEARCH QUESTION: Can pembrolizumab (pembro) dose frequency be reduced without compromising efficacy in advanced non-small cell lung cancer (NSCLC)? BACKGROUND: The anti-PD1 directed monoclonal antibody pembro is now given to ~3600 patients/year with advanced NSCLC in England alone. Treatment continues 6 weekly (wkly) for 2 years (yrs) or until disease progression/unmanageable toxicity. Up to 25% of patients remain on therapy at 2 yrs. However, pharmacodynamics and clinical studies suggest 6 wkly therapy results in overtreatment. Indeed, pembro remains bound to its target on T cells 100 days after a single dose [12], multiple trials have found no relationship between immunotherapy dose and efficacy [13–16] and long lasting anti-tumour and autoimmune effects are observed long after discontinuation [8,9,17,18]. Six wkly administration may therefore be unnecessarily costly, toxic and negatively impact patient quality of life. AIMS AND OBJECTIVES: Our primary goal is to determine the longest dose frequency of pembro with non-inferior survival compared to control 6 wkly therapy. Our secondary goals are to determine the cost, toxicity and quality of life benefits of reduced frequency therapy. METHODS: We propose a multicentre, randomised, phase III trial. We will recruit patients from 27-35 UK centres, with advanced NSCLC receiving first line pembro with or without chemotherapy, who are progression free at 6 months and planning to continue treatment. Using a novel DURATIONS design developed for this trial [21], patients will be recruited to control 6 wkly and experimental 9, 12, 15 and 18 weekly pembro arms. The frequency-response relationship will be modelled to determine the longest dose frequency that is non-inferior to control. Patients progressing on an experimental arm will be allowed to re-escalate to 6 wkly therapy. To assess the safety of this intervention, we will initially randomise patients 1:1 to 6 wkly (control) and 12 wkly arms. Once 150 patients are enrolled, an interim safety analysis will be performed. If the 12 wkly arm is not significantly less effective in terms of PFS, the remaining 9, 15 and 18 wkly reduced frequency arms will be opened. Simulation based sample size calculations indicate a requirement for 1750 patients (equally distributed across 5 arms) to identify whether the 18 weekly arm is non-inferior to control with 80% power at an alpha of 5%. Our primary outcome measure is landmark survival at 2 yrs post-initiation of pembro. Key secondary outcomes include time-to-event overall survival, progression free survival, quality of life (QoL) and cost savings. Collection of toxicity and quality of life data (using cancer EORTC QLQ C30, lung cancer EORTC QLQ LC13 and EQ-5D-5L questionnaires translated into QALYs) will enable a refined health-economic evaluation. TIMELINES: Total duration from setup to closure and publication will be 78 months. IMPACT AND DISSEMINATION: Our findings will potentially alter the landscape of NSCLC therapy, translate into significant cost savings for the NHS and health care systems worldwide and positively impact patient QoL. Our unique trial design will serve as a model for future immunotherapy studies a REFINE basket of trials across indications is in setup. Dissemination will be through a range of media including high impact peer-reviewed publication, national and international conference presentation, news media, NHS and departmental social media accounts.