Melanoma patients who have responded well to immunotherapy for a year will be randomly assigned to stop treatment or continue indefinitely, in a trial testing whether shorter therapy is just as effective. This matters because current practice is to keep patients on anti-PD1 drugs until their cancer progresses, but this can mean years of treatment with significant side effects and high costs. No large trial has yet established whether stopping at 12 months is safe for patients whose disease is stable or shrinking. If the trial shows that stopping at 12 months is non-inferior to continuing until progression, thousands of patients could avoid unnecessary drug toxicity and hospital visits. The NHS would also save substantial drug costs without compromising survival outcomes. The trial will also measure quality of life, toxicity, and cost-effectiveness, giving clinicians and funders concrete data to guide treatment decisions. The study will recruit 1,208 patients across 45 NHS cancer centres, with five years of follow-up. A multi-stage design allows early stopping if the shorter duration proves clearly inferior or if recruitment is not feasible.
View original technical description
Design A multicentre, randomised, controlled, multi-stage, non-inferiority trial comparing 12 months (m) of anti-PD1 therapy with standard duration until progression. The recruitment period is 5 years (y); each patient will be followed up for 4y. To ensure that both feasibility of the intervention & any lack of efficacy are identified early in the trial, we have incorporated 3 intermediate stages in the trial design with stop/go rules depending on randomisation & recruitment rate (stage 1 & 2) & an intermediate assessment of efficacy (stage 3). Providing these conditions are met, the trial will continue seamlessly enabling the stage 4 primary & secondary objectives to be attained in a timely manner. The stage 4 primary analysis will occur when all patients have completed 12 months of follow-up after randomisation. A long-term analysis (stage 5) will be conducted when all patients have completed 4y of follow-up. Randomisation (1:1) will be by minimisation with a random element incorporating BRAF status, prior BRAF/MEK inhibitor therapy for advanced disease, prior adjuvant immunotherapy, disease stage, presence of brain metastases, performance status, centre, treatment (pembrolizumab or nivolumab) & response after 1st 12m therapy (complete/partial response or stable disease). Patients will be followed for response & progression with standard 12 weekly imaging for 12m post-randomisation then 6-monthly until 4y post-randomisation (2 & 5y after start of anti-PD1 therapy). Setting 45 NHS cancer centres with expertise in treatment of advanced melanoma Target population Patients (18y or older) with advanced (unresectable or metastatic) melanoma who have commenced first line anti-PD1 therapy who are progression-free after 12m of treatment HTA technologies being assessed Anti-PD1 therapy for the standard duration of until progression will be compared against shorter duration of 12m Outcomes Primary PFS Secondary QoL (by 3-monthly EORTC QLQ-C30, QLQ-MEL38 & EQ5D until 18m post-randomisation) Response (by RECIST) OS Drug-induced toxicity Cost effectiveness (within trial & longer term modelling) Sample size Estimating the 2-year PFS rate in the control arm to be 51% for patients alive and progression-free at 12m & defining non-inferiority as a reduction in 2-year PFS of no more than 7.9% (a hazard ratio of 1.25), with 80% power and a one-sided significance level of 5%, 1,014 patients are required to test for this degree of non-inferiority using a one-sided log-rank test, assuming patients are followed for a fixed length of time (i.e. 12 months following randomisation) and that the hazard ratio is constant. To account for a 5% drop-out rate, 1,068 patients (534 per arm) will be required in total. SAMPLE SIZE REVISION AUGUST 2017 (Due to emergence of updated, longer-term follow-up data from the CheckMate-067 trial, which was used to inform the underlying assumptions of the DANTE sample size calculation) The PFS rate at 12 months post-randomisation (i.e. 2 years after the start of anti-PD1 therapy – ‘2-year PFS’) for patients in the control arm has been re-assessed; this is now estimated to be 86% rather than 51%. In addition the non-inferiority margin has been re-assessed. This was previously defined as a reduction in 2-year PFS of no more than 7.9%; this has now been reduced to 6%. As a result of these changes, the sample size is now 1208 randomised patients in total. Project timetables 1-12 months Project set-up 13-72 months Recruitment 73-120 months Follow-up, including primary analysis and write-up 121-124 months Long-term follow-up analysis & write-up Expertise in team Medical Oncology co-investigators who have conducted many clinical trials in patients with melanoma & have expertise in immunology & QoL. Clinical Trials Unit co-investigators for trial management, health economics, qualitative research & statistical expertise with extensive experience in cancer clinical trials.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know