Enhancing the design, conduct and analysis of Adaptive and Platform Trials through consensus-driven Statistical Analysis Plan guidance (APT-SAP)

We need efficient, innovative and robust trial designs to swiftly evaluate new treatments for safe and effective ones to reach patients quickly. Adaptive and platform trials (APTs) are such innovative approaches that have proved their efficiency and are being used more frequently. Adaptive designs allow pre-specified changes (adaptations) to aspects of the trial to be made by analysing interim data while making validity conclusions. Such adaptations include early stopping of treatment arm(s) as soon as sufficient evidence is gathered or allocating more patients to treatments showing greater benefits. In platform trials, new treatment arms can be added to an ongoing trial that is planned with or without interim analyses. Greater flexibility and statistical efficiency come at the cost of additional operational and statistical complexities. Adding new arms and/or using interim analyses to make adaptations complicates the design, conduct, analyses, and reporting. Adaptations should be accounted for in the design and all analyses – otherwise, misleading results can be produced. Importantly, research consumers need to be confident that researchers are not changing goalposts to suit them during the trial. Timely access to high-quality statistical analysis plans (SAPs), accompanied by protocols is vital to enhance public understanding and interpretation of findings. Existing SAP guidance does not adequately cover considerations for APTs. Statistical methods in some accessible APT-SAPs are insufficiently detailed, making it difficult to interpret results, reproduce methods and results, and raising suspicions or doubts about the trustworthiness of results. This undermines public confidence in research leading to research waste. In addition, there is no guidance on proportionate trial data/information/results that can be released from an ongoing APT while preserving trial integrity. Therefore, we aim to develop consensus-driven SAP guidance for APTs that can be used by researchers when conducting interim and final analyses. Specific objectives are to: develop a consensus-driven minimum set of essential items that should be included in the SAP through a Delphi process; provide practical useful examples to maximise immediate use of the guidance, with the help of an annotated APT-SAP template, to speed up the production of a high-quality SAP. Those examples will be based on specific improvements to SAPs across disease areas. The annotated SAP template will provide guidance notes on things to consider in different sections of an APT-SAP; understand current practice and provide recommendations for good practice to researchers on release of trial information/results/data from an ongoing APT while maintaining trial integrity; engage international multidisciplinary stakeholders in both the public and private sectors throughout the process and disseminate the research outputs to maximise the impact of the research.