MICA: Early Phase Dose-finding Trials: Development of reporting guidance to improve knowledge transfer

Clinical trials are research studies involving patients or healthy volunteers which aim to test whether a new treatment is safe and 'better' than current treatment. Clinical trials are grouped into stages (or phases) and researchers aim to answer different questions at each of the phases. The earliest of these, Phase I, aims to test whether a treatment is safe, to investigate side effects and to and to recommend a dose of the treatment for further testing.

Phase I trials involve small numbers of participants who may be healthy volunteers or patients. The first group of participants in a phase I trial are given a low dose of treatment and if shown to be safe and without many side effects, the next group of participants are given a higher dose. Further groups of participants are enrolled; with the dose, being raised for each successive group until a dose is reached that has too many negative side effects. The decision whether to give patients the same, a higher, or lower dose of treatment is carefully made before the treatment is given. These trials are called "dose-escalation" or "dose-finding" trials.

Once a clinical trial is complete, the results are reported to the clinical research community and to the public. To make sure that these reports are reliable and helpful for further research, a set of guidelines of the important items to be reported, Consolidated Standards of Reporting Trials (or CONSORT for short) have been published. It sets out a standard way for authors to report how the trial is designed, analysed and interpreted and has been instrumental in promoting transparent reporting. The original CONSORT guidelines were developed for specific types of trials and their design features often differ from dose-finding trials. This is important as currently trial reports of Phase I trials often miss out important information about how they were designed and conducted, which can make it difficult for the reader to interpret and gauge the validity of the trials. This wastes time and resources, but more importantly, may unethically expose participants to ineffective or even harmful interventions.

Making the best decisions in a Phase I trial, on whether to give a patient the same dose, or a higher or lower dose, is key to the success of the trial. The statistical methods used to guide these decisions have advanced significantly over the past 25 years, so that safer trials with more reliable results can be conducted. But these methods are often complex, and have additional transparency and reporting demands.

To address these problems, we will develop an 'extension' to the CONSORT guidelines, which is relevant to all early phase dose-finding designs in different diseases. The main CONSORT Statement has now been extended in 7 other design areas, primarily in later phase trials (www.consort-statement.org/extensions). A CONSORT extension for dose-finding trials is long overdue. To develop an internationally agreed way to report such trials, we have brought together a multi-disciplinary, international team of experts in the design, running and reporting of early phase trials who work in academic institutions or pharmaceutical industry, and those with expertise in developing reporting guidelines of trials.

Finally, to ensure that the developed CONSORT extension will be widely adopted, we will involve key groups throughout this research: the trial community, journal editors (who publish articles on dose-finding trials), peer reviewers (who assess research papers), regulators, patients and the public. We will publicise our outputs at relevant meetings and to local/national Patient and Pubic Involvement (PPI) groups, and will use social media to raise awareness. We will conduct practical workshops, highlighting common reporting flaws and how to use the new guideline. We will co-produce two lay papers with our PPI representative to effectively involve, engage and inform patients and the public about the importance of this work.

Technical Summary

Results from early phase dose-finding trials (Phase I or Phase I/II) directly influence decisions on further drug development and ultimately whether the selected dose(s) can be confirmed as safe and efficacious. Poorly reported dose-finding trials with inadequate rigour may lead to bias in reporting and lack of reproducibility. This wastes time and resources, but more importantly, may unethically expose participants to ineffective or even harmful interventions.

The main CONSORT Statement provides recommendations for reporting parallel group randomised trials, where participants are randomly assigned to different interventions. They do not address many methodological features such as interim dose-decisions (escalate, de-escalate, stay or stop) that arise in the design, conduct, analysis and interpretation of early phase dose-finding trials. The quality of reporting of such published trials is often inadequate.

To address this unmet need and reduce research waste, we will

1. Extend the CONSORT reporting guidance for dose-finding trials to improve reporting, using gold-standard methods for developing consensus-based guidelines recommended by the CONSORT Group (Moher et al 2010)

2. Actively involve, engage and disseminate outputs to stakeholders, including trialists (clinicians, statisticians, trial managers), funders, regulators, patients and public, journal editors and peer reviewers. Besides scientific publications, we will also produce lay summary papers for the public and patient community.

Motivated by the desire to capture diverse views of experts across sectors with multidisciplinary roles, we have strategically brought together a team of national and international statistical methodologists and trialists (clinicians with drug discovery expertise and statisticians) in early phase trials in both academia and pharmaceutical industry, a CONSORT Group representative, a patient and public representative (PPI), with oversight from an Independent Expert Panel.