Bringing Further Certainty into Platform Trials in Europe
New guidance to drill down on more specifics in hopes of harnessing the growth of these types of studies.
The European Medicines Agency (EMA) is continuing to forge ahead with its aim of improving the environment for clinical trials in Europe, and is now turning its attention to platform trials. It intends over the next year or so to deliver further guidance on their planning and conduct, particularly so that they can be used as pivotal trials. A reflection paper will cover some of the methodological issues these trials raise to bring greater consistency in scientific advice and regulatory assessment. Clarifying terminology and design features will help to allow reliable estimates of effect and complement existing guidance on issues such as multiplicity and adaptive design, it says.
The move is an explicit recognition of the continual growth in the use of platform trials, particularly for scientific advice (even if more rarely so far as part of marketing authorization applications). The new guidance will zero in on specifics, such as the introduction of new arms when the study is ongoing, and what sort of framework and constraints can generate reliable results. It will cover type I error rate or uncertainties such as the use of non-concurrently randomized controls and changes in allocation ratio, or the use of non-concurrent controls. It will also take account of the impact on external validity and potential bias resulting from inconsistencies over time in study population, or sequential dissemination of results. Other issues will include design characteristics that may risk interpretability of the results, and how requirements for type I error control at the level of a platform trial can lead to different operating characteristics when compared with an individual comparison within the trial results.
A core drafting group will be responsible for delivering the first outline, which will then be discussed with the agency's methodology working party and associated specialized expert and stakeholders.Within the agency, the agency’s main scientific committee, the Committee for Medicinal Products for Human Use (CHMP), and the scientific advice working party will be most closely involved in the process. Other agency bodies that will be consulted prior to release of a draft to the public include its committee for advanced therapies, the pediatric committee, the pharmacovigilance risk assessment committee, and the committee for orphan medicines. Input will also be sought from regulatory agencies in the US, UK, and Japan.
As is its custom, the agency is setting out its planning early, to acquaint the clinical trials community with its intentions, to seek early reactions to its thinking, and to indicate when and how it will be seeking further input into the deliberations. The detailed timetable foresees setting up a drafting group before the end of 2022, discussion in the CHMP by the end of2023, the release of a draft in spring 2024 with a midsummer deadline for comment, re-discussion within the agency in the autumn, and adoption by CHMP in December 2024. Right now, comments are invited on the entire concept before Jan. 31 2023.1
Meanwhile, closer scrutiny is now falling on the agency's poster-child, the mechanisms for the new European clinical trial framework that will be obligatory for all new trials in Europe from the end of January 2023. At the heart of the new system is the clinical trial information system that acts as a trial registry. This has been road-tested by one of the leading commentators on trial transparency, Nicholas DeVito, a key figure in the TranspariMED initiative that has recently been highlighting trial reporting failures across Europe and the US. In a late October assessment of the Clinical Trials Information System (CTIS) site, he found that the information on more than 10% of the trials listed is "extremely limited" and that it is difficult to identify the nature of the trials that appear. He says that more clarity is needed in the registry's presentation, so that casual users, or those with little European trial experience, are not going to be daunted by "the deep underlying regulatory mechanics that govern the registry." He found the options for exporting information about individual trials to be cumbersome and unfriendly, and concludes that "EMA hasn’t done anything yet to facilitate working with data from the new registry." And, in general, he finds that "there is no easy, simple way to have all the available information available in one scrollable screen."
His assessment is not all negative. He considers that a valuable area for researchers will open up with the trial protocol section—although for many trials this section will remain empty for up to five years after the end of the trial, owing to the way the legislation is framed. TranspariMED and other health campaigners have been calling for changes in this facility to oblige earlier publication of protocols and to end what it calls "protocol secrecy"—but that is a story for another occasion.
