An information day was held by the EMEA at the end of April, bringing together representatives of national Competent Authorities (CAs), pharma companies, CROs and non-profit stakeholder groups. The purpose of the day was to communicate the features and development timelines of the next few versions of the EudraCT database of clinical trials in the European Economic Area (EEA). These developments stem from EU Regulations 726/2004 and 1901/2006, which require certain information about clinical trials to be made available to the public.
Version 7 provides improvements to current function (ie, for sponsors to prepare information to submit to CAs) and is planned to go live on June 18th 2009. Version 8 is centred around making protocol information available to the public, and should be live in November 2009, while version 9 will publish information on the results of studies and is in early development, aiming towards the middle of 2010 for launch.
The first section of the day introduced the legal background and gave some statistics on the growth of EudraCT since its launch in 2004. To date, nearly 40,000 applications have been logged, corresponding to just over 19,000 studies. Of these, nearly 80% are commercially sponsored, 65% are multi-site, 60% include sites in multiple member states and 50% also includes sites outside the EEA.
Fergus Sweeney of the EMEA stated that for the current year he expects the EMEA to receive around 10,000 applications, that will equate to around 5,000 new trials.
The meeting was chaired by Fergus Sweeney and Brian Davis of the MHRA, with Fergus and IT Analyst Neil Cordwell giving the presentations.
EudraCT version 7 preview
Version 7 of the EudraCT database of clinical trials in the EU is expected to clear internal testing on June 1st 2009. User acceptance testing should take place from June 8th to 13th, in order to go live on June 18th.
At present, the database is designed for use by the Competent Authorities (CAs) throughout the EEA, with sponsor companies able to prepare study applications via a secure site, before making submissions to individual CAs by paper/CD-ROM. This basic purpose will not change in version 7.
Among the functions of the new version are inspection data and automatic alerting to all CAs on refusal, interruption, conclusion or termination of a study. These were not discussed in depth at the meeting, with more focus on external functionality for sponsors.
The database currently provides sponsors with an online form to enter information about a study. This information is then saved to a local computer as an XML file, which populates the application form and is also sent to the CA for uploading to the database itself.
Important new functions that were demonstrated include the validation of the XML file, to ensure that all fields in the form have been completed appropriately. This validation will be performed using the ISO Schematron standard, and details of the rules will be published so companies can build pre-validation into their internal systems. To further aid validation, much more explanatory text will be added to the online forms, to aid completion. The validation report can also be saved as a PDF for inclusion with the application submission, and is tagged with the same identifier as the relevant version of the XML file. It will be possible to make a submission even if the validation check is not entirely passed, to permit novel types of study; in such cases, it will be obvious which fields are not validated, and these can be discussed in more detail between the sponsor and CA.
Another new function is the preparation of an application package, collating the raw XML file, the populated application form PDF and the PDF validation report in a single zip file, for writing to CD-ROM for inclusion with an application. All pages of each file are labelled with the XML identifier for version control. The files will be named to include the EudraCT number for the study, the country being applied to and the date of preparation. One delegate highlighted that CAs have differing requirements for file names and structures on CD-ROM, but it was pointed out that the zipped files can easily be unzipped, renamed, moved etc. accordingly.
The third major change in functionality enables users to compare two XML files (eg, two versions of the same application, applications for the same study to two CAs etc.) This provides a short summary, which can be drilled through to see changes to individual data fields in a ‘tracked changes’-like format. In version 8, it will be possible to save this change report to a PDF file (although it can be simply printed out from version 7 for temporary, informal analysis).
The final area of change in version 7 will be an update to the terminology lists used to formalise usage throughout applications. In addition to updating MedDRA from version 9 to its current version (version 12), it will also include the EDQM lists of routes of administration and dosage, the ISO list of countries and the EudraVigilance list of active substances (eg, for comparators etc.) These will be updated monthly in version 7, and dynamically linked in version 8.
Look ahead to version 8
The meeting continued with a look ahead to future versions of EudraCT, which will enable protocol summaries and results of clinical trials to be made available to the general public, in accordance with EU Regulations 726/2004 and 1901/2006. These will be done in version 8 and version 9 respectively. Guidelines for the data fields to be published for protocol summaries and study results were finalised in July 2008 and February 2009 respectively.
Version 8 is intended to complete internal testing on September 1st 2009, with user testing in November to go live on November 30th 2009. It will make public specific fields from the protocol information as soon as a CA/REC decision has been input (ie, trials withdrawn before a decision will not be published), as initially entered (ie, mainly in English, although the coded terminology could be translated in the future). Data will still be entered to the database by the CAs with no direct sponsor access, although direct electronic submission by sponsors is being considered for a future version, as is ongoing maintenance of specific data items (eg, public contact point etc.).
Studies added from launch will be immediately available to the public, with a gradual backfill of older studies, although information not in the original submission will not be collected retrospectively, such as the public contact point that will be required from this version onwards. The new data fields for public accessibility have been published by the EC.
For sponsors using the system, it will move from a wizard system to being more menu-based, to enable use of modern ‘smart form’ logic to hide un-needed questions, collect additional information if required, enable re-use of core sponsor details between studies etc.
To accommodate all of this, a substantial change will be made to the underlying database structure, and to the XML schema. This will have significant implications for sponsors and national systems (eg, IRAS) which interact directly with this dataset. Drafts will be circulated as soon as possible so other systems can be updated accordingly.
The other major change in version 8 extends all of this functionality to accommodate paediatric studies being conducted in 3rd countries (ie, outside the EEA) but which are included in a Paediatric Investigation Plan (PIP).
In development for version 9
Looking further ahead to version 9, which is intended for user testing and launch around September 2010, this will make available information about the results of studies to the public. The details of this (ie, what level of summary/specifics) are still under discussion, although there is a desire to harmonise with similar initiatives outside the EU.
The regulations state that information on paediatric studies will be made public 6 months after completion of the study, with adult studies from Phase II onwards being published 12 months after completion (the regulation does not differentiate between Phase I studies with healthy volunteers and those with patients).
The programme of development to take EudraCT from its current v6 to version 9 over the next 18 months represents a significant amount of work, and a significant expansion of the usability and value of the EudraCT system, to Competent Authorities, sponsors and the general public alike.
A further meeting is planned for Autumn 2009 to discuss the ongoing development and intended launch of version 8, and we will continue to watch and report on this important area of development for clinical research being conducted in Europe.
