Why the changes?
The trend in clinical research is increasingly global. This means that, as well as multi-country trials becoming more common, there is competition between countries in terms of the speed and efficiency of their clinical trial systems.
The existing EU clinical trials regime has been heavily criticised for being onerous and expensive, and with too much divergence between EU states. As it stands, the system (based on Directive 2001/20/EC) leaves both ethical review and regulatory authorisation to individual EU states. The central European Medicines Agency (the EMA) only has a limited role, in maintaining the central EudraCT database, for example. So a clinical trial to be run in more than one country requires separate assessment in each. And although the law is harmonised to some extent, the detail of the rules differs from one country to another.
Dissatisfaction with the current system has resulted in declining numbers of trials being conducted in Europe. The EU Commission reported a drop of 25 per cent in the number of trials conducted in Europe from 2007 to 2011. The new law (Regulation 536/2014) is intended to reverse that trend by streamlining and standardising the system, while maintaining current standards of patient protection.
Another objective is to promote access to a larger population pool by bringing in patients from across the EU. This is increasingly important with the development of medicines tailored to genomic subgroups.
A single set of rules for the whole EU
Once it takes effect, the new system will apply automatically in all EU countries. By contrast, the current law was introduced through separate national implementation of the directive, with the result that the details vary considerably from one country to another.
A single application through the EU Portal
Under the new system, a sponsor will kick off the application process with a single application through a new, web- based EU Portal. This will require the development of a new IT system, and completion of the EU Portal will be necessary before the system can take effect.
Although a central application will be made, and a central channel of communication will be used between the trial sponsor and relevant states, the decision–making will still be in the hands of EU states rather than the EMA. One EU state will lead the process (as reporting member state) and this will carry out the lead analysis on the Part I aspects of the trial. These include analysis of benefits, risks, manufacturing requirements, labelling and investigator’s brochure. But each state in which trial activity will take place will be asked to consider and comment on the Part I assessment.
The assessment of Part II aspects (including informed consent, recruitment, data protection, site and investigator suitability and compensation arrangements) and the ethics committee review will be left entirely to the EU states in which it will be run.
This is often a sensitive subject in EU legislation. There is no uniform language requirement for communications through the EU Portal, and it will be up to individual states to determine language requirements. But the regulation does suggest that states should consider accepting ‘a commonly understood language in the medical field’ as the language for any documentation not destined for the subject. It remains to be seen whether this will result in a greater acceptance of material in English.
Division of trials into higher and lower risk categories
The regulation introduces a new concept of “low-intervention clinical trial”. This is where the additional risks for a patient taking part in a trial, compared with receiving a treatment in normal clinical practice, are negligible. The “low-intervention” category corresponds with categories A and B1 in the OECD’s 2012 Recommendation on the Governance of Clinical Trials. It applies where the investigational medicinal product is already authorised, and it is either used in accordance with the marketing authorisation or in a way that is supported by published scientific evidence, and with minimal additional risk to subjects compared to normal clinical practice.
If a trial is classified as low-intervention, it will benefit from lighter regulation, particularly relating to monitoring obligations, traceability of investigational medicinal products and requirements on insurance or other compensation.
Perhaps the greatest change in the new rules is to transparency. There is already an industry trend towards disclosure of clinical trial data, and the EMA is increasingly publishing clinical trial data on authorised medicines. The new system will take this further, establishing a new database that will store and as far as possible make available information on clinical trials.
All new trials will be registered on the database under a unique reference number. Information exchanged through the EU Portal will be included in the new database, with a readily searchable format and cross-linking to data on the same drug held elsewhere on the EMA’s systems. The general principle is that all data will be publicly available unless there are good reasons for confidentiality on the basis of protection of personal data; commercial confidentially (unless there is an overriding public interest in disclosure); confidential communications between EU states and to allow proper supervision of trials.
A summary of results, including a version in layperson’s language, must be uploaded within one year of a trial ending, for whatever reason. So data from unsuccessful trials will enter the public domain. Clinical study reports must also be added to the database within 30 days of grant of a marketing authorisation or withdrawal of an application for marketing authorisation.
The EMA’s approach to publication has been evolving through a period of consultation. Comments from industry on the risks to them of making available information on manufacturing process, for example, have been listened to and a more moderate approach to release of commercial information is now likely.
Tight timetable and tacit consent
The system will run to a tight timetable with a 45 day limit set for a standard assessment once the application is validated. Some still consider this to be too long, and there are possibilities for extension where additional information is requested, or in the case of more complex products, such as biologicals.
The regulation introduces the idea of tacit authorisation. Where the reporting member state fails to respond by the deadline on the completeness of an application, for example, the application will automatically be deemed complete. This cuts the other way – failure by a sponsor to respond to requests for comments or further material within the time allowed will result in the application lapsing.
Co-sponsorship and non-commercial sponsors
The rules allow for more than one sponsor, to take account of the increasing use of collaborative arrangements between a drug company and an academic institution, for example. Each sponsor will take on all the responsibilities of a sponsor, unless the sponsors allocate the responsibilities between them by written agreement. Generally, non-EU sponsors must have a legal representative in an EU state.
No help is offered to non-commercial sponsors, other than giving EU states the option of offering them reduced fees.
Non-EU clinical trials
Trials carried out in other countries can only be used to support an EU application for a marketing authorisation if they have met the same requirements. This includes the rules on transparency.
When will it take effect?
The new system will not take effect until, at the earliest, 28 May 2016. This date could be pushed back if the EU Portal takes longer than expected to implement.
Where applications for authorisation of clinical trials are submitted before the effective date, they will continue to be governed by the existing system for three years. Trials started within a year of the effective date can be authorised under the existing system.
Will it be a success?
Completion of the new portal and database present technical challenges that will have to be overcome before the system can get started. Making the bulk of the database truly accessible to the public, while protecting restricted information, will not be easy. Once it is up and running, a lot will depend on co-operation among EU states in implementing the system in an efficient way.