ICH guideline E6 (ICH-GCP) is, along with the Declaration of Helsinki, arguably the most important document in clinical research. Although neither has any direct status in the legislation of most countries where clinical research is conducted, their principles (and in many cases more substantive details) set the tone for how pretty much everyone conducts clinical research. Since its adoption in 1996 (in Europe; 1997 in the USA and Japan), ICH GCP has been the ‘bible’ for CRAs, auditors and other clinical research professionals worldwide. Since 1996…
The world of clinical research has moved on quite some way in the past 13 years, and even more so when you consider the period of several years that was taken for the drafting, consulting, reviewing and negotiating prior to the guideline’s finalisation. Other guidelines (most notably the Declaration of Helsinki) have been updated several times in the past decades, and have a timeline for regular review every few years.
So, following a remark made by
Professor Richard Gray when
speaking at the ICR Annual Conference earlier this year, we wondered whether ICH E6 should be reviewed and potentially revised. We put a poll on the front page of the ICR website, and were rather surprised by the result: over 80% thought that it should be reviewed (although from an admittedly small sample). Following on from that, we undertook a qualitative survey, asking all CSci and FICR members what elements of the guideline should be updated and/or what should be added that did not exist in 1996. Here we present some of the most interesting and provocative comments.
Assessing the need for change
Many of the respondents to our survey suggested specific areas where further clarification would reduce variability in interpretation, and aspects where science and working practice have moved on substantially since ICH GCP was adopted. These suggestions are discussed later in this article. First, though, comes the thorny question of whether we should even consider reforming the guideline. According to
Paul Chester MICR CSci, “
I think this discussion is overdue and very relevant. Whenever I talk about GCP, I feel it is important to point out that ICH GCP is now 13 years old and getting older every day, so it must be read in light of new regulation and guidance.”
Not everyone agrees on this question. There are two main arguments against attempting to update ICH GCP, which are themselves diametrically opposed. One is the widespread feeling that it is so successful, so entrenched after 13 years and so fundamental to modern clinical research that any attempt to change it would be futile and would produce more confusion and political turmoil than any changes might resolve.
Prof. David Hutchinson FICR CSci holds this view: “
I do not think that ICH GCP guidelines need to be changed. They have stood the test of time and remain perfectly reasonable. It should be remembered that these are a global standard and seem to fit the practices in most parts. The ICH GCP guidelines present clear requirements for sponsors, investigators and ethics committees. They protect subjects and they facilitate the collection of credible data backed up with good documentation.”
The other point of view is that ICH GCP’s era of dominance has in fact been ended, in the EU at least, by EU Directives and national legislation that call for the guideline’s principles to be taken into consideration but shy away from giving it any direct legal force. One person who took this line in a phone conversation with me was
Joan Perou HonFICR. However, Prof. Hutchinson reminds us that “
in both the UK/Europe and USA, the ICH GCP guidelines are still used as an inspection standard and are still very much part of our research. In Japan a slight modification to the ICH GCP is still the chosen standard.”
Interestingly,
Tina Barton HonFICR CSci suggests that “
the perception of issues with ICH GCP may be more related to the EU Clinical Trials Directive and/or the national legislation. The bigger issue may be the lack of a common view of users. Since its adoption, regulators (in particular the FDA and EMEA) are more closely aligned and in broad agreement of the application of this guidance, which has brought many benefits to the research community.”
Nigel Crossland FICR CSci holds the guideline in similarly high esteem, commenting that “
despite new requirements such as EU regulation, the ICH GCP Guideline has never been superseded,” but adds that he thinks “
consideration should be given to review and update of the ICH GCP Guideline.”
Dipti Amin MICR CSci agrees, saying “
I believe the fundamental premise of GCP is sound and still applicable. It serves as a good base for providing guidance around the processes and requirements for carrying out research particularly for those new to conducting research and those otherwise apt to cut corners. It helps achieve some degree of uniformity in expectations and sets the independently expected standard for all involved in clinical drug research.”
Rod Owen FICR CSci nicely sums up a pragmatic approach taken by many respondents: “
It is my view that the Principles of ICH GCP (ie, clauses 2.1-2.13) should remain unchanged as they are absolutely fundamental to the practice of internationally accepted ethical research. However, any minor ‘tweaks’ elsewhere within Guideline E6 would be perfectly acceptable and indeed logical”.
Mark Elsley MICR CSci has a similar view: “
ICH GCP has revolutionised the pharmaceutical industry to improve the quality of clinical data and assure patient safety. As such, I do not believe ICH GCP should undergo radical changes but there are some are some areas where I think further clarification is required”.
Dr Arun Bhatt HonFICR expressed this most strongly, saying “
The current ICH GCP guidelines are too broad and general to proactively manage today’s scientific and ethical issues. There is a need to make ICH GCP more explicit, specific and stringent”. However, perhaps we should be wary of making ICH GCP too prescriptive at the cost of making it overly complex, cautions
Jacqueline Briggs MICR CSci, “
I would not be keen to see ICH GCP changed into something more complex and confusing as I think its summary layout is helpful and practically applicable.”
Others argue that even keeping ICH GCP unchanged should be a conscious decision rather than a default. This sort of periodic review is the policy adopted by the World Medical Association with regard to the Declaration of Helsinki.
Anita Lindsay FICR CSci supports the process of regular consideration, “
I would be reassured to know that ICH GCP had been reviewed on a regular basis to ensure that it is still meeting the needs of patients, clinicians and the industry (even if there were no resultant changes, it would also be good to know that there was a process behind the scenes). A review would also raise the profile of ICH again.”
However,
Neil Sharpe MICR CSci reminds us that “
subject safety and data integrity should remain the founding principles of GCP: any changes made to reflect current practices should not compromise these principles.”
Specific areas for change
The ICR members who responded to this call for contributions had many suggestions of specific areas to change in a hypothetical revision of ICH GCP. Here is a selection of their recommendations:
Harmonisation & globalisation
One of the most significant changes in the way clinical trials are conducted since the early 1990s has been the opening up of many new countries as locations for clinical research, and the acceleration of ‘off-shoring’ large clinical trials to these locations. Dr Bhatt highlights one of the major issues arising as a consequence of this: “
There are also new concerns [to be met by ICH GCP, such as the] use of poor illiterate populations from developing countries as trial subjects.” Although outside the core ICH countries (EU, USA and Japan), ICH GCP plays a major role in these countries. Prof. Hutchinson reminds us that “
We have to think outside the national (UK)/European box and realise that researchers in many countries have only the ICH GCP principles to work with. There are some other GCP codes but these tend to be modelled on the successful ICH GCP standard.”
Clarifications
Perhaps more troublesome than any other aspect, the potential for differing interpretations of can lead to local variations, bureaucracy and confusion. While ICH GCP is generally considered relatively successful in avoiding this, there are areas that contributors think could be improved.
Tina Barton comments, “
Countries generally stick to the intent of ICH GCP, but there are different interpretations of some details and greater clarity around these would be helpful to assure a more consistent adoption of specifics, often increased by the legislative requirements and the interpretation of those into the everyday activities.”
Adam Jacobs FICR CSci notes that “
the section on informed consent of trial subjects, while undoubtedly well-intentioned, can be a bit over the top in some circumstances, resulting in information overload for subjects, which then compromises the validity of the whole informed consent process.”
Dipti Amin comments that “
from time to time the interpretation sometimes placed by those utilizing it in the conduct of clinical research causes unnecessary bureaucracy without adding anything to promote the safety, rights and well being of study participants or in ensuring the integrity of study data. Also, there could be more flexibility in terms of the wording around monitoring which currently suggests that central monitoring should be the exception whereas in reality it’s applicability depends on study design, indication, and study purpose as well as level of site experience in conducting research.”
Mark Elsley thinks that “
there should be more guidance about how much SDV to do as the definition in ICH-GCP is rather vague. Some companies do 100% and others take a risk-based approach, without any clear statement of what is acceptable.”
Dr Bhatt suggests that some of the terms used could be clarified to remove ambiguity, such as “
persistent non-compliance and protocol deviations vs violation”.
Scientific developments
Paul Chester, Dr Bhatt and Mark Elsley all highlight the area of clinical pharmacology, particularly from the point of view of integrating (or at least citing) the guidance on first-in-man studies developed after the TGN-1412 incident.
Anita Lindsay adds “
The big step forward since the mid-90s for me is personalised healthcare (including the requirements of the UK Human Tissue Act etc.); reference to human tissue collection is sparse in E6. Biologics have also emerged as a major component in drug development and E6 might need a refresh in this regard.” This latter point is also echoed by Paul Chester and Dr Bhatt.
Another correspondent who prefers to remain anonymous suggests that “
The definition of Standard Operating Procedure may need to be amended in view of the increasing use of business process software and workflows instead of written instructions.”
Data protection
Barbara Hepworth-Jones HonFICR feels that “
the area of greatest need is that ICH GCP is woefully lacking in data protection requirements. The increase in electronic data systems since ICH GCP came in to being and the potential for misuse of personally identifiable information mean that there is a real need for this to be added.”
Mark Elsley also notes the lack of provision for EDC: “
Many investigators now use electronic systems for recording source data. What should be the
minimum requirements for these systems and who is responsible for them?”
Barbara Hepworth-Jones adds “
I appreciate that in many countries there are data protection laws, inspectors and auditors can use those to supplement GCP, so in practice it can be covered. But if it is not covered in GCP, then subjects’ rights and well being are not totally protected by GCP.”
Essential documents
Karen Roy HonFICR CSci told me that exploring the area of essential documents is one area she’s closely involved with “
This is the focus of a DIA special interest group I co-chair with Lisa Mulcahy in the US. Basically, the group exists because the Essential Documents are not enough and because auditable documents are more than the Trial Master File (TMF). Our goal is to come up with a standard list of TMF and auditable documents, standard naming, standard structure and standard metadata (for eTMF). This could be adopted by industry worldwide.”
Nigel Crossland suggests that “
areas of improvement could include making section 8, which is used by many as an effective ‘checklist checklist’, better match up with the details in the preceding sections. There aren’t many, but there are several requirements which are not included in section 8. For example
- The requirement to have amongst the essential documents the investigator’s written application to the Ethics Committee (see 4.4.1)
- An IMP inventory at site showing details of quantities, batch numbers, expiry dates etc. of all IMP received (see 4.6.3)
- Details of SOPs used by the sponsor in the conduct of the trial (see 5.1.1)”
Other suggestions
The key area that Jacqueline Briggs would like to see discussed “
is recognition of the whole research team rather than just the PI. I totally understand that the PI is ultimately responsible, but in all aspects of healthcare the roles of other health professionals have developed and been recognised. It is important we recognise this in clinical research, and the first step of changing physician consent in the Declaration of Helsinki helps pave the way. Also I would like to see more focus on the study organiser and their protocol development requirements (ie, including people with on-the-ground expertise) and ensuring the protocol is deliverable, with an understanding that GCP needs to link with other policies and procedures in the many different organisations where it is applied.”
Adam Jacobs mentions that one area he considers “
ripe for an update is the section on data management (5.5). Although it’s not bad and covers the main points of data management, it’s a bit light on detail, and I have sometimes found it hard to argue with a client that their data management systems were not GCP compliant when I know that they were certainly way below best professional standards. There is also nothing at all on statistical analysis of data (even a reference to ICH E9 would be better than nothing ), and I also think the section on writing CSRs could do with being beefed up.”
Nigel Crossland adds that “
there has always been a shortage of details concerning equipment and testing apparatus used at the investigational site and laboratories used by the investigator. This whole area has been recognised by BARQA who issued a guideline some time ago on the requirements that should be in place in clinical laboratories. I have always thought that Clinical QA personnel should be familiar with the “pick and mix” approach to application of GLP standards in the clinical setting and to avoid areas being mistakenly omitted from clinical site audits.”
Reform & harmonisation of… something
Several contributors also suggested that revision of ICH GCP should be done in the context of a wider harmonisation of global guidelines and regulations, trying to ‘smooth out’ differences between the various other ‘flavours’ of GCP (eg, as defined in the EU GCP Directive etc.), the current version of the Declaration of Helsinki etc.
Paul Chester fears “
this would be much more difficult than the original production. Why? Mainly because we now have a legislative framework across Europe that was not in place at the time ICH GCP was implemented. Any changes would need to take into account the variations between countries. Furthermore the legislative framework covers all trials with IMPs while ICH GCP applies only to trials destined for use in licence applications.”
This is echoed by
Susan Ollier HonFICR CSci, who wonders “
why do we still not have a set of basic GCP principles, agreed internationally, that apply to all clinical research in humans? We currently have the globally accepted ICH-GCP which originally applied to commercial drug trials (and is now widely accepted for non-commercial drug trials), plus the various interpretations of GCP into national or regional laws (21CRF, EC/20/2001, Schedule Y etc.) which also largely apply to (interventional) drug trials. For device research, GCP is based upon ISO standards, not-for-profit organisations have occasionally written their own GCP standards (eg, the UK’s MRC GCP) and institutions setting standards for internal research governance again make their own interpretations. GCP would be easier to train upon, understand and use if we had a single guideline which covered the basic principles common to all human research; ethics, data quality issues, subject rights, the need to obtain applicable approvals, maintenance and storage of documentation and so on.”
However, as Tina Barton points out, “
changing ICH would only have ‘teeth’ if those changes were incorporated into CT Directive and national legislation.”
Conclusion
In conducting the research for this article, it has become clear to me that there are many areas in which the guidelines underpinning modern clinical research could (and, indeed, should) be updated to provide a more coherent environment for global clinical research, take account of developing science and operations/management practices, provide clear definitions and minimise the need for local interpretation.
We currently have a patchwork of guidelines, Directives, legislation and standard ways of working. In the past, it has proved politically impossible in some situations to agree key points between global regions, countries and even individual organisations. Looking at the situation pragmatically, it would seem equally impossible to move from this to a single, coherent guideline, and even more impossible to repeat the feat every few years to take account of further development in the field. As such, while we can each have our own idealised version of ICH GCP, it seems to me that the only way forward with this patchwork is to extend the patchwork further, filling in gaps and agreeing clarifications in subsidiary documents agreed by the stakeholders who need to use them. This could be done by ICH, but could equally be attempted by other groups of organisations, perhaps involving ICR.
Although this has been something of a theoretical exercise, it is still useful to consider which aspects of contemporary clinical research are poorly served by the current fragmented global network of regulation and guidelines, and how different ICH GCP would look if it were being created in 2009. Let’s continue this discussion: I’m interested to hear what you think…
