Putting the “good” into Good Clinical Practice

Current Good Clinical Practice guidelines are bureaucratic and should align with less burdensome examples of international trial policy

Clinical trials must be conducted in ways that protect participants and produce reliable results. Both are central tenets of the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) Good Clinical Practice (GCP) guideline.1 The ICH GCP guideline was developed to harmonise the conduct of trials across world regions and, since the mid‐1990s, its core principles have provided the bedrock for trial conduct. However, the devil is in the detail and, in the case of the ICH GCP guideline, that detail (and the interpretation of each word) has far‐reaching consequences.

The ICH GCP guideline was designed specifically for drug trials for regulatory submission but is widely applied to other trials. Although the guideline was revised in 2016, many critics, including supporters of the MoreTrials campaign (https://moretrials.net), have called for a more extensive overhaul.2,3,4,5

These calls have been heeded with plans for a “renovation” of GCP,6 and a concept paper on the proposed third revision was published in 2019.7 But the renovation will take years, and with trials already so mired in bureaucracy, interim solutions are needed. Importantly, full compliance with the ICH GCP guideline is not a legal requirement in many regions including the European Union, the United States and Australia (Box 1). Moreover, trial regulation and guidelines already exist that, by necessity, depart from the ICH GCP guideline to better emphasise important aspects of trial conduct.

In this article, we illustrate some of the shortcomings of the ICH GCP guideline in the hope that a better understanding of these shortcomings will deter policy writers from making full compliance with the ICH GCP guidelines mandatory. We also showcase less burdensome provisions and interpretations of the ICH GCP guideline found in regulations and guidelines from the United States, Europe and Australia.

Current guideline shortcomings

Lack of an empirical approach to guideline writing

Created (with little transparency) by an expert working group of regulators and the pharmaceutical industry, the ICH GCP guideline was developed through informal consensus that was neither inclusive (few, if any non‐industry trialists, funders or patient groups) nor evidence‐based.8 We illustrate the consequences with three examples.

Informed consent: The ICH GCP guideline requires inclusion of 20 separate elements in participant information leaflets; far above what is required by US and European Union trial regulation or Australia’s National Statement on Ethical Conduct in Human Research.9 The upshot is that participant information leaflets are often over 20 pages long, despite evidence suggesting that people are not likely to fully read documents longer than five pages.10

Lengthy, legalistic consent forms may protect sponsors and sites from liability but do little to aid patients’ comprehension of a trial’s key information. Although more research is required to determine the optimal content of participant information leaflets,11 evidence suggests that some elements (eg, the total number of participants) are not considered by patients as necessary for decision making.12

The absence of evidence of utility exacerbates criticisms levelled at trial consent as having become ritualistic — the document has become more important than the process.11

Indiscriminate safety reporting: Excessive provisions for safety reporting within the ICH GCP guideline neither increase the scientific quality of trials nor avert safety incidents.13 Investigators must review (sometimes dozens of) suspected unexpected serious adverse reactions even though sites are ill equipped to conduct a proper analysis of these reports.

By contrast, the EU Clinical Trials Regulation (536/2014)14 and Australia’s Safety Monitoring and Reporting Guidance 201615 no longer stipulate that investigators (or ethics committees) routinely receive single case suspected unexpected serious adverse reactions or line listings. Much more sensibly, it is the digested outputs of the sponsors’ ongoing safety analyses (eg, safety signals identified by a trial’s data monitoring committee or single case events that are truly significant safety issues) that are reported. In Australia, this has reduced both time and cost pressures on trial sites and sponsors. More importantly, it may well have reduced the risk to patient safety by ensuring key events or safety signals are not lost in the white noise created by indiscriminate reporting.

The pursuit of data accuracy rather than reliability: It has been argued that the fixation on data accuracy is misplaced and for many randomised trials, reliable results are not so much born out of accuracy as an absence of bias between the treatment and control groups.16 Despite recognition that centralised (statistical) monitoring may be appropriate, the ICH GCP guideline still requires monitors to verify that source documents are accurate. No surprise, therefore, that extensive source data verification is still widespread despite strong evidence to suggest that it has a negligible impact on data quality.17

By contrast, US Food and Drug Administration (FDA)‐supported work by the Clinical Trials Transformation Initiative and work by the European Medicines Agency promoting the principles of Quality by Design recognise an important fact that seems to have gone astray — quality is the absence of errors that matter.18 We also applaud the ICH for its thoughtful draft revision of ICH E8: ICH guidelines E8 (R1) on general considerations for clinical studies (https://www.ema.europa.eu/documents/scientific-guideline/draft-ich-guideline-e8-r1-general-considerations-clinical-studies-step-2b_en.pdf), which has a much greater emphasis on key scientific principles such as maintaining the integrity of the randomisation process, minimising dropout or loss to follow‐up and protocol adherence.

Ambiguity leading to widespread overinterpretation

When the stakes are so high (over $2 billion to develop a drug), it is hardly surprising that sponsors and contract research organisations take a conservative approach when interpreting GCP requirements. We discuss some examples of the lack of clarity within the ICH GCP guideline below.

To sign or not to sign: The ICH GCP guideline requires investigators to maintain a list of people with “significant trial‐related duties” but does not clarify the meaning of this phrase.1 This requirement is normally operationalised by trial staff signing a delegation log. Signing a log is not particularly burdensome, but inclusion invariably triggers a barrage of requests for financial disclosures, curricula vitae and training records. The issue is that sponsors’ interpretation of who should be included on a delegation log is becoming increasingly nonsensical with some requiring the inclusion of almost anyone involved in trial patients’ care, no matter how ancillary their role. Some sponsors even require external vendors such as optometrists or radiologists to sign the log even though the services they provide are contracted rather than delegated and vendor suitability is managed in other ways.

We suggest that principles promulgated in FDA guidance19 on completion of the 1572 form (a mandatory form completed by investigators for certain FDA‐regulated trials) could be applied more widely. The FDA guidance confirms that staff performing “ancillary or intermittent care but who do not make a direct and significant contribution to the clinical data” or with an “occasional role” need not be listed.

Who, exactly, needs training? The ICH GCP guideline’s lack of explanation on what it means to be “qualified by education, training, and experience”1 results in sponsors imposing a one‐size‐fits‐all approach to training; particularly GCP training, which is often mandated for anyone who so much as touches a trial patient. In their joint statement, the Medicines and Healthcare Products Regulatory Agency and the Health Research Authority take a common sense approach to training by recognising that some staff may simply require an awareness of GCP that is achieved through self‐directed learning or a written summary of any relevant requirements.20

Good clinical practice in other documents

Box 2 illustrates examples of regulation and guidance from the US, Europe and Australia that contain burden‐reducing provisions. These provisions have the potential to improve the regulatory environment by enhancing trial quality without unnecessarily complicating or delaying trial conduct.

Conclusion

The ICH GCP guideline is the established standard for trial conduct worldwide and one of the most influential trial documents ever written. However, it has inadvertently contributed to skyrocketing costs and wasteful practices while failing to prevent poor trial quality.4 The ensuing bureaucracy ultimately denies patients’ access to better treatments, hampering improvements to public health.

This article illustrates some of the ICH GCP guideline’s limitations and adds to growing calls for governments, regulators, policy writers and auditors to accept that where there is no legal imperative to follow every word of the ICH GCP guideline, trialists should be free to default to more sensible provisions. The documents we showcase provide a workable framework to operationalise the principles of the ICH GCP guideline and should be reflected in its planned renovation. Unfortunately, our experience is that even the smallest departure from these guidelines is considered inappropriate despite even its authors acknowledging its faults.6

To overcome these entrenched perceptions, we suggest change will require a top‐down approach. This should include training that engenders a common sense approach to trial oversight that encourages critical thinking over rigid and burdensome practices that do little to improve patient safety or data reliability. Similarly, regulators should do more to allay industry’s fears that a risk proportionate approach to trial management will jeopardise regulatory approval.

Australia has a rich history of high value trials conducted by clinical trials networks that collaborate nationally and internationally. However, the number of critical questions concerning health care still exceeds the trial community’s capacity to answer them. The renovation of the ICH GCP guideline has the potential to improve the efficiency of the trial enterprise and if successful, it will not have come soon enough.