Clinical trials are planned in phases to gather evidence, address specific aims and, ultimately, to assess whether a specific drug is efficacious in the patient group. This planning will usually include a protocol detailing the process from initial recruitment to post-trial drug access or follow up. However, when a clinical trial is terminated earlier than planned, this can result in less structured end-of-trial participation. This article looks at a range of international evidence to inform an at-a-glance review.
What is the incidence of early termination of clinical trials?
There are mixed reports on whether randomised controlled trials for children and young people are more or less likely to end prematurely (Schandelmaier et al, 2017; Dufetelle et al, 2018), which is confounded by a paucity of specific literature (Pica and Bourgeois, 2016; Schandelmaier et al, 2017). Owing to the dearth of evidence, and the international nature of paediatric clinical trials, research from around the world has been considered here.
Between 2008 and 2010, Pica and Bourgeois (2016) found that 11% of paediatric trials in the US, which included 8369 children or young people, were terminated after participant enrolment had begun. Problematic recruitment to the studies was the most frequently cited reason for termination. A further 29% of studies, which included 165 paediatric participants, were not published 2.5 years after completion. Similar percentages were reported in other specialist areas. Bernardez-Pereira et al (2014) reported 10.9% early termination in cardiovascular studies and Dufetelle et al (2018) found 8.5% early termination over all paediatric randomised controlled trials. In the wider population, 25% and 32% early trial termination was quoted by Briel et al (2017) and Stegert et al (2016), respectively. Thus, while early trial termination appears to be comparatively lower in paediatric than in adult studies, the fact remains that approximately 10% of paediatric trials end earlier than expected.
Why is this an issue?
The discontinuation of clinical trials earlier than proposed is problematic for a number of reasons (Knottneruss and Tugwell, 2016). One issue is the waste of resources it entails (Jitlal et al, 2012; Briel et al, 2016; 2017). These wasted resources could be financial or time specific, but could also include loss of trust and additional burdens shouldered by participants at the unplanned end of the trial (Wears, 2015). Pica and Bourgeois (2016) highlighted that the exposure of children and young people to study procedures without subsequent purpose or knowledge is contrary to the principles of the Declaration of Helsinki, which governs research on human subjects, and states that ‘Researchers have a duty to make publicly available the results of their research on human subjects … Negative and inconclusive as well as positive results must be published or otherwise made publicly available.’
If vulnerable groups are being subjected to procedures as part of a clinical trial, they are often doing so for the purposes of altruism, hopefulness or to add to knowledge (McGuinness and Cain, 2007; Peay et al, 2014). If the trial then ends early, and results or findings are not published, the ethics of the study can be called into question.
Non-publication of results, systematic reviews, lessons learned, or thorough reporting of the process overall also potentially subjects further participants to unnecessary procedures, harm and burden.
Why do studies end before the expected date?
Studies may end earlier than planned due to recommendations from a data monitoring committee. The recommendations commonly fall into one of four categories: unacceptable adverse events and/or risk to safety; medical futility; new external information; and overt benefit (Bassler et al, 2008).
A medical futility (Stegert et al, 2016) decision is usually made by a data monitoring committee based on the results available (Grant, 2004) where it is deemed that the outcomes observed are unremarkable and, therefore, continuation would be unethical (Bassler et al, 2008). However, this may mean that, on occasion, studies with modest clinical effects are stopped early (Jitlal et al, 2012). The converse can also be true, when results are deemed to be overtly beneficial and it would thus be unethical to withhold the treatment from those in the non-treatment arms of the study (Bassler et al, 2008).
Although the data monitoring committee recommendations may fall into the above categories, the most commonly cited reason for clinical trials terminating earlier than expected is low recruitment (Bernardez-Pereira et al, 2014; Stegert et al, 2016). Briel et al (2016) discussed the need to publish the specific reasons for poor recruitment in clinical trials, highlighting that the majority of the reasons given for discontinuation were foreseeable or preventable: overestimation of eligible participants was cited in 54% of studies, for example. This is a particular challenge in countries with smaller populations (Briel et al, 2017). It also highlights the need for researchers and funders to consider at the outset the minimum number of participants required to show efficacy, as well as anticipated numbers, and collaborative approaches (Wears, 2015).
Briel et al (2017) discussed lack of funding as a contributing factor in early clinical trial discontinuations. It is perhaps not surprising that Dufetelle et al (2018) found that multinational and industry-funded studies were less likely to be discontinued early. This is supported by the difficulties discussed by stakeholders (Briel et al, 2017). While initially this may be due to a funding insufficiency to begin the project or slower than predicted recruitment, Briel et al (2017) also highlighted the need for investment in progression and career opportunities for the retention of trained clinical research staff because high staff turnover may have a negative impact on the reliability of the data.
The concept of trained staff in conjunction with the high turnover discussed by Briel et al (2017) is a further issue owing to the relatively long time it can take to feel confident as a research nurse compared with other new positions taken by nurses (Tinkler et al, 2018). This is supported by studies that highlighted a waste of human, as well as monetary, resources, (Briel et al, 2017; Dufetelle et al, 2018).
Unnecessary participation in a trial can also be an issue. As inclusion criteria for paediatric clinical trials increasingly include ‘wash out’ periods in which the potential participant must not have been given another experimental compound within a certain time, prolonged participation in a drug trial with limited or questionable efficacy may risk participation in future trials, especially those with competitive recruitment at multiple sites.
Barriers to recruitment
There are barriers to recruitment on multiple fronts: participant, family, protocol and clinician. Briel et al (2017) discussed trial recruitment with a range of stakeholders, and potential participants often cited a lack of direct benefit, and the additional burden of participation, being reasons to decline to take part in a clinical trial.
Further to this, the risks and benefits of the clinical trial, perceived by professionals, can also change the recruitment approach, particularly where nurses feel caught between the needs of the study and their role in caring for the patient carer (Tinkler et al, 2018). The challenges and barriers nurses face as recruiters were further discussed by Zucchelli et al (2018) who found that nurses experienced challenges when talking to families about research when there was a language barrier or when families had complex social situations. Time constraints also compromised recruitment. While nurses or clinicians should not overstate the benefits of clinical trial participation (Camara, 2019), recruitment then becomes reliant on participants and families being interested and aware of the research.
Impact on participants, families and clinical research teams
The end of a clinical trial and return to standard clinical care can be a challenge for participants, families and staff. In clinical trials for cancer care it has been acknowledged that participants can experience feelings of abandonment, disappointment and anger at the planned end of a clinical trial (Wilson et al, 2007; Lawton et al, 2017).
Nurses and clinical teams, as well as participants, can experience complex emotions when a trial is ending (Wilson et al, 2007; Lawton et al, 2017). While the end of a clinical trial is expected to mean the transfer back to standard care, Lawton et al (2017) wrote of an expectation among clinicians that there would be ways to ‘fiddle it’ to continue the trial treatment for participants. There is a dearth of literature on the experience of families at the end of clinical trials for chronic conditions; however, if participants experience feelings of abandonment at a planned end it leads to questions for those whose trials are unexpectedly shortened.
Conclusion
The unexpected or early stopping of a clinical trial can be complicated and ethically difficult for research teams and participants. Integrated research and clinical care with a planned strategy for the end of clinical trials at any point may support staff and participants to manage the end of the trial. Clinical nurses can support research and, therefore, support the development of care for patients and their families. Adequate recruitment strategies and funding could contribute to the success of clinical trials in the future.