CLINICAL DEVELOPMENT IN ONCOLOGY
and incorporate this assumption into the planned sample size. However, when these metrics are higher than expected, this calculation may be inaccurate and the sample size available for analysis may not have sufficient power to detect a significant efficacy signal, contributing to lower success rates. In cases where the sample size is reduced due to recruitment difficulty or feasibility issues, elevated numbers of patients who dropout may further endanger the validity of the results and increase the likelihood of underpowered analysis. Clinical trials in oncology have been shown to
Credit: Halfpoint via
Shutterstock.com.
In this study, 21% of amendments and 33% of protocols providing these data reported a change in sample size. While this may include situations where the sample size was increased, on average the number of participants enrolled decreased between protocol approval and database lock regardless of changes in sample size. These findings suggest that future research evaluating changes in sample size during amendments, particularly for oncology clinical trials, are needed to better understand how recruitment difficulties are managed and to identify barriers to relaxing eligibility criteria. This may also help identify some of the root causes of increased amendment prevalence and frequency in oncology trials. The results of this study indicate that when
recruitment difficulties require an amendment to expand eligibility criteria or adjust the sample size in order to meet recruitment goals, that amendment can extend the treatment or enrollment period, resulting in increased participant burden and contributing to higher dropout rates. Dropout rates (not including deaths) in oncology were 35% higher than in non-oncology clinical trials. Oncology protocols where amendments were present had significantly higher dropout rates compared to oncology protocols without amendments (see chart on p6).
The protocol problem Most protocols and statistical analysis plans (SAPs) predict a certain percentage of screen failures and premature terminations (dropouts)
8 | Clinical Trials in Oncology
have higher complexity, narrower eligibility criteria, longer duration, and increased participant burden, all of which can contribute to increased frequency of protocol amendments. However, despite the stringent eligibility criteria often present in oncology and lower accrual rates, our analysis of protocol amendments overall and by disease area found that only a small percentage of amendments are specifically aimed at relaxing the eligibility criteria, and this percentage is lower among oncology trials. Instead, other amendment causes, such as Change in Study Strategy, were more common among oncology trials. Further, a significantly higher percentage of oncology amendments included a change in sample size, which may indicate that in some cases, a decrease in sample size may have been used to increase accrual feasibility instead of broadening eligibility criteria. This finding has implications for the level of resistance among clinical research scientists in oncology to loosen eligibility criteria. Eligibility criteria can be essential to
minimize confounding factors and ensure safety for participants. However, a number of studies have suggested that, as a common time-saving measure, eligibility criteria are copied from similar past protocols (Hauck, 2021). A more thorough review of the intended eligibility criteria prior to protocol approval may help avoid protocol amendments and recruitment delays. Intentional expansion of potential target populations in oncology clinical research may positively impact trial success rates in other ways, such as minimizing dropout rates and increasing statistical power.
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44
