Recent advances in methodology for clinical trials in small populations: the InSPiRe project

Abstract

Where there are a limited number of patients, such as in a rare disease, clinical trials in these small populations present several challenges, including statistical issues. This led to an EU FP7 call for proposals in 2013. One of the three projects funded was the Innovative Methodology for Small Populations Research (InSPiRe) project. This paper summarizes the main results of the project, which was completed in 2017.The InSPiRe project has led to development of novel statistical methodology for clinical trials in small populations in four areas. We have explored new decision-making methods for small population clinical trials using a Bayesian decision-theoretic framework to compare costs with potential benefits, developed approaches for targeted treatment trials, enabling simultaneous identification of subgroups and confirmation of treatment effect for these patients, worked on early phase clinical trial design and on extrapolation from adult to pediatric studies, developing methods to enable use of pharmacokinetics and pharmacodynamics data, and also developed improved robust meta-analysis methods for a small number of trials to support the planning, analysis and interpretation of a trial as well as enabling extrapolation between patient groups. In addition to scientific publications, we have contributed to regulatory guidance and produced free software in order to facilitate implementation of the novel methods.

Keywords: FP7 small populations methodology projects; Rare disease clinical trial; Statistical methods.

Fig. 1

Jittered boxplot of phase 2 and phase 3 trials with either actual (brown triangle) or anticipated (blue dot) sample size by prevalence class. Each symbol represents one observation and the mean sample size is indicated by the red diamond. Number of trials contributing to the plot is given at the top row, median sample size in the second row, first quartile in the third row and the third quartile in the last row of the bottom of each boxplot. Figure reproduced from [8] under CC BY 4.0 License [49]

Fig. 2

Optimal adaption rules of adaptive enrichment designs, optimized for a sponsor (left graph) and a societal perspective (right graph). Depending on the observed standardized treatment effects in the biomarker positive (plotted on the x-axes) and negative (plotted on the y-axes) population, the graph indicates the optimal second stage design option: futility stop (white area), enrichment design, recruiting biomarker positive patients only (red area), or partially enriched design (grey area). In addition, the second stage sample sizes are optimized (not shown in the graph). The optimisation is based on an a priori distribution on the effect sizes corresponding to the assumption that the treatment effect is either independent of the biomarker or that it is larger (or only present) in biomarker positive patients. See Ondra et al. [23] for details. Figure reproduced from [23] under CC BY-NC License [50]

Fig. 3

Meta-analyses of few studies are particularly challenging. Here, effect estimates from two studies in pediatric transplantation [51] are shown along with 5 different combined estimates based on several common approaches: two Bayesian analyses with different prior specifications, a normal approximation that is usually appropriate for large sample sizes, and two small-sample adjustments based on a Student-t distribution. We systematically investigated the long-run properties of popular meta-analysis procedures with a focus on few small studies [37, 38]. Figure reproduced from [38] under CC BY-NC-ND License [52]