Pediatric needs should be considered early in the clinical development lifecycle

“The pediatric drug development landscape today is vastly different from its unregulated beginnings” (Burckart and Kim 2020). Used to be called “therapeutic orphans” by Dr. Harry Shirkey in 1968, the societal imperative to develop drugs for children catalyzed the establishment of legislative provisions across regions. In the United States, the Pediatric Research Equity Act (PREA), which provides the requirement for studies, and the Best Pharmaceuticals for Children Act (BPCA), which provides the incentive of additional exclusivity for products of sponsors who conduct requested studies in the pediatric population, were enacted in 2003 and 2002, respectively. The European Union (EU) soon followed with the Pediatric Regulation, enacted in 2006 and effective in 2007. The Pediatric Regulation combines the requirement and the incentive into a single legislation. In the United States, RACE for Children’s Act was enacted in August 2020 to require early pediatric assessment for drugs directed at molecular targets observed in pediatric cancers (Barone et al. 2019).

The successful interplay between obligations and incentives stimulated pediatric research. Since then, how medicines are developed and labeled in many geographical regions has been transformed. In the United States (US) alone, the US Food and Drug Administration (FDA) has made labeling changes in 1049 applications including 204 from studies performed under Best Pharmaceuticals for Children Act (BPCA) alone, 164 under BPCA and Pediatric Research Equity Act (PREA), 606 under PREA only, and 49 under the Pediatric Rule (a precursor to PREA) since 2007 (FDA 2023). The European Medicines Agency (EMA) and its Paediatric Committee (PDCO) reported in 2017 that more than 260 new medicines for use by pediatric patients have been authorized, with most of them linked to the Paediatric Regulation’s requirements (Tomasi et al. 2017). Evidence for a more robust pediatric medicines development is also present in ClinicalTrials.gov, which reported that more than 770 industry-sponsored pediatric clinical trials are underway and that this number is increasing.¹

While achievements have been shown, the development strategies that can reduce the amount of evidence needed to make conclusions about the efficacy of a drug in children are still disjointed. Potentially because stakeholders are unclear about what is required in many circumstances or there is a lack of foresight in what development is achievable within a reasonable time. For example, the protracted negotiations on using extrapolation with regulatory agencies have defaulted to a checkbox exercise following a particular regulatory procedure and less thought-out clinical development plans. Between 2015 and 2019, the PDCO handled over 200 modifications of Paediatric Investigational Plans decisions per year. “The most common major [PIP] modifications involved changes to timelines (e.g., delays in the study completion), followed by sample size reduction, issues in the planning or conduct of studies, as well as the need to modify the number of design of studies, which may lead to late submission of data as well as trials that are insufficiently powered” (European Commission 2017).

Consideration of pediatric needs should occur early in the clinical development lifecycle to facilitate streamlined development including use of extrapolation, trial simulation, and other innovative trial designs. The current situation calls for careful planning of pediatric drug development as an integral part of the clinical development of drugs, particularly when the diseases for which the drug is being developed in adults also appear in children. The regulatory decision to require pediatric studies through US PREA and the EU Pediatric Regulations already implicitly assumes that the adult indication or condition exists in the pediatric population (i.e., the disease is “sufficiently similar” to establish a sufficient prospect of clinical benefit and justify the risks) and that clinical benefit observed in adults may apply to children as well. This observation provides background on why extrapolation should be a default strategy unless proven not warranted. This also provides the scientific justification for the use of aggregative analytic methods, e.g., information borrowing/bridging through Bayesian and meta-analytic methods, for obtaining efficacy conclusions in children.

In this issue, we have spent sufficient effort to create a holistic view of pediatric drug development and discuss avenues for feasibility and efficiency from the perspective of various statisticians in the industry. The issue is initially framed by a survey on the recent use of extrapolation in medicinal product development in the United States between 2015 and 2021 by Ye et al. (2023). Previous attempts to characterize extrapolation have been made by Dunne et al. (2011). This paper describes what has changed in pediatric extrapolation approaches since many laws to require or encourage pediatric development have been made permanent. Subsequently, there are various methodological papers that provide efficiency on trials through reduction in the extent of evidence generation in the pediatric cohort. One is on the use of win ratio as applied to a pediatric trial that balances benefit and risk as a primary objective (see Seifu et al. 2022). Win ratio has been shown to be efficient previously in composite endpoints, but this application takes it further to having endpoints that have incongruent component outcomes (see also Liao et al. 2022). Gamalo et al. (2023) discussed a novel borrowing of information in the frequentist sense through composite likelihoods and proposed statistical but clinically meaningful guardrails through borrowing bounds. Sebastien et al. (2022) proposed a method to construct prior distribution for the parameter of interest, not directly constructed from the efficacy results of the efficacious dose in adult patients but using pharmacodynamic modeling of a bridging biomarker using early phase pediatric data. Such is a clever approach and could pave the way to using validated physiological models to reduce the amount of data needed in pediatric trials. Majumdar et al. (2023) described a case study in pediatric acute lymphoblastic leukemia (ALL) where they examined the utility of propensity score-matched mixture and power priors in bringing appropriate external adult efficacy information into pediatric trial efficacy assessment and presented considerations for scaling fixed borrowing from external adult data. In a similar investigation, Spanakis et al. (2022) discussed a rigorous stepwise approach on how to address issues within the Bayesian framework, specifically a comprehensive quantitative framework is proposed to assess the extent, synergy, and impact of borrowing. There are also case studies and discussions on the use of transition designs (e.g. Yi et al. (2023)), platform trials (e.g., Uguen et al. (2023)), RWD (O’Connell et al. (2023), Lasky and Chakravarty (2022), starting dose selection (Ye et al. 2022), and clinical trial considerations in pediatric oncology trials (Cooner et al. (2023)). Additionally, perspectives on regulatory considerations on Bayesian methods by Travis et al. (2023).

In addition, a discussion of practical considerations and recommendations are provided on how to assess growth and development based on data collected from clinical trials in pediatric patients in a seminal paper by Choi et al. (2022). The endpoints and measures related to growth, sexual maturation, and neurocognitive development are discussed. Basic analysis approaches are also recommended. Gamalo et al. (2023) discussed safety review in pediatric trials, including specific types of safety assessments and precision on the estimation of event rates for specific adverse events (AEs) that can be achieved. They argue that the assessments which can provide a benchmark for the use of extrapolation of safety should focus on on-target effects. They also explore a unified approach for understanding precision using Bayesian approaches as the most appropriate methodology to describe or ascertain risk in probabilistic terms for the estimate of the event rate of specific AEs.

Lastly, collaboration with regulators to proactively demand for efficiency is needed when they think that a plan is not well thought out. A clinical development plan that does not propose efficiencies in operation (e.g., reducing patient burden and inclusion) and innovative approaches in trial design risks not being able to complete planned trials, or to complete them within a reasonable time. Both the sponsors and regulatory agencies need to assess the role for innovative approaches in every Pediatric Investigation Plan, Pediatric Study Plan, and Proposed Pediatric Study Request they submit and receive, or Written Requests they issue. For development proposals that already provide innovative approaches, encouragement is needed to make them more efficient and robust to inform on the appropriate use of therapies across pediatric populations. Examples of successful innovative plans and regulatory interactions can be a nudge toward further innovation. It is to the best interest of all to have an expedited pediatric drug development. Society needs to keep pediatric drug development aligned with the speed of innovation in medicines. Being innovative and efficient in pediatric drug development is good stewardship in this high unmet need field.

Conflict of interest statement

YounJeong Choi is an employee of Roche/Genentech.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.

Notes

1. ClinicalTrials.gov search using status = recruiting or not yet recruiting, age = child, study type = interventional, study phase = phase 3–4, funder type = industry. First posted on or before December 31, 2020