Adaptive Trial Design AI in Clinical Trials alcohol use disorder Alzheimer's Disease Article Artificial Intelligence in Clinical Trials Cell and Gene Therapy Clinical Operations Clinical Research Clinical Trial Design Clinical Trial Innovation Clinical Trial Operations Clinical Trial Quality Clinical Trials CLINOPS WATCHDOG CNS Trials Conference Coverage Data Analysis Data Management DCT Decentralized Clinical Trials DHT eClinical Tech News EMA Executive Interviews FDA Guidance FDA Regulatory Strategy GCP Compliance GLP-1 Receptor Agonists Immuno-Oncology Ketamine Therapy Neuromodulation Oncology Trials Opinion Pediatric Oncology Precision Oncology Protocol Design Rare Disease Real-World Evidence semaglutide site activation Site Operations site readiness study startup Treatment-Resistant Depression
Picture a pediatric oncologist in 2024 walking her patient — a nine-year-old with a relapsed solid tumor harboring an NTRK gene fusion — through the options. One of those options is larotrectinib (VITRAKVI), which received accelerated FDA approval in November 2018 on the strength of a histology-agnostic basket trial that enrolled both adults and children. She has a drug. But for every child sitting across from her with a different molecular target — one whose tumor biology matches a compound already approved for adults — the honest answer is: wait. Probably for years.
A call to industry published in Nature Reviews Drug Discovery puts a number on that wait: a median of 6.5 years longer than adults to access new drugs, with a range extending to 27.7 years between first-in-human and first-in-child trials. That figure is not a data artifact from an earlier era of neglect. It reflects how drug development is still being designed today — with pediatric populations treated as a downstream regulatory obligation rather than a co-primary population from the first protocol draft. The child waits because the trial architecture was never built to include her.
The structural problem runs deeper than compassion or corporate will. It lives inside the protocol itself.
The FDA’s FDARA Implementation Guidance for Pediatric Studies of Molecularly Targeted Oncology Drugs amended Section 505B of the FD&C Act to require sponsors of molecularly targeted oncology drugs to submit Pediatric Study Plans (iPSPs) early in development. The intent was correct: force sponsors to think about children before the adult NDA is already in review. The execution reveals a gap. An iPSP is a planning document. It does not require co-enrollment. It does not mandate molecular sub-study arms. It does not prevent a sponsor from submitting a plan that defers every pediatric study until after adult approval — and then discovering, years later, that the adult dose is pharmacokinetically incompatible with pediatric metabolism or that the tablet formulation is physically impossible for a five-year-old to swallow.
The counterintuitive read here is this: the regulatory machinery designed to protect children from being ignored has, in practice, given sponsors a compliant path to ignoring them. Submit the plan. Get the adult approval. Start the pediatric work. Wait 6.5 years. The FDA gets its checkbox; the child gets a wait.
Compare that to what happened when Bayer and Loxo Oncology designed larotrectinib’s pivotal basket trial. They enrolled pediatric patients alongside adults from early phases, treating the NTRK fusion as the population-defining characteristic rather than the histology. The result: a single trial that generated regulatory-grade efficacy evidence across multiple tumor types in both age groups. The FDA cleared it under accelerated approval for the broadest labeled population possible. That is not a lucky outcome. That is what happens when pediatric patients are embedded in the trial architecture rather than appended to it afterward.
The NCI-COG Pediatric MATCH trial extended this logic further into the adaptive design space. As a Phase 2 precision medicine platform trial, it matched children with relapsed or refractory solid tumors to targeted agents based on molecular profiling — continuously assigning patients to treatment arms based on genomic data rather than running separate histology-specific studies. The design itself functions as a standing infrastructure for evidence generation, one that can incorporate new agents without restarting from a blank protocol. The lesson has been available for years. Most industry sponsors have not absorbed it.
The strongest argument for integrating pediatric patients into adaptive and tumor-agnostic frameworks is not regulatory efficiency. It is biology. Pediatric cancers are predominantly driven by fusion oncoproteins and epigenetic dysregulation rather than the accumulated somatic mutations that characterize most adult malignancies. That means the molecular target — not the tumor’s tissue of origin — is often the most relevant treatment-selection criterion. A child with a ROS1 fusion in a lung primary and a child with a ROS1 fusion in a kidney primary share more biologically relevant similarities to each other than either shares to an adult with a histology-matched but fusion-negative tumor.
Basket trial designs exist precisely to exploit this logic. When a sponsor designs a histology-agnostic Phase 2 basket trial to support an adult NDA, adding a pediatric molecular cohort is not a separate trial. It is one more arm in existing infrastructure — sharing investigational site activations, central molecular profiling, data management architecture, and safety monitoring committees that are already running. The marginal cost is a fraction of a standalone pediatric study. The marginal regulatory value is substantial: the iPSP submission becomes a plan with data behind it rather than a promise of data to come.
The enrollment disparity makes this even more urgent. A CDC-sourced analysis of clinical trial participation at an NCI-designated comprehensive cancer center found that 54% of pediatric patients enrolled in some type of clinical study, compared to only 17% of older adults. Children, when given the opportunity, participate. The limiting factor is access to trials designed to include them — not willingness from families or investigators.
Sponsors who push back on early pediatric integration typically cite three objections: the adult dose has not been established yet, the formulation is not pediatric-appropriate, and the pediatric safety profile is unknown. Each of these is a real operational challenge. None of them is a structural reason to wait until adult approval. Adaptive master protocols can carry pediatric dose-finding cohorts in parallel with adult Phase 1/2 work. Age-appropriate formulation development can begin at IND submission, not after the NDA is filed. The FDA’s regulatory incentives — a six-month exclusivity extension under the Best Pharmaceuticals for Children Act for completing pediatric studies in response to a Written Request, plus Priority Review Vouchers for rare pediatric diseases — exist specifically to offset those development costs. A PRV alone, at recent market valuations exceeding $100 million, can fund the pediatric arm several times over.
The structural shift required is not primarily regulatory. The FDA has guidance. It has incentive mechanisms. What it cannot mandate is that a sponsor’s Phase 1 protocol writer opens a blank document in 2026 and writes “pediatric molecular cohort” into the eligibility criteria before the adult dose-finding work begins. That decision happens inside a company, at a very specific moment in early development, when the protocol team is deciding who gets enrolled. Once that document is finalized without a pediatric arm, the 6.5-year clock starts running.
The larotrectinib precedent and the Pediatric MATCH infrastructure together represent a proof of concept that the field already has. Histology-agnostic basket designs with co-enrolled pediatric cohorts can generate regulatory-grade evidence for both populations in a single study, with shared molecular profiling infrastructure, and within a timeline that closes — rather than extends — the access gap. Entrectinib (Rozlytrek), which also received FDA approval for NTRK fusion-positive tumors in adults and pediatric patients aged twelve and older, followed the same logic: molecular inclusion criteria that do not discriminate by age.
The biology has been pointing in this direction for a decade. The regulatory architecture is in place. The incentives exist. What happens in the next generation of molecularly targeted oncology programs will depend on whether protocol teams treat that nine-year-old’s NTRK fusion as a scheduling problem for post-approval — or as the enrollment criterion she was always supposed to be.
Moe Alsumidaie is Chief Editor of The Clinical Trial Vanguard. Moe holds decades of experience in the clinical trials industry. Moe also serves as Head of Research at CliniBiz and Chief Data Scientist at Annex Clinical Corporation.
Start typing and press Enter to search
