Paediatric clinical trials often face limitations that undermine their impact: small sample sizes, ethical constraints, high costs, and prolonged timelines. These challenges are particularly severe in neonatology and rare childhood diseases, where enrolling a sufficiently powered cohort can be nearly impossible. As a result, findings often lack statistical significance, which can delay or dilute clinical implementation.
Emerging digital innovations offer promising alternatives. Tools such as digital twins, synthetic patient data, and in-silico trials are reshaping the conceptualization and conduct of clinical trials in children.1 These technologies simulate real-world patient data and responses without subjecting actual children to experimental risks, thereby addressing both ethical and logistical barriers.
Digital twins are virtual replicas of individual patients built using multidimensional health data and AI-driven models. These patient-specific models can simulate disease progression and treatment responses, allowing for personalised virtual trials. In neonates, for instance, where physiological variability is vast, digital twins can provide nuanced simulations that reflect this diversity—something conventional control groups often fail to achieve.
Synthetic patient data, created through advanced generative algorithms, replicate the statistical properties of real datasets without exposing identifiable information. These data can supplement trial cohorts or serve as synthetic controls, especially when patient recruitment is difficult. This approach preserves privacy while improving trial feasibility and speed.
In-silico trials go a step further by simulating entire clinical trials using computational models. For rare paediatric diseases, where real-world trials are constrained by extremely low incidence, these simulations can predict outcomes with reasonable accuracy. They also enable the rapid testing of multiple scenarios, helping to refine trial designs before they are implemented live.
Despite the promise, challenges remain. Synthetic and simulated data cannot yet fully capture unanticipated adverse events. The placebo effect, which is significant in pediatric outcomes, may be absent in virtual models. Ethical and legal frameworks for consent, data ownership, and regulatory compliance are still evolving and must be addressed before widespread adoption can occur.
Nevertheless, these virtual tools offer the potential to make paediatric research faster, safer, and more equitable. By reducing reliance on traditional trials alone, they can accelerate evidence generation and expand access to tailored treatments for children. Realizing this potential will depend on collaborative efforts among clinicians, AI developers, ethicists, and regulators to ensure the safe, responsible, and effective integration of AI into clinical research.
Source: Pammi M, Shah PS, Yang LK, Hagan J, Aghaeepour N, Neu J. Digital twins, synthetic patient data, and in-silico trials: can they empower paediatric clinical trials? Lancet Digit Health. 2025;7:100851.
Please login to comment on this article