Original ResearchRevisiting the definition of dose-limiting toxicities in paediatric oncology phase I clinical trials: An analysis from the Innovative Therapies for Children with Cancer Consortium

- A new set of dose-limiting toxicities (DLTs) and recommended phase II dose (RP2D) definitions have been issued for molecularly targeted agents' adult phase I trials.
- We analysed the Innovative Therapies for Children with Cancer Consortium dose-finding trials and conducted a survey with among expert pediatricians.
- Several proposals for adult dose-finding trials are translated in paediatric studies.
- DLTs should be homogeneously defined across trials.
- Definition of RP2D should include delayed toxicities and supported by pharmacokinetics/pharmacodynamics data.

BackgroundDose-escalation trials aim to identify the maximum tolerated dose and, importantly, the recommended phase II dose (RP2D) and rely on the occurrence of dose-limiting toxicities (DLTs) during the first treatment cycle. Molecularly targeted agents (MTAs) often follow continuous and prolonged administrations, displaying a distinct toxicity profile compared to conventional chemotherapeutics, and classical DLT criteria might not be appropriate to evaluate MTAs' toxicity. We investigated this issue in children.MethodsThe Innovative Therapies for Children with Cancer Consortium (ITCC) phase I trials of novel anticancer agents between 2004 and 2015 were analysed. Data from investigational product, trial design, items defining DLT/RP2D were extracted. A survey on dose-escalation process, DLTs and RP2D definition was conducted among the ITCC clinical trials committee members.ResultsThirteen phase I trials with 15 dose-escalation cohorts were analysed. They explored 11 MTAs and 2 novel cytotoxics; 12 evaluated DLT during cycle 1. Definition of DLT was heterogeneous: Grade III-IV haematologic toxicities that were transient or asymptomatic and grade III-IV non-haematological toxicities manageable with adequate supportive care were often excluded, whereas some included dose intensity or grade II toxicities into DLT. None of the studies considered delayed toxicity into the RP2D definition.ConclusionDLTs should be homogeneously defined across trials, limiting the number of exceptions due to specific toxicities. Dose escalation should still be based on safety data from cycle 1, but delayed and overall toxicities, pharmacokinetic parameters and pharmacodynamic data should be considered to refine the final RP2D. The evaluation of long-term toxicity in the developing child cannot be adequately addressed in early trials.