Background: Running clinical trials in relapsed neuroblastoma is challenging due to its rarity and involve a long process from concept to publication. Given the paucity of reliable evidence, there remains a need to identify optimal conventional therapies as well as evaluating novel agents quickly as they become available.
Design: Most randomised controlled trials (RCT) consist of two arms, while there have been many single arm studies in neuroblastoma. If several treatments are of interest, rather than just selecting one (possibly an ineffective one) for evaluation against the standard, a MAMS design permits several treatments to be assessed in a single trial; insufficiently active treatments in terms of intermediate outcome (e.g. response) are dropped after Phase II; recruitment continues in the remaining experimental arms for Phase III evaluation of event-free (EFS) and overall survival. Such adaptive designs are very efficient and, being randomised and comparative, avoid the biases of single-arm Phase II trials and historical controls.
Results: A prospective MAMS design is planned for the next RCT in relapsed neuroblastoma (BEACON-2). Once through Phase 1 dose-finding in children, agents given top priority by the Neuroblastoma New Drug Development Strategy project will be evaluated, with stratification by molecular profile where appropriate. Potential agents may target: ALK, BET, TORC, MDM2, MEK, CHK1, BIRC5, PD-1/PD-L1, CDK4/6. Optimal chemotherapy will continue to be identified. Ancillary biological studies will enable biomarker analyses. Preliminary design parameters and patient numbers per arm (80% power) are: Phase II, one-sided alpha=0.2, n=70 assuming baseline response rate 25% and 15% improvement; Phase III, two-sided alpha=0.05, n=130 assuming baseline 6-month EFS 50% and 17.5% difference.
Conclusion: The adaptive design of BEACON-2 will allow multiple agents/regimens to be evaluated quickly and efficiently, with ineffective ones being dropped and new ones brought in as they become available, in a rolling long-term programme of research.