Childhood Cancer Drug Development: Why Alternative Designs for Drug Development are Required

In the US, the regulatory environment, through the Research to Accelerate Cures and Equity for Children Act (RACE for Children Act) requires FDA to develop a list of molecular targets and molecular targets of new drugs and biologics in development, that are determined to be substantially relevant to the growth and progression of pediatric cancer, and that may trigger the requirement for pediatric investigations. The intent is to engage Pharma in pediatric testing at an early stage in drug development. Under the RACE for Children Act, FDA may now require pediatric assessments when molecular targets under FDA review are substantially relevant to childhood cancer. This may require evaluation of up to 200 entities each year. However, it is clear that because a cancer has a particular ‘molecular driver’ it does not necessarily predict tumor response to a targeted agent. For example, some patients with low‐grade BRAF‐driven glioma have dramatic responses to MEK inhibitors, whereas others have some disease stabilization. Thus, factors other than the driver‐mutation appear important to tumor response. Consequently, to encompass clinical heterogeneity (genetic/epigenetic) that may determine response to treatment, a large number of models having similar molecular characteristics will have to be developed to allow adequate pre‐clinical evaluation of new targeted entities. Previously, the Pediatric Preclinical Testing Program (PPTP) conducted a retrospective analysis of over 2100 studies using 83 PDX and xenograft models and 67 drug entities undertaken using conventional testing (10 mice/group for solid tumor PDX models, 8 mice/group for leukemia PDX models). This analysis showed that use of one mouse per treatment group would accurately predict the group median response in 78% of experiments. Allowing a deviation of + or − one response category, the accuracy was 95% (Murphy B. et al, Cancer Res. 2016). Here we have prospectively tested the ‘single mouse’ experimental design to evaluate both small molecule and antibody‐drug conjugate drugs. We conclude that the ‘single mouse’ approach is highly accurate, and yields more information than traditional experimental approaches, particularly with respect to identifying biomarkers that associate with tumor response. Further, within the current resource constraints, this design allows up to 20 PDX models to be used for every one model used in the conventional testing design. Thus, the single mouse design potentially allows incorporation of disease heterogeneity that may be required for evaluating new agents under the RACE for Children Act. Support or Funding Information NO1‐CM42216, UO1CA199297, CA165995 from NIH, and CPRIT RP160716