Development of Novel Therapy for Childhood Cancer: Promise and Challenges

Only a limited number of phase 1 and phase 2 trials of new agents can be undertaken in the childhood cancer population. The Pediatric Preclinical Testing Program (PPTP) is an experiment to determine whether using xenografts derived from childhood cancers can be used to identify agents, or combinations of agents that may have significant clinical activity.

Thirty one drugs or biological agents were tested against the PPTP in vivo panel. Tumors were characterized by expression and SNP profiling.

By expression profiling xenografts clustered with histotype and their clinical counterpart. By SNP analysis gains and losses of copy number correlated with clinical samples. Thirty one agents have been evaluated against the in vivo primary screen (45 solid tumors and 8 ALL xeonografts). Known clinically active drugs (vincristine, cyclophosphamide, cisplatin, topotecan) were identified as active in the models (30.7% response rate). In contrast the objective response rate for 'experimental' agents was 9.4%. Notable activity was observed for the Auroroa kinase A inhibitor (MLN8237), Seneca Valley virus (NTX010), and a CENP inhibitor (GSK92395A).

Molecular characterization of the PPTP models indicates high fidelity with the clinical samples. The in vivo screen identifies known active agents, and prospectively has identified agents that have significant antitumor activity that is histotype-selective. The challenge will be to use the molecular characterization data to rationally select agents with significant activity. One such example is the MEK1 inhibitor in BRAF-mutant pilocytic astrocytomas.