High-dose DFMO alters protein translation in neuroblastoma.

Abstract

DFMO has been studied as a cancer therapeutic at doses ranging from 500 to 9,000 mg/m2/day. Lower doses are favored for cancer prevention studies while higher doses, often with chemotherapy, are studied in refractory cancers. DFMO inhibits the rate-limiting enzyme in polyamine synthesis, ornithine decarboxylase (ODC), an oncogene transcriptionally regulated by MYC. MYC genes are the principal oncogenic drivers of neuroblastoma, and ODC1 is co-amplified in a subset with dismal outcome, so DFMO is a rational therapeutic candidate. Low-dose DFMO has now been FDA-approved for high-risk patients though the mechanisms for its anti-tumor activity, and the exposures required to elicit them, remain obscure. We sought to define biomarkers of activity across exposures achieved in the clinic with low through high-dose DFMO. Polyamines support protein translation by providing spermidine, which is essential to hypusinate (and activate) the elongation factor, eIF5A. Selective binding of polyamines with tRNA and rRNA provide eIF5A-independent mechanisms of translation support. We show that low-dose DFMO does not extend survival in mouse models in vivo nor alter translation biomarkers in vitro. High-dose DFMO consistently extends survival in neuroblastoma models, and, in a subset of neuroblastoma cell lines, inhibits eIF5A hypusination and global translation at achievable concentrations. However, the concentration required to engage these changes across many cell lines exceeded that achievable even with high-dose DFMO. No correlation was seen among MYCN and/or ODC1 copy number and sensitivity to DFMO. Combining high-dose DFMO with additional agents to further deplete tumor polyamines may be necessary to fully engage polyamine-depletion effects on tumors, and more granular measures of translation, including codon-resolution ribosome profiling, may be required to define these effects.