With the recent expanse of technology
With the recent expanse of technology to evaluate synthetic lethality, new combinations with DHODH inhibitors may be uncovered (O\'Neil, Bailey, & Hieter, 2017). Many oncogenic backgrounds share synthetic lethality with DHODH. Cells with mutant PTEN have increased glutamine metabolism and are sensitive to DHODH inhibition (Mathur et al., 2017). Teriflunomide was synergistic with the BRAF(V600E) inhibitor PLX4720 in melanoma M871 (White et al., 2011). KRAS mutant cells were found to be more sensitive to DHODH inhibition over KRAS WT cells (Koundinya et al., 2018). Interestingly, DHODH inhibitors were found to increase p53 synthesis and were synergistic with an MDM2 inhibitor (Ladds et al., 2018). While more studies are needed, these results suggest that targeting pyrimidine biosynthesis with DHODH inhibitors may be a pivotal point of exploitation in specific genomic backgrounds.
Future DHODH-targeted therapy may be improved by identifying patient populations that will be responsive to DHODH inhibition. The recent decrease in the cost of genomic profiling makes effective personalized medicine increasingly feasible and may be used to develop and test biomarkers that predict DHODH sensitivity. Several biomarkers have been identified, including mTORC1. In fludarabine-resistant cells, mTORC1 is overexpressed, and these cells were surprisingly sensitive to both leflunomide and other inhibitors of the de novo pyrimidine pathway (PALA) (Sharma et al., 2014). A similar response was observed with inhibitors of de novo purine biosynthesis suggesting a potential dependence of cells expressing mTORC1 on de novo nucleotide biosynthesis pathways (Valvezan et al., 2017). In addition to mTORC1, PTEN may also be a valuable biomarker. Cell lines with mutant PTEN, a known hallmark of resistant prostate and breast cancer, were remarkably sensitive to both brequinar and leflunomide (Keniry & Parsons, 2008; Mathur et al., 2017). While more data are needed, DHODH-targeted therapy will be more effective if validated biomarkers are available to predict responsive patient populations.
DHODH-targeted therapy may also be better suited for non-solid tumors. Previous clinical trials focused primarily on solid tumors and showed minimal objective responses. Brequinar, for example, was never evaluated clinically in AML patients. As myelosuppression was a common side effect of brequinar therapy, DHODH inhibition may be more efficacious in patients with leukemia (de Forni et al., 1993; Schwartsmann et al., 1990). In fact, the extent of myelosuppression observed during clinical trials was enough to consider brequinar as a potential immunosuppressant (Chastanet et al., 1998; D\'Silva et al., 1996; Makowka, Sher, & Cramer, 1993; Wang, Qu, Stepkowski, Chou, & Kahan, 1996). DHODH inhibition is known to suppress the immune system as two DHODH inhibitors, leflunomide and teriflunomide, are FDA approved for autoimmune diseases, including rheumatoid arthritis (Sanders & Harisdangkul, 2002). Therefore, DHODH-targeted therapy may be more effective in leukemia, and current clinical trials are investigating this hypothesis.
Conflict of interest statement
Acknowledgments Support for this work was provided by the following sources: UMCCC Core Grant (P30 CA046592). C.R.C. is a trainee of the University of Michigan Pharmacological Sciences Training Program (PSTP, T32-GM007767). We thank Professors David Lombard, Costas Lyssiotis, and Daniel Wahl for their critical reading of the manuscript.
We have developed a novel series of DHODH inhibitors based on a lead, which came out of a docking procedure using 4Scan technology and medicinal chemistry exploration. The activity of the initial lead was improved by a QSAR method and yielded low nanomolar inhibitors. Such compounds promise to have great potential as treatment for autoimmune diseases such as rheumatoid arthritis, graft versus host disease, and multiple sclerosis. To further explore the scope of DHODH inhibitors of types – (), a replacement of the cyclopentene ring with other small aromatic systems was envisaged. Our X-ray structure indicated that the cyclopentene ring lies virtually planar in a pocket close to the FAD cofactor. Flat small aromatic rings might therefore be of benefit for binding and activity.