Abstract
Introduction: Deleterious variants in the DPYD gene result in a deficiency of the dihydropyrimidine dehydrogenase (DPD) enzyme. This increases the risk of life threatening normal tissue toxicity after administration of the anticancer drug 5 fluorouracil (5-FU). The relationship between DPYD pharmacogenes and metabolism of 5-FU by human liver has not been extensively reported.
Methods: Thirteen human livers were characterised for DPYD gene variants using Sequenom iPLEX, DPD protein expression by immunoassay, and endogenous uracil concentrations by LC/MS. Liver cytosol samples (5 mg/ml) were incubated in replicate (n= 4) with 5-FU (2 microM) and progress curves used to determine as degradation of 5-FU using HPLC analysis.
Results: A wide range of 5-FU metabolism was found across the livers, including five null-low activity livers and one very high activity liver. One liver was heterozygous for a known deleterious function variant (IVS14+1G>A, rs3918290). This liver had no detectable activity for 5-FU metabolism. However, no deleterious variants were detected in the other null function livers. Six livers were carriers of C29R (rs 1801265) and one liver was a carrier of M166V (rs 2297595). No obvious pattern of metabolism was associated with these variants. Levels of uracil also varied widely across the livers (range 7.89-214 microM). When the livers were categorised as either null, intermediate or high activity for 5-FU degradation there was a significant negative association with hepatic uracil concentrations (P <0.05).
Conclusions: The extremely high endogenous levels of uracil in some livers relative to the pharmacologically relevant 5-FU concentration appears to lead to almost complete inhibition of DPD activity. The ability of endogenous uracil to cause phenoconversion resulting in a "poor metaboliser" phenotype should not be overlooked as this could have substantial impact on accuracy of genetic testing for DPYD variants to predict risk of 5-FU toxicity.
