Page 305 - Personalised medicine of fluoropyrimidines using DPYD pharmacogenetics Carin Lunenburg

Introduction
Fluoropyrimidines, including 5-fluorouracil (5-FU) and capecitabine, represent the backbone of chemotherapeutic regimens used to treat solid tumours, such as colorectal and breast cancer. Severe (grade ≥3) fluoropyrimidine-induced toxicity can occur in up to 30% of the patients, depending on the treatment regimen and may even be lethal in up to 1% of the patients experiencing toxicity.1,2 Common fluoropyrimidine-induced adverse events are diarrhoea, mucositis, hand-foot syndrome and myelosuppression.1,3 Dihydropyrimidine dehydrogenase (DPD) plays a key role in the degradation of 5-FU into inactive metabolites4 and is encoded by the gene DPYD. Both DPD and genetic variants in DPYD have been widely investigated to explain severe fluoropyrimidine-induced toxicity. Recently, we have shown that prospective genotyping and dose reduction based on four variants in DPYD (DPYD*2A, c.2846A>T, c.1679T>G and c.1236G>A) reduces severe fluoropyrimidine-induced toxicity in these DPYD variant allele carriers.5 These four variants were selected based on previous studies and meta-analyses in which the association with fluoropyrimidine-induced toxicity was reported.6-13 Nonetheless, severe toxicity did still occur in 23% of patients wild-type for these four variants, showing that other genetic variants or non-genetic factors may play a role in the onset of severe toxicity.5
Variants in genes other than DPYD could also play a role in the onset of severe fluoropyrimidine-induced toxicity. Previously, research to identify genetic variants has been conducted based on the pharmacological background of fluoropyrimidines, for example in pathway analyses or candidate gene studies. Several variants in CDA (cytidine deaminase), CES1 (carboxylesterase 1), TYMS (thymidylate synthase), MTHFR (methylenetetrahydrofolate reductase), ENOSF1 (enolase superfamily, member 1), SLC22A7 (solute carrier family 22, member 7), UMPS (uridine monophosphate synthase) and TYMP (thymidine phosphorylase) genes were previously identified and associated with severe fluoropyrimidine-induced toxicity.1,14-21 However, genome-wide association studies (GWAS) have the potential to identify novel variants without making assumptions based on a pharmacological background. Previously, O’Donnell et al. executed a GWAS on 503 cell lines to identify novel single nucleotide polymorphisms (SNPs) associated with capecitabine sensitivity.22 Five variants showed genome-wide significance in this cell-line based GWAS, but replication in 268 patients only showed an association with sensitivity for capecitabine for ADCY2 rs4702484.23 Fernandez-Rodzilla et al. analysed data of 221 colorectal cancer patients treated with 5-FU or FOLFOX (folinic acid, 5-FU and oxaliplatin).24 Seven SNPs were associated with adverse drug reactions, yet none reached the genome-wide significance level. Low et al. executed a GWAS on 13,220 patients in total, of which 1,460 patients received 5-FU, focussing on neutropenic and leukopenic toxicities.25 For 5-fluorouracil, they identified four SNPs associated to neutropenia and leukopenia, yet none reached the genome-wide significance level. We conducted a GWAS to discover novel genetic variants associated with the onset of severe fluoropyrimidine-induced toxicity.
12
Genome-wide association study
303