Page 82 - Read Online

Page 124 De Mattia et al. Cancer Drug Resist 2019;2:116-30 I http://dx.doi.org/10.20517/cdr.2019.04

5,10-methylenetetrahydrofolate reductase
MTHFR, is another key enzyme for intracellular folate homeostasis and metabolism, catalyzing the irreversible
conversion of 5,10-methylentetrahydrofolate, required for DNA synthesis, to 5-methyltetrahydrofolate, the
primary methyl donor indispensable for nucleic acid methylation [Figure 1] [31,32,60,61] . Two non-synonymous
MTHFR variants, rs1801133 and rs1801131, are among the most studied genetic markers for identifying
predictors of FPs-related severe toxicity. These polymorphisms, in high LD, were associated with a decreased
enzyme activity, an increased level of homocysteine and an altered distribution of folate, with a synergic
effect on the haplotype arrangement [31,62] . It could be supposed that a reduced MTHFR functionality due
to the polymorphic variants might lead to an increased 5,10-methyleneTHF concentration that, enhancing
the formation and stability of the ternary inhibitory complex, might increase the cytotoxic effect of 5-FU or
capecitabine and therefore the rate of adverse effects. This hypothesis was tested by some pharmacogenetic
studies, though without generating the robust and validated evidences necessary for supporting the use
of rs1801133 and rs1801131 markers in the clinical practice. Particularly, three meta-analyses [39,54,55] were
performed integrating the data derived from published studies, that present mainly a small cohort size, in
order to clarify the role of MTHFR polymorphisms in predicting toxicity risk in CRC patients receiving
FPs. Globally, MTHFR rs1801133 and rs1801131 variants were not found to have any association with FPs-
induced toxicity, with an exception of a small protective effect of rs1801133 on the risk of neutropenia after
[63]
bolus 5-FU monotherapy in Caucasian patients. However, the more recent study by Pellicer et al. has
revived the interest about a potential role of MTHFR polymorphisms as predictors of FPs-related toxicity. In
fact, this study using a novel approach based on exome sequencing of samples derived from CRC patients
exhibiting extreme toxicity phenotypes, and a subsequent validation of the most promising markers on a
large set of capecitabine-treated patients, has reported thatMTHFR rs1801133, among other genetic markers,
is still one of the most significantly associated with the delayed administration of chemotherapy due to
toxicity. In another recent comprehensive analysis of genetic variants in folate-mediated one-carbon pathway
on more than one thousand CRC patients receiving 5-FU-containing regimes, the same MTHFR rs1801133
[64]
polymorphism has been also reported to be associated with OS . These data claim further investigations
with an adequate statistical power and a homogeneous cohort of patients and treatment modality to
definitively clarify the utility ofMTHFR polymorphisms as parameters for toxicity risk stratification.

Other pharmacogenes
Preliminary results highlighted a potential role of the genetic variants in thymidine phosphorylase (TYMP)
and uridine monophosphate synthase (UMPS), related to the folate homeostasis; ABCB1 and SLC22A7
involved in the drug transmembrane translocation; and cytidine deaminase (CDA) enzyme necessary for
capecitabine activation, as toxicity markers in FP-treated patients [39,59,63,65,66] . Also variants in carboxylesterase
1 (CES1) and CES2, required for capecitabine metabolism, were considered as FP pharmacogenetic markers,
but without generating positive results [39,58,63] .

The protein product of UMPS gene is the orotate phosphoribosyltransferase enzyme, which catalyzes the
conversion of 5-FU into fluorouridine monophosphate, a common substrate for the production of cytotoxic
metabolites that target RNA and DNA. Studies on Japanese CRC patients receiving 5-FU as adjuvant
chemotherapy reported that the functional missense variant rs1801019 in UMPS was associated with more
severe symptoms of GI toxicity such as grade 3 diarrhea and grade 2-3 anorexia [65,66] . However, the predictive
[67]
role of this UMPS marker on GI side effects was not reproduced in a group of Caucasian CRC patients .

The work of García-González et al. evidenced a potential predictive role on capecitabine-related toxicity
[59]
also for other genetic markers. Particularly, the ABCB1*1 haplotype (rs1128503-C, rs2032582-G, and
rs1045642-C) was associated with a high risk of grade > 2 diarrhea and overall toxicity (any adverse drug
reactions classed as grade 3 or higher) as well as with a delay in drug administration or capecitabine dose
reduction, while the CDA rs2072671-A was associated with a high risk of overall toxicity. A score based
on ABCB1-CDA polymorphisms was also developed and showed to efficiently predict patients at high risk