Page 163 - Read Online

Page 8 of 25 Battaglin et al. J Cancer Metastasis Treat 2018;4:12 I http://dx.doi.org/10.20517/2394-4722.2018.04

leading to a decreased enzymatic activity have been found to correlate with the risk of 5-FU and capecitabine
severe toxicities in several pharmacogenetic studies. Over 30 single nucleotide polymorphisms (SNPs) in
the DPYD gene have been studied over the last 20 years, although many of these variants did not appear to
have any functional effect. Among the most well-known, the c.2846 A>T and c.1679 T>G variants, alongside
the G>A mutation (DPYD*2A) of the invariant splice site in exon 14 (IVS14+1G>A), coding for a truncated
protein with no enzymatic activity, have been consistently associated with decreased DPD activity and a
[114]
4-fold increase of risk of developing 5-FU related toxicities . DPYD*2A is the most frequent SNPs in
the Caucasian population, nevertheless its incidence is low (about 1%-2% for the heterozygote genotype)
and shows substantial ethnic variations. Homozygous for DPYD*2A have been associated with cases of
lethal toxicities in patients treated with fluoropyrimidine-based chemotherapy [115,116] . More recently a large
[117]
meta-analysis from Meulendijks et al. confirmed the predictive role for drug-related toxicities for four
DPYD variants: DPYD*2A, c.2846A>T, c.1679 T>G and c.1236G>A/haplotype B3. Data from retrospective
pharmacogenetic analyses from the Italian adjuvant TOSCA trial confirm the role of DPDY*2A as a risk
factor for fluoropyrimidine-related toxicities . Additionally, a prospective study enrolling 2,038 patients
[118]
candidate to receive a fluoropyrimidine-based chemotherapy demonstrated the feasibility and cost-
effectiveness of upfront DPYD*2A genotyping before treatment start. DPYD*2A variant allele carriers were
treated with a reduced dose-intensity leading to a significant reduction of the risk of grade ≥ 3 toxicity
(28% vs. 73% in historical controls, P < 0.001) and a reduction of drug-induced death from 10% to 0% .
[119]
The low frequencies of the aforementioned risk alleles, however, cannot fully explain the estimated risk of
DPD-linked fluoropyrimidine-related adverse events, underlining the complex multi-level modulation of
DPD activity, involving both transcriptional and post-transcriptional mediators, and the need to investigate
additional DPYD risk variants. Nevertheless, available data support the role of DPYD testing as a pre-
treatment screening in patients undergoing 5-FU and capecitabine treatment in order to improve the safety
of fluoropyrimidine-based therapies and potentially allow genotype-guided dose adaptations, as recently
recommended by the clinical pharmacogenetics implementation consortium .
[120]
Evidence on the role of DPD deficiency as a toxicity biomarker led the FDA to include a warning annotation on
the label of fluorouracil for patients with low or absent DPD activity, recommending to withheld or permanently
discontinue fluorouracil in patients with evidence of acute early-onset or unusually severe toxicity, which may
indicate near complete or total absence of DPD activity. On the other hand, latest published ESMO clinical
practice guidelines on metastatic colorectal cancer management suggest for the first-time pre-treatment DPYD
testing as an option . This indication, however, is focused on those patients who experience severe 5-FU toxicity
[9]
before 5-FU re-introduction and routine testing is not recommended, despite the authors stating that patients
with known partial DPD deficiency benefit from dose adaptation of 5-FU/capecitabine therapy to avoid severe
toxicity, while in patients with complete DPD deficiency fluoropyrimidines should be avoided and an alternative
treatment offered. The lack of recommended standardized assessment techniques represents an additional issue
to the introduction of routine DPD testing.

The predictive role of genetic variants in other key genes involved in the folate pathway, such as TYMS and
5,10-methylenetetrahydrofolate reductase, has not been validated and their use in clinical practice is not
recommended.

UDP-Glucuronosyltrasferase A1
Irinotecan, a topoisomerase I inhibitor, is another key drug in the chemotherapy treatment of mCRC, which
can be used as a monotherapy or in combination with 5-FU and/or other agents in different treatment
lines [9,10] . This agent is administered as a pro-drug which is metabolized to its active form, SN-38, via
carboxylation. SN-38 catabolism and excretion are subsequently dependent on conversion to its inactive
form, SN-38G, operated by hepatic UDP-Glucuronosyltrasferases (UGT) such as UGT1A1 . Additionally,
[121]
the pharmacokinetics of irinotecan involves several other enzymes, such as CYP3A4, which control its