5-fluorouracil, Capecitabine & Raltitrexed
5-fluorouracil has been used in CRC patients for almost 50 years. In combination with leucovorin (folinic acid), it still remains the basis of CRC treatment, both in adjuvant and palliative settings.[8,9] Different treatment regimens (including bolus intravenous injections, continuous infusions, different doses and different schedules) of 5-FU exist; the response rates vary from 20 to 30%. The main toxicities that occur under 5-FU treatment are leucopenia, stomatitis and diarrhea. In addition, in rectal cancer patients 5-FU is used as the drug of choice for both neoadjuvant and adjuvant chemoradiation therapy. Disease-free survival has increased in rectal cancer patients treated with a combination of 5-FU and radiotherapy versus radiotherapy alone.
Activation of 5-FU is supported by orotate phosphoribosyltransferase (OPRT), catalyzing the conversion of 5-FU to 5-fluorouridinemonophosphate (FUMP), which is then converted to active 5-fluorodesoxyuridinemonophosphate (FdUMP).[13,14] After being metabolized to FdUMP, the drug exerts its cytotoxic effect via inhibition of thymidylate synthase (TS), an enzyme essential for DNA synthesis. FdUMP forms an inactive ternary complex with TS and the cofactor 5,10-methylentetrahydrofolate (5,10-MTHF), blocking the essential conversion of desoxyuridinemonophosphate (dUMP) to desoxythymidinemonophosphate (dTMP), thus inhibiting DNA synthesis.[15,16] A further mechanism is through the incorporation of metabolites of 5-FU into RNA and DNA.
Leucovorin, often administered in addition to 5-FU, potentiates the 5-FU-induced inhibition of TS by stabilizing the complex of FdUMP TS MTHF. Synergistic anticancer effects of 5-FU and leucovorin were shown.
Capecitabine, an oral prodrug of 5-FU, is activated to 5-FU, catalyzed by the tumor-associated angiogenic factor thymidine phosphorylase. Therefore, higher concentration levels of 5-FU can be achieved in tumor tissue than in nontumor tissue. Capecitabine has shown comparable or superior results concerning overall survival or response, respectively, and is much more convenient for patients owing to its oral mode of administration.[21,22]
Raltitrexed, also acting as a TS inhibitor, can be given alternatively to 5-FU in patients suffering from severe side effects under 5-FU therapy. Although raltitrexed is not approved by the US FDA, the drug was licensed in Canada and some European countries.
Several genetic variants involved in 5-FU-related pathways have been examined with respect to outcome in CRC patients who received 5-FU based chemotherapy, and in rectal cancer patients who received 5-FU-based chemoradiation ( Table 1 & Table 2 ).
Thymidylate synthase is the main target enzyme of 5-FU. Some studies found high protein levels of TS to be associated with a decreased response to 5-FU; however, results are still inconsistent.[24–26] According to this hypothesis, high TS levels may be associated with decreased benefit from 5-FU treatment, including decreased response, increased overall mortality and increased risk of recurrence and progression, but also lower probabilities of suffering from severe side effects.
Three common polymorphisms have been associated with altered TS gene expression. The variable number of tandem repeats (VNTR) polymorphism mainly leads to double or triple tandem repeats (2R or 3R) of a 28-bp sequence. In our analysis, we set the 2R/2R genotype as the reference group. Four, five or nine tandem repeats have also been described for this polymorphism. The homozygous 2R/2R genotype has been associated with low TS levels in vitro and in tumor tissue samples.[27,28] An improved response has, however, been associated with both the 2R/2R genotype[19,21,29] and the 3R/3R genotype,[8,9,30] respectively. These diverging results have been attributed to the possibly different effects of 5-FU, either by incorporation in RNA or inhibition of TS, depending on the mode of administration. Consistent with the hypothesis of decreased 5-FU benefits associated with the 3R/3R genotype, patients carrying this genotype were found to have a significantly increased overall mortality in one study, and nonsignificantly increased mortality in two others.[9,29] Nevertheless, contradictory results have been reported in two studies that observed a significantly decreased overall mortality for patients with the 3R/3R genotype[32,33] and a nonsignificantly decreased overall mortality in one study, whereas three studies reported no associations between this genotype and overall mortality.[30,35,36] One of two studies presented a significantly decreased risk of recurrence for carriers of the 3R/3R genotype,[33,37] whereas three studies did not present any significant associations between VNTR polymorphism and the risk of progression.[8,34,38] Regarding toxicity, two out of six studies confirmed the hypothesis of a significantly decreased risk for the 3R/3R genotype,[9,39] and one study presented a nonsignificantly decreased risk for carriers of the 3R allele to suffer from severe side effects. One study in a Japanese population did not observe significant effects; however, it revealed large differences in genotype frequencies in the Japanese study population when compared with Caucasians.
In rectal cancer patients treated with combined chemoradiation, the 2R allele was presented as a predictive marker for tumor downstaging. No further significant associations between treatment outcome and VNTR polymorphism have been reported, but a tendency for a decreased response and an increased risk of recurrence for patients with the 3R/3R genotype has been presented Table 2 .[40–42]
In the second repeat of the 3R allele, a SNP resulting in a G>C change has been observed. The G allele of this SNP has been linked to increased gene expression and protein levels. Therefore, based on in vitro activity data, the 2R/3G, 3C/3G, 3G/3G genotype has been considered as a high-expression genotype, and compared with a low-expression genotype (2R/2R, 2R/3C, 3C/3C).[32,43,44] The high-expression genotype has been significantly associated with decreased response, and both increased overall mortality and increased risk of progression, in one study. This genotype has also been associated with decreased overall mortality and an increased risk of progression in two other studies; however, these findings were not significant.[32,43] A tendency for an increased response of patients carrying the high-expression genotype has been presented in rectal cancer patients treated with chemoradiation therapy.
A deletion of 6 bp in the TS gene has been associated with decreased TS expression. Inconsistent and nonsignificant results concerning the deletion polymorphism and clinical outcome in CRC patients after 5-FU-based chemotherapy have been reported. The homozygous deleted genotype has been associated with decreased response,[9,19] increased overall mortality[33,34] and increased risk of recurrence and progression, but also with decreased overall mortality and decreased risk of progression. Only one study investigated the impact of the deletion polymorphism on clinical outcome in rectal cancer patients treated with chemoradiation, and observed a nonsignificantly increased risk of recurrence for carriers of the homozygous deleted genotype.
Methylenetetrahydrofolate reductase (MTHFR) is an important enzyme in the folate pathway converting 5,10-MTHF to 5-methyltetrahydrofolate. Reduced enzyme activity has been linked to increased levels of 5,10-MTHF available for inhibiting TS, and therefore leading to increased efficacy of 5-FU.
Two SNPs in the MTHFR gene have been associated with altered enzyme activity. The C677T polymorphism leads to an amino acid change of alanine to valine. Reduced enzyme activity for the T allele has been reported. Two out of five studies indeed found significantly increased response to 5-FU treatment associated with the number of T alleles,[8,45] whereas the nonsignificant associations reported by the other studies were inconsistent.[46–48] Overall, no significant effects on overall mortality, disease-specific mortality, sex-specific mortality and risk of progression were found.[8,36,38,46–50] Contrary to the hypothesis of improved 5-FU treatment benefit for variant allele carriers, one study in rectal cancer patients treated with combined chemoradiation reported a significantly decreased response for patients carrying the T allele.
The A1298C polymorphism, resulting in an amino acid change of glutamine to alanine, also leads to reduced enzyme activity of the minor allele; however, this is to a lesser extent than for the 677T allele.[8,50] Although three studies did not observe significant associations between this polymorphism and treatment outcome,[8,38,48] one study found a better survival, but a decreased response in patients carrying the homozygous wild-type genotype. The findings regarding survival were confirmed by another study, although restricted to female patients. Only one study observed the associations between treatment outcome and chemoradiation therapy in rectal cancer patients and presented a nonsignificantly increased response for carriers of the variant C allele.
Dihydropyrimidine acts in the degradation of 5-FU. Reduced DPD activity has been associated with reduced clearance of 5-FU, leading to both increased toxicity and increased efficacy of the drug. Common genetic polymorphisms in the DPD gene (DPD-C1896T, DPD-T85C, DPD-A496G, DPD-A557G, DPD-A1627G, DPD-T3351C, DPD-G3649A, DPD-A3844G and DPD-T3856C) resulting in altered enzyme activity were examined in two studies, but no significant associations between DPD genotype and toxicity or response have been reported.[18,19]
Orotate phosphoribosyltransferase catalyzes the conversion of the prodrug 5-FU into its active metabolite. The alanine allele of the OPRT-Gly213Ala polymorphism has been associated with increased enzyme activity in in vitro experiments, and may lead to increased inhibition of TS. One study accordingly observed increased risks of severe diarrhea among 5-FU-treated Japanese patients carrying the alanine allele. The association between OPRT genotype and treatment outcome has not yet been investigated in Caucasians. According to the dbSNP database, the minor allele frequency (MAF) in Asians differs from that in Caucasians (MAFAsians = 0.273 versus MAFCaucasians = 0.150). Thus, larger studies in Caucasian populations will be required to observe a similar association between treatment outcome and Gly213Ala polymorphism.
Tumor suppressor gene p53 plays a crucial role in apoptosis and cell proliferation. Different mutations (exons 4, 5, 6, 7, 8 and 9), including stop mutations and nonstop mutations, have been observed and linked to altered 5-FU activity. Significant associations were found between stop mutations of p53 and cancer-related mortality, and between general mutations and overall mortality and risk of recurrence, respectively. These findings have not yet been replicated in further studies.
Pharmacogenomics. 2008;9(8):1079-1099. © 2008 Future Medicine Ltd.
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
No writing assistance was utilized in the production of this manuscript.
Cite this: Pharmacogenetics in Colorectal Cancer: A Systematic Review - Medscape - Aug 01, 2008.