5-fluorouracil pharmacogenomics: still rocking after ... - Future Medicine

4 downloads 36 Views 1MB Size Report
Several studies have tried to investigate the rela- tionship between the presence of mutations in enzymes linked to 5-fluorouracil (5-FU) activ- ity (Figure 1), such ...
Review For reprint orders, please contact: [email protected]

5-fluorouracil pharmacogenomics: still rocking after all these years? The 5-fluorouracil (5‑FU) metabolic pathway is mainly dependent on the activity of several intracellular enzymes. Among them, four in particular; thymidylate synthase, methylenetetrahydrofolate reductase, dihydropyrimidine dehydrogenase and thymidine phosphorylase are considered the key points in determining sensitivity or resistance to this drug. These enzymes are needed to metabolize the drug in its active form (thymidylate phosphorylase) or to drop the concentration of the active drug in the cell (dihydropyrimidine dehydrogenase) or both (thymidylate synthase and methylenetetrahydrofolate reductase). Several different studies have tried to investigate the relationship between the presence of mutations in these enzymes and a reduced/improved activity of treatment based on 5‑FU or its derivatives. In this article, we will focus on the often contradictory results of these studies. KEYWORDS: 5-fluorouracil n chemotherapy n dihydropyrimidine dehydrogenase n DPD n methylenetetrahydrofolate reductase n MTHFR n pharmacogenomics n predictive n prognostic n survival n thymidine phosphorylase n thymidylate synthase n TP n TS

Several studies have tried to investigate the relationship between the presence of mutations in enzymes linked to 5-fluorouracil (5‑FU) activity (Figure 1) , such as thymidylate synthase (TS), methylenetetrahydrofolate reductase (MTHFR), dihydropyrimidine dehydrogenase (DPD) and thymidine phosphorylase (TP) and activity or toxicity of treatment based on 5‑FU or its derivatives. We will discuss the evidence available on the impact of 5‑FU pharmacogenomics on the clinical outcome of cancer patients treated with 5‑FU-based chemotherapy.

Thymidylate synthase Thymidylate synthase is the enzyme responsible for converting uridine into thymidine, which is subsequently incorporated into DNA. 5‑FU exhibits an anti-tumoral effect by blocking thymidylate synthase through the binding of the metabolite 5-fluorodeoxyuridine monophosphate (5-FdUMP) to TS itself. TS complexed with 5-FdUMP is unable to continue its activity of conversion. Studies on TS mutations or polymorphisms have been based on the assumption that an alteration of the enzyme structure may cause the binding of 5-FdUMP to TS to be either hampered or more probable, thus changing the efficacy of 5‑FU treatment. An increasing number of known alterations of the TS gene have been analyzed. However, particular attention has been paid to a few

alterations that seem to occur more frequently (Table 1) . Among these, a relevant role has been suggested for the presence of different 28-bp variable number tandem repeats (VNTR) in the 5´‑UTR of TS. According to Kawakami and Watanabe classification [1] , different numbers of repeats determines two different ‘alleles’; 2R and 3R, with three common genotypes (2R/2R, 2R/3R and 3R/3R). Data of different trials suggest that 3R allele yields fourfold higher TS mRNA levels in tumor tissue obtained from patients with metastatic colorectal cancer then when compared with patients who carry the 2R variant (p C SNP within the second repeat of the 3R allele TS promoter enhancer region (TSER) tandem repeats was associated with a decreased force of binding of USF‑1 within the repeat, therefore resulting in decreased transcriptional activity of the 3R TS gene variant [3] . A lower transcription rate of the TSER 3RC allele in vitro is also observable upon comparison with TSER 3RG, ­comparable with the TSER 2R/2R genotype [3,4] .

10.2217/PGS.10.167 © 2011 Future Medicine Ltd

Pharmacogenomics (2011) 12(2), 251–265

Mario Scartozzi†1, Elena Maccaroni2, Riccardo Giampieri2, Mirco Pistelli2, Alessandro Bittoni2, Michela Del Prete1, Rossana Berardi1 & Stefano Cascinu1 Clinica di Oncologia Medica, AO Ospedali Riuniti-Ancona, Università Politecnica delle Marche, Ancona, Italy 2 Scuola di Specializzazione in Oncologia, Università Politecnica delle Marche, Ancona, Italy † Author for correspondence: Tel.: +39 071 596 3834 Fax: +39 071 596 4291 [email protected] 1

ISSN 1462-2416

251

Review

Scartozzi, Maccaroni, Giampieri et al.

5-FU

DHFU

Extracellular

DPD

5-FU

TP

FdUrd

TK

dUMP

Purine synthesis

5-10CH=FH4

5-CHOFH4 (FA)

Cytoplasm

FdUMP

TS

dTMP

5-10CH2FH4

FH2

MTHFR

DHFR

5-CH3FH4

MS

Nucleus

FH4

Vitamin B12 Homocysteine

DNA synthesis

DNA methylation Methionine

Figure 1. Main modalities of the action of 5-fluorouracil. 5‑FU enters the cell using a facilitated transport mechanism and is converted intracellularly to several active metabolites. An activation pathway involves the TP-catalyzed conversion of 5‑FU to FdUrd, which is then phosphorylated by TK to FdUMP. FdUMP binds to the nucleotide-binding site of TS, thus inhibiting its activity in creating dTMP that is crucial in DNA synthesis. Damages to DNA and RNA are also caused by FdUTP and FUTP incorporation into DNA or RNA, respectively. To catalyze the reaction of conversion of dUMP into dTMP, the cofactor 5-10CH2FH4 is needed. MTHFR catalyzes a unidirectional reaction that lowers the levels of 5-10CH2FH4 by raising levels of 5-CH3FH4 which is used for biological methylation. Other factors, such as vitamin B12 and homocysteine, are involved in biological methylation processes. The rate-limiting enzyme in 5‑FU catabolism is DPD, which converts 5‑FU to DHFU in normal and tumor cells; up to 80% of administered 5‑FU is broken down by DPD in the liver, where this enzyme is abundantly expressed. 5-10CH=FH4: 5-10 methenyltetrahydrofolate; 5-10CH2FH4: 5-10 methylenetetrahydrofolate; 5-CH3FH4: 5-methyltetrahydrofolate; 5-CHOFH4: Formyltetrahydrofolate; 5‑FU: 5-fluorouracil; DHFR: Dihydrofolate reductase; DHFU: Dihydrofluorouracil; DPD: Dihydropyrimidine dehydrogenase; dTMP: Deoxythymidine 5´-monophosphate; dUMP: Deoxyuridine 5´-monophosphate; FA: Folinic acid; FdUMP: 5-fluorodeoxyuridine 5´-monophosphate; FdUrd: 5-fluorodeoxyuridine; FH2: Dihydrofolate; FH4: Tetrahydrofolate; MS: Methionine synthase; MTHFR: Methylenetetrahydrofolate reductase; TK: Thymidine kinase; TP: Thymidine phosphorylase; TS: Thymidylate synthase.

Another common alteration, frequently described in various studies, is the presence of a 6-bp deletion (1494del6) in the TS gene in the 3´‑UTR region. This mutation seems to be responsible for an alteration in RNA stability, resulting in a decreased level of mRNA. Indeed, in a relatively recent report by Ulrich et al. [5] , in which 43 patients were analyzed for mRNA expression, patients homozygous for the 6-bp deletion had a threefold lower level of steadystate TS mRNA than patients homozygous for the 6-bp insertion allele (p = 0.017). Different polymorphisms in the TS gene have been studied as prognostic factors in various tumor cancer types. Results of these studies are often plagued with different and often contradictory results. 252

Pharmacogenomics (2011) 12(2)

„„ Adjuvant setting In a study by Suh et al. the role of the 3R allele as a prognostic factor has been evaluated [6] . The assessment was performed on tumor samples of 121 patients resected for stage II or III colon cancer and treated with adjuvant 5‑FU according to Mayo clinic regimen. In particular, 3R/3R patients demonstrated a significantly worse 5‑year actuarial survival when compared with either 2R/3R or 2R/2R patients (58 vs 80%; p = 0.048). In stage II patients, this difference suggested a trend towards a worse survival, but it was not statistically significant (p = 0.1678). In stage III patients, the difference was greater and statistically significant (41 vs 77%; p = 0.0414). In a study by Gosens et al., 38 patients treated with adjuvant 5‑FU-based chemotherapy for future science group

5-fluorouracil pharmacogenomics: still rocking after all these years?

resected lymph nodal involvement (N)+ colon adenocarcinoma were tested for TS genotyping in tumor tissue [7] . Findings from this study suggested a statistically significant difference in terms of overall survival (OS) between the two groups (i.e., high and low TS expression), but only upon testing those patients who did not relapse during adjuvant treatment (34 patients). Patients with the low TS expression genotype exhibited a lower recurrence rate than the patients with high TS genotype (p  =  0.038). Cancer-specific survival rate was significantly improved in patients with low TS genotype versus those with the high TS genotype (p = 0.021). Authors also compared genotyping ana­lysis results with standard histological parameters such as parietal tumour invasion (T) or N status and concluded that TS genotyping could have a better gain in terms of prognostic value than histological features. Fernandez-Contreraz recorded completely different results in a similar study [8] . In detail, when analyzing whether the genotype subset (2R/2R vs 2R/3R vs 3R/3R) was related to disease-free survival (DFS) in 201 patients treated with adjuvant 5‑FU postsurgical intervention for colon or rectal cancer, no significant difference between the various groups (with estimated 5‑year DFS rate among the groups respectively of 18.0 vs 20.5 vs 18.6 months; p = 0.997) was evident. The ana­lysis conducted on tumor samples, according to high and low TS-expression genotypes also failed to demonstrate any statistically significant difference (p = 0.731), thus discouraging its role as a prognostic factor. A similar conclusion, though reached using a different method of evaluation, was made by Gusella et al. [9] . The authors analyzed the blood samples of 122 patients treated with 5‑FU in order to determine whether the VNTR or the G/C SNP had a correlation with both

Review

progression-free survival (PFS) or OS. The odds ratio for PFS between those harboring the guanine in place of cytosine nucleotide polymorphism versus wild-type status was 0.99 (95% CI: 0.48–2.047; p = 0.98). The overall response rate between the high TS expression genotype versus the low expression group was 0.86 (and also not statistical significant, with a 95% CI of 0.41–1.77; p  =  0.68). Authors concluded that traditional predictive/prognostic factors such as clinicopathological features, were still major determinants of the outcome prediction of 5‑FU-based chemotherapy. „„ Neoadjuvant setting In a recent study by Paez et al., blood samples were collected from 51 patients before chemoradiation therapy for advanced rectal cancer [10] . An interesting statistically significant correlation between the 3R/3R genotype and a higher response rate was observed, suggesting a role of 3R/3R as a positive predictive factor of higher response rate. Data showed 62% of pathological complete response rate for 3R/3R genotype versus 22% of 2R/3R and 2R/2R subtype (p = 0.013). In addition, the same study also demonstrated that the expression of the 3R/3R genotype was linked to a better OS regardless of treatment, thus suggesting a possible prognostic role (p C SNP) is all thanks to the mutational status or either to the effect of the other drugs in combination. The landscape is clearer in neoadjuvant setting for rectal cancer patients future science group

5-fluorouracil pharmacogenomics: still rocking after all these years?

Review

Table 2. TS polymorphisms and their impact on clinical outcome in gastrointestinal cancer patients. Study

Treatment setting

Patients TS (n) polymorphism analyzed

Suh et al.

Adjuvant CRC

121

Gosens et al.

Adjuvant CRC

8

FernandezContreras et al. Gusella et al.

Adjuvant CRC

201

Adjuvant CRC

122

Paez et al.

Neoadjuvant 51 RC Neoadjuvant 65 RC Metastatic CRC 80 (liver only)

Villafranca et al. Graziano et al.

Marcuello et al.

Metastatic CRC 89

Stoehlmacher et al. Cui et al.

Metastatic CRC 106 Metastatic CRC 68 or GC

Type of assessment

Prognostic

TS 2R/3R repeat

Tumor samples

[6]

TS 2R/3R repeat G/C SNP TS 1494del6 TS 2R/3R repeat, G/C SNP TS 2R/3R repeat G/C SNP TS 1494del6 TS 2R/3R repeat

Tumor samples

Tumor samples

Worse 5‑year survival NE in TS 3R/3R patients Better DFS and OS in NE patients with low TS expression genotype NS NE

Blood samples

NS

[9]

Blood samples

TS 2R/3R repeat

Tumor samples

Better OS in 3R/3R patients NS

TS 2R/3R repeat G/C SNP TS 1494del6 Low-expression group: TSER 2R/2R; 2R/3RC; 3RC/3RC High-expression group: TSER 2R/3G; 3RC/3RG; 3RG/3RG TS 2R/3R repeat TS 1494del6 TS 2R/3R repeat

Blood samples

Blood samples

Predictive

NE

Higher RR in 3R/3R patients No response in 3R/3R patients Worse TTP in 2R/3G, Worse ORR in 3C/3G and 3G/3G 2R/3G, 3C/3G and patients 3G/3G patients Worse OS in patients Better ORR in with high TS patients with expression genotypes low TS expression genotypes

Ref.

[7]

[8]

[10] [11] [12]

[13]

Blood samples

NS

NE

[14]

Blood samples

Better PFS in 2R/3R patients

NS

[15]

CRC: Colorectal cancer; DFS: Disease-free survival; GC: Gastric cancer; NE: Not evaluated; NS: Not significant; ORR: Overall response rate; OS: Overall survival; PFS: Progression-free survival; RC: Rectal cancer; RR: Response rate; TS: Thymidylate synthase; TTP: Time to progression.

who undergo chemoradiation with 5‑FU as the only drug in combination, with almost all data pointing out at a major role as predictive factor in this subset of patients.

MTHFR Methylenetetrahydrofolate reductase is a key regulatory enzyme involved in intracellular folate metabolism. This enzyme catalyzes the irreversible conversion of 5–10 methylenetetrahydrofolate (CH2FH4), required for purine and thymidine synthesis, into 5-methyltetrahydro­folate (CH3FH4), directing the folate pool towards the remethylation of homocysteine to methionine and representing the main determinant of intracellular concentration of CH2FH4. Several studies have demonstrated that an elevated concentration of CH2FH4 is a crucial factor in 5‑FU induced cytotoxicity [16] . In fact, 5‑FU exerts its anticancer effect after the conversion of FdUMP, which inhibits TS and subsequent DNA synthesis through the formation of an inactive ternary future science group

complex between TS, FdUMP and CH 2FH4. High levels of CH 2FH4 can enhance the stability of this ternary complex. Clinical studies have confirmed this observation demonstrating higher efficacy of 5‑FU when associated with a CH2FH4 precursor as folinic acid [17] . Different MTHFR gene polymorphisms have been described. The two most common polymorphisms linked to altered enzymatic activity are 677C>T (Ala to Val at codon 222, exon 4) and 1298A>C (Glu to Ala at codon 428, exon 7). The frequency of both polymorphisms varies by geographical origin, with a prevalence of approximately 25–45% in Caucasian and Asian populations which decrease to approximately 10% in African populations [18] . Furthermore, about 15–23% of Caucasian patients have been found to be heterozygous for both polymorphisms (677T/1298A). The 677T and 1298C polymorphisms were demonstrated to affect the activity of MTHFR reducing its enzymatic activity and thus increasing intracellular level www.futuremedicine.com

255

Review

Scartozzi, Maccaroni, Giampieri et al.

of CH2FH4 [19,20] . Then it can be hypothesized that MTHFR polymorphisms may increase 5‑FU cytotoxic activity increasing intracellular CH2FH4 concentrations and enhancing inhibition of TS. A possible predictive role of MTHFR polymorphisms on tumor sensitivity to 5‑FU was reported both by preclinical and clinical studies and studies on tumor cell lines showed a higher sensitivity to 5‑FU in cell lines bearing 677T or 1298C alleles [21,22] . „„ MTHFR & clinical outcome Several clinical studies, the majority conducted on advanced colorectal cancer patients, have tried to clarify the possible predictive role of MTHFR polymorphisms on 5‑FU activity and clinical outcome with conflicting results. A study by Cohen et  al. was the first to report a link between the 677T variant and response to 5‑FU chemotherapy in 43 metastatic colorectal cancer patients [23] . Subsequently, a retrospective ana­ lysis by Etienne et al. on 98 patients with metastatic colorectal cancer treated with 5‑FU-folinic acid, analyzed the correlation between 677T and 1298C genotypes and clinical outcome in terms of response rate and survival [24] . Response rate was not linked to 1298C genotype but was significantly related to 677T allele (response rate: 40, 21 and 56% in 677 CC, CT and TT respectively; p = 0.04). Surprisingly, 1298CC variant was significantly linked with poorer OS. This detrimental effect of 1298CC polymorphism on patients’ prognosis is probably not related to 5‑FU activity while it could reflect an unfavorable tumor biology: the higher levels of CH2FH4 related to deficient MTHFR polymorphisms could enhance the purine synthesis, leading to a faster tumor proliferation rate. In a similar study by Jakobsen et al. on metastatic colorectal cancer patients treated with 5‑FU and folinic acid, response rate was significantly higher in patients with the 677T allele (66 vs 21 vs 33% in 677 TT, CT and CC respectively; p = 0.04) [25] . In addition, in this ana­lysis no correlation demonstrated between the 1298 A>C polymorphism and response. By contrast, a study by Marcuello et al. failed to demonstrate any predictive value of MTHFR polymorphisms in 5‑FU treated patients [26] . The study included 94 advanced colorectal cancer patients treated with fluoropyrimidinebased first-line chemotherapy containing 5‑FU combined with either irinotecan or oxaliplatin. No correlation between 677T or 1298C allele and response was reported. This lack of predictive value was not modified by the addiction 256

Pharmacogenomics (2011) 12(2)

of folinic acid. No difference in OS and PFS were found when patients bearing both MTHFR polymorphisms were compared with hetero­ zygous or wild-type patients. Accordingly, a study by Ruzzo et al. reported no correlation between 677T and 1298C polymorphisms and response in a large cohort of 166 colorectal cancer patients treated with 5‑FU, folinic acid and oxaliplatin (FOLFOX) in first-line chemo­ therapy [27] . Furthermore, in this study 677T and 1298C exhibited no influence on PFS. Interestingly, the predictive role of MTHFR polymorphisms has been described only by studies on patients treated with 5‑FU monochemotherapy while this was not confirmed in studies where 5‑FU was combined with other agents. This observation suggests that the addition of active drugs such as irinotecan or oxaliplatin may mask the effect of MTHFR polymorphisms on tumor response. A recent meta-ana­lysis has tried to overcome the disagreements among studies and assess the effect of the most common MTHFR polymorphisms on response to 5‑FU treatment [28] . Data from ten studies performed on colorectal cancer patients were included in the meta-ana­ lysis and OR of responders versus nonresponders to chemotherapy for the two different alleles of 677C>T and 1298A>C polymorphisms were calculated. The results demonstrated that the two MTHFR polymorphisms evaluated were not predictive of 5‑FU response with an OR of 0.93 (95%  CI: 0.72–1.20) for 677C>T and 1.13 (95% CI: 0.82–1.56) for 1298A>C. Nevertheless, as stated by the authors, the results of this ana­lysis should be interpreted cautiously, given the heterogeneity of the studies included in terms of chemotherapy regimen administered and outcome definition and the unavailability of individual patient data. Recently, a prospective study has strengthened the hypothesis of the predictive role of MTHFR polymorphisms in colorectal cancer. The study, an ancillary study part of the OPTIMOX 2 trial, included 117 advanced colorectal cancer patients treated with FOLFOX-7 chemotherapy as firstline treatment. Several germinal gene polymorphisms were analyzed, including MTHFR 677C>T and 1298A>C. The authors demonstrated a statistically significant link between the 677T and 1298C alleles and tumor response (p = 0.042 and p = 0.004, respectively). In particular, tumor response was demonstrated to increase with the number of favorable MTHFR alleles present, with response rates of 37.1, 53.3, 62.5 and 80% (p = 0.04) in patients bearing future science group

5-fluorouracil pharmacogenomics: still rocking after all these years?

either none, one, two or three favorable alleles, respectively (no homozygous patient for both variants was found). Toxicity was not related to any MTHFR polymorphism evaluated. MTHFR polymorphisms were not correlated to either PFS or OS [29] . In Table 3, the results of the studies on MTHFR polymorphisms and their correlation with response to chemotherapy, toxicity and survival have been summarized. „„ MTHFR & toxicity The role of MTHFR polymorphisms in 5‑FU-related toxicity is similarly controversial. As expected, some studies have demonstrated a correlation between MTHFR genotype and 5‑FU toxicities. In a study by Lu et al. on 75 gastric cancer patients, the incidence of grade 3–4 toxicities was found to be higher in patients bearing the 677TT variant compared with other variants [30] . In addition, an Italian study demonstrated a correlation between MTHFR polymorphisms and toxicity: in this ana­lysis, performed on 130 resected colorectal cancer patients treated with 5‑FU/leucovorin adjuvant chemotherapy, the 677CC genotype was demonstrated to be protective against grade 3–4 toxicities (p = 0.040)  [9] . By contrast, no correlation between MTHFR polymorphisms and toxicity was found in the studies by Cohen and Ruzzo, as discussed before. Several aspects, both patient and diseaserelated, may explain the conflicting results reported by clinical studies. Intracellular levels of CH2FH4 are not linked only to MTHFR enzymatic activity, but depend on other factors, such as dietary folate intake. It is possible that the effect of MTHFR polymorphisms could be more evident in patients with low levels of folate in the diet. In addition, epigenetic interactions could account for the lack of a direct genotype–phenotype relationship. CH2FH4 levels were not always related to a favorable MTHFR genotype.

Review

DPD More than 80% of the administered 5‑FU is primarily catabolized in the liver where the enzyme DPD is mainly expressed [31] . DPD mediates 5‑FU catabolism, which results in the formation of inactive dihydrofluorouracil (DHFU) that is further degraded to fluoroureidopropionate (FUPA) and the inactive fluoro-b-alanine (F-BAL), which is excreted in the urine. DPD is the rate-limiting enzyme in the 5‑FU catabolism [32] . DPD is encoded by a gene (DPYD) that is located in human chromosome 1p22 and that is comprised of 23 exons encompassing approximately 950 kb [33] . In the recent years, several researches investigated the genetic variations of DPYD and the level expression of DPD in tumor cells with the aim to predict both 5‑FU toxicity and efficacy. „„ DPD & toxicity In the case of DPD enzyme deficiency, blood levels of 5‑FU and its active metabolites increase. Thus, patients with DPD enzyme deficiency risk the development of serious toxicity after the 5‑FU administration. It has been demonstrated that alterations in DYPD coding for the keyenzyme DPD of fluoropyrimidines catabolism, contribute to the development of serious 5‑FU toxicity. Conditions resulting in a mutant DPYD allele include base substitutions, splicing deficits and frameshift mutations [34–36] . To date, 17 variant alleles have been identified, one of them being by far the most frequent one (DPYD*2A); it consists of a splice-site mutation, IVS14 + 1G>A. In this case, exon 14 is skipped as a result of the G>A translocation at intron 14, and an inactive enzyme is formed. The hetero­zygote genotype is characterized by severe toxicity. However, with the relevant exception of grade IV neutropenia, no substantial differences have been described in either (other than

Table 3. MTHFR polymorphisms analyzed and their clinical impact on outcome, response to chemotherapy and toxicity in gastrointestinal cancer patients. Study

Treatment setting

Patients Type of (n) assessment

Cohen et al. Etienne et al.

Metastatic CRC 43 Metastatic CRC 98

Blood samples 677T Liver metastases 677T + 1298C

Jakobsen et al. Marcuello et al. Ruzzo et al. EtienneGrimaldi et al.

Metastatic CRC Metastatic CRC Metastatic CRC Metastatic CRC

Blood samples Blood samples Blood samples Blood samples

51 94 166 117

MTHFR PFS polymorphism

677T + 1298C 677T + 1298C 677T + 1298C 677T + 1298C

OS

Lower (1298C) NS NS NS

NS NS NS

Response to Toxicity chemotherapy Higher Higher (677T) Higher (677T) NS NS Higher (677T, 1298C)

NS

Ref. [23] [24] [25] [26] [27]

NS

[29]

CRC: Colorectal cancer; NS: Not significant; OS: Overall survival; PFS: Progression-free survival.

future science group

www.futuremedicine.com

257

Review

Scartozzi, Maccaroni, Giampieri et al.

neutropenia) hematological or gastrointestinal toxicity, flu-like symptoms, or other types of side effects. On the contrary the homozygote genotype is characterized by mental disorders. In the general population, homozygote and heterozygote DPYD dysfunction is estimated to be between 0.1 and 3–5%, respectively. Other rare mutations (C, 496A>G, 1627A>G, 2194G>A, 2846A>T), may have the same consequences. These individuals have no symptoms in the absence of drug treatment (5‑FU); however, they could be at increased risk for toxicity if exposed to 5‑FU during chemotherapy. DPD deficiency has been demonstrated in approximately 60% of patients developing a severe 5‑FU toxicity, whereas a DPYD*2A polymorphism was identified in 50% of patients with grade IV neutropenia. Nevertheless (and surprisingly), DPD enzyme activity resulted unaltered in most patients with severe 5‑FU toxicity [37] . It is then conceivable that other factors such as age, high serum alkaline phosphatase and elapsed time during 5‑FU infusion may also play a crucial role in toxicity [38] . We can also speculate that some genomic polymorphisms or rearrangement in DPYD are involved in the development of 5‑FU toxicity; more than 40 different polymorphisms related to this gene have been reported. Probably only some of these alterations are related to the development of serious fluoropyrimidines related toxicity (Table 4) . Recently, evidence that genetic variations also located outside the coding regions of DYPD are potentially predictive of severe toxicity in 5‑FU-based chemotherapy is of interest. On this basis, a haplotype containing three novel intronic polymorphisms (IVS5+18G>A, IVS6+139G>A, IVS9–51T>G) and a synonymous mutation (c.1236G>A) has been proposed as an alternative mechanism for DPD deficiency and thus, as a cause of severe toxicity in patients Table 4. Some variant alleles of DYPD gene involved in the toxicity of antimetabolites. DYPD mutation

DPD alteration

Consequence

Allele frequency (%)

IVS14+1G>A

Del (exon 14)

No expression

1

85T>C 496A>G 1627A>G 2194G>A 2846A>T

Cys29arg Met166val Ile543val Val732ile Asp949val

Decreased expression Decreased expression Decreased expression Decreased expression Decreased expression

A mutation in the dihydropyrimidine dehydrogenase gene of patients with severe 5-fluorouracil associated toxicity. Pharmacogenetics 12, 555–558 (2002). 37 Van Kuilenberg ABP, Haasjes J, Richel DJ

et al.: Clinical implications of dihydropyrimidine dehydrogenase (DPD) deficiency in patients with severe 5-fluorouracil-asociated toxicity: identification of new mutations in the DPD gene. Clin. Cancer Res. 6, 4705–4712 (2000). 38 Sumi T, Katsumata K, Tsuchida A et al.:

Correlations of clinicopathological factors with protein expression levels of thymidylate synthase, dihydropyrimidine dehydrogenase and orotate phosphoribosyltransferase in colorectal cancer. Chemotherapy 56(2), 120–126 (2010). 39 Amstutz U, Farese S, Aebi S et al.:

Dihydropyrimidine dehydrogenase gene variation and severe 5-fluorouracil toxicity: a haplotype assessment. Pharmacogenomics 10(6), 931–944 (2009). 40 Van Kuilenburg A, Meijer J, Mul A et al.:

Intragenic deletions and a deep intronic mutation affecting pre-mRNA splicing in the dihydropyrimidine dehydrogenase gene as novel mechanisms causing 5-fluorouracil toxicity. Hum. Genet. 128(5), 529–538 (2010). 41 Amstutz U, Farese S, Aebi S et al.:

Hypermethylation of the DPYD promoter region is not a major predictor of severe toxicity in 5-fluorouracil based chemotherapy. J. Exp. Clin. Cancer Res. 27, 54 (2008). 42 Salonga D, Danenberg KD, Johnson M et al.:

Colorectal tumors responding to 5-fluorouracil have low gene expression levels of dihydropyrimidine dehydrogenase, thymidylate synthase, and thymidine phosphorylase. Clin. Cancer Res. 6(4), 1322–1327 (2000).

future science group

5-fluorouracil pharmacogenomics: still rocking after all these years?

43 Vallböhmer D, Yang DY, Kuramochi H et al.:

thymidylate synthase mRNA expression in primary colorectal tumors-correlation to tumor histopathology and clinical follow-up. Int. J. Colorectal Dis. 21, 238–247 (2006).

DPD is a molecular determinant of capecitabine efficacy in colorectal cancer. Int. J. Oncol. 31(2), 413–418 (2007). 44 Yanagisawa Y, Maruta F, Iinuma N et al.:

45

Meropol NJ, Gold PJ, Diasio RB et al.: Thymidine phosphorylase expression is associated with response to capecitabine plus irinotecan in patients with metastatic colorectal cancer. J. Clin. Oncol. 24(25), 4051–4053 (2006).

Prognostic significance of thymidylate synthase, dihydropyrimidine dehydrogenase and thymidine phosphorylase protein expression in colorectal cancer patients treated with or without 5-fluorouracil-based chemotherapy. Ann. Oncol. 19, 915–919 (2008). 55

46 Vallböhmer D, Kuramochi H, Shimizu D et al.:

Molecular factors of 5-fluorouracil metabolism in colorectal cancer: ana­lysis of primary tumor and lymph node metastasis. Int. J. Oncol. 28(2), 527–533 (2006).

Predictive value of thymidylate synthase and dihydropyrimidine dehydrogenase protein expression on survival in adjuvantly treated stage III colon cancer patients. Ann. Oncol. 16, 1646–1653 (2005). 57 Terashima M, Fujiwara H, Takagane A et al.:

Role of thymidine phosphorylase and dihydropyrimidine dehydrogenase in tumour progression and sensitivity to doxifluridine in gastric cancers patients. Eur. J. Cancer 38, 2375–2381 (2002).

48 Koizumi W, Okayasu I, Hyodo I et al.:

Prediction of the effect of capecitabine in gastric cancer by immunohistochemical staining of thymidine phosphorylase and dihydropyrimidine dehydrogenase. Anticancer Drugs 19, 819–824 (2008).

58 Nishina T, Hyodo I, Miyaike J et al.: The

ratio of thymidine phosphorylase to dihydropyrimidine dehydrogenase in tumour tissues of patients with metastatic gastric cancer is predictive of the clinical response to 50-deoxy-5-fluorouridine. Eur. J. Cancer 40, 1566–1571 (2004).

49 Kornmann M, Schwabe W, Sander S et al.:

Thymidylate synthase and dihydropyrimidine dehydrogenase mRNA expression levels: predictors for survival in colorectal cancer patients receiving adjuvant 5-fluorouracil. Clin. Cancer Res. 9, 4116–4124 (2003).

59 Grau JJ, Caballero M, Monzò M et al.:

Dihydropyrimidine dehydrogenases and cytidine-deaminase gene polymorphisms as outcome predictors in resected gastric cancer patients treated with fluoropyrimidine adjuvant chemotherapy. J. Surg. Oncol. 98, 130–134 (2008).

50 Ciaparrone M, Quirino M, Schinzari G et al.:

Predictive role of thymidylate synthase, dihydropyrimidine dehydrogenase and thymidine phosphorylase expression in colorectal cancer patients receiving adjuvant 5-fluorouracil. Oncology 70(5), 366–377 (2006). 51

52

Oi K, Makino M, Ozaki M et al.: Immunohistochemical dihydropyrimidine dehydrogenase expression is a good prognostic indicator for patients with Dukes’ C colorectal cancer. Anticancer Res. 24(1), 273–279 (2004). Jensen SA, Vainer B, Sorensen JB: The prognostic significance of thymidylate synthase and dihydropyrimidine dehydrogenase in colorectal cancer of 303 patients adjuvantly treated with 5-fluorouracil. Int. J. Cancer 120, 694–701 (2007).

53 Lassmann S, Hennig M, Rosenberg R et al.:

Thymidine phosphorylase, dihydropyrimidine dehydrogenase and

future science group

Hotta T, Takifuji K, Taniguchi K et al.: The relationship between survival and the expression of dihydropyrimidine dehydrogenase in patients with colorectal cancer. Oncol. Rep. 16(1), 177–182 (2006).

56 Westra JL, Hollema H, Schaapveld M et al.:

47 Kodera Y, Ito S, Fujiwara M et al.: Gene

expression of 5-fluorouracil metabolic enzymes in primary gastric cancer: correlation with drug sensitivity against 5-fluorouracil. Cancer Lett. 252, 307–313 (2007).

63 Koopman M, Venderbosch S,

van Tinteren H et al.: Predictive and prognostic markers for the outcome of chemotherapy in advanced colorectal cancer, a retrospective ana­lysis of the Phase III randomised CAIRO study. Eur. J. Cancer 45, 1999–2006 (2009).

54 Soong R, Shah N, Salto-Tellez M et al.:

Modified irinotecan/5FU/leucovorin therapy in advanced colorectal cancer and predicting therapeutic efficacy by expression of tumorrelated enzymes. Scand. J. Gastroenterol. 42(4), 477–484 (2007).

60 Tokunaga Y, Takahashi K, Saito T: Clinical

61

Review

64 Nii A, Shqwsedrftimada M, Ikegami T

et al.: Significance of dihydropyrimidine dehydrogenase and thymidylate synthase mRNA expressions in hepatocellular carcinoma. Hepatol. Res. 39(3), 274–281 (2009). 65

Nakahara O, Takamori H, Tanaka H et al.: Clinical significance of dihydropyrimidine dehydrogenase and thymidylate synthase expression in patients with pancreatic cancer. Int. J. Clin. Oncol. 15(1), 39–45 (2010).

66 Li Y, Mizutani Y, Shiraishi T et al.: The

significance of the expression of dihydropyrimidine dehydrogenase in prostate cancer. BJU Int. 99(3), 663–668 (2007). 67 Tsunoda Y, Suzuki K, Sakamoto MA et al.:

[Evaluation of 5-fluorouracil-related genes in breast cancer to predict the effect of adjuvant therapy with CMF]. Gan To Kagaku Ryoho 36(1), 51–55 (2009) . 68 Miyadera K, Sumizawa T, Haraguchi M

et al.: Role of thymidine phosphorylase activity in the angiogenic effect of platelet derived endothelial cell growth factor/ thymidine phosphorylase. Cancer Res. 55, 1687–1690 (1995). 69 Soong R, Diasio RB: Advances and

challenges in fluoropyrimidine pharmacogenomics and pharmacogenetics. Pharmacogenomics 6, 835–847 (2005). 70 Hua D, Huang ZH, Mao Y et al.:

Thymidylate synthase and thymidine phosphorylase gene expression as predictive parameters for the efficacy of 5-fluorouracilbased adjuvant chemotherapy for gastric cancer. World J. Gastroenterol. 13(37), 5030–5034 (2007).

role of thymidine phosphorylase and dihydropyrimidine dehydrogenase in colorectal cancer treated with postoperative fluoropyrimidine. Hepatogastroenterology 52(66), 1715–1721 (2005).

71 Fox SB, Engels K, Comley M et al.:

Tokunaga Y, Sasaki H, Saito T: Clinical role of orotate phosphoribosyl transferase and dihydropyrimidine dehydrogenase in colorectal cancer treated with postoperative fluoropyrimidine. Surgery 141(3), 346–353 (2007).

72 Marchetti S, Chazal M, Dubreuil A,

62 Gustavsson B, Kaiser C, Carlsson G et al.:

Molecular determinants of efficacy for 5‑FU-based treatments in advanced colorectal cancer: mRNA expression for 18 chemotherapy-related genes. Int. J. Cancer 124(5), 1220–1226 (2009).

www.futuremedicine.com

Relationship of elevated tumour thymidine phosphorylase in node-positive breast carcinomas to the effects of adjuvant CMF. Ann. Oncol. 8, 271–275 (1997). Fischel JL, Etienne MC, Milano G: Impact of thymidine phosphorylase surexpression on fluoropyrimidine activity and on tumour angiogenesis. Br. J. Cancer 85, 439–445 (2001).

265