Abstract
To identify the genes and clinical parameters associated with efficacy of uracil and tegafur/leucovorin (UFT/LV) chemotherapy in colorectal cancer (CRC), we compared the levels of reduced folate in tumors between patients receiving LV and those not receiving LV (study I), and explored the changes in the expression levels of 14 genes after two weeks of UFT/LV chemotherapy in 73 patients with CRC (study II). In study I, the reduced folate levels after LV administration were approximately 3-fold higher than those without LV administration in patients with right-sided tumors or aged >75 years old (p=0.019). In study II, the reduction in γ-glutamyl hydrolase (GGH) expression in responders and in patients with right-sided tumors or aged >75 years old was significantly greater than that in non-responders (p<0.001) and in other patients, respectively (p=0.003). Increase of reduced folate and reduction in GGH expression were associated with response to UFT/LV chemotherapy in patients with right-sided tumors or aged >75 years old.
The therapeutic regimen consisting of 5-fluorouracil (5-FU) with leucovorin (LV) combined with oxaliplatin or irinotecan has been established as a standard regimen in chemotherapy for colorectal cancer (CRC) (1-6). Because the orally active derivative of fluoropyrimidine is highly convenient to use, such combined chemotherapy regimens with 5-FU with LV replaced by oral uracil and tegafur (UFT) with LV, capecitabine or S-1 are now extensively prescribed (7-9).
As predictive biomarkers of the efficacy of 5-FU/LV or UFT/LV chemotherapy in CRC, low thymidylate synthetase (TYMS) gene expression (10-13), high uridine monophosphate synthetase (UMPS) gene expression (14), low dihydropyrimidine dehydrogenase (DPYD) gene expression (15), and low thymidine phosphorylase (TYMP) gene expression levels (16) have been reported. All these enzymes are pyrimidine metabolism-related enzymes. Reduced folate LV, when used in combination with 5-FU, enhances the 5-FU-induced inhibition of TYMS. Therefore, folate metabolism-related enzymes in addition to pyrimidine metabolism-related enzymes may also be candidate predictors of the efficacy of 5-FU/LV or UFT/LV chemotherapy. Actually, we have reported that gene expression levels of folylpolyglutamate synthase (FPGS) and γ-glutamyl hydrolase (GGH) in tumors could predict levels of reduced folate after LV administration in the tumor tissue (17).
UFT/LV chemotherapy in patients with CRC has been used in the adjuvant setting. In order to identify factors predictive of survival in patients with CRC treated with UFT/LV adjuvant chemotherapy, a long-term investigation is required. However, Fujii et al. reported that pathological tumor necrosis in response to preoperative short-term UFT administration could be used to predict the response of CRC to postoperative adjuvant chemotherapy with UFT (18). Therefore, the histological response to neoadjuvant chemotherapy with UFT/LV can be a surrogate endpoint of a response to UFT/LV adjuvant chemotherapy. Based upon these results, we investigated the predictive and responsive factors of neoadjuvant chemotherapy with UFT/LV in patients with CRC. We have already reported that the TYMP gene expression levels among the responders were significantly higher than those among the non-responders, in patients who received a two-week UFT/LV administration (19). Right-sided tumors with high TYMP gene expression levels were associated with a significantly higher response rate to UFT/LV chemotherapy than left-sided tumors (19).
Changes in the tumor gene expression profile during chemotherapy are an important issue related to the antitumor effects of anticancer drugs because patients with CRC are often treated for more than several months. Several studies have reported changes in the tumor expression levels of pyrimidine metabolism-related genes in patients with adenocarcinoma receiving neoadjuvant therapy including 5-FU. Significant increases in the TYMS mRNA level in tumor tissues after 5-FU-based neoadjuvant chemotherapy were reported in patients with esophageal cancer and CRC (20, 21). In contrast, one study reported a significant reduction in the tumor tissue TYMS mRNA levels after 5-FU-based neoadjuvant chemotherapy in patients with esophageal cancer (22). In regard to DPYD, both a significant increase (20) and significant decrease (21, 22) in tumor mRNA levels after 5-FU-based neoadjuvant chemotherapy have been reported in patients with esophageal cancer or CRC. Langer et al. (23) reported a significant post-therapeutic reduction in the expression levels of TYMP in patients with esophageal adenocarcinoma receiving 5-FU-based neoadjuvant chemotherapy. Thus, controversial results have been obtained from several studies conducted previously.
In this study, we first analyzed the relationship between levels of reduced folate in tumor tissues after administration of a single dose of oral LV and clinical parameters. Next, we analyzed the association of the efficacy of UFT/LV with the changes in the expression levels of pyrimidine and folate metabolism-related genes after UFT/LV neoadjuvant chemotherapy.
Patients and Methods
Study I
Patients. The retrospective analysis was performed using the data of our previous study (17). In the previous study, 60 patients with resectable, histologically-confirmed colonic or rectal adenocarcinoma were enrolled and randomly assigned to the following four groups: a control group in which the patients did not receive LV (n=30); and three other groups (n=10 for each) in which the patients received a single 25 mg dose of oral LV 4, 12 or 18 h before surgery, by the method of simple randomization. In this study, we excluded the 4-h LV group from analysis because the levels of reduced folate in this group did not increase after LV administration compared with the control group. The 12-h LV and 18-h LV groups were defined as the LV-treated group (n=20). Informed consent was obtained from all of the patients prior to the enrollment. This study was conducted with the approval of the Ethics Committees of Tokai University School of Medicine (06-019), and Taiho Pharmaceutical Co., Ltd. (S06-009). All patients were kept nil by mouth for 24 h before the surgery. The resected CRC tissues were collected within 10 min of excision. Some parts of these tissues were immediately frozen and stored at −80°C. The characteristics of the 50 patients included are shown in Table I.
Measurement of reduced folate in the tumor tissues. The level of reduced folate, as a combined pool of tetrahydrofolate (THF) and 5,10-CH2THF, in the tumor tissues was measured by the 5-fluoro-2’-deoxyuridine-5’-monophosphate (FdUMP)-TYMS binding assay using [6-3H]FdUMP (Moravek Biochemicals Inc., CA, USA) and recombinant human TYMS (Taiho Pharmaceutical Co., Ltd.) (24). The methods have been described in detail previously (17). Values under the lower limit of quantitation in a liquid-scintillation counter (TRI-CARB 2000CA, Packard Instruments, Meriden, CT, USA) were counted as 0 for analysis.
Patients' characteristics in study I.
Statistical analysis. After dividing the samples into two groups on the basis of clinical parameters (namely, sex, age, primary tumor site and histological type), the differences in the folate levels between the control and LV-treated groups were evaluated using Welch's t-test.
Study II
Patients. Seventy-six patients with clinical stage II or stage III colorectal adenocarcinoma who were scheduled to undergo surgery at the Tokai University Hospital between July 2009 and February 2011 were enrolled in this study, as described previously (19). Patients with multiple-organ primary cancer, who had been previously treated by chemotherapy or radiotherapy, who had severe comorbidity, who had poorly-differentiated or mucinous adenocarcinoma, or who did not provide informed consent were excluded. Informed consent was obtained from all the patients prior to the enrollment. The following three patients were excluded because the gene expression data of post-chemotherapy tumor samples were not available. For one patient with a tumor regression grading score of grade 4, no tumor tissue was found in the resected specimens. For two patients, post-chemotherapy tumor samples were not obtained. Thus, in total, 73 patients were investigated in this study. This study was conducted with the approval of the ethics committees of the Tokai University School of Medicine (09R-036) and Taiho Pharmaceutical Co., Ltd (S09-006). The characteristics of the 73 patients are shown in Table II.
Patients' characteristics and pathological tumor responses to uracil and tegafur/leucovorin (UFT/LV) neoadjuvant chemotherapy in Study II.
The patients received oral treatment with UFT (300 mg/m2/day, t.i.d.) and LV (75 mg/day, t.i.d.) for two weeks, with surgery undertaken three days after the final administration. Endoscopic biopsy samples were obtained prior to drug administration. The tumor biopsy samples or surgically resected tumor specimens were immediately immersed in RNAlater solution (Life Technologies, Grand Island, NY, USA) and incubated overnight at 4°C. The tissues were then removed from the RNAlater solution and stored at −80°C.
Evaluation of the antitumor effect. The pathological response to UFT/LV chemotherapy was evaluated using the Tumor Regression Grading Score (25-28) as follows: grade 0, no regression; grade 1, dominant tumor mass with obvious fibrosis and/or vasculopathy; grade 2, dominant fibrotic changes with few tumor cells or groups (easy to find); grade 3, very few (difficult to find microscopically) tumor cells in fibrotic tissue with or without mucous substance; grade 4, no tumor cells, only fibrotic mass (total regression or response). Patients with a grade 3 or 4 response were defined as responders, and those with a grade 0-2 response as non-responders.
Gene expression analysis. The mRNA expressions of 5-FU-related enzymes (six genes: DPYD, UMPS, ribonucleotide reductase M1 (RRM1), thymidine kinase 1 (TK1), TYMP, and TYMS) and of reduced folate-related enzymes (eight genes: 5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase/IMP cyclohydrolase (ATIC), dihydrofolate reductase (DHFR), folate receptor 1 (FOLR1), folate receptor 2 (FOLR2), FPGS, aldehyde dehydrogenase 1 family, member L1 (ALDH1L1), phosphoribosylglycinamide formyltransferase (GART), and GGH) were quantitatively evaluated by a reverse transcription-polymerase chain reaction (RT-PCR) assay, as described previously (19). The methods are briefly described below. Real-time RT-PCR was performed using an ABI PRISM 7900HT sequence detection system (Life Technologies) and a real-time RT-PCR array (TaqMan Array; Life Technologies). The gene expression levels were normalized to that of the reference gene, beta actin (ACTB) (29, 30). The relative gene expression levels were calculated using the delta threshold cycle (Ct) method according to the formula shown below. The expression levels of the target genes were expressed as 2−(delta Ct) ×1000 to simplify the calculation.
Statistical analysis. The Fisher's exact test was used to analyze the associations between clinical parameters (sex, age, histological type, and primary tumor site) and the tumor response. The differences in the gene expression ratios (post-chemotherapy to pre-chemotherapy) between responders and non-responders were evaluated using Welch's t-test. The differences in the gene expression ratios (post-chemotherapy to pre-chemotherapy) between the two groups according to the clinical parameters were also evaluated using Welch's t-test. All the analyses were performed using the JMP 7.0.1 statistical software and the SAS statistical package, version 9.1.3 (SAS Institute Inc., Cary, NC, USA). Differences were considered significant when p<0.05.
Results
Study I
Table III displays the comparison of the tumor levels of reduced folate between patients receiving and those not receiving LV. The tumor tissue levels of reduced folate were significantly higher in LV-treated patients with right-sided colon cancer as compared with the control group not receiving LV (p=0.039). The tumor tissue levels of reduced folate were more than 3-fold higher in the LV-treated patients who were elderly (over 75 years of age) as compared with the levels in the control group not receiving LV, although the inter-group difference did not attain statistical significance. Moreover, the tumor tissue level of reduced folate in LV-treated patients who had right-sided tumors or were aged >75 years was significantly higher compared with the levels in the control group (p=0.019, Table III and Figure 1). On the other hand, the tumor tissue level of reduced folate in LV-treated patients who had left-sided tumors and were aged ≤75 years did not significantly differ from levels of the control group (Table III and Figure 1).
Comparison of the tumor levels of reduced folate between patients receiving and not receiving leucovorin (LV).
Study II
The relationships between patients' characteristics and the therapeutic responses are shown in Table II. There were 14 responders (19.2%) with a grade 3 tumor regression and 59 non-responders (80.8%) with grade 0, 1 or 2 tumor regression, according to the Tumor Regression Grading system (Table II). The response rate was 37.0% in patients with right-sided colon cancer and 8.7% in patients with left-sided colon cancer, being significantly higher in the patients with right-sided tumors (p=0.005).
Changes in the expression levels of the pyrimidine and folate metabolism-related genes following UFT/LV chemotherapy as compared to the baseline values are presented in Table IV. An approximately 2-fold or greater increase of the gene expression levels of DPYD, TYMP and TYMS among pyrimidine-metabolizing enzymes, and of DHFR, FOLR1, FOLR2 and ALDH1L1 among folate-metabolizing enzymes was observed following the treatment. GGH was the only gene whose expression level was decreased to less than 50% of the baseline value following treatment.
Fold changes (post/pre-treatment ratio) in the expressions of these genes were compared between the responder and non-responder groups (Table IV). The increase in the gene expression level of DPYD and FOLR2 was more than 2-fold greater in the responders compared to the non-responders. The gene expression level of GGH, on the other hand, decreased by more than 2-fold in the responders compared to the non-responders.
Comparison between right-sided tumors and left-sided tumors revealed GGH as the only gene whose fold change (post/pre-treatment ratio) significantly differed between the two groups; the decrease in the expression level of this enzyme was about 2-fold greater in the right-sided tumor group as compared to the left-sided tumor group (Table V). In regard to the relationship of the decrease in the GGH gene expression level with other clinical parameters (Table V), the reduction in GGH gene expression after two weeks' chemotherapy was significantly greater in female patients, in patients aged >75 years old, and patients with right-sided colon cancer, as compared with those in the other group, respectively. Analysis of combined primary tumor site and age data revealed that the extent of decrease in the GGH gene expression levels following chemotherapy were significantly higher in patients who had right-sided tumors or were >75 years old than patients who had left-sided tumors and were under 75 years of age (p=0.003).
Comparison of the level of reduced folate in tumor in control and leucovorin (LV)-treated groups. A: patients who had right-sided tumors or were aged >75 years; B: patients who had left-sided tumors and were aged ≤75 years. In the diamond plots, the central horizontal line represents the group mean. The vertical extremities represent the lower and upper 95% confidence intervals.
Comparison of the changes in the gene expression levels between the responders (Res) and non-responders (Non-res).
In the patients who had right-sided tumors or were >75 years old, the decrease in the gene expression level of GGH was significantly greater in the responders, compared to the non-responders (p=0.007, data not shown).
Discussion
Study I showed that the expression levels of almost all genes assayed showed a more than expected increase after two weeks of UFT/LV chemotherapy. In particular, the mRNA expression of genes encoding three (DPYD, TYMP and TYMS) out of the six pyrimidine-metabolizing enzymes, and four (DHFR, FOLR1, FOLR2 and ALDH1L1) out of the eight folate-metabolizing enzymes were up-regulated by 2-fold or more. Previous studies that investigated the changes in the expression levels of genes encoding pyrimidine-metabolizing enzymes after 5-FU-based neoadjuvant chemotherapy for esophageal adenocarcinoma or CRC reported no appreciable increase in the gene expressions of these metabolizing enzymes (20-23). This discrepancy between the present study and previous reports could be attributable to differences in the drug regimens employed (UFT/LV chemotherapy versus 5-FU based-chemoradiotherapy, 5-FU bolus, etc.), duration of treatment (two weeks versus 1 day-10 weeks), duration of post-treatment (3 days versus 2 h–5 weeks), etc.
Associations between clinical parameters and reduction in γ-glutamyl hydrolase (GGH) gene expression.
The expression level of TYMP increased by 4.16-fold after UFT/LV treatment. Our previous report documented that the tumor response to UFT/LV chemotherapy was related to the pre-treatment tumor expression level of TYMP (19), suggesting the possibility that elevation of the TYMP gene expression following continued treatment might lead to further enhancement of the tumor response to UFT/LV chemotherapy.
Meanwhile, GGH was the only folate-metabolizing enzyme for which the mRNA expression level decreased by more than 50% following two weeks of UFT/LV chemotherapy. GGH is a lysosomal enzyme that acts as an endopeptidase and exopeptidase to remove the terminal glutamates of the polyglutamated folates synthesized by FPGS (31, 32). The polyglutamate forms of folate are more easily retained within cells (31). We previously reported that the siRNA down-regulation of GGH mRNA increased both the intracellular folate level after LV treatment and the cellular sensitivity to FdUrd plus LV (33). We also reported that the expression level of GGH was negatively correlated with the level of reduced folate in CRC tissues (17). Therefore, it is possible that the reduction in GGH mRNA observed in the present series might lead to elevation of the level of folate in tumor tissue, thereby contributing to an enhancement of the antitumor effect of UFT/LV chemotherapy.
The results of study I suggest that following single-dose LV administration, the level of folate in tumor was more likely to increase in patients with right-sided CRC than in patients with left-sided CRC. In study II, on the other hand, the response rate to UFT/LV chemotherapy was significantly higher in patients with right-sided CRC than in patients with left-sided CRC, and the degree of GGH reduction was also conspicuously greater in patients with right-sided CRC. From these results it could be inferred that following administration of LV, reduced folate is more likely to accumulate in the tumor tissues in patients with right-sided CRC, which may account, at least in part, for the significantly higher response rate to two weeks of UFT/LV chemotherapy in patients with right-sided CRC.
Analysis of combined primary tumor site and age data revealed that after administration of a single dose of LV, the level of folate in tumor was more likely to increase in patients with CRC who had right-sided tumors or were aged >75 years than in other patients (study I). The analysis also revealed that the reduction in GGH gene expression following UFT/LV chemotherapy was also more likely to occur in the aforementioned patient groups (study II). This suggests that during UFT/LV chemotherapy, reduced folate is more likely to accumulate in the tumor cells of these patients; in other words, these patient subgroups are likely to represent suitable subgroups for UFT/LV chemotherapy.
It has been reported that methylation of CpG islands occurs at a higher frequency (34-36) in proximal colon cancer as compared to distal colon cancer. Hawkins et al. documented that CpG island methylation occurred at a higher frequency in CRC in elderly patients (36). The level of reduced folate in normal colonic mucosa has been reported to decline with advancing age in mice, and the folate transporter ABCC3 and folate retention enzyme GGH have been shown to be down-regulated in aged mice as compared to young mice (37). Therefore, the reduction in tumor expression of GGH during UFT/LV chemotherapy in patients who had right-sided tumors or were aged >75 years in this study may be associated with CpG island methylation.
We previously reported significantly higher TYMP mRNA expression levels in right-sided colon cancer than in left-sided colon cancer (19). The present results suggest that the reduction in GGH gene expression levels during treatment differ depending on the site of tumor and the patient's age, which seem likely to have an intimate bearing upon therapeutic responses to UFT/LV regimens. Further clinical investigation including a DNA methylation assay is needed to clarify the mechanism of the reduction in GGH gene expression in greater detail.
In this study, increase of reduced folate and reduction in GGH expression were associated with the response to UFT/LV chemotherapy in patients with right-sided tumors or aged >75 years old. However, this study was a retrospective pilot study. Prospective investigation of these changes over longer periods of time in patients receiving chemotherapy is needed since the duration of chemotherapy usually extends for several months or even longer, in contrast to the 2-week duration of UFT/LV treatment in this study.
Footnotes
-
Conflicts of Interest
S. Sadahiro, T. Suzuki, A. Tanaka, K. Okada and A. Kamijo have reported no potential conflicts of interest. H. Nagase and J. Uchida are employees of Taiho Pharmaceutical Co., Ltd.
- Received May 14, 2013.
- Revision received June 13, 2013.
- Accepted June 14, 2013.
- Copyright© 2013 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
