Introduction
Gastric cancer was the fifth most common in new cases of cancer (1,033,701 cases) and the third most common in cancer-related deaths (781,631 deaths) in 2018 (Bray et al.
2018). Curative resection and postoperative adjuvant chemotherapy is the standard treatment for locally advanced gastric cancer in Japan. Since Adjuvant Chemotherapy Trial of S-1 for Gastric Cancer (ACTS-GC), adjuvant chemotherapy with S-1 is one of the standard treatments for pathological TNM stage II/III gastric cancer (except pT1N2-3/pT3N0) for the prevention of recurrence (Sakuramoto et al.
2007; Japanese gastric cancer treatment guidelines 2014 (ver. 4)
2017). Moreover, CLASSIC and JACCRO-07 trials have been used to confirm the effectiveness of capecitabine plus oxaliplatin (CapeOX) therapy and S-1 plus docetaxel (DS) therapy; fluoropyrimidine is still the key drug used in the postoperative adjuvant chemotherapy for gastric cancer (Sakuramoto et al.
2007; Bang et al.
2012; Noh et al.
2014). However, the outcomes of treatment are still insufficient, as recurrence occurs in 20–60% of patients, even after a complete resection and the administration of the appropriate adjuvant therapy (Maehara et al.
2000; Rivera et al.
2007). The personalization of postoperative adjuvant chemotherapy treatment using biomarkers is a promising strategy to improve the survival of patients with localized advanced gastric cancer. However, a specific biomarker for use in predicting the therapeutic effects of fluorinated pyrimidine and the long-term outcomes of patients with locally advanced gastric cancer has yet to be identified (Pizzorno et al.
1988; Shubbar et al.
2013; Melling et al.
2017).
Folic acid is a molecule which is necessary for cell proliferation, DNA synthesis, and repair. Gamma-glutamyl hydrolase (GGH) and folylpolyglutamate synthetase (FPGS) are enzymes which regulate intracellular folate concentrations (Bailey
2010). GGH promotes the production of monoglutamyl acid folate, a metabolite of folic acid required for DNA synthesis. On the other hand, FPGS catalyzes the hydrolysis of monoglutamate folate into polyglutamate, which has a high intracellular retention (Bailey
2010). Thus, GGH and FPGS strongly influence DNA synthesis in cancer cells.
Previous studies have reported that high GGH expression is a risk factor for the prognosis of various malignancies (Shubbar et al.
2013; Melling et al.
2017). Moreover, the expression of GGH and FPGS was reported to affect thymidylate synthase (TS) inhibition of 5-fluorouracil (5-FU) in a range of malignancies (Cheradame et al.
1997; Sakamoto et al.
2008; Kim et al.
2013). Recently, it was reported that high GGH expression is a risk factor of lymph node recurrence after surgery in patients with stage II/III gastric cancer, using clinical specimens of ACTS-GC (Terashima et al.
2017). However, the relationship between the expression of GGH and FPGS and the effect of S-1 in adjuvant chemotherapy has not yet been evaluated. In this study, we examined the clinical significance of GGH and FPGS expression in patients with stage II/III gastric cancer who were undertaking postoperative adjuvant chemotherapy with S-1.
Discussion
In this study, we measured the levels of GGH and FPGS mRNA expression in cancer tissues and adjacent normal mucosa in patients with stage II/III gastric cancer. We then examined the relationship between the expression levels of these genes and the clinicopathological features and long-term outcomes to evaluate the clinical significance of GGH and FPGS mRNA expression in gastric cancer tissue in patients with Stage II/III gastric cancer undergoing postoperative adjuvant chemotherapy with S-1.
First, the expression levels of
GGH and
FPGS mRNA in gastric cancer tissue and adjacent normal mucosa were compared. Several previous studies have compared the relative expression levels of
GGH and
FPGS mRNA between various types of cancer tissue and adjacent normal tissue (Kidd et al.
2005; Pollard et al.
2009; Shubbar et al.
2013). These studies reported that
GGH mRNA expression was higher in cancer tissue compared to normal tissue in breast cancer (Shubbar et al.
2013) and bladder cancer (Pollard et al.
2009). Our results are consistent with these findings, as the expression levels of
GGH mRNA were found to be significantly higher in the gastric cancer tissue than in the paired adjacent normal mucosa. As for
FPGS expression, it has been previously reported that the expression of
FPGS was higher in cancer tissue than normal mucosa in colorectal cancer (Odin et al.
2003; Kidd et al.
2005). However, in our study, there was no significant difference in the expression levels of
FPGS mRNA between the cancer tissue and adjacent normal tissue.
Next, the relationship between the expression levels of
GGH and
FPGS mRNA and the clinicopathological features in patients with stage II/III gastric cancer were examined. It was reported that GGH expression was significantly associated with a high histological tumor grade (BRE grade III,
P < 0.001), as well as ER/PR receptors in patients with breast cancer (Shubbar et al.
2013). As for FPGS, to the best of our knowledge, there are no reports. In this study, high levels of
GGH mRNA expression were significantly correlated with age, histological type, and vascular invasion. High levels of
FPGS mRNA expression, on the other hand, were found to be related to vascular invasion and lymphatic invasion.
The relationship between the expression levels of
GGH and
FPGS mRNA in cancer tissue and long-term outcomes in patients with stage II/III gastric cancer was then assessed. Several studies have previously reported that patients with high levels of
GGH mRNA expression in cancer tissue resulted in significantly poor outcomes compared to patients with low levels of
GGH mRNA expression. Melling et al. (
2017) reported that 10-year recurrence-free survival was significantly higher in patients with high levels of
GGH expression compared to patients with low level of
GGH expression in ERG negative prostate cancer (
P = 0.0002). Shubber et al. (
2013) reported that 8-year disease-specific survival (DSS) was significantly different between patients expressing GGH (39%) and patients whose tumors were GGH-negative (68%,
P = 0.037) in invasive breast cancer. In addition, univariate analysis showed that GGH expression exhibited a lower DSS probability, with a 2.7-fold increase in risk of death (
P = 0.04) (Shubbar et al.
2013). As for FPGS, it was reported that 5-year tumor-specific survival (TSS) was significantly better in patients with high levels of
FPGS mRNA expression (75%) than patients with low levels of
FPGS mRNA expression (35%,
P = 0.002). In addition,
FPGS expression has been previously reported to be an independent significant prognostic factor of TSS in patients with colorectal cancer (Odin et al.
2003).
In present study, the 5-year OS was significantly poorer in patients with high levels of GGH mRNA expression than in those with low expression levels. In patients with stage II/III gastric cancer who were administrated adjuvant chemotherapy with S-1, although there was no significant difference in the 5-year OS between patients with high and low levels of FPGS mRNA expression, the 5-year OS in patients with the combination of high GGH mRNA and low FPGS mRNA expression levels in cancer tissue was significantly poorer than that in the other patients.
The elucidation of the mechanism whereby high levels of
GGH expression and low levels of FPGS expression could be used as prognostic biomarkers in patients with locally advanced gastric cancer for which the administration of postoperative adjuvant chemotherapy with S-1 is not currently sufficient. Previous reports have suggested the following mechanism: 5-FU is an active ingredient of S-1 that inhibits the action of thymidylate synthase (TS) and suppresses DNA synthesis and cell proliferation by forming a trimer with 5-fluorodeoxyuridylate (Fd-UMP), which is a metabolite of 5-FU (Longley et al.
2003; Wilson et al.
2014). The polyglutamate folate is produced by the folate metabolism and induces the formation of the trimer (Moran
1999). GGH promotes the production of monoglutamyl acid folate, a metabolite of folic acid required for DNA synthesis. Meanwhile, FPGS catalyzes the hydrolysis of monoglutamylate folate into polyglutamate, which has a high intracellular retention (Bailey
2010). Thus, high
GGH and/or low
FPGS activity reduces 5,10-methylenetetrahydrofolic acid, as well as the TS, which can bind to 5,10-methylenetetrahydrofolic acid. As a result, it is possible that a small amount of 5-FU can exert a TS inhibitory effect in patients with high
GGH and/or low
FPGS activity (Moran
1999). These mechanisms indicated that the outcomes of patients with high levels of
GGH expression and low levels of
FPGS expression in cancer tissue were poorer than the other patients. Moreover, when combined with our results, these findings suggest that a combination of high
GGH expression and low
FPGS expression in cancer tissue could be used as prognostic biomarkers in patients with locally advanced gastric cancer undergoing postoperative adjuvant chemotherapy with S-1 after curative resection.
This study has several limitations. First, the study examined only mRNA expression in gastric cancer tissues. Considering its clinical utility as a biomarker, future studies should examine both mRNA and protein expression levels in the same specimen. Second, there was an issue regarding the heterogeneity of the gastric cancer tissue. The sample from which the mRNA was extracted was a 5-mm square stomach cancer tissue, including the deepest part, which, however, did not completely represent the entire tumor. Third, since GGH and FPGS have contrasting effects, the prognosis of the high GGH/low FPGS group could be predicted to be the worst, and, inversely, that of the low GGH/high FPGS group the best. The present study demonstrated that a significant difference was observed when comparing the OS in the four groups based on different combinations of GGH and FPGS mRNA expression. In addition, there was an identically significant difference in the OS of the four groups with patients that had received S-1 adjuvant treatment and the analysis of the two groups with high or low GGH. Although the obtained result was different from the hypothesis, which stated that the prognosis of the low GGH/high FPGS group would be the best, the number of low GGH/low FPGS patients was nine, which may be the cause of the lack of statistical power. As such, it is possible that the combination of GGH/FPGS mRNA may have an enhanced accuracy as a biomarker.
In conclusion, high levels of GGH mRNA expression and low levels of FPGS mRNA expression in cancer tissue may be useful predictive biomarkers for the survival of patients with stage II/III gastric cancer undergoing postoperative adjuvant chemotherapy with S-1 after radical resection. It would be interesting for future studies to verify the possibility of personalizing treatments by selecting the appropriate regimen based on the expression levels of GGH and FPGS for patients with locally advanced gastric cancer in a clinical trial.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.