Unresectable pancreatic cancer is known to have a poor prognosis, with most patients dying within several months of diagnosis. However, recent progress in chemotherapy using gemcitabine (GEM) for this disease has improved patient survival. A number of phase III clinical trials have been performed to determine the GEM regimens that lead to the greatest increases in survival compared with GEM monotherapy. To date, only one regimen has been shown to yield significantly longer survival periods than GEM alone in phase III studies: GEM with erlotinib, an epidermal growth factor receptor (EGFR)-targeting agent [¹].

S-1 is an oral fluoropyrimidine derivative that contains tegafur (a 5-FU prodrug) and a reversible competitive dihydropyrimidine dehydrogenase (DPD) inhibitor, 5-chloro-2,4-dihydrogenase (CDHP). As DPD is a rate-limiting enzyme that degrades 5-FU, CDHP is expected to enhance the cytotoxicity of 5-FU by prolonging high 5-FU concentrations in blood and tumor tissues [²]. In Japan, S-1 has been clinically used as a first-line chemotherapeutic agent for pancreatic cancer since being approved for national health insurance coverage in 2006. A phase II study of S-1 for 40 patients with metastatic pancreatic cancers resulted in the response rate of 37.5% and the overall survival time of 9.2 months [³].

As the efficacy of S-1 monotherapy against pancreatic cancer is not satisfactory, numerous studies using S-1 combined with GEM have been conducted. Two phase I studies and two phase II studies of the combination therapy showed promising efficacy and acceptable adverse events [^4-7]. A phase III study comparing GEM+S-1 vs. S-1 monotherapy vs. GEM monotherapy for metastatic pancreatic cancer (GEST study) has been underway in Japan and Taiwan since 2007. In contrast to the large number of clinical trials regarding GEM+S-1, pharmacokinetic studies to investigate the interaction between the two agents have been very limited. This is the first study to compare the plasma pharmacokinetics (PK) of GEM and 5-FU after GEM+S-1 to those after single administration of individual drugs in the same patients.

For the last decade, GEM monotherapy has been the standard chemotherapy regimen to treat advanced pancreatic cancer. The drug has an approximately 5% response rate and improves MST to less than 6 months [¹⁰]. Clinical trial data has demonstrated a response rate of 44-48% and an MST of 10.1-12.5 months when S-1 is administrated with GEM [⁶,⁷]. The efficacy of this combination therapy, including our regimen, thus appears to be better than GEM monotherapy, although a true evaluation requires data from the ongoing phase III trial (GEST study).

Our results demonstrated that pre-administration of S-1 did not increase Cmax, AUCinf or T1/2 of plasma GEM (Table 1, Figure 2). Nakamura et al. performed a PK study of GEM with S-1; S-1 was given orally at a dose of 30 mg/m² twice daily for 14 consecutive days, followed by a 1-week rest. GEM 1000 mg/m² was given in a 30-min i.v. on day 8 and day 15. In six patients with metastatic pancreatic cancer, the PK parameters of Cmax and AUCinf for GEM were examined on day 8. It was concluded that their data were similar to those of GEM single-administration, as determined in a phase I study [¹¹] carried out by other investigators [¹²].

The sample size affects the statistical accuracy, however, the ethical matters limit the sample size. There have been some reports statistically comparing the PK parameters between two groups composed of five or six patients [¹³,¹⁴]. In our study on six patients, the statistical analysis was done to detect the relative change of the PK parameters in individual patients using the paired Student's t-test. In this analysis, the statistical power depends on the intra-individual variance and not on the inter-individual variance.

Correale et al. reported that pre-administration of GEM had an effect on the plasma PK of 5-FU [¹⁵]. In their study, 20 patients with metastatic gastroenteric carcinomas were treated with 30 min i.v. of 5-FU 400 mg/m² and folinic acid (FA) 100 mg/m² at 1 h after 30 min i.v. of GEM 1000 mg/m². The control group (5-FU/FA group) consisted of 16 patients with gastroenteric carcinomas receiving 30 min i.v. of 5-FU 400 mg/m² and FA 100 mg/m². The AUC of plasma 5-FU in GEM+5-FU/FA group was approximately twice as high as that in 5-FU/FA group. The Cmax and T1/2 of 5-FU in GEM+5-FU/FA group were higher than those in 5-FU/FA group. The enhanced 5-FU systemic exposure in the presence of GEM may induce severe adverse events as well as high levels of antitumor activity. In fact, a clinical phase I/II trial testing GEM+5-FU/FA for 51 patients with gastroenteric cancers reported frequent grade 4 gastroenteric toxicity and two treatment-related deaths [¹⁵].

In contrast to the study by Correale et al., in our examination, the plasma Cmax, AUCinf and T1/2 of 5-FU after co-administration of S-1 with GEM showed no increases when compared to those after S-1 single-administration (Table 2, Figure 3). Although significant differences were not shown, the mean values of Cmax and AUCinf of 5-FU at day 15 were lower than those at day 3 (Table 2). The reason is obscure, however, continuous administration of S-1 might affect 5-FU pharmacokinetics.

In the catabolic pathways, 5-FU is degraded by DPD. As S-1 contains a very strong DPD inhibitor, CDHP, the Cmax and AUCinf of plasma 5-FU after S-1 administration may reach the limit expected by the amount of tegafur present in S-1. The affects of GEM metabolites on Cmax and AUC of plasma 5-FU after S-1 administration may be little lower than expected based on the presence of CDHP in plasma. The above-mentioned mechanism may explain our results that PK parameters of plasma 5-FU after S-1 administration did not differ with and without GEM administration. Moreover, no enhancement of 5-FU systemic exposure after S-1 administration in the presence of GEM may be an advantage in reducing the frequency of adverse events [¹⁶].

The synergistic effects of S-1 and GEM may be explained by the following mechanism occurring in tumor cells. S-1 is converted into 5-FU. An active metabolite of 5-FU is fluorodeoxyuridine monophosphate (FdUMP), which inhibits DNA synthesis by forming of ternary complex with 5,10-methylene tetrahydrofolate and thymidylate synthase. GEM inhibits ribonucleotid reductase, a key enzyme in the salvage pathway of pyrimidine biosynthesis. Consequently, GEM reduces the synthesis of deoxyuridine monophosphate, a major competitor of FdUMP, resulting enhancement of 5-FU cytotoxicity [¹⁷]. Another potential mechanism is that 5-FU leads to an increase in cell surface human equilibrative nucleoside transporter 1 (hENT1) [¹⁸,¹⁹]. The most active GEM uptake is via hENT1. Thus, increased hENT1 expression by 5-FU may augment GEM cytotoxicity by increasing GEM concentrations in tumor cells.

In conclusion, the present study obtained by the limited number of patients demonstrated the combination chemotherapy of S-1 with GEM did not affect the PK of each drug. As S-1 combined with GEM may be a promising regimen, further investigations should be carried out to elucidate the synergistic mechanisms between the two drugs.

GEM: gemcitabine; 5-FU: 5-fluorouracil; PK: pharmacokinetics; Cmax: maximum concentration; AUCinf: area under the curve from the drug administration to the infinite time; T1/2: elimination half-life; Tmax: time required to reach Cmax; EGFR: epidermal growth factor receptor; DPD: dihydropyrimidine dehydrogenase; CDHP: 5-chloro-2,4-dihydrogenase; MST: median survival time; PS: performance status; i.v.: intravenously; HPLC: high-performance liquid chromatography; LC/MS: high-performance liquid chromatographic-mass spectrometry; FdUMP: fluorodeoxyuridine monophosphate; hENT1: human equilibrative nucleoside transporter 1.

The authors declare that they have no competing interests.

BN have made substantially contribution to conception, design, data analysis, interpretation of data, and drafting the manuscript. RA, SN, TT, OS, TH, and YO have made substantial contributions to patients sample collection and acquisition of data. NY and KH have made contributions to revising the manuscript critically for important intellectual content. All authors read and approved the final manuscript.