Gemcitabine plus S-1: a hopeful frontline treatment for Asian patients with unresectable advanced pancreatic cancer

Transcription

1 Japanese Journal of Clinical Oncology, 2015, 45(12) doi: /jjco/hyv141 Advance Access ublication Date: 30 October 2015 Original Article Original Article emcitabine plus S-1: a hopeful frontline treatment for Asian patients with unresectable advanced pancreatic cancer Chunxiang Cao 1,2,, Meng Kuang 1,,WeiXu 1, Xunlei Zhang 3, Jinfei Chen 1, *, and Cuiju Tang 1, * 1 Department of Oncology, Nanjing First Hospital, Nanjing Medical University, Nanjing, 2 Department of Radiation Oncology, The First Affiliated Hospital of Soochow University, Soochow University, Suzhou, and 3 Department of Oncology, Nantong Tumor Hospital, Nantong *For reprints and all correspondence: Cuiju Tang and Jinfei Chen, Department of Oncology, Nanjing First Hospital, Nanjing Medical University, 68 Changle Road, Nanjing ,. R. China. tangcuiju2013@163.com; jinfeichen@sohu.com Cao Chunxiang and Kuang Meng contributed equally to the writing of this article. Received 29 June 2015; Accepted 24 August 2015 Abstract Objective: emcitabine-based chemotherapy is widely used for unresectable advanced pancreatic cancer which contains locally advanced and metastatic pancreatic cancer. We performed metaanalysis to examine whether gemcitabine plus S-1 could improve treatment efficacy as first-line chemotherapy for those patients when compared with gemcitabine alone. Methods: STATA was used to estimate the summary hazard ratios or odds ratios and their 95% confidence intervals. Heterogeneity among trials was examined by Cochran s χ 2 test. ublication bias was evaluated by Begg s and Egger s tests. Subgroup analysis based on the extent of disease was performed. Results: Four randomized controlled trials including 878 Asian patients were analyzed. In total metaanalysis, gemcitabine plus S-1 significantly improved overall survival (hazard ratio, 0.82; 95% confidence interval, ; = 0.015), progression-free survival (hazard ratio, 0.64; 95% confidence interval, ; < 0.001), overall response rate (odds ratio, 3.00; 95% confidence interval, ; < 0.001) and disease control rate (odds ratio, 1.78; 95% confidence interval, 1.32 to 2.39; < 0.001), and was associated with more but manageable hematologic (leukocytopenia, neutropenia, thrombocytopenia) and non-hematologic (diarrhea, stomatitis, nausea, rash) adverse events. In subgroup analysis, gemcitabine plus S-1, comparing with gemcitabine, significantly improved overall survival in locally advanced patients (hazard ratio, 0.69; 95% confidence interval, 0.48 to 0.99; = 0.022) but not in metastatic patients (hazard ratio, 0.75; 95% confidence interval, ; = 0.256). Conclusion: This meta-analysis confirmed the survival benefits of gemcitabine plus S-1 as first-line treatment for unresectable advanced pancreatic cancer at least in Asia, while good Eastern Cooperative Oncology group performance status was warranted. Importantly, we highlighted the significant overall survival benefit of gemcitabine plus S-1 in locally advanced patients but not in metastatic patients. Key words: pancreatic cancer, gemcitabine, S-1, chemotherapy, meta-analysis The Author ublished by Oxford University ress. All rights reserved. For ermissions, please journals.permissions@oup.com 1122

2 Jpn J Clin Oncol, 2015, Vol. 45, No Introduction ancreatic cancer (C) remains a deadly disease characterized by its low incidence but insidious onset, late diagnosis and bad survival rate (1). It is currently the fourth leading cause of cancer-related mortality in the world (2). Only 20 25% of C patients can be treated with surgical resection and subsequent adjuvant chemotherapy, because most patients are diagnosed with unresectable advanced stages at first presentation. Chemotherapy is the most common option for unresectable advanced pancreatic cancer (UAC) patients (3). Since gemcitabine was proved to be superior to fluorouracil as first-line treatment for advanced C patients in the randomized controlled trial (RCT) [median overall survival (OS): 5.65 vs months], monotherapy of gemcitabine had been approved as the firstline standard regimen in the USA by the food and Drug Administration (FDA) (4). From then on, many groups combined gemcitabine with other drugs, such as fluoropyrimidines, platinum analogs, irinotecan, docetaxel, and reported survival benefits using these combination regimens. While the survival results from RCTs which compared these combination with gemcitabine alone were inconsistent (5 11). A meta-analysis by Heinemann concluded that a significant survival benefit was observed when gemcitabine combined with either fluoropyrimidines or platinum analogs, but such benefit was not found in other combination regimens when compared with gemcitabine alone (12). S-1 is a fourth generation oral fluoropyrimidine, which encompasses tegafur/gimeracil/oteracil potassium at a molar ratio of 1.0:0.4:1.0 (13). S-1, which has been used as a single agent in UAC treatment, is comparably effective and well tolerated to gemcitabine monotherapy (14 16). A preclinical study indicated that S-1 and gemcitabine had synergistic effects on each other (17). Several hase II RCTs indicated that gemcitabine combined with S-1 was effective, tolerable and safe in advanced C patients (18 22). Until now, four English published RCTs have been performed to evaluate the survival efficacy between gemcitabine plus S-1 () and gemcitabine monotherapy as first-line chemotherapy in UAC patients (23 26). Whether could improve treatment efficacy for UAC was controversial in these four trials. Three of them reported OS data in locally advanced pancreatic cancer (LAC) and metastatic pancreatic cancer (MC) patients, respectively, but the results were inconsistent between these trials. Therefore, we conducted this comprehensive meta-analysis to further examine the survival benefit of combination therapy for UAC, and we also performed subgroup analysis based on the extent of disease to select the favorable population for the regimen. Methods Search strategy We searched ubmed (up to 10 May 2015) using various combinations of different terms unresectable, advanced, metastatic, pancreatic cancer, first-line treatment, gemcitabine and S-1. We also searched posters published in the annual meetings of the American Society of Clinical Oncology (ASCO) and the European Society of Medical Oncology in the last decade. The search was limited to RCTs in English language, and reference lists of the selected trials were scanned for any other relevant trials. Selection criteria RCTs fulfilling the following selection criteria were included in this meta-analysis: (i) vs. gemcitabine alone in first-line treatment for patients with LAC or MC; (ii) patients with pathologically or cytologically proven pancreatic adenocarcinoma or adenosquamous carcinoma who had not previously received chemotherapy or radiotherapy; (iii) sufficient data for extraction or have Kaplan Meier curve. hase I trials, retrospective trials, comments, reviews and case reports were excluded. Two independent reviewers (C.C. and K.M.) assessed the eligibility of the clinical trials for inclusion in our meta-analysis. This study was carefully double-checked by another reviewer (X.W.). Data extraction and quality assessment Two reviewers (C.C. and K.M.) extracted the data independently. Data extracted from the studies included publication details, characteristics of patients and information of trial (such as sample size, chemotherapy regimens, treatment line and outcome measures). If we could not extract the hazard ratios (HRs) of OS or progression-free survival (FS) from the original reports directly, we deciphered them from the survival curve as reported by armar et al. (27). The quality of the study was assessed using the five-point Jadad score (28), and only studies with a score of at least 2 were included in the analysis. End points The primary outcome endpoint was OS. The secondary outcome endpoints were FS, overall response rate (ORR) and disease control rate (DCR). Definitions are as follows: OS, defined as the time from randomization to death from any case; FS, defined as the time between randomization and progression, or death without progression, or last date of follow-up for censored patients; ORR, defined as the sum of partial and complete response rates; DCR, defined as the sum of partial response, complete response and stable disease rates. For safety profile, data on rade 3 4 adverse events were extracted and analyzed. Statistical analysis STATA version 10.0 software (Stata Corporation, College Station, Texas, USA) was utilized to estimate the summary HRs and their 95% confidence intervals (CIs) for OS and FS. We also estimated the summary odds ratios (ORs) and their 95% CIs for ORR, DCR and toxicity. A statistical test with a value <0.05 was considered significant. All values were two sided. Cochran s χ 2 test (Q test) was used to test for heterogeneity among trials, and the I 2 index, which expresses the proportion of variability of the results due to heterogeneity, was calculated (29). Significant heterogeneity was considered to be present for heterogeneity 0.1 in the Q test or for I 2 > 50%. If heterogeneity existed, data were analyzed using a random-effects model. In the absence of heterogeneity, a fixedeffects model was used. The presence of publication bias was evaluated by using the Begg s and Egger s tests (30,31). Results Characteristics of included trials Four eligible trials were identified (23 26). Characteristics of these trials, including chemotherapy regimens and outcome measures, are summarized in Table 1. The analysis was performed on the data of the 878 Asian patients with UAC, and they were randomly assigned to receive (432 patients) or gemcitabine alone (439 patients), respectively. Almost all trials were conducted in Japan and all the studies were fully published. The eligibility criteria for enrollment into these four trials were accordant: LAC or MC, histologically or cytologically proven diagnosis of adenocarcinoma or adenosquamous carcinoma, no prior

3 1124 emcitabine plus S-1 for UAC Table 1. Characteristics of enrolled trials, including treatment regimens and survival data Study Country hase No. Treatment Regimen Median OS (months) Median FS (months) ORR (%) DCR (%) Jadad Score Nakai et al. (23) Ozaka et al. (24) Ueno et al. (25) Sudo et al. (26) Japan II 106 : 1000 mg/m 2 over 30 min on days 1 and 15; +S-1: twice daily according to the BSA (BSA 1.25 m 2, 80 mg/day; 1.25 m 2 BSA 1.5 m 2, 100 mg/day; BSA 1.5 m 2, 120 mg/day) for 2 weeks. Cycle: 4. : 1000 mg/m 2 over 30 min on days 1, 8 and 15. Cycle: 4. Japan II 117 : 1000 mg/m 2 during 30 min on days 1 and 8; +S-1: 40 mg/m 2 twice daily for 2 weeks. Cycle: 3. : 1000 mg/m 2 during 30 min on days 1, 8 and 15. Cycle: 4. Japan, Taiwan China chemotherapy or radiotherapy for C, age of years, an Eastern Cooperative Oncology group performance status (ECO S) score of 0 1 (99.43% of patients) or 2 (0.57% of patients), and adequate organ functions. atient characteristics were well balanced, and more details are shown in Table 2. All adverse events were assessed according to the Common Terminology Criteria for Adverse Events, version 3.0. Computed Tomography or magnetic resonance imaging scans were performed at the baseline. The response was assessed by the investigators according to the Response Evaluation Criteria in Solid Tumors (RECIST), version 1.0 (23 26). Total meta-analysis III 554 : 1000 mg/m 2 over 30 min on days 1 and 8; +S-1: twice daily according to the BSA (BSA 1.25 m 2, 60 mg/day; 1.25 m 2 BSA 1.5 m 2, 80 mg/day; BSA 1.5 m 2, 100 mg/day) for 2 weeks. Cycle: 3. : 1000 mg/m 2 over 30 min on days 1, 8 and 15. Cycle: 4. Japan NR 101 : 1000 mg/m 2 during 30 min on days 8 and 15; +S-1: 30 mg/m 2 twice daily on days Cycle: 3. : 1000 mg/m 2 during 30 min on days 1, 8 and 15. Cycle: 4. OS and FS All trails provided OS and FS data directly or indirectly (Table 1). ooled analysis indicated that was associated with clinically substantial and statistically significant reduction by 18% in the hazard for death and 36% in the hazard for progression when compared with gemcitabine alone (HR, 0.82; 95% CI, , Fig. 1A; < < < NR , gemcitabine; No., number of patients; BSA, body surface area;, value; OS, overall survival; FS, progression-free survival; ORR, overall response rate; DCR, disease control rate; NR, not reported. HR, 0.64; 95% CI, , Fig. 1B), respectively. There was no significant heterogeneity (OS: I 2 = 0.0%, heterogeneity = 0.453; FS: I 2 = 0.0%, heterogeneity = 0.920), and the pooled HRs for OS and FS were performed by using fixed-effects models. ORR and DCR We extracted data on response rate and DCR from the reports (Table 1). The results of our analysis demonstrated that a substantial and significant increase of 200% in the OR for response and of 78% in the OR for disease control were attributable to regimen compared with gemcitabine monotherapy (OR, 3.00; 95% CI, , Fig. 2A; OR, 1.78; 95% CI, , Fig. 2B). No statistically significant heterogeneity in the ORs for ORR or DCR was observed (ORR: I 2 =0.0%, heterogeneity =0.640;DCR:I 2 =0.0%, heterogeneity = 0.475). Therefore, fixed-effects models were performed. Toxicity Four trials including 851 patients provided toxicity-profile results (Table 3). Both rade 3 4 hematologic and non-hematologic

4 Jpn J Clin Oncol, 2015, Vol. 45, No Table 2. Characteristics of the patients enrolled in the meta-analysis Study Nakai et al. (23) Ozaka et al. (24) Ueno et al. (25) Sudo et al. (26) Characteristic (n = 53) (n = 53) (n = 53) (n = 59) (n = 275) (n = 277) (n = 51) (n = 50) Age, years >20 20 and 80 >20 and < Sex, n (%) Male 42 (79.2%) 33 (62.3%) 32 (60.4%) 35 (59.3%) 158 (57.5%) 170 (61.4%) 27 (52.9%) 34 (68%) Female 11 (20.8%) 20 (37.8%) 21 (39.6%) 24 (40.7%) 117 (42.5%) 107 (38.6%) 24 (47.1%) 16 (32%) ECO S, n (%) 0 31 (58.5%) 32 (60.4%) 44 (83.0%) 45 (76.3%) 172 (62.5%) 181 (65.3%) 35 (68.6%) 35 (70%) 1 22 (41.5%) 20 (37.7%) 9 (17.0%) a 14 (23.7%) a 103 (37.5%) 96 (34.7%) 16 (31.4%) 15 (30%) (1.9%) Extent of disease, n (%) Locally advanced 15 (28.3%) 13 (24.5%) 13 (24.5%) 18 (30.5%) 68 (24.7%) 66 (23.8%) 18 (35.3%) 19 (38%) Metastatic 38 (71.7%) 40 (75.5%) 40 (75.5%) 41 (69.5%) 207 (75.3%) 211 (76.2%) 33 (64.7%) 31 (62%) Tumor location, n (%) Head 20 (37.7%) 18 (34.0%) NR NR 116 (42.2%) 122 (44.0%) 22 (43.1%) 18 (36%) Body 10 (18.9%) 12 (22.6%) NR NR 102 (37.1%) 88 (31.8%) 29 (56.9%) b 32 (64%) b Tail 23 (43.4%) 23 (43.4%) NR NR 66 (24.0%) 68 (24.5%) Metastatic sites, n (%) Liver 24 (45.3%) 25 (47.2%) 28 (52.8%) 30 (50.8%) NR NR 23 (45.1%) 19 (38%) Lymph node 21 (39.6%) 20 (37.7%) 6 (11.3%) 10 (16.9%) NR NR NR NR eritoneum 7 (13.2%) 10 (18.9%) 12 (22.6%) 7 (11.9%) NR NR 9 (17.6%) 9 (18%) Lung 1 (1.9%) 3 (5.7%) 8 (15.1%) 3 (5.1%) NR NR 4 (7.8%) 6 (12%) CA19 9, U/ml Median NR NR , gemcitabine plus S-1;, gemcitabine; n, number; ECO S, Eastern Cooperative Oncology group performance status; NR, not reported. a Total number of patients with an ECO S of 1 and 2; b Total number of patients with tumor located at body and tail of pancreas. Figure 1. ooled HRs (95% CI) of OS (A) and FS (B) associated with combination therapy compared with gemcitabine monotherapy. CI, confidence interval; OS, overall survival; FS, progression-free survival;, gemcitabine plus S-1. toxicities were analyzed. atients in the group had significantly higher incidences of rade 3 or 4 adverse events, including leukocytopenia, neutropenia, thrombocytopenia, diarrhea, stomatitis, nausea and rash (all ORs > 1 and values <0.05). All adverse events were tolerated and manageable. In total, only seven and four treat-related deaths at most were reported in two groups without significant difference ( = 0.376). Subgroup meta-analysis In total, 230 LAC and 641 MC patients were included in our research. OS data based on these two different populations were provided, respectively, in all trials except the trial by Sudo et al. (23 26). The median OS time reported in each trial was longer in group than that in corresponding gemcitabine group (Table 4). HRs of OS for LAC or MC patients were provided in EST study.

5 1126 emcitabine plus S-1 for UAC Figure 2. ooled ORs (95% CI) of ORR (A) and DCR (B) associated with combination therapy compared with gemcitabine monotherapy. OR, odds ratio; ORR, overall response rate; DCR, disease control rate. Table 3. Summary of rade 3 or worse adverse events Study Nakai et al. (23) Ozaka et al. (24) Ueno et al. (25) Sudo et al. (26) Total AEs (%) (n = 51) (n = 52) (n = 53) (n = 59) (n = 267) (n = 273) (n = 50) (n = 46) (n = 421) (n = 430) Hematologic Leukocytopenia Neutropenia Thrombocytopenia Anemia Non-hematologic Fatigue Anorexia Nausea/vomiting Diarrhea Stomatitis Rash neumonitis Febrile neutropenia AST ALT Bilirubin I hemorrhage Constipation Treat-related death AEs, adverse events. represents none or not reported. value Kaplan Meier curves of OS in these two different populations were provided in C-01 study, and we deciphered the HRs of OS from the survival curves. Finally, we pooled the HRs of OS in each subgroup. FS and other survival data in these two populations were not adequate to perform subgroup analysis. LAC The HR for mortality in patients with LAC was 0.69 (95% CI, ). It was marginal but significant (Fig. 3A). There was no significant heterogeneity (I 2 = 0.0%, heterogeneity = 0.708), so the pooled HR for OS in this subgroup was performed by fixed-effects model. MC The HR for mortality was 0.75 in the subgroup with MC (95% CI, ), which indicated a 25% decrease in the HR for mortality, but it was not significant (Fig. 3B). Because we found significant heterogeneity when pooled HRs for OS (I 2 = 74.7%, heterogeneity = 0.704), random-effects model was finally used.

6 Jpn J Clin Oncol, 2015, Vol. 45, No Table 4. Survival data in subgroup analysis (month) LAC MC value HR value HR Nakai et al. (23) Median OS Median FS Ozaka et al. (24) Median OS (0.26, 2.77) a (0.35, 0.87) a Median FS Ueno et al. (25) Median OS (0.46, 0.99) (0.76, 1.15) Median FS HR, hazard ratio of OS or FS. represents not reported. Sudo et al. did not report survival data on each population, so their study was not presented. a Deciphered from the survival curve as reported by armar et al. Figure 3. Subgroup analysis of OS: patients with locally advanced pancreatic cancer (A) and patients with metastatic pancreatic cancer (B). ublication bias Funnel plots (Fig. 4) did not show any evidence of publication bias, although the number of included trials was relatively small. Discussion Our meta-analysis demonstrated that was associated with superior OS, FS, ORR and DCR compared with gemcitabine in UAC Asian patients. Although patients in arm were more likely to experience severe hematologic and non-hematologic adverse events, the adverse events were predictable, manageable and acceptable. Accordingly, we conclude that the regimen would be accepted as a standard treatment option for patients with UAC at least in Asia, while ECO S 1 was warranted. Three of the four trials failed to demonstrate significant improvement in OS (23,25,26). As discussed in these trials, second-line chemotherapy mainly with S-1 in the gemcitabine group might have contributed to the prolonged survival. In total, 60% of patients used S-1 as second-line treatment after gemcitabine failure in our analysis (data not shown). In spite of this crossover, there was a statistically significant improvement in OS in the group by pooling the data of the four trials together. With respect to FS, the HR for FS in the C-01 study (24), which was not provided directly in the article, was deciphered by the Kaplan Meier curve in our analysis (HR, 0.57; 95% CI, ). All the four trials demonstrated significant improvements in FS in favor of regimen, and was associated with a marked and statistically significant reduction by 36% in the hazard for progression as pooling data. For ORR and DCR, some trials did not show a higher significant response or disease control rate in the group compared with gemcitabine group (23,26), but a substantial improvement of ORR or DCR in arm was eventually observed after pooling data together. Because locally advanced and metastatic diseases are considered as two different clinical entities with its own biological and clinical characteristics, it is recommended that locally advanced disease should be separately studied from metastatic disease (32). In our analysis, all enrolled trials included patients with LAC as well as those with MC (Table 1). In total, 73.6% of participants had metastatic cancer. It is necessary to confirm survival benefit of in each population. Three studies performed subgroup analysis in OS based on the two different clinical entities (23 25). In the EMSA study, median OS in

7 1128 emcitabine plus S-1 for UAC Figure 4. Begg s funnel plots of publication bias test: OS (A); FS (B). locally advanced or metastatic disease was longer in arm than that in gemcitabine arm (LAC: 23.9 vs months; MC: 8.9 vs. 7.9 months), but the difference was not significant in each subgroup (LAC: = 0.297; MC: = 0.311). But HRs for OS in this subset were not provided (23). Interestingly, the C-01 and EST study came to a contrary conclusion in each subset. In detail, the C-01 trial reported was favorable in metastatic disease, while the EST study favors in locally advanced disease (24,25). In the C-01 trial, the HR in OS of each subgroup was deciphered by the Kaplan Meier curve as reported by armar et al. (27). As a result, OS in the subgroup of patients with LAC was significantly improved in favor of the regimen in our analysis, which was in accordance with the results reported by Yanagimoto et al. (33) (HR of OS: our study, 0.69 [95% CI, ]; Yanagimoto et al s study: 0.67 [95% CI, ]). Besides analysis on OS, Yanagimoto et al. reported that also improved FS and ORR in LAC patients over gemcitabine alone, confirming survival benefit of in LAC population. In addition, Yanagimoto et al performed pooled analysis instead of meta-analysis which means they utilized individual data that we could not get directly from published trials. This superiority of their study could help them to draw a more reliable conclusion. However, they did not study MC patients in these two chemotherapy regimens. To some degree, our results in MC subgroup made up for this area. As far as our knowledge, we are the first to report that there was no significant difference between and gemcitabine regimen in MC patients. Of course, the limitation of our subgroup analysis especially the analysis in MC patients should be paid attention. First, we only analyzed OS in MC patients because of the limited published data. FS and ORR in MC population in each regimen were not reported in publications. Second, published data but not individual and updated data were used which means the results in MC patients should be interpreted with care and doubt. Third, there were only two studies included in the subgroup analysis indicating the limited ability to estimate the difference. While, our negative results in MC patients could probability arouse interest in conducting more RCTs to study regimen in MC, and texting the necessity of studying LAC separately from MC in the future RCTs. Importantly, surgical resection is the main goal for borderline resectable C, while palliative systemic management is the treatment option for those patients with advanced disease. The chance of resection for borderline resectable C and LAC after chemotherapy or radiotherapy always raises great attention (34). Among the four trials, only one trial conducted by Sudo et al. reported the number of LAC who were eligible for resection after chemotherapy. There were three patients obtained the chance for surgery in the arm and two of them achieved R0 resection, but no patient in the gemcitabine arm underwent surgery. However, the other three studies did not report the eligibility patients for resection after participation. Therefore, we could not give a comprehensive evaluation on this subject. While, we speculate that the regimen could result in tumor shrinkage and could be considered and deserve further study in the neoadjuvant chemotherapy for borderline resectable C and LAC patients. Apart from the extent of disease, there are several other factors that may also affect the results of meta-analysis. A meta-analysis by Heinmennan demonstrated that patients with a good ECO S (S: 0 or 1) had a survival benefit from gemcitabine combination, whereas those with a S of 2 did not. atients with a S of 2 are probably suitable for gemcitabine monotherapy. erformance status is confirmed to be a critical factor for gemcitabine-based combination regimens (12). In our meta-analysis, 99.43% of patients were in good condition (S: 0 or 1). Thus, the results from our meta-analysis could not be simply extrapolated to patients with poor S (S 2). Additionally, patients were also stratified by enrolling center, with or without pathological diagnosis,andpresentorabsentofpainduetocancerinsomeincluded studies. However, data in these cases were inadequate to obtain pooled results. More trials are needed to further evaluate these influence factors. Several other limitations should also be considered in our analysis. First, the paucity of hase II or III trials currently available for meta-analysis is the major limitation of this study. Small sample size with 878 patients could not confirm the conclusion. Second, the schedules and planned dose intensity in groups were somewhat different among the trials which might contribute to increase the clinical heterogeneity of the meta-analysis (Table 1). The first-line regimen for LAC patients was the same with that for MC patients in each trial, while the second-line chemotherapy for these two populations might be different which would affect OS. Third, the proportion of LAC and MC may be an important factor affecting the survival outcomes. 73.6% of participants had MC but only 26.4% of participants had LAC, which might give a lower estimation for OS. Finally, only the EST study assessed the quality of life and reported the results of quality-adjusted life-years (QALYs), which are important issues in determining the true value of chemotherapy. The EST

8 Jpn J Clin Oncol, 2015, Vol. 45, No study indicated that group was associated with significantly better QALY value when compared with gemcitabine group ( < 0.001) (25). FOLFIRINOX (oxaliplatin and irinotecan plus fluorouracil and leucovorin) and regimen of weekly nab-paclitaxel (brand name Abraxane) plus gemcitabine emerged to demonstrate marked survival improvements with more severe but acceptable adverse events in the first-line treatment of UAC in comparison with gemcitabine monotherapy (35,36). While an ECO S 1 or a KS of was warranted in these two effective regimens. In our analysis, regimen also showed promising anti-cancer activity and acceptable toxicities in the patients with S 1. regimen should be compared and evaluated with above two regimens in RCTs to help clinicians, and oncologists make proper decisions for suitable patients. In summary, when compared with gemcitabine monotherapy, combination therapy was associated with significantly increased survival benefits with its increased but acceptable toxicities for Asian patients with UAC. While, survival advantages from was only applicable to patients with ECO S 1. In terms of OS, LAC patients benefited from regimen in comparison with gemcitabine alone in our analysis, while metastatic patients did not. regimen raises hope for UAC patients and more multicenter RCTs should be conducted to further confirm its survival superiority and to select the proper populations. regimen also deserves further investigation in neoadjuvant chemotherapy for LAC patients. Funding This study was supported by National Basic Science Research Development rogram of China (2013CB911302). Conflicts of interest statement None declared. References 1. Shi S, Yao W, Xu J, Long J, Liu C, Yu X. Combinational therapy: new hope for pancreatic cancer? Cancer Lett 2012;317: Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, Jemal A. lobal cancer statistics, CA Cancer J Clin 2015;65: udjonsson B. Cancer of the pancreas. 50 years of surgery. Cancer 1987;60: Burris HA III, Moore MJ, Andersen J, et al. Improvements in survival and clinical benefit with gemcitabine as first-line therapy for patients with advanced pancreas cancer: a randomized trial. J Clin Oncol 1997;15: Berlin JD, Catalano, Thomas J, Kugler JW, Haller D, Benson AB III. hase III study of gemcitabine in combination with fluorouracil versus gemcitabine alone in patients with advanced pancreatic carcinoma: Eastern Cooperative Oncology roup Trial E2297. J Clin Oncol 2002;20: Di Costanzo F, Carlini, Doni L, et al. emcitabine with or without continuous infusion 5-FU in advanced pancreatic cancer: a randomised phase II trial of the Italian oncology group for clinical research (OIRC). Br J Cancer 2005;93: Louvet C, Labianca R, Hammel, et al. emcitabine in combination with oxaliplatin compared with gemcitabine alone in locally advanced or metastatic pancreatic cancer: results of a ERCOR and ISCAD phase III trial. J Clin Oncol 2005;23: oplin E, Feng Y, Berlin J, et al. hase III, randomized study of gemcitabine and oxaliplatin versus gemcitabine (fixed-dose rate infusion) compared with gemcitabine (30-minute infusion) in patients with pancreatic carcinoma E6201: a trial of the Eastern Cooperative Oncology roup. J Clin Oncol 2009;27: Colucci, Labianca R, Di Costanzo F, et al. Randomized phase III trial of gemcitabine plus cisplatin compared with single-agent gemcitabine as firstline treatment of patients with advanced pancreatic cancer: the I-1 study. J Clin Oncol 2010;28: Kulke MH, Tempero MA, Niedzwiecki D, et al. Randomized phase II study of gemcitabine administered at a fixed dose rate or in combination with cisplatin, docetaxel, or irinotecan in patients with metastatic pancreatic cancer: CALB J Clin Oncol 2009;27: Stathopoulos, Syrigos K, Aravantinos, et al. A multicenter phase III trial comparing irinotecan-gemcitabine (I) with gemcitabine () monotherapy as first-line treatment in patients with locally advanced or metastatic pancreatic cancer. Br J Cancer 2006;95: Heinemann V, Boeck S, Hinke A, Labianca R, Louvet C. Meta-analysis of randomized trials: evaluation of benefit from gemcitabine-based combination chemotherapy applied in advanced pancreatic cancer. BMC Cancer 2008;8: Shirasaka T, Nakano K, Takechi T, et al. Antitumor activity of 1 M tegafur-0.4 M 5-chloro-2,4-dihydroxypyridine-1 M potassium oxonate (S-1) against human colon carcinoma orthotopically implanted into nude rats. Cancer Res 1996;56: Ueno H, Okusaka T, Ikeda M, Takezako Y, Morizane C. An early phase II study of S-1 in patients with metastatic pancreatic cancer. Oncology 2005;68: Okusaka T, Funakoshi A, Furuse J, et al. A late phase II study of S-1 for metastatic pancreatic cancer. Cancer Chemother harmacol 2008;61: Schultheis B, Strumberg D, Bergmann L, et al. Results of a phase II trial of S-1 as first-line treatment of metastatic pancreatic cancer (CESAR-study group). Invest New Drugs 2012;30: Yoshizawa J, Takizawa A, Takeuchi O, et al. Experimental study of combination therapy with S-1 against pancreatic cancer. Cancer Chemother harmacol 2009;64: Song H, Han B, ark CK, et al. hase II trial of gemcitabine and S-1 for patients with advanced pancreatic cancer. Cancer Chemother harmacol 2013;72: Ueno H, Okusaka T, Furuse J, et al. Multicenter phase II study of gemcitabine and S-1 combination therapy ( Therapy) in patients with metastatic pancreatic cancer. Jpn J Clin Oncol 2011;41: Lee W, Kim HJ, Ju JH, et al. hase II trial of S-1 in combination with gemcitabine for chemo-naive patients with locally advanced or metastatic pancreatic cancer. Cancer Chemother harmacol 2009;64: Oh DY, Cha Y, Choi IS, et al. A multicenter phase II study of gemcitabine and S-1 combination chemotherapy in patients with unresectable pancreatic cancer. Cancer Chemother harmacol 2010;65: Nakamura K, Yamaguchi T, Ishihara T, Sudo K, Kato H, Saisho H. hase II trial of oral S-1 combined with gemcitabine in metastatic pancreatic cancer. Br J Cancer 2006;94: Nakai Y, Isayama H, Sasaki T, et al. A multicentre randomised phase II trial of gemcitabine alone vs gemcitabine and S-1 combination therapy in advanced pancreatic cancer: EMSA study. Br J Cancer 2012;106: Ozaka M, Matsumura Y, Ishii H, et al. Randomized phase II study of gemcitabine and S-1 combination versus gemcitabine alone in the treatment of unresectable advanced pancreatic cancer (Japan Clinical Cancer Research Organization C-01 study). Cancer Chemother harmacol 2012; 69: Ueno H, Ioka T, Ikeda M, et al. Randomized phase III study of gemcitabine plus S-1, S-1 alone, or gemcitabine alone in patients with locally advanced and metastatic pancreatic cancer in Japan and Taiwan: EST study. J Clin Oncol 2013;31: Sudo K, Ishihara T, Hirata N, et al. Randomized controlled study of gemcitabine plus S-1 combination chemotherapy versus gemcitabine for unresectable pancreatic cancer. Cancer Chemother harmacol 2014;73: armar MK, Torri V, Stewart L. Extracting summary statistics to perform meta-analyses of the published literature for survival endpoints. Stat Med 1998;17:

9 1130 emcitabine plus S-1 for UAC 28. Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996;17: Zintzaras E, Ioannidis J. Heterogeneity testing in meta-analysis of genome searches. enet Epidemiol 2005;28: Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for publication bias. Biometrics 1994;50: Egger M, Davey Smith, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315: hilip A, Mooney M, Jaffe D, et al. Consensus report of the national cancer institute clinical trials planning meeting on pancreas cancer treatment. J Clin Oncol 2009;27: Yanagimoto H, Ishii H, Nakai Y, et al. Improved survival with combined gemcitabine and S-1 for locally advanced pancreatic cancer: pooled analysis of three randomized studies. J Hepatobiliary ancreat Sci 2014;21: Werner J, Combs SE, Springfeld C, Hartwig W, Hackert T, Buchler MW. Advanced-stage pancreatic cancer: therapy options. Nat Rev Clin Oncol 2013;10: Vaccaro V, Sperduti I, Milella M. FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. NEnglJMed2011;365:768 9; author reply Von Hoff DD, Ervin T, Arena F, et al. Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine. N Engl J Med 2013;369: