ABSTRACT
BACKGROUND:
The 5-year survival of pancreatic adenocarcinoma with surgery and adjuvant chemotherapy is below 25%. The original Gastrointestinal Tumor Study Group (GITSG) adjuvant study demonstrated a survival benefit attributed to weekly intravenous boluses of 5-fluorouracil (5-FU) for 2 years in addition to chemoradiation compared to surgery alone. In theory, the prolonged exposure to therapy could maintain pressure on dormant cancer cells that remain in G0 arrest and kill them as they infrequently enter the G1/S phase. We retrospectively evaluated outcomes in patients who were treated with adjuvant chemotherapy and maintenance capecitabine compared with those who received only adjuvant chemotherapy.
METHODS:
Patients who had undergone surgical resection with curative intent and received adjuvant chemotherapy were analyzed. Those who subsequently received maintenance capecitabine therapy were compared to those who received adjuvant chemotherapy only. The primary end points were disease recurrence and all-cause mortality.
RESULTS:
The median overall survival (OS) of patients receiving maintenance capecitabine was greater than 48.4 months (the exact estimate was not available, since the survival probability curve does not cross 0.5). It was 22.0 months (95% confidence interval [CI], 16.6–29.2) in patients who received adjuvant chemotherapy only (P < .001 by log-rank test). The median recurrence-free survival (RFS) was also longer in the maintenance capecitabine group: 54.3 (95% CI, 22.2–Inf) compared to 14.1 (95% CI, 11.6–16.7) months (P < .001, by log-rank test).
CONCLUSIONS:
In this retrospective study, patients with resected pancreatic adenocarcinoma who received adjuvant chemotherapy had improved OS and RFS with additional maintenance therapy with capecitabine. These findings should be confirmed with a randomized, controlled trial.
Pancreatic cancer has an altogether poor prognosis and is the fourth leading cause of cancer deaths in the United States.1 Even in patients in whom the disease is resectable, recurrence is common, and the most common cause of mortality is systemic disease.2 The 5-year survival of pancreatic adenocarcinoma with surgery and adjuvant chemotherapy remains less than 25%.1
The original Gastrointestinal Tumor Study Group (GITSG) adjuvant study evaluated postoperative therapy with radiation and concurrent and subsequent weekly bolus 5-fluorouracil (5-FU). The latter was administered for 2 years. The experimental arm demonstrated a survival benefit of chemoradiation compared to surgery alone.3 Therefore, chemoradiation became the standard adjuvant therapy for pancreatic cancer in the United States. The benefit of chemoradiation was called into question by trials that showed reduced survival in patients who received chemoradiation.4 Further studies attempting to improve outcomes with adjuvant therapy in resected pancreatic cancer focused initially on chemoradiation, and more recently, on chemotherapy alone, typically for about 6 months, but the potential benefits of prolonged or maintenance chemotherapy were not subsequently explored.
There are logistic difficulties in prolonged administration of intravenous chemotherapy, but prolonged exposure to maintenance bolus 5-FU, as in the GITSG study, could maintain pressure on dormant cancer cells that remain in G0 arrest and kill them as they infrequently enter the G1/S phase. Capecitabine (Xeloda, Roche) is an oral prodrug that is enzymatically converted to 5-FU in a 3-step process that includes metabolism in the liver and the tumor.5 Randomized clinical trials have demonstrated that capecitabine has an efficacy similar to that of 5-FU in numerous cancers, including colorectal and gastric cancers.6–8 Because of its similar efficacy to intravenous 5-FU and its greater convenience as an oral medication, capecitabine may be a more acceptable alternative as a maintenance therapy than weekly bolus 5-FU.
We retrospectively evaluated patients with resected pancreatic adenocarcinoma who were treated with adjuvant chemotherapy and maintenance capecitabine compared with patients who received adjuvant chemotherapy only.
METHODS
We conducted a retrospective chart review of patients who were treated at the Lombardi Comprehensive Cancer Center (LCCC), Georgetown University Medical Center. Patients were found by searching for the International Classification of Diseases (ICD) code for pancreatic adenocarcinoma via the LCCC electronic medical records system (Aria, Varian). Patients were selected who had pathology-proven pancreatic adenocarcinoma, had undergone surgical resection with curative intent between 2003 and 2010, and had received adjuvant chemotherapy with or without chemoradiation. Patients who had recurrence of disease while on adjuvant chemotherapy were excluded. Of the patients who met the inclusion criteria, those who received maintenance capecitabine after adjuvant therapy were selected as the experimental group, and those who did not receive maintenance capecitabine became the control group.
Those patients who received maintenance capecitabine were typically started at 1000 mg orally twice daily, most commonly scheduled Monday through Friday. In patients who showed toxicity to capecitabine, the dose was adjusted at the discretion of their primary oncologist. Charts were reviewed to determine adverse events attributed to capecitabine.
Primary end points for analysis were disease recurrence and all-cause mortality, determined using time from surgery to disease recurrence or death, respectively. The vital status of all patients was confirmed by using the Social Security Death Index (SSDI). The study was approved by our institutional review board (IRB).
Descriptive statistics (eg, mean, median, range, and proportion) were used to describe patients' demographic information. The chi-square test or Fisher's exact test was used to compare categorical variables. The 2-sample t-test was used to compare continuous variables, the Kaplan-Meier method was used to calculate the survival probabilities at different time points, and the log-rank test was used to compare the overall survival (OS) experience between groups. Cox regression models were fit to assess the relationship between independent risk factors and survival. All tests were 2-tailed, and P < .05 was considered statistically significant. SAS software (version 9.3; SAS Institute Inc., Cary, NC) was used to perform the analyses.
RESULTS
From 2003 through 2010, there were 21 patients who received maintenance capecitabine (maintenance group) after adjuvant therapy and 58 who did not (adjuvant-only group). One patient in the maintenance group left the country and was lost to follow-up after 13 months, before which the patient was on capecitabine for 6 months. Thirteen patients from the adjuvant-only group were excluded because of documented disease recurrence while receiving adjuvant chemotherapy. The median follow-up duration was 38 months in the maintenance arm and 17 months in the adjuvant-only arm.
The characteristics of the study groups were comparable, with some exceptions. The adjuvant-only group was slightly older and had fewer R0 resections, a statistically significant shorter time of follow-up, and more poorly differentiated tumors; fewer of them received postoperative chemoradiation (Table 1). The majority of patients in both groups received gemcitabine for adjuvant therapy. Of those who did not, 3 patients in the adjuvant-only group had enrolled in clinical trials in which other adjuvant therapy regimens were used (detailed in Table 1), and 1 had an unknown adjuvant regimen.
Table 1.
Patient and tumor characteristics
| Characteristics | Received adjuvant therapy and maintenance capecitabine (%) | Received adjuvant therapy only (%) |
|---|---|---|
| Patients: | ||
| Total | 21 (100) | 45 (100) |
| Male | 12 (57) | 27 (60) |
| Female | 9 (43) | 18 (40) |
| Median age at surgery in years (range) | 60.1 (38.0-80.2) | 64.7 (42.9-83.2) |
| Surgery type | ||
| Whipple (pancreaticoduodenectomy) | 17 (81) | 38 (84) |
| Total pancreatectomy | 1 (5) | 0 (0) |
| Distal pancreatectomy | 3 (14) | 7 (16) |
| R0 | 19 (90) | 33 (73) |
| R1 | 2 (10) | 8 (18) |
| R2 | 0 (0) | 1 (2) |
| Unknown | 0 (0) | 3 (7) |
| Tumors: | ||
| Histology | ||
| Adenocarcinoma/ductal | 20 (95) | 43 (96) |
| Adenosquamous | 0 (0) | 1 (2) |
| Ampullary | 1 (5) | 1 (2) |
| IPMN | 3 (14) | 14 (31) |
| Differentiation* | ||
| Poor | 1 (5) | 15 (33) |
| Moderate to poor | 4 (19) | 3 (7) |
| Moderate | 13 (62) | 20 (44) |
| Well to moderate | 1 (5) | 2 (4) |
| Well | 2 (10) | 2 (4) |
| Undifferentiated/anaplastic | 0 (0) | 1 (2) |
| Unknown | 0 (0) | 2 (4) |
| T-stage | ||
| T1 | 2 (10) | 2 (4) |
| T2 | 5 (24) | 7 (16) |
| T3 | 14 (67) | 33 (73) |
| T4 | 0 (0) | 2 (4) |
| T unknown | 0 (0) | 1 (2) |
| N-stage | ||
| N0 | 7 (33) | 14 (31) |
| N1 | 14 (67) | 30 (67) |
| N unknown | 0 (0) | 1 (2) |
| M-stage | ||
| M0 | 21 (100) | 45 (100) |
| M1 | 0 (0) | 0 (0) |
| Stage | ||
| Stage IA | 1 (5) | 1 (2) |
| Stage IB | 4 (19) | 4 (9) |
| Stage IIA | 2 (10) | 10 (22) |
| Stage IIB | 14 (67) | 25 (56) |
| Stage III | 0 (0) | 1 (2) |
| Stage IV | 0 (0) | 0 (0) |
| Stage unknown | 0 (0) | 4 (9) |
| Treatment characteristics: | ||
| Median follow-up in months (range)* | 38.4 (17.6-81.3) | 17.1 (2.8-71.2) |
| Neoadjuvant therapy | 0 (0) | 3 (7) |
| Adjuvant therapy | ||
| Gemcitabine-based | 21 (100) | 43 (96) |
| Gemcitabine only | 21 (100) | 39 (87) |
| Gemcitabine with GlobeImmune GI-4000 vaccine | 0 (0) | 3 (7) |
| Gemcitabine, bevacizumab, and cetuximab | 0 (0) | 1 (2) |
| 5-FU | 0 (0) | 2 (4) |
| J0810 vaccine with cyclophosphamide | 0 (0) | 1 (2) |
| Unknown | 0 (0) | 1 (2) |
| Chemoradiation* | 17 (81) | 24 (53) |
Clinical and tumor characteristics of patients in each group, including treatment regimens administered in all cases. Statistics in parentheses indicate the percentage of total patients in each category (except for the age at surgery and follow-up categories, in which time range in months is indicated). IPMN = intraductal papillary mucinous neoplasm.
Statistically significant difference between groups (P < .05).
Of the 21 patients in the maintenance capecitabine group, 11 were alive at the end of the analysis period, 9 had died, and 1 was lost to follow-up. Ten of the patients had documented disease recurrence, and 11 did not. Of the 45 patients in the adjuvant-only group, only 6 were alive at the end of the analysis period, 39 had died, and 38 had documented recurrence.
Median OS was longer than 48.4 months in the maintenance group (the exact estimate was not available because the survival probability curve does not cross 0.5) compared to 22.0 months in the adjuvant-only group (95% confidence interval [CI], 16.6–29.2). Median recurrence-free survival (RFS) was also longer in the maintenance group: 54.3 compared to 14.1 months (95% CI, 22.2–Inf and 11.6–16.7, respectively; see Table 2). Kaplan-Meier survival plots of OS and RFS demonstrate a significant survival advantage for maintenance capecitabine (P < .001 for both from log-rank tests; see Figures 1 and 2).
Table 2.
Overall survival (OS) and recurrence free survival (RFS) for both study cohorts
| Result | Received adjuvant therapy and maintenance capecitabine | Received adjuvant therapy only |
|---|---|---|
| Median length of maintenance capecitabine therapy in months (range) | 14 (2–52) | 0 |
| Number surviving | 11 (55%) | 6 (13%) |
| Median OS in months (range) | >48.4 (13.0–81.3) | 22.0 (3.4–71.2) |
| 2-year OS % (95% CI) | 95 (81–100) | 44 (30–59) |
| 5-year OS % (95% CI) | 50 (28–73) | 14 (5–27) |
| Number with recurrence | 10 (48%) | 38 (84%) |
| Median RFS in months (range) | 54.3 (13.0–71.8) | 14.1 (2.8–66.0) |
| 2-year RFS % (95% CI) | 70 (49–88) | 24 (12–38) |
| 5-year RFS % (95% CI) | 36 (12–65) | 10 (2–21) |
CI = confidence interval.
Figure 1.
OS estimates. Kaplan-Meier survival plot of OS comparing the maintenance capecitabine group (dashed line) with the adjuvant-only group (solid line) (P = .0005, log-rank test).
Figure 2.
RFS estimates. Kaplan-Meier survival plot of RFS comparing the maintenance capecitabine group (dashed line) with the adjuvant-only group (solid line) (P < .001, log-rank test).
A Cox proportional hazards model further demonstrates reduced risk of death and recurrence in patients who received maintenance capecitabine (hazards ratio [HR], 0.29, 95% CI, 0.14–0.61, and P = .001 for OS; HR, 0.26, 95% CI, 0.12–0.52, and P < .001 for RFS). Independent factors for OS that were also statistically significant included age at surgery (P = .017), N-stage (P = .019), resection type (R0 vs. other; P = .004), and tumor differentiation (P = .007). For RFS, N-stage (P = .031), resection type (P = .008), and tumor differentiation (P = .020) were also statistically significant. After accounting for other significant covariates in the multivariate Cox models, the risk reduction induced by maintenance therapy was sustained (Tables 3 and 4).
Table 3.
Proportional hazards models for OS
| HR (95% CI) | P | |
|---|---|---|
| Univariate Cox model | ||
| Maintenance capecitabine | 0.29 (0.14–0.61) | 0.001* |
| Age at surgery (per year) | 1.04 (1.01–1.07) | 0.017* |
| Received chemoradiation | 0.56 (0.31–1.00) | 0.050 |
| Gender (male vs. female) | 1.53 (0.84–2.82) | 0.168 |
| T-stage (relative to T1) | 0.518 | |
| N-stage (N1 relative to N0) | 2.21 (1.14–4.27) | 0.019* |
| Stage (relative to IA) | 0.287 | |
| Resection type (R1/R2 relative to R0) | 2.99 (1.42–6.28) | 0.004* |
| IPMN | 0.92 (0.48–1.77) | 0.802 |
| Surgery type (distal/total relative to Whipple) | 0.80 (0.37–1.71) | 0.559 |
| Differentiation (relative to poor) | 0.003* | |
| Poor | Reference | |
| Moderate to poor | 0.55 (0.20–1.52) | 0.245 |
| Moderate | 0.30 (0.15–0.58) | < 0.001* |
| Well to moderate | 0.22 (0.03–1.67) | 0.142 |
| Well | 0.27 (0.06–1.18) | 0.081 |
| Histology (adenosquamous/ampullary relative to adenocarcinoma/ductal) | 1.62 (0.50–5.26) | 0.423 |
| Neoadjuvant therapy | 2.13 (0.65–6.94) | 0.209 |
| Multivariate Cox model | ||
| Maintenance capecitabine | 0.40 (0.18–0.91) | 0.020* |
| Resection type (R1/R2 relative to R0) | 3.08 (1.26–7.53) | 0.03* |
Data are derived from Cox proportional hazards models for OS. IPMN = intraductal papillary mucinous neoplasm.
Statistically significant (P < .05).
Table 4.
Proportional hazards models for RFS
| HR (95% CI) | P | |
|---|---|---|
| Univariate Cox model | ||
| Maintenance capecitabine | 0.26 (0.12–0.52) | < 0.001* |
| Age at surgery (per year) | 1.02 (0.99–1.05) | 0.239 |
| Received chemoradiation | 0.65 (0.36–1.17) | 0.864 |
| Gender (male vs. female) | 1.05 (0.59–1.89) | 0.864 |
| T-stage (relative to T1) | 0.126 | |
| N-stage (N1 relative to N0) | 2.08 (1.07–4.04) | 0.031* |
| Stage (relative to IA) | 0.425 | |
| Resection (R1/R2 relative to R0) | 2.79 (1.30–5.97) | 0.008* |
| IPMN | 0.91 (0.47–1.75) | 0.772 |
| Surgery type (distal/total relative to Whipple) | 0.98 (0.48–2.04) | 0.966 |
| Differentiation (relative to poor) | 0.020* | |
| Poor | Reference | |
| Moderate to poor | 0.63 (0.24–1.64) | 0.354 |
| Moderate | 0.36 (0.18–0.71) | 0.004* |
| Well to moderate | 0.17 (0.02–1.32) | 0.091 |
| Well | 0.14 (0.49–5.17) | 0.058 |
| Histology (adenosquamous/ampullary relative to adenocarcinoma/ductal) | 1.60 (0.49–5.17) | 0.437 |
| Neoadjuvant therapy | 0.80 (0.11–5.82) | 0.824 |
| Multivariate Cox model | ||
| Maintenance capecitabine | 0.25 (0.11–0.58) | 0.001* |
| N-stage (N1 relative to N0) | 1.86 (0.81–4.25) | 0.144 |
| Resection type (R1/R2 relative to R0) | 2.41 (1.07–5.44) | 0.034* |
Data are derived from Cox proportional hazards models for RFS.
Statistically significant (P < .05).
Capecitabine was generally well tolerated, with the most common toxicity being hand–foot syndrome (HFS): 6 patients had grade 1–2 HFS. Four patients discontinued capecitabine because of neutropenic fever, fatigue, diarrhea, or weight loss, and 1 patient had a gastrointestinal hemorrhage due to an interaction with warfarin (Table 5).
Table 5.
Adverse events
| Adverse event | Grades 1-2 | Grades 3-5* |
|---|---|---|
| Neutropenic fever | 0 | 1† |
| Fatigue | 3 | 1† |
| Diarrhea | 0 | 1† |
| Weight loss | 0 | 1† |
| Gastrointestinal hemorrhage | 0 | 1‡ |
| Hand-foot syndrome | 6 | 0 |
| Dry skin | 1 | 0 |
No grade 5 events were noted.
This event caused discontinuation of capecitabine.
This event was due to an interaction with warfarin.
DISCUSSION
Our retrospective analysis suggests that maintenance therapy with capecitabine after resection of pancreatic adenocarcinoma followed by adjuvant therapy improves RFS and OS. Capecitabine was generally given weekly on Monday-to-Friday schedules and was well tolerated. Only 4 patients of 21 discontinued the medication due to adverse effects, despite prolonged administration (a median of 14 months and up to 52 months) at the time of our analysis.
Although retrospective, our data are hypothesis generating and clearly support further investigation of maintenance therapy in pancreatic cancer. The findings of our study are limited by a relatively small sample size and potential imbalances between the 2 groups. The maintenance capecitabine group had more patients who had undergone R0 resections and were more likely to receive adjuvant chemoradiation than the adjuvant-only group. Patients in the adjuvant-only group were more likely to have poorly differentiated tumors. Ampullary histology was underrepresented in both study groups, as only 1 patient in each group had this histology. Also, there may have been additional physician bias in selecting patients who would receive maintenance capecitabine. The use of chemoradiation appears to have had an OS benefit, although the results were not statistically significant.
Although there is an emerging consensus that optimal therapy for resectable pancreatic cancer requires multimodality therapy, including surgery and chemotherapy, there remains uncertainty about the optimal approach. Trials conducted by the European Organization for Research and Treatment of Cancer (EORTC) and the European Study Group for Pancreatic Cancer 1 (ESPAC-1) demonstrated a negative impact on OS and RFS for patients receiving chemoradiation.2,4 However, the ESPAC-1 study also suggested that this finding could be due to a delay in patients receiving adjuvant bolus 5-FU/leucovorin chemotherapy, as adjuvant chemotherapy alone demonstrated a survival advantage compared to that obtained with chemoradiation followed by chemotherapy.4 The CONKO-001 (Charité Onkologie) study showed that adjuvant therapy with gemcitabine for 6 months improved OS compared with that obtained with surgery alone.9 Radiation Therapy Oncology Group (RTOG) Study 9704 showed an improved OS in patients receiving chemoradiation plus gemcitabine, compared to those receiving chemoradiation plus 5-FU (20.6 and 16.9 months, respectively), although the results were not statistically significant.10 However, the ESPAC-3 trial did not show improved OS with gemcitabine compared to 5-FU plus folinic acid and in fact suggested that the 2 agents have equal benefit in the adjuvant setting.11 Based on these data, there is a general consensus that chemotherapy is the key component of adjuvant therapy for pancreatic adenocarcinoma, with gemcitabine as the preferred agent because of tolerability. However, the optimal duration of therapy and the potential role and timing of radiation remain uncertain.
In the original GITSG study, patients who underwent surgical resection were randomly assigned to a control group that received no adjuvant therapy or to a combination therapy group that received both chemotherapy and radiation. The combination group received 2 courses of 2000 rad (20 Gy) 2 weeks apart, as well as intravenous 5-FU bolus injections with each course of radiation followed by a once-weekly bolus 5-FU for 2 years, unless there was intolerable toxicity or recurrence.3 Although the study was small, involving only 43 patients, a significant improvement in median survival with chemoradiation (20 months) was clearly demonstrated when compared with observation alone (11 months); 2-year OS rates were also improved (42% compared with 15%).3
Subsequent studies attempting to improve outcomes in resected pancreatic adenocarcinoma largely focused on the potential role of radiation. However, an alternative potential explanation for the notably improved outcomes that were demonstrated in the GITSG study comes from the prolonged administration of weekly bolus 5-FU maintenance therapy. It has been postulated that there are residual tumor cells after resection that are mostly in the G0 phase of the cell cycle. Periodically, these cells enter the G1/S phase to replicate. The theory is that, by continuing chemotherapy for a long duration, pressure is maintained on these cells, and they are killed when they enter the G1/S phase.
At the 2013 American Society for Clinical Oncology (ASCO) Annual Meeting, a small phase II study was presented that involved the use of an oral fluoropyrimidine (S-1) in combination with gemcitabine in resected pancreatic cancer. The trial arms included S-1 alone (days 1–14, with cycles repeated every 3 weeks, for 16 cycles) for 48 weeks, gemcitabine alone for 12 cycles, or the 2 in combination.12 Two-year disease-free survival was similar in the gemcitabine and S-1 arms (24.2% and 28.1%, respectively), with 34.4% in the combination arm. The median OS was 21, 26, and 27.9 months, respectively. The S-1 arm, especially the gemcitabine/S-1 arm, performed better than expected relative to historical data, although prolonged gemcitabine did not perform better than in previous studies (22 months OS in CONKO-001 and 23.6 months in ESPAC-3).9,11 These results suggest that a longer course of maintenance chemotherapy, or perhaps the different mechanism of action of the fluoropyrimidine and gemcitabine, played an important role in obtaining the results that we have reported.
Further supporting the role of maintenance chemotherapy in advanced gastrointestinal malignancies, the CAIRO-3 study in metastatic colorectal cancer was also reported at the 2013 ASCO Annual Meeting. This study assessed the role of maintenance therapy in patients with responding or stable disease after 6 cycles of initial chemotherapy with oxaliplatin/capecitabine/bevacizumab.13 In that study, 558 patients were randomized, and those who were randomly assigned to continue capecitabine/bevacizumab had a longer median progression-free survival (7.4 months) and median survival (21.7 months) than those who were initially assigned to observation after 4 months of chemotherapy (4.1 and 17.9 months, respectively). These results lend additional support to the notion that maintenance therapy improves outcomes in advanced gastrointestinal malignancies, including pancreatic cancer.
The improved outcomes we report may be the result of giving 2 active agents in sequence, rather than prolonged maintenance therapy. The ongoing ESPAC-4 trial randomizes patients to adjuvant gemcitabine or gemcitabine plus capecitabine for 24 weeks, and its data may elucidate the survival benefits of combination therapy.14 To assess the role of maintenance therapy, future studies could test 3 arms: gemcitabine alone, gemcitabine followed by 6 months of capecitabine, and gemcitabine followed by indefinite capecitabine. Also, given new data with 5-FU-based front-line therapy FOLFIRINOX as well as gemcitabine and nab-paclitaxel in the metastatic setting, adjuvant treatment may shift away from single-agent gemcitabine as the primary treatment, especially for patients with good performance status.15,16
Our results show that benefit may be gained from maintenance capecitabine therapy after surgical resection and adjuvant chemotherapy. These approaches should be further investigated in randomized clinical trials in which patients who have had surgical resection and adjuvant chemotherapy are randomized to receive maintenance therapy with capecitabine for a finite or indefinite period or to surveillance.
Acknowledgment
Dr. Marion L. Hartley provided editorial assistance for this article.
Footnotes
Disclosures of Potential Conflicts of Interest
Dr. Marshall is a consultant for Genentech. Dr. Hwang is a consultant and speaker for Roche/Genentech. The remaining authors indicated no potential conflicts of interest.
REFERENCES
- 1. Howlader N, Noone A, Krapch M, et al. : SEER Cancer Statistics Review, 1975-2008. National Cancer Institute, Bethesda, MD, 2010 [Google Scholar]
- 2. Klinkenbijl JH, Jeekel J, Sahmoud T, et al. : Adjuvant radiotherapy and 5-fluorouracil after curative resection of cancer of the pancreas and periampullary region: phase III trial of the EORTC gastrointestinal tract cancer cooperative group. Ann Surg 230:776–782, 1999 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Kalser MH, Ellenberg SS: Pancreatic cancer: adjuvant combined radiation and chemotherapy following curative resection. Arch Surg 120:899–903, 1985 [DOI] [PubMed] [Google Scholar]
- 4. Neoptolemos JP, Stocken DD, Friess H, et al. : A randomized trial of chemoradiotherapy and chemotherapy after resection of pancreatic cancer. N Engl J Med 350:1200–1210, 2004 [DOI] [PubMed] [Google Scholar]
- 5. Miwa M, Ura M, Nishida M, et al. : Design of a novel oral fluoropyrimidine carbamate, capecitabine, which generates 5-fluorouracil selectively in tumours by enzymes concentrated in human liver and cancer tissue. Eur J Cancer 34:1274–1281, 1998 [DOI] [PubMed] [Google Scholar]
- 6. Hoff PM, Ansari R, Batist G, et al. : Comparison of oral capecitabine versus intravenous fluorouracil plus leucovorin as first-line treatment in 605 patients with metastatic colorectal cancer: results of a randomized phase III study. J Clin Oncol 19:2282–2292, 2001 [DOI] [PubMed] [Google Scholar]
- 7. Cassidy J, Clarke S, Diaz-Rubio E, et al. : Randomized phase III study of capecitabine plus oxaliplatin compared with fluorouracil/folinic acid plus oxaliplatin as first-line therapy for metastatic colorectal cancer. J Clin Oncol 26:2006–2012, 2008 [DOI] [PubMed] [Google Scholar]
- 8. Cunningham D, Starling N, Rao S, et al. : Capecitabine and oxaliplatin for advanced esophagogastric cancer. N Engl J Med 358:36–46, 2008 [DOI] [PubMed] [Google Scholar]
- 9. Oettle H, Neuhaus P, Hochhaus A, et al. : Adjuvant chemotherapy with gemcitabine and long-term outcomes among patients with resected pancreatic cancer: the CONKO-001 randomized trial. JAMA 310:1473–1481, 2013 [DOI] [PubMed] [Google Scholar]
- 10. Regine WF, Winter KA, Abrams RA, et al. : Fluorouracil vs gemcitabine chemotherapy before and after fluorouracil-based chemoradiation following resection of pancreatic adenocarcinoma: a randomized controlled trial. JAMA 299:1019–1026, 2008 [DOI] [PubMed] [Google Scholar]
- 11. Neoptolemos JP, Stocken DD, Bassi C, et al. : Adjuvant chemotherapy with fluorouracil plus folinic acid vs gemcitabine following pancreatic cancer resection: a randomized controlled trial. JAMA 304:1073–1081, 2010 [DOI] [PubMed] [Google Scholar]
- 12. Yoshitomi H, Shimizu H, Yoshidome H, et al. : A randomized phase II trial of adjuvant chemotherapy with S-1 versus S-1 and gemcitabine (GS) versus gemcitabine alone (GEM) in patients with resected pancreatic cancer (CAP-002 study). J Clin Oncol 31, 2013. (abstr 4056) [Google Scholar]
- 13. Koopman M, Simkens LHJ, Ten Tije AJ, et al. : Maintenance treatment with capecitabine and bevacizumab versus observation after induction treatment with chemotherapy and bevacizumab in metastatic colorectal cancer (mCRC): The phase III CAIRO study of the Dutch Colorectal Cancer Group (DCCG). J Clin Oncol 31, 2013. (abstr 3502) [Google Scholar]
- 14. Neoptolemos J: European study group for pancreatic cancer (ESPAC). Trial 4: combination versus single agent chemotherapy in resectable pancreatic ductal and periampullary cancers. Cancer Research UK Liverpool Cancer Trials Unit. EUDRACT No: 2007-004299-38. ISRCTN96397434 [Google Scholar]
- 15. Conroy T, Desseigne F, Ychou M, et al. : FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med 364:1817–1825, 2011 [DOI] [PubMed] [Google Scholar]
- 16. Von Hoff DD, Ervin TJ, Arena FP, et al. : Results of a randomized phase III trial (MPACT) of weekly nab-paclitaxel plus gemcitabine versus gemcitabine alone for patients with metastatic adenocarcinoma of the pancreas with PET and CA 19-9 correlates. J Clin Oncol 31, 2013. (abstr 4005) [Google Scholar]