Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Original article Annals of Oncology 13: 1576–1582, 2002 DOI: 10.1093/annonc/mdf274 Phase II trial of 5-fluorouracil plus eniluracil in patients with advanced pancreatic cancer: a Southwest Oncology Group study M. L. Rothenberg1*, J. K. Benedetti2, J. S. Macdonald3, T. E. Seay4, M. A. Neubauer5, C. S. George6, M. S. Tanaka Jr7, J. K. Giguere8, B. T. Pruitt9 & J. L. Abbruzzese10† 1 Vanderbilt University Medical Center, Nashville, TN; 2Southwest Oncology Group Statistical Center, Seattle, WA; 3St Vincent’s Comprehensive Cancer Center, New York, NY; 4Atlanta Regional Community Clinical Oncology Program, Atlanta, GA; 5Kansas City Community Clinical Oncology Program, Kansas City, MO; 6 Columbus Community Clinical Oncology Program, Columbus, OH; 7University of California, Davis, Sacramento, CA; 8Greenville Community Clinical Oncology Program, Greenville, SC; 9Don and Sybil Harrington Cancer Center, Amarillo, TX; 10University of Texas, M.D. Anderson Cancer Center, Houston, TX, USA Received 31 January 2002; revised 4 April 2002; accepted 24 April 2002 Background: Eniluracil is a potent, irreversible inactivator of dihydropyrimidine dehydrogenase, the major catabolic enzyme for 5-fluorouracil (5-FU). Pretreatment with eniluracil significantly increases plasma half-life, plasma concentration and oral bioavailability of 5-FU. This multicenter phase II trial was designed to estimate the 6-month survival rate in patients with metastatic adenocarcinoma of the pancreas treated with 5-FU and eniluracil. Patients and methods: One hundred and sixteen patients (61 with no prior chemotherapy and 55 with prior chemotherapy) were registered for treatment with eniluracil 50 mg (total dose) p.o. on days 1–7 and 5-FU 20 mg/m2/day p.o. on days 2–6 of a 28-day treatment cycle. Results: In 106 patients evaluable for survival, the 6-month survival rate was 34% [95% confidence interval (CI) 22% to 47%, median survival 3.6 months] for patients who had not been treated previously with chemotherapy and 29% (95% CI 16% to 42%, median survival 3.4 months) for those who had received prior chemotherapy. For those patients with measurable disease, the confirmed response rates were 8% and 2%, respectively. The most common grade 3–4 toxicities were neutropenia (29% of patients) and diarrhea (12% of patients). Overall, 69% of patients experienced a grade 3 or worse adverse event during treatment. Conclusions: These results suggest that the combination of a 7-day course of eniluracil and a 5-day course of oral 5-FU has limited activity in patients with advanced pancreatic cancer, and is associated with a high frequency of clinically significant adverse events. Key words: 5-fluorouracil, clinical trial, eniluracil, oral chemotherapy, pancreatic cancer, phase II Introduction Pancreatic cancer is one of the leading causes of cancer-related death in the United States; it is estimated that 28300 people died from cancer of the pancreas in 2000 [1]. Eighty to 90% of patients have unresectable disease at the time of diagnosis. For many years, 5-fluorouracil (5-FU) was the primary therapy for patients newly diagnosed with advanced stage disease. The activity reported for 5-FU in this setting has been highly variable, with objective response rates ranging from 0% to *Correspondence to: Dr M. L. Rothenberg, Vanderbilt University Medical Center, 777 Preston Research Building, Nashville, TN 37232-6307, USA. Tel: +1-615-343-8422; Fax: +1-615-343-7602; E-mail: [email protected] †Reprint requests to: Southwest Oncology Group (SWOG-9629), Operations Office, 14980 Omicron Drive, San Antonio, TX 78245-3217, USA. © 2002 European Society for Medical Oncology >20% in phase II trials [2]. Despite numerous attempts to improve the efficacy of 5-FU through biochemical modulation, the addition of other cytotoxic drugs or an alteration to the drug administration schedule, no maneuver consistently resulted in improved overall survival above that attained with single-agent 5-FU. More recently, the use of 5-FU in this setting has declined due to the results of a phase III study that demonstrated the superiority of gemcitabine over 5-FU in the treatment of patients with newly diagnosed, advanced stage pancreatic cancer in terms of survival (median survival 5.65 months compared with 4.41 months in 5-FU-treated patients) and relief from tumor-related symptoms (clinical benefit response rate 23.8% compared with 4.8%) [3]. The 6-month survival rate was 46% for patients who received gemcitabine and 31% for those treated with 5-FU. Eniluracil (5-ethynyluracil, BW776C85, GW776C85, 776C85; GlaxoSmithKline, Collegeville, PA) is a potent, irrevers- 1577 ible inactivator of dihydropyrimidine dehydrogenase (DPD), the rate-limiting enzyme in the breakdown of 5-FU [4]. Pretreatment with eniluracil significantly increases the plasma half-life as well as the plasma concentration and duration (i.e. area under the concentration–time curve) of 5-FU. Production of inactive catabolites of 5-FU that may interfere with the antitumor activity of 5-FU, such as α-fluoro-β-alanine and 5-H2-FU, are reduced by pretreatment with eniluracil [4]. Eniluracil converts 5-FU from a drug that is enzymatically catabolized to one that is dependent upon renal clearance as its primary mode of drug elimination. Preclinical studies demonstrated that eniluracil markedly improved the complete remission rate from 14% to 100%, and improved the therapeutic index of 5-FU from 1.0 to 6.0 in rats bearing Ward colorectal carcinomas [5]. Given this enhanced activity against a murine tumor with minimal sensitivity to single agent 5-FU, we felt that it would be of interest to evaluate the combination of eniluracil and 5-FU in patients with advanced adenocarcinomas of the pancreas. Two schedules of the 5-FU + eniluracil combination were brought forward into phase II testing: (i) a shorter schedule consisting of a 5-day course of 5-FU with a 7-day course of eniluracil and (ii) a longer schedule consisting of a 28-day course of both drugs [6, 7]. The shorter schedule was designed to mimic the 5-day course of intravenous 5-FU (Mayo Clinic schedule) that was popular in the United States in the 1980s and 1990s. It was this shorter schedule that was selected for evaluation in this clinical trial. Phase II trials of new agents are often designed around a primary end point of objective response rate as a method of identifying agents with sufficient activity to warrant further clinical evaluation. Unfortunately, pancreatic cancer does not fit easily into this paradigm. Pancreatic cancer is notoriously difficult to measure precisely in anatomic imaging studies, using CT or MRI for example [8]. This is probably one of the main reasons why high response rates in some phase II trials fail to translate into an improvement in survival in phase III trials. Published survival data for patients treated with gemcitabine in phase II and III trials, as well as in the investigational new drug (IND) treatment program, has been fairly consistent [3, 9, 10, 20]. The median survival in patients with advanced pancreatic cancer who received gemcitabine as first-line chemotherapy is 5.1–6.3 months, and the 6-month survival rate is 46–52% [3, 10, 20]. For those patients who receive gemcitabine as second-line chemotherapy, median survival is 3.9–4.4 months, with a 6-month survival rate of ∼31% [9, 10]. This consistency in survival data, coupled with the unreliability of anatomical imaging studies, prompted us to design this phase II trial of oral 5-FU and eniluracil around a primary end point of 6-month survival rate rather than objective response rate. Patients and methods Patients were required to meet the following eligibility criteria: (i) cytologically or pathologically verified diagnosis of pancreatic adenocarcinoma that was metastatic at the time of registration; (ii) bidimensionally measurable or evaluable disease; (iii) Southwest Oncology performance status of 0–2; (iv) calculated creatinine clearance of ≥50 ml/min using the Cockroft–Gault formula; (v) no more than one prior chemotherapy regimen for metastatic pancreatic cancer; and (vi) the ability to swallow and retain oral medications. Prior treatment with 5-FU or mitomycin-C as a single agent or as part of a combination regimen was not permitted. If the patient had been treated previously for pancreatic cancer, there must be current evidence of progressive disease. Prior radiotherapy and surgery were permitted if completed at least 4 weeks prior to registration. Informed consent was obtained from all patients in accordance with institutional and federal guidelines prior to the initiation of treatment. Treatment Treatment consisted of eniluracil 50 mg (total dose/day) p.o. qd (days 1–7) and 5-FU 20 mg/m2/day p.o. qd (days 2–6) as part of a 28-day treatment cycle. Patients took eniluracil tablets immediately before 5-FU tablet(s) on days 2–6. Patients fasted for 1 h prior to the dose of eniluracil and for 1 h after the dose of 5-FU. To begin the next course of treatment, patients were required to have a platelet count ≥75 000/mm3, granulocytes ≥1500/mm3, and resolution of clinically significant non-hematologic toxicities to no more than grade 1. Dose modification of 5-FU was based upon the maximum intensity of drug-related toxicities occurring in the previous course [21]. Dose was calculated as a percentage of the previous dose of 5-FU and rounded to the nearest 5 mg. The dose of 5-FU was reduced by 20% for grade 3–4 hematologic toxicity, except in the case of grade 4 neutropenia with fever or infection, or grade 4 thrombocytopenia, when the dose of 5-FU was reduced by 40%. Treatment was not given to any patient whose creatinine clearance fell to <40 cc/min. Patients with a creatinine clearance of 40–49 cc/min received 80% of their previous dose. 5-FU was reduced by 20% for any other grade 2–3 non-hematologic toxicity, and reduced by 40% for any grade 4 non-hematologic toxicity. Dose reductions for nausea, vomiting or diarrhea were made only if they occurred despite appropriate supportive measures. Patients who were dose reduced and developed no grade 2 or worse toxicity could receive the next cycle of treatment at a 20% dose escalation. The highest dose of 5-FU permitted was the original dose. No modification was made to the eniluracil dose for any toxicity. Evaluation Patients were monitored with a complete blood count and toxicity assessment every 2 weeks, a physical examination every 2 weeks during the first cycle of therapy and then every 4 weeks thereafter, and estimated creatinine clearance, total bilirubin and AST (SGOT) or ALT (SGPT) every 4 weeks just prior to a new cycle of treatment. Tumor assessment was performed every 8 weeks. Treatment continued until disease progression, the appearance of unacceptable toxicity, a delay in initiation of a new cycle of therapy by ≥3 weeks due to toxicity, withdrawal of patient consent, or death. Statistical analysis The primary objective of this study was to evaluate the 6-month overall survival rate in patients with pancreatic cancer treated with oral 5-FU + 1578 eniluracil. Because it was assumed that patients who had received prior chemotherapy would have a poorer prognosis, accrual proceeded independently for those patients who were chemotherapy naïve and those who had received prior treatment with gemcitabine. In the chemotherapy naïve group, it was assumed that the combination of oral 5-FU and eniluracil would not be of further interest if the 6-month survival rate was ≤35%, but of considerable interest if it was ≥55%. These assumptions were based on the 46% 6-month survival rate observed for gemcitabine-treated patients in the original phase III trial and a similar survival rate achieved in previously untreated patients who received gemcitabine under the IND treatment [3, 10]. The accrual goal was 55 evaluable patients. Twenty-six or more patients surviving past 6 months would be considered evidence of the need for further study of this regimen. This design has a significance level (i.e. probability of declaring an agent with a 35% true 6-month survival rate as warranting further study) of 0.04 and a power (probability of correctly declaring an agent with a 55% true 6-month survival rate as warranting further study) of 0.90. Among patients who received 5-FU and eniluracil as second-line chemotherapy, it was assumed that this regimen would not be of further interest if the true 6-month survival rate was ≤20%, but of considerable interest if it was ≥40%. These assumptions were based on the 31% 6-month survival rate observed in the phase II trial of second-line gemcitabine and a similar rate in patients who received gemcitabine under the IND treatment [9, 10]. The accrual goal for this stratum was 44 eligible patients. Fourteen or more patients surviving past 6 months would be considered evidence of the need for further study of this agent. This design has a significance level of 0.04 and a power of 0.90. Descriptive factors obtained at baseline in all patients included: (i) prior surgery, yes or no; (ii) prior radiation therapy, yes or no; (iii) liver metastases, yes or no; and (iv) performance status: 0–1 or 2. Results One hundred and sixteen patients were registered on this trial; 61 of whom had received no prior chemotherapy and 55 of whom had received prior treatment with gemcitabine. Eight patients were ineligible: five due to missing documentation and three due to violation of eligibility criteria. Two patients never received protocol treatment. Overall, 106 patients were eligible and evaluated for toxicity and efficacy, 58 of whom were previously untreated and 48 of whom had received previous chemotherapy. Patient characteristics Patient characteristics are summarized in Table 1. The median age of patients enrolled on this study with and without prior chemotherapy was 59.0 years (range 38–81 years) and 64.0 years (range 36–80 years), respectively. The majority of patients were male and Caucasian. Seventy-nine percent of patients had a performance status of 0 or 1. Eighty-eight percent had liver involvement and 42% had previous surgery. Efficacy Previously untreated patients All 58 previously untreated patients are now off treatment. The most common reasons for treatment discontinuation were tumor progression in 40 patients (70%) and toxicity or side- Table 1. Patient characteristics Characteristic No previous chemotherapy (n = 58) Previous chemotherapy (n = 48) Median age (range) 64.0 years (36–80 years) 59.0 years (38–81 years) Male 41 (71%) 29 (60%) Female 17 (29%) 19 (40%) 46 (79%) 43 (91%) Gender Race White Black 6 (10%) 3 (6%) Hispanic 2 (3%) 2 (4%) 4 (7%) 0 (0%) 0–1 47 (81%) 37 (77%) 2 11 (19%) 11 (23%) Yes 51 (88%) 42 (88%) No 7 (12%) 6 (12%) Yes 23 (40%) 21 (44%) No 34 (60%) 27 (56%) Asian/Pacific islander Performance status a Liver metastases Prior surgeryb a Southwest Oncology performance status. Including pancreaticoduodenectomy or palliative choledocojejunostomy with or without gastrojejunal bypass. b 1579 Figure 1. Kaplan–Meier survival curves for chemotherapy-naïve (solid line) and previously treated (dashed line) patients with advanced pancreatic cancer, treated with 5-fluorouracil and eniluracil. effects in six (11%). The 6-month survival rate for patients who received oral 5-FU + eniluracil as first-line treatment for metastatic pancreatic cancer was 34% [95% confidence interval (CI) 22% to 47%] (Figure 1). The 1-year survival rate was 16% (95% CI 6% to 25%). Median survival was 3.6 months. In the 51 patients with measurable disease, one patient (2%) had a complete response and three patients (6%) had confirmed partial responses. Previously treated patients Of the 48 previously treated patients, one remains on treatment. The most common reason for discontinuation of treatment was tumor progression in 29 patients (62%). The 6-month survival rate for patients who received oral 5-FU + eniluracil as second-line therapy was 29% (95% CI 16% to 42%) (Figure 1). The 1-year survival rate for this group of patients was 10% (95% CI 6% to 15%). Median survival was 3.4 months. All 48 patients had measurable disease. One patient (2%) had a confirmed partial response. One additional patient had an unconfirmed response. Grade 3–4 neutropenia occurred in 19 out of 58 (33%) previously untreated patients and 12 out of 48 (25%) previously treated patients. Neutropenic fever was rare and occurred in three (5%) and two (4%) patients, respectively. There was one case of fatal neutropenic infection in each treatment arm. Grade 3 or worse diarrhea occurred in six out of 58 (10%) previously untreated patients and in eight out of 48 (17%) previously treated patients. One patient who received oral 5-FU + eniluracil as second-line therapy died of dehydration caused by treatment-related diarrhea. Grade 3 nausea was observed in nine out of 58 (16%) previously untreated patients and in two out of 48 (4%) previously treated patients. Grade 3–4 vomiting occurred in seven out of 58 (12%) and one out of 48 (2%) patients, respectively. Although not a prospective end point of the study, it appears that previously treated patients tolerated therapy as well as previously untreated patients. Grade 3 or worse toxicity occurred in 41 out of 58 (71%) previously untreated patients and 31 out of 48 (65%) previously treated patients. Discussion Toxicity The most common toxicities experienced by patients treated with oral 5-FU + eniluracil are listed in Table 2. Anemia was the single most common toxicity, with 40 out of 58 (69%) previously untreated patients and 38 out of 48 (79%) previously treated patients experiencing anemia while on study. However, many of these patients had pre-existing anemia. Grade 3 or worse anemia occurred in only four patients (4%) overall and all of these patients had received prior chemotherapy. For the past 5 years, new treatment strategies for patients with advanced stage pancreatic cancer have focused on optimization of gemcitabine-based treatment. This interest was prompted by the demonstration that first-line treatment with gemcitabine was superior to bolus weekly i.v. 5-FU in terms of both survival and reduction of tumor-related symptoms in patients with advanced stage disease [3]. Although singleagent 5-FU was found to be therapeutically inferior to gemcitabine in that phase III trial, this should not be interpreted as 1580 Table 2. Toxicities Toxicity First-line therapy (n = 58) grade Second-line therapy (n = 48) grade U 0 1 2 3 5 U 0 1 2 4 5 Anemia 0 18 28 12 0 4 0 0 0 10 20 14 3 4 0 0 Neutropenia 0 34 0 5 8 11 0 0 31 4 1 5 7 0 Thrombocytopenia 0 44 6 6 2 0 0 0 36 7 3 1 1 0 Abdominal pain 2 50 2 3 1 0 0 3 29 4 6 5 1 0 Diarrhea 1 38 10 3 6 0 0 0 25 7 8 5 2 1 Fatigue 1 35 12 8 2 0 0 0 25 8 5 10 0 0 Infection with grade 3–4 neutropenia 0 – – – 1 1 1 0 – – – 1 0 1 Nausea 1 23 16 9 9 – – 0 22 17 7 2 – – Stomatitis 1 42 9 5 1 0 0 0 34 6 5 2 0 1 Vomiting 1 33 14 3 6 1 0 1 37 5 4 1 0 0 U, unknown. meaning that 5-FU lacks utility in this disease. Median survival is 5 months for patients treated with single-agent 5-FU, and 5–6 months for patients treated with 5-FU-based combination regimens in phase III clinical trials [11]. More recently, at least five phase II trials have evaluated the combination of gemcitabine with 5-FU (± leucovorin) as first-line therapy in patients with advanced pancreatic cancer [12–16]. Objective response rates ranged from 4% to 22%, median survival ranged from 4.4–11 months, and 1-year survival rates ranged from ∼10% to 40%. In a recently reported phase III trial, Berlin and colleagues compared single agent 5-FU with the combination of 5-FU + gemcitabine [17]. This trial demonstrated a 1.3 month difference in median survival (6.7 compared with 5.4 months) in favor of the two-drug combination, a difference that approached, but did not reach, statistical significance (P = 0.09). These results suggest that there may be incomplete clinical cross-resistance between 5-FU and gemcitabine, and that 5-FU has activity that may be complementary to that of gemcitabine. DPD reduces 5-FU to 5-fluoro-5,6-dihydrouracil (5-FUH2) [4]. In preclinical studies, eniluracil inactivated >99% of hepatic DPD activity, thereby blocking the formation of this and other metabolites, such as α-fluoro-β-alanine (FBAL), which may interfere with the antitumor activity of 5-FU [4, 20]. Indeed, in preclinical models, eniluracil increased the therapeutic index of 5-FU from <1.0 to 6.0, enhanced the oral bioavailability of 5-FU to 100%, and improved its antitumor activity [4, 5]. In phase I clinical studies, eniluracil itself had no appreciable toxicities, but significantly increased the plasma half-life of 5-FU from 15 min to 5 h and reduced 5-FU clearance by 18-fold [18]. The plasma concentration–time curve (AUC) for an oral dose of 5-FU of 20 mg/m2 that was preceded by eniluracil approximated that of a 600 mg/m2 dose of 5-FU administered intravenously without eniluracil [18]. Partial responses were observed in several patients with advanced colorectal cancer that had progressed on intraven- ous 5-FU + leucovorin [18]. Taken together, the preclinical, pharmacological and clinical data on eniluracil + oral 5-FU was sufficiently intriguing to warrant evaluation of this combination in patients with advanced pancreatic cancer. Our study evaluated eniluracil, administered on days 1–7, in combination with oral 5-FU, administered on days 2–6, in patients with metastatic pancreatic cancer. Patients were drawn from two populations: those who were previously untreated for metastatic pancreatic cancer, and those who had received gemcitabine as initial treatment for metastatic disease and experienced disease progression. Because the beneficial effects of therapy for patients with adenocarcinoma of the pancreas may not be accurately reflected by tumor shrinkage (i.e. response rate), the primary end point selected for this trial was survival. Expectations for this therapy were influenced by clinical experience with gemcitabine. As first-line treatment in patients with locally advanced or metastatic pancreatic cancer, gemcitabine is associated with a median survival of 5.1–6.3 months and a 6-month survival rate of 46–52% [3, 10, 19, 20]. The phase II study and IND treatment experience with gemcitabine as second-line therapy in patients with advanced pancreatic cancer reported median survivals of 3.9–4.4 months and 6-month survival rates of ∼31% [9, 10]. The parameters of this study were set such that the combination of oral 5-FU + eniluracil would be pursued further if the 6-month survival rate was consistent with a true rate in excess of 55% when used in the front-line setting, and 40% when used in the second-line setting. Neither goal was achieved. Oral 5-FU + eniluracil was associated with a 6-month survival rate of 34% (95% CI 22% to 47%) when used as first-line chemotherapy, and 29% (95% CI 16% to 42%) when used as second-line chemotherapy. Although prospective evaluation of the impact of this therapy on tumor-related symptoms was not performed due to the exploratory nature of this study, it is notable that 68% of patients treated with oral 5-FU and eniluracil experienced a grade 3 or worse adverse event while on study. It should be 1581 noted that some of the serious adverse events were more likely to be related to the effects of progressive pancreatic cancer than to the effects of the chemotherapy. Nevertheless, the combination of 6-month survival rates that fall below the lower limits of interest and the high frequency of grade 3 or worse adverse events suggest that the combination of day 2–6 oral 5-FU + day 1–7 eniluracil does not warrant further evaluation in patients with metastatic adenocarcinoma of the pancreas. There are several possible explanations for these results. The first is that the addition of eniluracil was not sufficient to enhance the limited clinical activity of 5-FU. Recent clinical trials in patients with advanced colorectal cancer suggest that this is the most likely explanation. Promising activity was observed with this regimen (with or without leucovorin) in phase II trials, including partial response rates of 21–25% and median survival of 13.6 months in patients with previously untreated metastatic colorectal cancer [21, 22]. However, none of the 25 patients previously treated with intravenous 5-FU and leucovorin responded to the combination of oral 5-FU + eniluracil, suggesting a high degree of clinical crossresistance between the two therapies. In addition, preliminary results from two phase III trials indicate that oral 5-FU + eniluracil is no better, and may indeed be inferior to intravenous 5-FU + leucovorin in patients with previously untreated metastatic colorectal cancer. Another possible explanation is that patient selection may have influenced our results as much as, if not more than, the treatment. Most clinical trials in the advanced pancreatic cancer setting include a mixture of stage II, III and IV patients. Our survival estimates were based on figures derived from those studies. Stage of disease is a consistent and strong prognostic indicator for survival in patients with pancreatic cancer [10]. Our study required patients to have metastatic disease at the time of enrolment. In the IND treatment experience with gemcitabine, for instance, the median survival for patients with stage IV disease was only 4.4 months [8]. However, even taking this into account, the 3.6-month median survival observed for previously untreated patients who received oral 5-FU + eniluracil in the current trial falls short of this threshold. A third possible explanation is that the dose and/or schedule of therapy was suboptimal. In fact, a phase II study of the 5-day regimen of oral 5-FU and the 7-day regimen of eniluracil in patients with metastatic colorectal cancer used a 25 mg/m2/day dose of 5-FU rather than the 20 mg/m2/day dose used in this study [21]. However, this dose was not tolerated as six of the first 20 patients treated in that trial were hospitalized with grade 4 neutropenia and sepsis [21]. A 28-day-on/7-dayoff treatment schedule has also been evaluated in patients with metastatic colorectal cancer [22]. The 25% objective response rate and 5.2-month progression-free survival observed for this schedule was very similar to the 21% response rate and 4.4-month progression-free survival observed using the 5-day schedule [22]. This same 28-day-on/7-day-off treatment has been evaluated in patients with advanced pancreatic cancer, and preliminary results reported good tolerability and objective responses in three out of 41 (7.3%) patients [23]. Survival data, however, are not available for that study. There are several possible explanations for the high rate of grade 3 or 4 adverse events observed in this trial. The first is that patients with pancreatic cancer represent a relatively frail patient population and, as such, are more prone to toxicities than patients enrolled in phase I trials. The second is that there is evidence that the recommended dose of 25 mg/m2/day from phase I studies was not well tolerated in phase II. In one study, six of the first 20 patients with colorectal cancer treated with this dose required hospitalization for neutropenia and sepsis [21]. A third is that we chose to use the term ‘adverse events’ rather than ‘toxicities’ to reflect the fact that, in some cases, these events may have been due to factors other than the drug, such as tumor-related symptoms, side-effects of other medications or intercurrent illnesses. In summary, the combination of oral 5-FU + eniluracil does not demonstrate sufficient clinical activity, as reflected by 6-month survival rate, to warrant further study as first- or second-line therapy in patients with metastatic adenocarcinoma of the pancreas. In addition, the high frequency of grade 3 or worse adverse events makes it unlikely that this regimen could be safely combined with other drugs for the treatment of patients with advanced pancreatic cancer. Acknowledgements This investigation was supported in part by the following Public Health Service Cooperative Agreement grants, awarded by the National Cancer Institute, Department of Health and Human Services: CA38926, CA32102, CA45450, CA35176, CA35261, CA46441, CA67663, CA-22433, CA12644, CA35281, CA58415, CA58416, CA58882, CA52386, CA35192, CA37981, CA35431, CA20319, CA45461, CA58723, CA35119, CA58861, CA76448, CA45807, CA63844, CA42777, CA45377, CA35178, CA12213, CA-45560, CA63850, CA35128, CA14028, CA04919, and an unrestricted grant from GlaxoSmithKline (Collegeville, PA, USA). References 1. Greenlee RT, Murray T, Bolden S et al. Cancer statistics, 2000. CA Cancer J Clin 2000; 50: 7–33. 2. Berlin JD, Rothenberg ML. Chemotherapy for resectable and advanced pancreatic cancer. Oncology (Huntington) 2001; 15: 1241– 1254. 3. 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: 2403–2413. 4. Baccanari DP, Davis ST, Knick VC, Spector T. 5-Ethynyluracil (776C85): a potent modulator of the pharmacokinetics and antitumor 1582 efficacy of 5-fluorouracil. Proc Natl Acad Sci USA 1993; 90: 11064– 11068. 5. Cao S, Rustum YM, Spector T. 5-ethynyluracil (776C85): modulation of 5-fluorouracil efficacy and therapeutic index in rats bearing advanced colorectal carcinoma. Cancer Res 1994; 54: 1507–1510. 15. 6. Levin J, Hohneker J. Clinical development of eniluracil/fluorouracil: an oral treatment for patients with solid tumors. Invest New Drugs 2000; 18: 383–390. 16. 7. Baker SD, Khor SP, Adjei AA et al. Pharmacokinetic, oral bioavailability, and safety study of fluorouracil in patients treated with 776C85, an inactivator of dihydropyrimidine dehydrogenase. J Clin Oncol 1996; 14: 3085–3096. 17. 8. Rothenberg ML, Abbruzzese JL, Moore M et al. A rationale for expanding the endpoints for clinical trials in advanced pancreatic cancer. Cancer 1996; 78: 627–632. 9. Rothenberg ML, Moore MJ, Cripps MC et al. A phase II trial of gemcitabine in patients with pancreas cancer refractory to 5-FU. Ann Oncol 1996; 7: 347–353. 10. Storniolo AM, Enas NH, Brown CA et al. An Investigational New Drug (IND) treatment program for patients with gemcitabine: results for over 3000 patients with pancreatic cancer. Cancer 1999; 85: 1261–1268. 11. DiMagno EP, Reber HA, Tempero MA. AGA technical review on the epidemiology, diagnosis, and treatment of pancreatic ductal adenocarcinoma. Gastroenterology 1999; 117: 1464–1484. 12. Cascinu S, Silva RR, Barni S et al. A combination of gemcitabine and 5-fluorouracil in advanced pancreatic cancer: a report from the Italian Group for the Study of Digestive Tract Cancers (GISCAD). Br J Cancer 1999; 80: 1595–1598. 13. Hidalgo M, Castellano D, Paz-Ares L et al. Phase I–II study of gemcitabine and fluorouracil as a continuous infusion in patients with pancreatic cancer. J Clin Oncol 1999; 17: 585–592. 14. Cascinu S, Frontini L, Labianca R et al. A combination of a fixed dose rate infusion of gemcitabine associated to a bolus 5-fluorouracil in advanced pancreatic cancer, a report from the Italian Group for the 18. 19. 20. 21. 22. 23. Study of Digestive Tract Cancer (GISCAD). Ann Oncol 2000; 11: 1309–1311. Berlin JD, Adak S, Vaughn DJ et al. A phase II study of gemcitabine and 5-fluorouracil in metastatic pancreatic cancer: an Eastern Cooperative Oncology Group study (E3296). Oncology 2000; 58: 215–218. Oettle H, Arning M, Pelzer U et al. A phase II trial of gemcitabine in combination with 5-fluorouracil (24-hour) and folinic acid in patients with chemonaive advanced pancreatic cancer. Ann Oncol 2000; 11: 1267–1272. Berlin JD, Catalano P, Thomas JP et al. Phase III study of gemcitabine in combination with fluorouracil versus gemcitabine alone in patients with advanced pancreatic cancer: Eastern Cooperative Oncology Group trial E2297. J Clin Oncol 2002; 20: 3270–3275. Baker SD, Diasio RB, O’Reilly S et al. Phase I and pharmacologic study of oral fluorouracil on a chronic daily schedule in combination with the dihydropyrimidine dehydrogenase inactivator eniluracil. J Clin Oncol 2000; 18: 915–926. Casper ES, Green MR, Kelsen DP et al. Phase II trial of gemcitabine (2′2′-difluorodeoxycytidine) in patients with adenocarcinoma of the pancreas. Invest New Drugs 1994; 12: 1229–1234. Carmichael J, Fink U, Russell RCG et al. Phase II study of gemcitabine in patients with advanced pancreas cancer. Br J Cancer 1996; 73: 101–105. Schilsky RL, Bukowski R, Burris H III et al. A multicenter phase II study of a five-day regimen of oral 5-fluorouracil plus eniluracil with or without leucovorin in patients with metastatic colorectal cancer. Ann Oncol 2000; 11: 415–420. Mani S, Hochster H, Beck T et al. Multicenter phase II study to evaluate a 28-day regimen of oral fluorouracil plus eniluracil in the treatment of patients with previously untreated metastatic colorectal cancer. J Clin Oncol 2000; 18: 2894–2901. Wilki N, Glimelius B, Schueller J et al. Promising early results of oral 5-fluorouracil plus eniluracil (776C85) in patients with advanced adenocarcinoma of the pancreas. Ann Oncol 1998; 9 (Suppl 4): 46–47.