Front-line treatment of inoperable or metastatic pancreatic cancer with gemcitabine and capecitabine: an intergroup, multicenter, phase II study.

PURPOSE: To evaluate the efficacy and toxicity of gemcitabine (GEM) combined with capecitabine (CAP) in untreated patients with inoperable or metastatic pancreatic cancer. PATIENTS AND METHODS: Fifty-three patients with pancreatic cancer (85% stage IV) were enrolled. Patients were treated with GEM 1000 mg/m2 on days 1 and 8 and CAP 1300 mg/m2 per day PO (per os), divided into two equal doses on days 1-14, in 21-day cycles. RESULTS: In an-intention-to-treat analysis, 10 (18.9%) objective partial responses were achieved (95% confidence interval 8.33% to 29.4%). Twenty-two (42%) patients had stable disease and 15 (28%) had progressive disease. The median response time was 3 months (range 1.5-7.0) and the median time to tumor progression was 6.5 months (range 3.5-15.5). Median overall survival time was 8 months (range 1.0-15.5) and 1-year survival was 34.8%. Pain improvement during treatment was observed in 23 of 43 (53%) patients, and eight of 18 (44%) patients who had been receiving opioids discontinued their use. Weight gain was observed in 12 of 33 (36%) patients. Grade 3 anemia occurred in five (9%) patients and grade 3-4 thrombocytopenia occurred in three (6%). Grade 3-4 neutropenia occurred in 13 (25%) and five (9%) patients, respectively, and two (4%) developed febrile neutropenia. Non-hematological toxicity was mild. CONCLUSION: In patients with pancreatic cancer, the combination of GEM with CAP is an active and well tolerated regimen that merits further evaluation in prospective randomized studies.