Purpose: This study was designed to evaluate the effects of preoperative chemoradiation on resectability, response, local control, and survival in patients with local or local-regional involvement from carcinoma of the pancreas or cancer of the duodenum and to assess the associated toxicity of such treatment.
Methods and materials: This prospective pilot study of preoperative chemoradiation was initiated in 1986 for patients with clinical evidence of adenocarcinoma of the pancreas or duodenum without evidence of distant metastases. Radiation was given at 1.8 Gy per day to a total dose of 50.4 Gy. Two cycles of chemotherapy were given concurrent with radiation. On days 2-5 and 29-32, 5-fluorouracil (1 gm/m2/24 h x 4 days) was given, while mitomycin-C (10 mg/m2) was given on day 2 only. Surgical resection was 4-6 weeks following completion of chemoradiation. Thirty-one patients (17 male and 14 female) were entered on the protocol with a median potential follow-up of 4.5 years (range 6 months to 7.5 years). The median age was 64 years (range 32-73 years). Twenty-seven patients had pancreatic cancer (25 head, two body), while four patients had carcinoma arising from the duodenum. Twenty-one patients were initially judged to be unresectable and ten potentially resectable prior to chemoradiation.
Results: Twenty-nine of 31 patients completed the entire course of radiation and both cycles of chemotherapy. Acute toxicity from chemoradiation consisted of nausea, vomiting, diarrhea, stomatitis, or hematologic suppression which was moderate to severe (Grade 3 or 4) in seven patients (23%). One patient died of sepsis following the first week of therapy. Seventeen patients (55%) underwent curative resection with subtotal or total pancreatectomy or Whipple resection (four duodenum, 13 pancreas) and two (2/17) had pathologic nodal involvement, while (0/17) none had involved margins. A complete pathologic response was seen in all four (4/4) patients with duodenal cancer and in none (0/13) with pancreatic cancer who underwent resection. The median postoperative hospitalization stay was 22 days (range 4-144 days). Of 17 patients who underwent curative resection, there were two postoperative mortalities (12%). Late complications have included abscess, one; and nonmalignant ascites, five. Ten of the 31 patients are alive. For patients with pancreatic cancer the median survival is 9 months, while survival at 1 year and 3 years are 36% and 19% overall and 60% and 43% at 1 and 3 years for those undergoing resection. Six of the 27 patients (22%) with pancreatic cancer are alive without recurrence. All four patients with duodenal cancer are alive without recurrence (12 months, 23 months, 35 months, 90 months).
Conclusion: Preoperative chemoradiation for cancer of the pancreas and duodenal region was relatively well-tolerated and enhanced resectability and downstaging of nodal metastases were suggested. The 3-year survival, particularly in patients who underwent resection, was high. For these reasons the applicability of this treatment regimen for pancreatic cancer is presently being studied in a group-wide multi-institutional Phase II trial. Chemoradiation for duodenal cancer has produced a complete pathologic response in all patients and survival has been excellent, suggesting efficacy of this regimen for duodenal cancer.