Limitations of conventional doses of chemoradiation for unresectable biliary cancer.

PURPOSE: To determine, in a retrospective review, the limitations of definitive chemoradiation in the treatment of patients with unresectable extrahepatic cholangiocarcinoma and generate testable hypotheses for future prospective clinical trials. METHODS AND MATERIALS: Between 1957 and 2000, 52 patients with localized, unresectable cholangiocarcinoma were treated with radiotherapy (RT) with or without concurrent chemotherapy. Unresectable disease was defined, by evidence on imaging studies or at surgical exploration, as localized tumor abutting or involving the main portal vein, tumor involvement of secondary biliary radicals, or evidence of nodal metastases. Patients were grouped according to the RT dose: 27 patients received a total dose of 30 Gy (Group 1), 14 patients received 36-50.4 Gy (Group 2), and 11 patients received 54-85 Gy (Group 3). 192Ir intracavitary boosts (median 20 Gy) were delivered in 3 patients, and an intraoperative boost (20 Gy) was used in 1 patient. Of the 52 patients, 38 (73%) received concomitant protracted venous infusion of 5-fluorouracil (200-300 mg/m2 daily, Monday through Friday). Kaplan-Meier analysis was used to calculate the actuarial 1-year and median overall survival (OS), radiographic local progression, symptomatic progression, and distant failure. Treatment-related variables and prognostic factors were evaluated using the log-rank test. RESULTS: The first site of disease progression was local in 72% of cases. The actuarial local progression rate at 12 months for all patients was 59%. The median time to radiographic local progression was 9, 11, and 15 months in Groups 1, 2, and 3, respectively (p = 0.48). Fifteen percent of all patients developed metastatic disease (1-year OS rate 18%). The median survival rate for all patients was 10 months (1-year OS rate 44%). The RT dose, use of concurrent chemotherapy, histologic grade, initial extent of liver involvement, and extent of vascular involvement had no influence on radiographic local progression or OS. Grade 3 or greater toxicity was similar in all dose groups (22% vs. 14% vs. 27%, p = 0.718). CONCLUSION: The primary limitation of definitive chemoradiation was local progression. Although the small patient numbers limited the statistical power of this study, a suggestion of improved local control was found with the use of higher RT doses. To address this pattern of failure, future prospective investigation using high-dose conformal RT with novel cytotoxic and/or biologic agents with radiosensitizing properties is warranted.