A phase I/II study of external beam radiation, brachytherapy and concurrent chemotherapy in localized cancer of the esophagus (RTOG 92-07): Preliminary toxicity report

Purpose: A multi-institutional, prospective study was designed to determine the feasibility and tolerance of external beam irradiation plus concurrent chemotherapy and esophageal brachytherapy (EB) in a potentially curable group of patients with adenocarcinoma or squamous cell carcinoma of the esophagus. Methods and Materials: Planned treatment was 50 Gy external beam radiation (25 fractions/5 weeks) followed 2 weeks later by EB [either high dose rate (HDR) 5 Gy, weeks 8, 9, and 10, for a total of 15 Gy, or low dose rate (LDR) 20 Gy, week 8]. The protocol was later revised to delete the LDR alternative, owing to poor accrual, and to decrease the HDR dose to 10 Gy (i.e. 5 Gy, weeks 8 and 9). Chemotherapy was given weeks 1, 5, 8, and 11 with cisplatin 75 mg/m(2) and 5-fluorouracil 1000 mg(2)/m per 24 h, 96-h infusion. The study closed in January 1995 after 56 patients had been entered on the HDR arm. Six patients were declared ineligible owing to tumor extension to the gastroesophageal junction (three patients) or involved celiac lymph nodes (three patients). Of the 50 eligible patients, the planned EB dose was 15 and 10 Gy in 40 and 10 patients, respectively. Forty-six (92%) of the eligible patients had squamous histology, and three (6%) adenocarcinoma. Results: Life-threatening toxicity or treatment-related death occurred in 13 (26%) and 4 (8%) of the 50 eligible patients, respectively. Treatment-related esophageal fistulas occurred in three patients (12% overall, 14% of patients starting EB) at 0.5-6.2 months from the first day of brachytherapy, leading to death in three. The fourth death was secondary to renal toxicity and infection attributed to chemotherapy. No correlation was found between the development of fistula and location of primary tumor, brachytherapy active length or applicator diameter. So far, 5 of the 6 treatment-related fistulas have occurred following 15 Gy EB. The other fistula occurred after only 5 Gy of a planned 15 Gy was delivered. Conclusion: Thirty-five patients (70%) were able to complete external beam, FB, and at least two courses of chemotherapy. Estimated survival rate at 12 months is 48%, with an estimated 11-month median survival rate. Survival following external beam radiation plus concurrent chemotherapy and EB does not appear to be significantly different from survival seen following external beam radiation and chemotherapy only. The development of six fistulas in the 35 patients completing EB is of concern. Based on the high incidence of fistulas, we urge extreme caution in employing EB as a boost following concurrent external beam radiation and chemotherapy. (C) 1997 Elsevier Science Inc.