A randomised trial of radiotherapy compared with cisplatin chemo-radiotherapy in patients with unresectable squamous cell cancer of the esophagus

Background and purpose: Following our phase 11 experience, a randomised trial was undertaken to evaluate the efficacy of adding chemotherapy to radiotherapy in patients with unresectable squamous cell cancer of the esophagus. Patients and methods: Patients randomised to the RT group received 50 Gy/25 fx/5 weeks of teletherapy followed 1-2 weeks later with 12 Gy/2 fx of high-close-rate intra-lumenal brachytherapy spaced a week apart. Following the first 3 years of recruitment, due to unexpected late morbidity, brachytherapy was excluded and the protocol modified to 66 Gy/33 fx/6.5 weeks. The CRT group received identical radiotherapy with concurrent weekly cisplatin at 35 Mg/M2 for 6-7 cycles. Results: Between April 1999 and December 2005, 125 patients were randomised to a RT (n = 60) or CRT group (n = 65). Radiotherapy treatment was completed in 78% (47/60) of the RT group and 89% (58/65) of the CRT group (P = 0. 10). Six or more cycles of cisplatin could be delivered in 63% (41/65), which resulted in RTOG grade 3 neutropenia of 3%. Late morbidity in the form of ulcers (5% vs. 15% odds ratio 0.29, 95% Cl 0.08-1.11, P = 0.08) and strictures (113% vs. 28%, odds ratio 0.40, 95% Cl 0. 16-1.012 P = 0.05) was observed in the RT and CRT groups, respectively. At a median follow up of 23 months of all patients alive (range 6-82 months) and with 95/125 events, the median, 1, 2 and 5 year projected survival was 7.1 months, 32.3%, 22.8% and 13.7% vs. 13.4 months, 57.6%, 38.9% and 24.8% for the RT and CRT groups, respectively (hazard ratio 0.65, 95% Cl 0.44-0.98, P = 0.038). Conclusions: The addition of concurrent cisplatin to radiotherapy resulted in a modest improvement in survival and was associated with manageable additional acute and late morbidity. (C) 2007 Elsevier Ireland Ltd. All rights reserved.