Full-dose reirradiation for unresectable head and neck carcinoma: experience at the Gustave-Roussy Institute in a series of 169 patients.

Abstract

To review our experience using full-dose external reirradiation given with a curative intent for patients with unresectable head and neck carcinoma (HNC). Between January 1980 and December 1996, 169 patients who presented with unresectable nonmetastatic HNC in a previously irradiated area were included in this series. The median time between the first and the second irradiation was 33 months. Reirradiation protocols were as follows: radiotherapy alone (65 Gy over 6.5 weeks at 2 Gy/d), 27 patients; Vokes protocol, ie, five to six cycles of radiotherapy (median total dose, 60 Gy; 2 Gy/d) with simultaneous fluorouracil (5-FU) and hydroxyurea, 106 patients; and bifractionated radiotherapy (median total dose, 60 Gy; 2 x 1.5 Gy/d) with concomitant mitomycin, 5-FU, and cisplatin, 36 patients. The median cumulative dose of the two irradiations was 120 Gy. Eighty-five percent of the tumors were squamous cell carcinoma, 14% undifferentiated carcinoma of nasopharyngeal type, and 1% adenocarcinoma. Forty-four percent were local recurrences, 23% nodal recurrences, 14% both local and nodal, and 19% second primary tumors. Mucositis grade 3 (World Health Organization [WHO]) was found in 32% and grade 4 in 14% of cases. Four patients presented with neutropenia or thrombocytopenia (grade 3 or 4 WHO). Late toxicities (> 6 months) were as follows: cervical fibrosis (grade 2 to 3 Radiation Therapy Oncology Group [RTOG]), 41%; mucosal necrosis, 21%; osteoradionecrosis, 8%; and trismus, 30%. Five patients died of carotid hemorrhage, apparently in complete remission. Six months after the onset of reirradiation, 37% of patients were in complete response. Patterns of failure were local only (53%), nodal only (20%), metastatic only (7%), and multiple (20%). Median follow-up time was 70 months. Overall survival rate (Kaplan-Meier) was 21% (95% confidence interval [CI], 15% to 29%) at 2 years and 9% (95% CI, 5% to 16%) at 5 years. Median survival time was 10 months for the entire population. Thirteen patients, of whom 12 were treated with the Vokes protocol, were long-term disease-free survivors. In a multivariate analysis, the volume of the second irradiation was the only factor significantly associated with the risk of death: relative risk=1.8 (95% CI, 1.13 to 5.7) (P=.01). Full-dose reirradiation combined with chemotherapy was feasible in patients with inoperable HNC. The incidence and severity of late toxicity was markedly increased in comparison to that observed after the first irradiation. Median survival was better than that generally obtained using palliative chemotherapy alone. A small proportion of patients were long-term disease-free survivors.