Pilot study of postoperative reirradiation, chemotherapy, and amifostine after surgical salvage for recurrent head-and-neck cancer

Purpose: Salvage surgery alone after radiotherapy (RT) failure for locally advanced head-and-neck cancer is frequently unsuccessful because of subsequent recurrence. We designed a prospective protocol to determine the feasibility, toxicity, and preliminary efficacy of a regimen of postoperative reirradiation, chemotherapy and the radioprotector amifostine after salvage head-and-neck surgery. Methods and Materials: Eligible patients had biopsy-proven locally advanced, but resectable, recurrence without distant metastases > 6 months after previous RT. After adequate healing from surgery, patients underwent RT to 54-66 Gy within 5-5.5 weeks to the resection bed (lifetime RT dose, 100-130 Gy). The fractionation was 1.5 Gy b.i.d. within 2 weeks, followed by a 1-week break, followed by 1.5 Gy b.i.d. to a total of 54-66 Gy. Chemotherapy consisted of cisplatin 25 mg/m(2)/d three times and 5-fluorouracil 500 mg/m(2)/d continuous infusion for 4 days, for two cycles (Weeks 1 and 5). Amifostine (500 mg i.v.) was administered daily, 30 min before either the morning or the afternoon RT. Results: Between 1998 and 2001, 16 patients were enrolled and studied. Two patients had gross residual disease after surgery; all other patients underwent complete surgical resection but had high-risk features (rT3-T4 and/or N+ disease). Three patients (19%) had serious acute toxicity events (nonneutropenic infections) that were reversible. The median follow-up was 35 months. The actuarial locoregional control rate was 81% at 3 years. Three patients developed isolated distant metastases and one developed a fatal second primary cancer (hepatoma). The 2- and 3-year actuarial event-free survival rate was 81% and 50%, respectively. The 2- and 3-year actuarial overall survival rate was 81% and 63%, respectively. Both patients who had gross residual disease after surgery had early recurrence; if these patients were excluded from analysis, the 3-year actuarial survival and event-free survival rate was 67% and 59%, respectively. Of the 16 patients, 6 (38%) developed Grade 3+ late toxicity, including one fatal stroke and two life-threatening major vessel necrosis and/or bleeding events. Conclusion: This regimen of postoperative reirradiation/chemotherapy plus amifostine is feasible and was well tolerated acutely, with encouraging oncologic efficacy. However, the incidence and severity of late effects was significant and suggests that modifications are necessary for future studies in this patient population. (C) 2004 Elsevier Inc.