Randomized study of a short course of weekly cisplatin with or without amifostine in advanced head and neck cancer. EORTC Head and Neck Cooperative Group

Abstract

Background: Cisplatin is one of the most active cytotoxic agents available for the treatment of patients with head and neck cancer. In a previous phase II study with weekly administration of cisplatin, a response rate of 51% was achieved. However, only in a minority of the patients the planned high dose intensity of 80 mg/m2/week could be reached because of toxicity, mainly thrombocytopenia and ototoxicity. Amifostine is a cytoprotective drug that can diminish the toxicity of alkylating agents and platinum compounds. Therefore the effect of amifostine on toxicity and activity of weekly cisplatin was investigated in a randomized study.

Patients and methods: Patients with locally advanced, recurrent or metastatic head and neck cancer were eligible. Patients were randomized to weekly cisplatin 70 mg/m2 for six cycles preceded by amifostine 740 mg/m2, or cisplatin only. Cisplatin was administered in hypertonic saline (3% NaCl) as a one-hour infusion; amifostine was administered as a 15-minute infusion directly before the administration of cisplatin.

Results: Seventy-four patients were entered in the study. The median number of cisplatin administrations was 6 (range 2-6), equal in both arms. In both treatment arms the median dose intensity of cisplatin achieved was the planned 70 mg/m2/week. In the cisplatin only arm 6 out of 206 cycles were complicated by thrombocytopenia grade 3 or 4 versus 1 of 184 cycles in the amifostine arm (P = 0.035). Hypomagnesaemia grade 2 + 3 was significantly less observed in the amifostine arm (P = 0.04). Neurotoxicity analyzed by serial vibration perception thresholds (VPT) showed a diminished incidence of subclinical neurotoxicity in the amifostine arm (P = 0.03). No protective effect on renal and ototoxicity could be shown. Hypotension was the main side effect of amifostine but only of relevance in one patient. The antitumor activity of cisplatin was preserved as 63% of the evaluable patients in the amifostine arm responded compared to 50% of the evaluable patients in the cisplatin alone arm.

Conclusion: Our study indicated that in combination with weekly administered cisplatin amifostine reduced the risk of thrombocytopenia, hypomagnesemia as well as subclinical neurotoxicity, but did not result in a higher dose intensity of cisplatin. Addition of amifostine did not compromise the antitumor effect of cisplatin.