Between 1983 and 1986, a multicenter randomized study was conducted to compare a sequential program of induction chemotherapy (CT) followed by radiotherapy (RT), Arm A, with an alternation of 4 cycles of CT with 3 courses of RT (20 Gy/10 fractions up to a total dose of 60 Gy), Arm B, in advanced head and neck cancer patients. The same CT (VBM: Vinblastine, Bleomycin, Methotrexate) was used on both arms; one hundred and sixteen patients (pts) entered the study, 55 in Arm A, 61 in Arm B. Fourty-five pts had stage III and 71 stage IV cancers. The two arms are fully comparable. Up to October 1987, 116 pts are evaluable for survival, while 112 are evaluable for toxicity and 105 for response. In 21 patients (10 in Arm A, 11 in B) the association CT-RT was followed by surgery. Response analysis shows 14 complete responses in Arm A and 30 in Arm B (p less than or equal to 0.03). The median disease-free survival and median overall survival are also statistically different, with an advantage for Arm B (33 vs 22 weeks, p less than or equal to 0.0007, and 59 vs 38 weeks, p less than 0.03 respectively). The actuarial overall survival of complete responders at 50 months is 43% (B) and 21% (A). Toxicity (mainly stage III-IV mucositis) is superior in Arm B (30% vs 4%). Our experience demonstrates the advantages of alternate over sequential CT-RT. A comparison of this cyclic association with RT alone is in progress.