Randomized comparison of neoadjuvant cisplatin and fluorouracil infusion followed by radiation versus concomitant treatment in advanced head and neck cancer.

Abstract

PURPOSE To compare two published schedules of cisplatin plus fluorouracil (5-FU) infusion and radiation as either sequential or concomitant treatment for toxicity and efficacy in patients with unresectable head and neck cancer. PATIENTS AND METHODS This was a randomized trial between cisplatin 100 mg/m2 over 15 minutes on day 1 plus 5-FU 1.0 g/m2 by continuous infusion on days 1 to 5, repeated every 3 weeks for three cycles, followed by 70 Gy of radiation in 7 to 8 weeks, versus cisplatin 60 mg/m2 over 15 minutes on day 1 plus 5-FU 800 mg/m2 by continuous infusion on days 1 to 5 plus radiation 2 Gy on days 1 to 5, repeated every other week for seven cycles. Unresectable head and neck squamous cancer patients not previously treated with radiation or chemotherapy and with a performance status of 0 to 2 were stratified by tumor (T) and node (N) groupings and performance status and randomized. RESULTS Two hundred fifteen patients were entered and 214 analyzed, 107 on each arm. After all treatment, overall response rates were different (P = .003), with similar complete response rates, but more partial responses and fewer patients with no change or progression with concomitant treatment. Cox regression analysis for progression-free survival identified concomitant treatment (P = .003), Radiation Therapy Oncology Group (RTOG) stage III grouping (P < .0001), performance status (P = .0002), concomitant treatment (P = .003), and treating institution (P = .006) as significant. The sequential and concomitant treatments showed similar distant failure patterns (10% and 7%, respectively), but divergent regional failure rates (55% and 39%). Severe and worse toxic events were similar between the treatment programs, but radiation-induced mucositis combined with cisplatin-induced water-losing nephropathy, in the concomitant arm only, demanded more supportive care. Survival duration was similar between the treatment arms, but significantly more patients in the sequential arm died of their cancer (P = .011). CONCLUSION Concomitant treatment offered improved disease control, predominantly of regional disease, but benefit was dependent on the experience of the treating institution. Translation of this benefit into improved survival is not yet evident, with an excess of deaths from other causes in the concomitant arm.