The new england journal of medicine original article Postoperative Irradiation with or without Concomitant Chemotherapy for Locally Advanced Head and Neck Cancer Jacques Bernier, M.D., Ph.D., Christian Domenge, M.D., Mahmut Ozsahin, M.D., Ph.D., Katarzyna Matuszewska, M.D., Jean-Louis Lefèbvre, M.D., Richard H. Greiner, M.D., Jordi Giralt, M.D., Philippe Maingon, M.D., Frédéric Rolland, M.D., Michel Bolla, M.D., Francesco Cognetti, M.D., Jean Bourhis, M.D., Anne Kirkpatrick, M.Sc., and Martine van Glabbeke, Ir., M.Sc., for the European Organization for Research and Treatment of Cancer Trial 22931 abstract background We compared concomitant cisplatin and irradiation with radiotherapy alone as adju- From the Department of Radio-Oncology, vant treatment for stage III or IV head and neck cancer. Oncology Institute of Southern Switzerland, Bellinzona, Switzerland (J. Bernier); the methods Department of Radio-Oncology, Institut Gustave Roussy, Villejuif, France (C.D., J. After undergoing surgery with curative intent, 167 patients were randomly assigned to Bourhis); the Department of Radio-Oncolo- receive radiotherapy alone (66 Gy over a period of 61⁄2 weeks) and 167 to receive the gy, Centre Hospitalier Universitaire Vaudois, same radiotherapy regimen combined with 100 mg of cisplatin per square meter of Lausanne, Switzerland (M.O.); the Depart- ment of Oncology and Radiotherapy, Med- body-surface area on days 1, 22, and 43 of the radiotherapy regimen. ical University of Gdansk, Gdansk, Poland (K.M.); the Ear, Nose, and Throat Depart- results ment, Centre O. Lambret, Lille, France After a median follow-up of 60 months, the rate of progression-free survival was sig- (J.-L.L.); the Department of Radio-Oncolo- gy, Inselspital, Bern, Switzerland (R.H.G.); nificantly higher in the combined-therapy group than in the group given radiotherapy the Department of Radio-Oncology, Hospi- alone (P=0.04 by the log-rank test; hazard ratio for disease progression, 0.75; 95 per- tal General Vall d’Hebron, Barcelona, Spain cent confidence interval, 0.56 to 0.99), with 5-year Kaplan–Meier estimates of progres- (J.G.); the Department of Radio-Oncology, Centre G.F. Leclerc, Dijon, France (P.M.); sion-free survival of 47 percent and 36 percent, respectively. The overall survival rate the Department of Radio-Oncology, Centre was also significantly higher in the combined-therapy group than in the radiotherapy R. Gauducheau, St. Herblain, France (F.R.); group (P=0.02 by the log-rank test; hazard ratio for death, 0.70; 95 percent confidence the Department of Radio-Oncology, Cen- tre Hospitalier Universitaire A. Michallon, interval, 0.52 to 0.95), with five-year Kaplan–Meier estimates of overall survival of 53 Grenoble, France (M.B.); Istituto Regina El- percent and 40 percent, respectively. The cumulative incidence of local or regional re- ena, Rome (F.C.); and the European Orga- lapses was significantly lower in the combined-therapy group (P=0.007). The estimated nization for Research and Treatment of Cancer Data Center, Brussels (A.K., M.G.). five-year cumulative incidence of local or regional relapses (considering death from oth- Address reprint requests to Dr. Bernier at er causes as a competing risk) was 31 percent after radiotherapy and 18 percent after the Department of Radio-Oncology, San combined therapy. Severe (grade 3 or higher) adverse effects were more frequent after Giovanni Hospital, Oncology Institute of Southern Switzerland, CH-6504 Bellinzo- combined therapy (41 percent) than after radiotherapy (21 percent, P=0.001); the types na, Switzerland, or at jacques.bernier@ of severe mucosal adverse effects were similar in the two groups, as was the incidence of hcuge.ch. late adverse effects. N Engl J Med 2004;350:1945-52. conclusions Copyright © 2004 Massachusetts Medical Society. Postoperative concurrent administration of high-dose cisplatin with radiotherapy is more efficacious than radiotherapy alone in patients with locally advanced head and neck cancer and does not cause an undue number of late complications. n engl j med 350;19 www.nejm.org may 6, 2004 1945 The New England Journal of Medicine Downloaded from nejm.org at WELCH MEDICAL LIBRARY-JHU on December 13, 2010. For personal use only. No other uses without permission. Copyright © 2004 Massachusetts Medical Society. All rights reserved. The new england journal of medicine ocal or regional recurrences and mography of the site of the primary tumor and the ldistant metastases are frequent after surgical neck was highly recommended. treatment of stage III or IV squamous-cell To be eligible, patients had to have previously un- carcinoma of the head and neck. The risk of failure treated, histologically proven squamous-cell carci- is particularly high in patients with inadequate re- noma arising from the oral cavity, oropharynx, hy- section margins, extranodal spread, or multiple in- popharynx, or larynx, with a tumor (T) stage of pT3 volved lymph nodes.1-4 In patients with such locally or pT4 and any nodal stage (N), except T3N0 of the advanced tumors, surgery is usually followed by ad- larynx, with negative resection margins, or a tumor juvant radiotherapy. The advantage of postoperative stage of 1 or 2 with a nodal stage of 2 or 3 and no radiotherapy is well documented2-5 and compares distant metastasis (M0). Patients with stage T1 or favorably with the benefit afforded by preoperative T2 and N0 or N1 who had unfavorable pathological irradiation.6,7 findings (extranodal spread, positive resection mar- Several studies have demonstrated that concur- gins, perineural involvement, or vascular tumor rent treatment with radiotherapy and chemotherapy embolism) were also eligible, as were those with is a promising approach for locally advanced squa- oral-cavity or oropharyngeal tumors with involved mous-cell carcinoma that is not amenable to sur- lymph nodes at level IV or V, according to the ana- gery,8,9 justifying tests of the efficacy of chemother- tomical lymph-node distribution proposed by Rob- apy plus radiotherapy as postoperative (adjuvant) bins et al.15 treatment.10 Indeed, sequential adjuvant treatment Patients had to be at least 18 years of age and no with chemotherapy and radiotherapy significantly older than 70 years, with a performance status of 0, reduced the probability of nodal failure and distant 1, or 2, according to the scale of the World Health metastasis, and this improvement was directly Organization; they also had to have a serum creati- linked to the levels of clinical and pathological nine concentration of 1.36 mg per deciliter (120 risk.11 Moreover, in early randomized trials, con- µmol per liter) or less, a white-cell count of at least comitant postoperative radiotherapy and chemo- 4000 per cubic millimeter, a platelet count of at least therapy significantly improved local or regional con- 100,000 per cubic millimeter, and a hemoglobin trol but had no effect on overall survival.12,13 concentration of at least 11.0 g per deciliter (6.8 In 1994, the European Organization for Re- mmol per liter). Aminotransferase values and biliru- search and Treatment of Cancer (EORTC) began a bin values could not exceed twice the upper limit of randomized trial (EORTC trial 22931) to test the normal. Patients who had a history of invasive or hypothesis that adjuvant chemotherapy and radio- synchronous cancer (except nonmelanoma skin therapy improves progression-free survival, overall cancer), had previously received chemotherapy, or survival, and local or regional control more than had known central nervous system disease were ex- does radiotherapy alone in patients with stage III or cluded from the study. IV head and neck cancer. The study protocol was accepted by the indepen- dent review committee of each participating center. methods Informed consent was obtained from all patients in accordance with institutional guidelines. patient population and eligibility criteria The main objective of this study was to determine surgery whether the addition of cisplatin to high-dose ra- All eligible patients underwent primary surgery per- diotherapy after radical surgery increases progres- formed with curative intent. The extent of surgical sion-free survival in patients at high risk for recur- resection of the primary tumor and neck-dissection rent cancer. Secondary end points included overall procedures followed accepted criteria for adequate survival, relapse, and acute and late adverse effects. excision, which depend on the volume and location In this multicenter study, the stage of the tumor of the tumor. If the tumor was within 5 mm of the was determined on the basis of the histologic find- surgical margins, the resection margins were con- ings and classified according to the criteria of the sidered to be close. Union Internationale contre le Cancer.14 All patients underwent a full endoscopic examination during radiotherapy which a diagram was made of the extent of disease. All patients received postoperative radiotherapy con- Chest radiography, serum chemical analyses, and a sisting of conventionally fractionated doses of 2 Gy complete blood count were obtained. Computed to- each in five weekly sessions. Maximal and minimal 1946 n engl j med 350;19 www.nejm.org may 6, 2004 The New England Journal of Medicine Downloaded from nejm.org at WELCH MEDICAL LIBRARY-JHU on December 13, 2010. For personal use only. No other uses without permission. Copyright © 2004 Massachusetts Medical Society. All rights reserved. postoperative chemoradiotherapy in head and neck cancer target-volume doses and the maximal dose to the mary end point, was defined as the time from ran- spinal cord were recorded. Treatments were con- domization to any type of progression or death from ducted on linear accelerators of 4 to 6 MV with the any cause. Overall survival was defined as the time use of isocentric techniques. A large volume encom- from randomization to death from any cause.