Patterns of failure, prognostic factors and survival in locoregionally advanced head and neck cancer treated with concomitant chemoradiotherapy: a 9-year, 337-patient, multi-institutional experience

Background: Locoregionally advanced, stage IV head and neck cancer has traditionally carried a poor prognosis. We sought to assess changes in patterns of failure, prognostic factors for recurrence, and overall outcome, using two different strategies of chemoradiotherapy conducted in prospective, multi-institutional phase II trials. Patients and methods: Three hundred and thirty-seven stage IV patients were treated from 1989 to 1998. We compared locoregional and distant recurrence rates, overall survival and progression-free survival from two different treatment strategies: intensive induction chemotherapy followed by split-course chemoradiotherapy (type 1, n = 127), or intensified, split-course, hyperfractionated multiagent chemoradiotherapy alone (type 2, n = 210). Univariate and multivariate analyses of 12 chosen covariates were assessed separately for the two study types. Results: The pattern of failure varied greatly between study types 1 and 2 (5-year locoregional failure of 31% and 17% for study types 1 and 2, respectively, P = 0.01; 5-year distant failure rate of 13% and 22% for study types 1 and 2, P = 0.03). Combined 5-year overall survival was 47% [95% confidence interval (CI) 41% to 53%) and progression-free survival was 60% (95% CI 55% to 66%). Both treatment strategies yielded similar survival rates. Poor overall survival and distant recurrence were best predicted by advanced nodal stage. Locoregional recurrence was extremely rare for patients with T0-T3 tumor stage, regardless of lymph-node stage. Conclusions: This analysis suggests that pattern of failure in primary head and neck cancer may be dependent upon treatment strategy. Randomized clinical trials of induction chemotherapy are warranted as a means to determine if a decrease in distant metastases can lead to an increase in survival rates in the setting of effective chemoradiotherapy for locoregional control. Additionally, this analysis provides impetus for randomized clinical trials of organ preservation chemoradiotherapy in sites outside the larynx and hypopharynx.