Efficacy of adjuvant chemotherapy for patients with resectable head and neck cancer: a subset analysis of the Head and Neck Contracts Program.

Abstract

To evaluate the efficacy of adjuvant chemotherapy for patients with advanced head and neck squamous cancer, the Head and Neck Contracts Program conducted a three-arm study comparing standard surgery and radiation, induction chemotherapy (cisplatin and bleomycin) plus standard therapy, and induction chemotherapy plus standard therapy followed by maintenance cisplatin for 6 months. As previously reported, this trial of 462 patients demonstrated no significant difference in disease-free survival or survival, but a significantly lower metastatic rate in the maintenance arm. To determine whether particular subgroups may have benefited from adjuvant therapy, we evaluated results based on primary site, and tumor (T) and node (N) stage. Of the 192 patients with oral cavity cancer, those on the maintenance arm had a significantly improved 3-year disease-free survival (67%) compared with the standard arm (49%) or induction arm (44%) (overall P = .05). For hypopharyngeal and laryngeal cancers there was no marked overall benefit. For the 106 patients with T1 plus T2 disease, there was marginal improvement in disease-free survival for the maintenance group (72%) compared with the standard group (47%) or induction group (43%) (overall P = .09). There was no advantage for patients with T3 and T4 disease. There was superior disease-free survival for patients with N1 disease on the maintenance arm (70%) compared with the standard arm (42%) (P = .024). The same was true for disease-free survival in 109 patients with N2 disease: standard (52%), induction (30%), maintenance (84%) (overall P less than .001). There was no benefit for N3 disease. A significant survival advantage with maintenance chemotherapy was only seen for N2 disease (overall P = .04). Since head and neck cancer patients are a heterogeneous group, there may be particular sites and stages for which adjuvant chemotherapy would be advantageous, and subset analysis can help indicate directions for new trials.