Importance: Head and neck basaloid squamous cell carcinoma (BSCC) has been considered a more aggressive variant of squamous cell carcinoma (SCC) with a poorer prognosis, although case-control studies have reached conflicting conclusions.
Objective: To examine the prognostic significance of head and neck BSCC on overall survival in a large population-based registry.
Design and setting: Retrospective data review of a population-based registry from the Surveillance, Epidemiology, and End Results database.
Participants: Individual case data for 34,196 patients treated between January 2004 and December 2009 with head and neck primary SCC (n = 33,554) and BSCC (n = 642) of the oral cavity, oropharyx, larynx, or hypopharynx. Patients with metastatic disease, incomplete staging information, and those who did not receive surgery or radiation were excluded.
Interventions: Patients had been treated with surgery, radiation, or both.
Main outcomes and measures: Distribution of patient characteristics between patients of each histology. Hazard ratios, 3-year overall survival, subgroup, and multivariate analysis of patient and treatment characteristics were investigated.
Results: Across each cohort, patients with BSCC more often had high-grade tumors and treatment with lymph node dissection. Multivariate analysis found that group stage, T stage, N stage, size, lymph node dissection, and age statistically significantly influenced overall survival. In multivariate analysis, the hazard ratio for death for patients with BSCC in the oral cavity and larynx and hypopharynx was not statistically significantly different from that for SCC. In the oropharynx, the hazard ratio for death for BSCC histology compared with SCC histology was 0.73 (P = .03).
Conclusions and relevance: Compared with SCC, BSCC is not an independent adverse prognostic factor for patients with head and neck cancer. The Surveillance, Epidemiology, and End Results analysis has limits, including lack of information regarding chemotherapy, but after controlling for disease and treatment variables, including neck dissection and radiotherapy, BSCC histology did not have an independent adverse prognostic effect on overall survival. The reported association between human papillomavirus and BSCC histology may explain the lower hazard ratio for death in patients with oropharynx BSCC.