Survival analysis and risk factors for recurrence in oral squamous cell carcinoma: does surgical salvage affect outcome?

PURPOSE: The purpose of this retrospective study was to review the outcomes and recurrence rates of subjects with oral cavity squamous cell carcinoma treated at a single institution by primary surgical resection, with or without adjuvant radiation or chemotherapy, to identify factors that affect locoregional control and determine whether surgical salvage affects survival.

MATERIALS AND METHODS: The records of 157 subjects diagnosed with oral cavity squamous cell carcinoma treated at a single institution from 1997 to 2007 were identified. Data on demographics, site, clinical stage, pathologic stage, treatment, recurrence, and survival were collected. Defined outcome measures were overall survival, disease-free survival, and length of survival after recurrence. Analysis of the data was performed by use of the Cox proportional hazards model. Kaplan-Meier survival curves were created for disease-free survival, as well as survival by histologic grade, nodal status, recurrence, and tumor stage.

RESULTS: We identified 157 subjects, with 155 meeting the inclusion criteria. The overall 5-year survival rate was 48%, with a disease-free survival rate of 42% (95% confidence interval, 36%-53%). Survival was found to be influenced by stage (P = .0001), nodal status (P = .0025), and histologic grade (P = .04). There were 24 subjects with recurrence (15%). Of these, 11 had local recurrence (46%), 9 had regional recurrence (37%), 2 had distant recurrence (8%), 1 had both local and regional recurrence (4%), and 1 had both local and distant metastasis (4%). Recurrence was not found to be significantly affected by pathologic stage (P = .71), clinical stage (P = .6), histologic grade (P = .178), postoperative radiation therapy (P = .54), postoperative chemotherapy (P = .66), N-positive status (P = .71), or whether the subject underwent a neck dissection (P = .984). Surgery significantly increased both overall survival time (P = .009) and survival time after recurrence (P = .006). Radiation therapy (P = .4) and chemotherapy (P = .82) did not have a survival benefit as therapy for recurrence.

CONCLUSIONS: Survival is influenced by stage at presentation, nodal status, and histologic grade. No variables were found to influence recurrence rates. Surgery significantly increased overall survival time, and salvage surgery increased survival after recurrence.