Zachary G Schwam1, Zain Husain2, Benjamin L Judson3. 1. Yale University School of Medicine, New Haven, USA. Electronic address: Zachary.schwam@yale.edu. 2. Department of Therapeutic Radiology, Yale University School of Medicine, New Haven, USA. 3. Department of Surgery, Section of Otolaryngology, Yale University School of Medicine, New Haven, USA.
Abstract
BACKGROUND AND PURPOSE: Administering postoperative radiotherapy (PORT) is associated with improved survival and slower disease progression in select head and neck cancer patients. Predictive factors for PORT refusal have not been described in this population. MATERIALS AND METHODS: Retrospective analysis of 6127 head and neck cancer patients who received or refused PORT in the National Cancer Database (2003-2006) was performed. Statistical analysis included Chi-square, multivariable logistic regression, Kaplan-Meier, and Cox proportional hazards analysis. RESULTS: In total, 247 patients (4.0%) refused PORT. Three-year overall survival was 62.8% versus 53.4% for those who received and refused PORT, respectively. PORT refusers were more likely to have negative nodes than those who underwent PORT (37.4% versus 20.1%, p<.001). In multivariate analysis, predictive factors for refusing PORT included living far from the treatment facility (OR 1.92), having negative nodes (OR 2.14), and Charlson score of ⩾ 2 (OR 2.14) (all p ⩽.001). PORT refusal was associated with increased mortality (hazard ratio 1.20, p=.044). CONCLUSIONS: A significant proportion of head and neck cancer patients refused PORT; this was associated with compromised overall survival. Predictive factors for PORT refusal included socioeconomic, demographic, and pathologic variables. Elucidating root causes of refusal may lead to interventions that improve long-term outcomes.
BACKGROUND AND PURPOSE: Administering postoperative radiotherapy (PORT) is associated with improved survival and slower disease progression in select head and neck cancerpatients. Predictive factors for PORT refusal have not been described in this population. MATERIALS AND METHODS: Retrospective analysis of 6127 head and neck cancerpatients who received or refused PORT in the National Cancer Database (2003-2006) was performed. Statistical analysis included Chi-square, multivariable logistic regression, Kaplan-Meier, and Cox proportional hazards analysis. RESULTS: In total, 247 patients (4.0%) refused PORT. Three-year overall survival was 62.8% versus 53.4% for those who received and refused PORT, respectively. PORT refusers were more likely to have negative nodes than those who underwent PORT (37.4% versus 20.1%, p<.001). In multivariate analysis, predictive factors for refusing PORT included living far from the treatment facility (OR 1.92), having negative nodes (OR 2.14), and Charlson score of ⩾ 2 (OR 2.14) (all p ⩽.001). PORT refusal was associated with increased mortality (hazard ratio 1.20, p=.044). CONCLUSIONS: A significant proportion of head and neck cancerpatients refused PORT; this was associated with compromised overall survival. Predictive factors for PORT refusal included socioeconomic, demographic, and pathologic variables. Elucidating root causes of refusal may lead to interventions that improve long-term outcomes.