Upfront surgery versus definitive chemoradiotherapy in patients with human Papillomavirus-associated oropharyngeal squamous cell cancer

• Published in: Oral oncology 2018-04, Vol.79, p.64-70
• Main Authors: Kelly, Jacqueline R., Park, Henry S., An, Yi, Yarbrough, Wendell G., Contessa, Joseph N., Decker, Roy, Mehra, Saral, Judson, Benjamin L., Burtness, Barbara, Husain, Zain
• Format: Article
• Language: English
• Subjects: Aged; Alphapapillomavirus - isolation & purification; Chemoradiotherapy; Female; Human papillomavirus; Humans; Male; Middle Aged; Neoplasm; Oral cancer; Oropharyngeal Neoplasms - surgery; Oropharyngeal Neoplasms - therapy; Oropharyngeal Neoplasms - virology; Oropharynx; Prospective Studies; Radiotherapy; Squamous cell cancer; Squamous Cell Carcinoma of Head and Neck - surgery; Squamous Cell Carcinoma of Head and Neck - therapy; Squamous Cell Carcinoma of Head and Neck - virology; Surgery; Survival Analysis
• ISSN: 1368-8375; 1879-0593
• DOI: 10.1016/j.oraloncology.2018.02.017

•The NCDB was used to determine the effectiveness of primary surgery vs. CRT.•3-year OS was 90.8% in CRT patients vs. 93.6% in surgery patients (p = 0.27).•65.4% of surgical patients received trimodal therapy with adjuvant CRT.•An additional 22.7% of surgical patients received adjuvant RT. Currently, human papillomavirus-associated oropharyngeal squamous cell carcinoma (HPV-A OPC) is managed with either primary surgery or definitive chemoradiotherapy (CRT), despite the lack of supporting randomized prospective data. We therefore assessed the outcomes of each treatment strategy using the National Cancer Database (NCDB). The NCDB was used to identify patients diagnosed with cT1 N2a-2b or cT2 N1-2b HPV-A OPC from 2010 to 2013 who underwent treatment with primary surgery or CRT. Demographic and clinicopathologic predictors of treatment were analyzed by the chi-square test and logistic regression. Overall survival (OS) was evaluated using multivariable Cox proportional hazard regression, Kaplan-Meier, log-rank test, and propensity score-matched analysis. We identified 3063 patients; 1576 (51.5%) received CRT and 1487 (48.5%) underwent primary surgery. Median follow up was 32 months. 972 (65.4%) surgical patients received adjuvant CRT. On multivariable Cox regression, 3-year OS was comparable between surgery and CRT (hazard ratio [HR] 1.08, 95% confidence interval [CI] 0.83–1.41, P = 0.58). Inferior OS was significantly associated with increasing clinical T and N stage, older age, and non-private insurance. Propensity score-matching yielded a 2526 patient cohort and redemonstrated similar OS (HR, 1.09; 95% CI 0.81–1.47; P = 0.55). Comparable outcomes persisted in a subset analysis of patients with margin-negative resection, with 3-year OS 90.8% in CRT patients vs. 93.6% in surgery patients (log-rank P = 0.27). Upfront surgery and CRT yielded comparable 3-year OS outcomes in this cohort. In this national sample, 65.4% of surgical patients received trimodal therapy with adjuvant CRT, highlighting the need for improved patient selection for primary surgery.