Impact of Smoking and Primary Tumor Subsite on Recurrence in HPV-Associated Oropharyngeal Squamous Cell Carcinoma

Abstract

Objective To describe risk of recurrence and recurrence characteristics between ever- and never-smoking patients with human papillomavirus–associated oropharyngeal squamous cell carcinoma (HPV+ OPSCC) when stratified by primary tumor subsite. Study Design Retrospective observational study. Setting Tertiary care center. Methods Retrospective chart review of 171 patients with HPV+ OPSCC with primary treatment between 2008 and 2019. Five-year recurrence-free survival and risk of recurrence were evaluated through Kaplan-Meier curves with log-rank test and Cox proportional hazards models, respectively. Results Of 171 patients with HPV+ OPSCC, 81.9% were male, and the average age was 63.9 years. Eighty patients (46.8%) had a smoking history (average, 17.7 pack-years), including 4 current smokers. Recurrence occurred in 31 patients (18.1%), 19 of whom were ever smokers. The recurrence rate for ever smokers with primary base of tongue (BOT) cancer was 41.7%, while 5.1% of never smokers with BOT primaries had recurrence. For primary tonsillar disease, 9.1% of ever smokers had recurrence versus 19.2% of never smokers. Five-year recurrence-free survival for BOT primaries was lower in ever smokers than never smokers ( P = .001) but did not differ between ever and never smokers for tonsillar primaries ( P = .215). In multivariable analysis across this period, ever-smoking status was associated with higher risk of recurrence than never-smoking status in BOT primaries (adjusted hazard ratio, 7.36; 95% CI, 1.61-33.68; P = .010) but with lower risk of recurrence after tonsillar primaries (adjusted hazard ratio, 0.23; 95% CI, 0.06-0.89; P = .033). Conclusion Smoking may uniquely interact with tumor subsites within the oropharynx to influence recurrence risk. Understanding the association between smoking and HPV+ OPSCC recurrence could lead to personalized, evidence-based treatments to improve oncologic outcomes.