Detailed Analysis of Clinicopathologic Factors Demonstrate Distinct Difference in Outcome and Prognostic Factors Between Surgically Treated HPV-Positive and Negative Oropharyngeal Cancer

Background: Oropharyngeal cancers (OPC) secondary to human papillomavirus (HPV) infections likely represent a completely different disease compared with conventional head and neck cancers. Our objective was to analyze a surgically treated cohort to determine predictors of outcome in HPV-positive versus HPV-negative patients. Methods: HPV positivity was inferred based on p16-immunohistochemistry. Data was available for 201 patients with OPC treated with surgical resection with/without adjuvant radiotherapy between 1985 and 2005. Subsite distribution was: 66 (33 %) tonsil, 46 (23 %) soft palate, and 89 (44 %) tongue base. Patients were classified into low-, intermediate-, and high-risk groups based on p16 status and smoking history. Outcomes stratified by p16 status and risk groups were determined by the Kaplan–Meier method. Factors predictive of outcome were determined by univariate and multivariate analyses. Results: In this cohort, 30 % had locally advanced disease (pT3/T4) and 71 % had nodal metastasis. The 5-year overall (OS), disease-specific, and recurrence-free survival rates were 60, 76, and 66 %, respectively. There were 22 % low-, 34 % intermediate-, and 44 % high-risk patients. Patients who were p16-positive had better survival compared with p16-negative (OS, 74 vs. 44 %; p < .001). Similarly, low-risk group patients had a better survival compared with intermediate- and high-risk groups (OS, 76, 68, 45 %, respectively, p < .001). Independent predictors of survival in p16-negative patients included margin status, lymphovascular invasion, pN status, and extracapsular spread. In contrast, none of these were predictive in p16-positive patients. Conclusions: Surgically treated patients with p16-positive OPC have superior survival compared with p16-negative patients. Outcomes in p16-positive and p16-negative OPC are determined by different prognostic factors supporting the notion that these are very different diseases. These should be incorporated into future clinical trials design.