Evaluation of proposed staging systems for human papillomavirus-related oropharyngeal squamous cell carcinoma.

BACKGROUND: Patients with human papillomavirus (HPV)-related oropharyngeal cancer (OPC) have improved survival when compared with those with HPV-negative OPC. Unfortunately, the American Joint Committee on Cancer seventh edition (AJCC-7ed) staging system does not account for the prognostic advantage observed with HPV-positive OPC. The purpose of the current study was to validate and compare 2 recently proposed staging systems for HPV-positive OPC.

METHODS: Patients treated for HPV-positive OPC from 2005 to 2015 at Johns Hopkins Hospital (JHH) were included for analysis. The International Collaboration on Oropharyngeal cancer Network for Staging (ICON-S) and The University of Texas MD Anderson Cancer Center (MDACC) staging systems were applied and survival was calculated using Kaplan-Meier methods. Cox proportional hazard regression was used to determine the relationship between stage of disease and survival. Models were compared using the Akaike information criterion (AIC).

RESULTS: A total of 435 patients were eligible for analysis. There was a dramatic shift in lymph node category and overall stage of disease when ICON-S and MDACC stage were applied to the JHH cohort. There was superior stratification of overall survival and progression-free survival by ICON-S stage. Both proposed models had an improved fit based on AIC scores (P<.001 for both) over the AJCC-7ed. The ICON-S staging system had the lowest AIC score, and thus a better fit within the JHH population.

CONCLUSIONS: The current analysis provides external validation for both staging systems in an independent and heterogeneously treated patient population. Although the MDACC staging system is an improvement over the AJCC-7ed, the ICON-S stage provides superior stratification of overall and progression-free survival, thereby supporting its use as the updated AJCC staging system for OPC. Cancer 2017;123:1768-1777. © 2017 American Cancer Society.