Characterization of Loss of Prognostic Data with Updated Pathologic Nodal Staging System for p16+ Oropharyngeal Squamous Cell Carcinoma

Abstract

<h3>Purpose/Objective(s)</h3> Although the AJCC 8^th edition introduced a distinct pathologic nodal staging system for p16+ OPSCC, the removal of multiple factors in the 7^th edition, including nodal size, quantity, extracapsular extension (ECE), and laterality may limit valuable risk stratification and prognostication. This study aimed to characterize whether the simplification of the new staging system resulted in the loss of prognostic value by using the National Cancer Database (NCDB). <h3>Materials/Methods</h3> The NCDB was queried for patients diagnosed with node-positive p16+ OPSCC. Patients without staging information, metastatic disease, who did not receive definitive surgery, had unknown quantity of positive nodes, or had unknown follow-up were excluded. The optimal nodal quantity threshold was explored using a sensitivity analysis with univariable Cox proportional hazards regression. The prognostic impact of nodal size, nodal laterality, ECE, and the previously identified nodal quantity on OS were then assessed using survival analysis with the Kaplan Meier method, univariable, and multivariable Cox proportional hazards regression. <h3>Results</h3> A total of 5,186 patients met the inclusion criteria. On Kaplan Meier analysis, patients with more than four positive lymph nodes had significantly inferior OS (p<0.001; 5-yr OS: 76% vs 90%). Patients with ECE also had inferior outcomes (p<0.001; 5-yr OS: 83% vs 92%). A nodal size of >6 cm was also associated with inferior OS (p<0.001; 5-yr OS: 81% vs 89%). Lastly, patients with contralateral or bilateral nodal involvement also had inferior OS (p<0.001; 5-yr OS: 81% vs 89%). The nodal count threshold of four (p<0.001; HR [95% CI]: 2.75 [2.27-3.33]), as used in the AJCC 8^th edition, was also validated as the optimal prognostic value using the sensitivity analysis, in comparison to thresholds of three (p<0.001; HR [95% CI]: 2.33 [1.94-2.80]) or five (p<0.001; HR [95% CI]: 2.86 [2.33-3.53]). On multivariable Cox regression, having more than four positive nodes (p<0.001; HR [95% CI]: 1.82 [1.47-2.26]), ECE (p<0.001; HR [95% CI]: 1.60 [1.32-1.95]), and a node >6 cm (p<0.001; HR [95% CI]: 1.78 [1.04-3.02]) remained associated with inferior OS. Node laterality was no longer significant (p=0.412). Other factors associated with OS included age, Charlson Deyo score, insurance type, T stage, and lymphovascular invasion (LVSI). <h3>Conclusion</h3> Patients with a nodal size of >6 cm, more than four positive lymph nodes, ECE, or contralateral or bilateral nodal involvement have inferior OS. While AJCC 8^th edition serves to improve pathologic staging criteria, the associated simplifications result in loss of prognostic information in nodal staging. Although all of these factors needn't necessarily be included in the staging system, work evaluating the impact of the updated pathologic nodal staging for p16+ OPSCC prognostication and treatment selection is warranted, particularly given that factors such as ECE do influence adjuvant treatment.