Correlation of pathologic tumor grade with survival in oral cavity squamous cell carcinoma.

Abstract

e18557 Background: Pathologic tumor grade is a well-established prognostic risk factor that impacts staging and standard-of-care treatment decisions across multiple cancer types. However, the significance of tumor grade in cancers of the head and neck is less certain. Even in oral cavity squamous cell carcinoma (OCSCC), the head and neck cancer subsite with largest body of literature regarding the predictive value of tumor grade is, the prognostic significance of tumor grade remains controversial. Thus, we sought to better elucidate the prognostic importance of tumor grade in OCSCC. Methods: Patients with OCSCC diagnosed from 2004-2015 and undergoing primary surgery with or without adjuvant treatment in the National Cancer Data Base (NCDB) were identified. Overall survival (OS) was estimated using the Kaplan-Meier method with and without propensity score matching (PSM). Univariate and multivariable survival analyses were performed using Cox regression. Analyses were adjusted for multiple clinicopathologic factors, including age, sex, comorbidity status, year of diagnosis, pathologic staging, margin status, number of lymph nodes (LN) examined/positive, extranodal extension (ENE), lymphovascular invasion (LVI), adjuvant radiation, and concomitant chemotherapy. Results: Median follow-up was 40.7 months. Of 13,941 patients with OCSCC, 2,883 had low-grade tumors, 8,716 had intermediate-grade tumors, and 2,342 had high-grade tumors. Higher tumor grade was strongly associated with decreased survival. Specifically, five year OS was 62.7%, 52.8%, and 42.5% in low-grade (LG), intermediate-grade (IG), and high-grade (HG) OCSCC, respectively (p-value < 0.001). In PSM cohorts, OCSCC patients with high-grade had significantly worse 5 year OS (47.7% vs. 57.7%, p < 0.001) in comparison to those with LG OCSCC. Similarly, patients with IG tumors has worse 5-year OS (55.6% vs. 60.3%, p = 0.001) than patients with LG tumors in PSM cohorts. In multivariable analysis, both HG (HR 1.38, 95% CI 1.25-1.52, p < 0.001) and IG (HR 1.17, 95% CI 1.08-1.26, p < 0.001) OCSCC was associated with worse survival than what was observed in LG tumors. The magnitude of the independent effect of tumor grade in multivariable analysis was greater than or equal to what was observed with other well-established prognostic factors like margin positivity (HR 1.34), ENE (HR 1.35), and LVI (HR 1.18). Conclusions: Pathologic tumor grade is a strong predictor of survival among patients with OCSCC. Tumor grade should be considered when making therapeutic recommendations for OCSCC.