Depth of Invasion Alone is Not an Independent Prognostic Factor in Early-Stage Node-Negative Oral Cavity Carcinoma

Abstract

The use of depth of invasion (DOI) as a prognostic factor and criterion for neck dissection (ND) in patients with oral cavity carcinoma (OCC) has been the topic of multiple clinical analyses. However, these studies are heterogeneous, frequently including patients of any stage and with multiple other histologic risk factors for recurrence. In addition, patients often received adjuvant treatment in the form of chemotherapy and/or radiation. Thus, it is difficult to draw conclusions on the importance of DOI as a true prognostic factor. We created a unique population of patients with DOI as the sole risk factor for recurrence to accurately evaluate whether it is an independent prognostic factor for progression in patients with early-stage node-negative OCC. All patients were drawn from an institutional review-board approved database of OCC patients treated at a single academic institution between 2003 and 2013. A total of 560 patients were retrospectively reviewed. Patients were included in the final analysis if they had pathologic T1-2N0 OCC treated with surgical resection with or without neck dissection (ND). Pathology specimens were independently reviewed by 2 board-certified pathologists to confirm proper evaluation of DOI and other histologic parameters. Patients were excluded if they received adjuvant treatment, chemotherapy for other cancer diagnoses in the follow up period, or if other high-risk pathologic features were present, including: perineural invasion, lymphovascular invasion, positive lymph nodes, and close or positive margins. Kaplan-Meier and Cox proportional hazards regression analyses were performed to identify factors predictive for local (LR) and regional recurrence (RR), as well as progression free survival (PFS). A total of 126 patients with T1-2 N0 OCC were analyzed. Median follow up for the entire cohort was 42.5 months and median overall survival (OS) was 46 months. Median DOI for all sites was 4 mm, with 60 (47.6%) patients with DOI <4 mm and 66 (52.4%) patients with DOI ≥4 mm. There was no significant difference in local (p=0.95), regional (p=0.81), or distant (p=0.96) recurrence between patients with DOI <4 mm and ≥4 mm. In total, 50 (42.9%) patients underwent ND. There was no significant difference in LR among patients with DOI <4 mm and ≥4 mm (p=0.95). In patients with DOI <4 mm, 5-year RR was 0% if ND was performed and 22% if ND was omitted. In the subset of patients with oral tongue carcinoma and DOI <4mm (n=36), there were no regional or distant recurrences in those patients who underwent ND compared to 20% and 4% in patients who did not undergo ND. On multivariate analysis including T stage, gender, age, DOI, and multiple head and neck primaries, the only factor significant for PFS was ND (HR 0.25, p=0.006). In this study, we show that DOI ≥4 mm alone does not predict for worse local, regional, or distant recurrence. ND appeared to improve the RR for even those with DOI <4mm.