Related Experiment Video
Updated: Aug 6, 2026
Intraoperative Assessment of Resection Margins in Oral Cavity Cancer: This is the Way
Published on: May 10, 2021
Margin Status and Local Recurrence in Early-Stage Oral-Cavity Squamous Cell Carcinoma: Analysis of Reresection
Hannah J Brown1, Naomi C Wang1, Rahul Alapati1
1Department of Otolaryngology-Head and Neck Surgery, The University of Kansas Medical Center, Kansas City, Kansas, USA.
Objective:
To assess the impact of surgical margin status and re-resection on local recurrence (LR) in early-stage oral cavity squamous cell carcinoma (OCSCC).
Study Design:
Retrospective analysis of 352 patients with cT1-2N0 OCSCC.
Setting:
Tertiary care center from 2005 to 2024.
Methods:
Patients were stratified by margin status: initial negative (n = 325), re-resected to negative (RRN) (n = 20), and positive (n = 7). Initial margin status was determined intraoperatively with frozen sections using specimen-based or tumor bed-based sampling. Primary outcome was 3-year LR. Chi-square tests and Cox proportional hazards models evaluated differences across groups, accounting for potential confounders, for example, adjuvant treatment.
Results:
The most common subsite was the tongue (60.2%). 75.9% underwent elective neck dissection. 5.7% had an initial positive margin RRN. 4.8% and 22.7% received adjuvant chemoradiotherapy and radiotherapy, respectively. LR rates were 10.5% for initial negative, 35.0% for RRN, and 42.9% for positive margins despite re-resection. Compared to patients with initial negative, those with positive (hazard ratio [HR] 5.67, 95% CI 1.74-18.61) and RRN (HR 3.32, 95% CI 1.47-7.49) margins showed higher rates of LR. There was no significant difference in LR rates between positive versus RRN margins (P = .374). After controlling for adjuvant therapy and margin assessment method, positive (aHR 8.32, 95% CI 2.31-29.9) and RRN (aHR 3.82, 95% CI 1.68-8.68) margins remained associated with higher LR. Adjuvant therapy trended toward a protective effect (aHR 0.53, 95% CI 0.24-1.17, P = .117).
Conclusion:
RRN margins should not be considered oncologically equivalent to initial negative margins, as demonstrated by higher 3-year LR rates.