Omission of the Pathologic Node-Negative Neck From Postoperative Radiation Therapy for Oral Cavity Cancer

Abstract

<h3>Purpose/Objective(s)</h3> Elective neck dissection has been shown to confer a survival benefit in the setting of oral cavity squamous cell cancer (OC-SCC). Post-operative radiation therapy (RT) is recommended after primary surgical management of OC-SCC for adverse pathologic features. Current guidelines recommend coverage of the operative bed, including the pathologic node-negative (pN0) neck, which is associated with worse late toxicity and quality of life. The risk of regional failure with omission of the pN0 neck from the RT treatment volume is not well described. <h3>Materials/Methods</h3> Patients (pts) with OC-SCC who underwent primary surgical management and were treated with post-operative RT at a single institution between 2011 and 2019 were retrospectively analyzed. Intensity-modulated RT was used in all cases. Post-operative RT was delivered once daily. Post-operative chemoradiation therapy (CRT) typically consisted of concomitant fluorouracil, hydroxyurea, with or without paclitaxel in a week-on/week-off fashion with either once- or twice-daily RT. The Kaplan-Meier method was used to estimate rates of local failure-free survival (LFFS), omitted neck failure-free survival (ONFFS), distant metastasis-free survival (DMFS), and overall survival. Cox regression analysis was used to examine the relationships between clinicopathologic variables, and ONFFS. <h3>Results</h3> 47 pN0 hemi-necks in 36 pts were omitted from the post-operative RT treatment volume. The median follow-up was 53.5 months. Primary site was oral tongue in the majority (N = 16). Other subsites included floor of mouth (N = 7), alveolus (N = 6), buccal (N = 3), retromolar trigone (N = 1), and palate (N = 1). Median number of lymph nodes removed from the RT-omitted pN0 ipsilateral and contralateral hemi-neck were 24 (IQR: 19-27) and 18.5 (IQR: 9-24), respectively. Perineural invasion was present in 63% and lymphovascular invasion was present in 18%. T-stage was T1 (N = 4), T2 (N = 10), T3 (N = 7), and T4 (N = 15). Overall N-stage (including the pathologic node-positive hemi-neck, if present) was N0 (N = 23), N1 (N = 5), N2a (N = 1), N2b (N = 2), and N3b (N = 5). Median RT dose was 66 Gy (IQR: 60-66). 70% of pts received concurrent CRT. 89% of pts received once-daily RT and 11% received twice-daily RT. Overall, three hemi-necks (6%) in three pts (8%) were found to have omitted neck relapse. Two-year LFFS, ONFFS, DMFS, and OS were 97%, 91%, 94%, and 92%, respectively. Four-year LFFS, ONFFS, DMFS, and OS were 93%, 91%, 91%, and 85%, respectively. On Cox regression, floor of mouth subsite (FOM) (HR 11.1 [C.I.: 1.0-126.2], p=0.05) was associated with an increased risk of ONFFS. <h3>Conclusion</h3> In this largest reported series of OC-SCC pts treated with post-operative RT omitting the pN0 neck, low rates of regional failures occurred overall.