OBJECTIVE: For patients with clinical nodal-negative (cN0) maxillary oral squamous cell carcinoma (MOSCC), neck dissection (ND) and clinical observation are the main two management strategies for the neck. However, the indications corresponding to these two options remain controversial. This study aimed to elucidate the clinical factors affecting ND treatment and to identify clinical characteristics of the population that may benefit from ND based on a retrospective analysis of cN0 MOSCC patient data from the Surveillance, Epidemiology, and End Results (SEER) database.
METHODS: 8846 MOSCC patients were identified in the SEER database from 2000 to 2020. The Kaplan-Meier method was utilized to examine overall survival (OS) and disease-specific survival (DSS), while the hazard ratio (HR) was estimated using the stepwise multivariate Cox regression model. Furthermore, multi-subgroup analyses of DSS and OS were performed to compare ND and No ND.
RESULTS: We included 2,512 cN0 MOSCC patients. Basic survival analysis and Cox regression modeling showed that ND was an independent prognostic factor that promoted DSS and OS. Additional subgroup analyses revealed that the primary site and T-stage might influence the efficacy of ND modality. Moreover, patients with T3/T4 stage of upper gingival squamous cell carcinoma (UGSCC) (DSS p = 0.009, OS p = 0.004), hard palate squamous cell carcinoma (HPSCC) (DSS p = 0.001, OS p < 0.001), and soft palate squamous cell carcinoma (SPSCC) (p = 0.029) showed a better survival benefit with ND in OS and DSS. Nonetheless, no differences were observed in OS and DSS between ND and No ND at the T1/T2 stage of the abovementioned primary tumor sites. Additionally, the DSS outcomes for T1/T2 stage upper lip squamous cell carcinoma (ULSCC) patients were significantly worse in the ND group than in the No ND group (p = 0.018). However, no significant differences were noted in OS (p = 0.140) as well as OS (p = 0.248) and DSS (p = 0.627) for T1/T2 and T3/T4 patients, respectively.
CONCLUSIONS: Active surveillance might be a feasible strategy for managing all T-staged ULSCC as well as early-stage (T1/T2) UGSCC, SPSCC, and HPSCC, provided regular and meticulous follow-up is performed. Hence, concurrent ND is recommended for patients with intermediate to advanced (T3/T4) stage UGSCC, SPSCC, and HPSCC.

The neck management of clinical-nodal negative (cN0) oral squamous cell carcinoma (OSCC) remains controversial. Elective neck dissection (END) and observation are the main strategies, but it is still not clear who could benefit the most from END. The purpose of this study was to clarify the potential clinical factors that affect the therapeutic value of END and to explore the actual characteristics associated with benefit from END.
Patients with cN0 OSCC were identified in the SEER database from 2000 to 2019. 5-year Overall survival (OS) and disease-specific survival (DSS) were analyzed using the Kaplan‒Meier method, and the hazard ratios (HRs) for survival were estimated using the Cox regression model. Multiple subgroup analyses of DSS and OS among different factors, comparing END and No END, were performed.
A total of 17,019 patients with cN0 OSCC were included. The basic survival analysis and Cox regression model showed that END increased the probability of 5-year DSS and OS and was an independent prognostic factor. However, among patients who underwent only primary tumor surgery, no significant differences were found between the END and No END groups in 5-year DSS (P = 0. 585) and OS (P = 0.465). Further subgroup analysis showed that primary sites and T stage, but not other factors, might influence the benefit of END. Significant differences were found for T1 (P < 0.001 for OS) and T2 (P = 0.001 for DSS and < 0.001 for OS) tongue squamous cell carcinoma (TSCC) but not for other primary tumor sites.
This large-scale retrospective population-based cohort study suggests that not all patients with cN0 OSCC could benefit from END. Patients with cN0 TSCC are recommended to undergo END, especially with early-stage tumors.