Introduction: Adenoid cystic carcinoma (ACC) is traditionally thought to have low risk for cervical lymph node metastasis. However, recent work demonstrates that the rate of positive nodes is higher than previously thought and that nodal metastasis is associated with increased risk of distant metastasis, and worse survival. Despite a mechanistic rationale for neck dissection (ND), the survival benefit is currently unclear. The objective of our study was to examine the impact of ND on survival in patients with major salivary gland ACC. We focused on two patient groups: 1) patients with clinical N0 disease and 2) patients with clinical N1-3 disease.
Methods: This was a retrospective cohort study using data from the National Cancer Data Base (NCDB) of the American College of Surgeons. The study cohort included patients with major salivary gland ACC diagnosed between 2004 and 2014, treated with surgery. Univariable analysis was performed using Kaplan-Meier analysis. Multivariable analysis was performed using Cox proportional hazard regression analysis, adjusting for the following covariates: tumor site, distant metastasis, margin status, radiotherapy (RT), age, sex, race, Charlson comorbidity score, and year of diagnosis.
Results: The study included 1,545 subjects that met inclusion criteria. ND was performed in 38.4% of clinical N0 necks and 73.9% of clinical N+ necks. Of the 55 clinical N0 patients who underwent ND, 70 had positive lymph nodes on pathologic assessment. In clinical N0 patients, univariable analysis showed no association between ND and survival [5-year OS 82% (95% CI 75% to 85%) for ND vs. 83% (95% CI 79% to 86%) for no ND, 10-year OS 61% (95% CI 53% to 67%) for ND vs. 64% (95% CI 53% to 73%) for no ND, p = 0.5]. Multivariable analysis showed that ND was not associated with survival in the overall cohort (HR = 0.95, 95% CI 0.73 to 1.24, p = 0.73). Stratified multivariable analysis showed that ND was associated with improved survival for T3-4 tumors (HR = 0.70, 95% CI 0.49 to 0.97, p = 0.03), but not for T1-2 tumors (HR = 1.19, 95% CI 0.79 to 1.80, p = 0.42). In clinical N1-3 patients, univariable analysis showed no association between ND and survival [5-year OS 43% (95% CI 34% to 54%) for ND vs. 56% (95% CI 39% to 70%) for no ND, 10-year OS 26% (95% CI 1% to 42%) for ND vs. 13% (95% CI 1% to 20%) for no ND, p = 0.6]. Multivariable analysis showed that ND was not associated with survival (HR = 1.08, 95% CI 0.59 to 1.95, p = 0.81).
Conclusions: In our study cohort, ND for clinical N0 disease was associated with improved survival in patients with T3-4 tumors, but not T1-2 tumors. ND was not associated with survival in patients with clinical N1-3 disease. There are several potential explanations for this finding. Patients with clinical nodal disease may be more likely to receive high-dose RT. They are also at increased risk of developing distant metastasis. These factors may minimize the effect of ND on survival.