The Number of Positive Lymph Nodes Instead of Extranodal Extension and Nodal Diameter is an Independent Predictor of Survival and Treatment Outcome of High-Grade Parotid Gland Carcinoma

Background

There is emerging evidence that current 8^th edition of the American Joint Committee staging system is not sensitive enough to predict parotid gland carcinoma (PGC) survival outcomes.

Purpose

The present study aimed to analyze pathological nodal factors related to survival and treatment outcomes in a cohort of patients surgically treated with PGC.

Study Design, Setting, Sample

We performed a retrospective cohort study of consecutive patients surgically treated with PGC at the authors' institution from January 1993 to December 2018. The inclusion criteria were as follows: confirmed high-grade parotid gland malignancy on histopathology first surgical treatment of the parotid cancer with neck dissection with curative intent at the study clinic; and sufficient data for review. The exclusion criteria were previous treatment in another institution, low-grade carcinomas, cases where neck dissection was not performed, incurable local disease and distant metastases at the time of first diagnosis, and patients lost to follow-up.

Predictor Variable

Predictor variable comprised pathological nodal factors grouped as the number of cervical node metastases, extranodal extension (ENE), largest diameter of nodal metastasis, and involvement of parotid lymph nodes.

Main Outcome Variables

Outcomes evaluated were overall survival (OS), disease-specific survival (DSS), recurrence-free survival (RFS), locoregional recurrence-free survival, and distant metastasis-free survival (DMFS). OS was calculated from the day of biopsy or definitive surgery to the last known hospital follow-up date or the date of death found in the hospital records or social security data. DSS was calculated from the day of biopsy or definitive surgery until the last known follow-up or death from PGC reported in the patient record. Patients who died from causes other than the disease being studied are not counted in this measurement. Treatment outcome was evaluated through the occurrence of locoregional relapse of the disease or development of distant metastases. RFS was defined as the time from the date of biopsy or definitive surgery to the date of locoregional recurrence free survivalor DMFS reported in the patient record.

Covariates

Covariates were composed of a set of heterogeneous variables grouped into the following categories: demographic, pathologic, and clinical.

Analyses

Unadjusted and adjusted hazard ratios for each variable were calculated with univariate and multivariable Cox regression. The tatistical significance was defined at a P value of < .05.

Results

The cohort of 112 patients included 62 males (55%) and 50 (45%) females. The mean age of the patients was 60.52 ± 15.22 years. The median follow-up time was 59 months (3-221 months). Adenoid cystic carcinoma was the most common tumor type with the incidence of 45%. Cumulative OS for the 5-and 10-year follow-up period was 75 and 61%, respectively. Locoregional recurrences occurred with 27 patients (24%), distant metastases occurred with 25 patients (22%), and both were diagnosed with 5 patients (6%). The number of metastatic nodes was the most important nodal prognostic factor related to OS (P = .02; HR = 2.67; CI = 0.03-6.35), DSS (P = .011; HR = 2.55; CI = .61-6.83), and DMFS (P = .005; HR = 2.85; CI = 0.12-4.76). The presence of pathological parotid nodes was associated with poorer RFS (P = .015; HR = 3.45; CI = 0.25-6.02).

Conclusion and Relevance

The number of metastatic lymph nodes, instead of ENE and largest nodal diameter, was the contributing factor associated with survival and treatment outcomes of surgically treated patients with high-grade PGC. Since the main function of staging system is to predict outcomes, the significance of ENE and nodal dimension in salivary gland cancer staging system requires further clarification. An important finding in the present study was that the presence of positive parotid lymph nodes was associated to locoregional treatment failure.