Is Perineural Invasion a Reasonable Indicator for Neck Dissection in cT1N0M0 Squamous Cell Carcinoma of the Oral Cavity?

ABSTRACT

PURPOSE: Perineural invasion (PNI) is considered an adverse histological feature in oral squamous cell carcinoma (OSCC). Controversy exists regarding elective neck dissection (END) in cT1N0M0 OSCC with PNI as the only risk factor. The purpose of this study was to evaluate PNI as an indicator for END, as well as its utility when combined with the depth of invasion (DOI) as a second indicator. PATIENTS AND METHODS: cT1N0M0 (AJCC8) OSCC patients treated from August 1998 to July 2017 in the Department of Oral and Maxillofacial Surgery, University of Michigan (MI), and Beijing Stomatological Hospital, Capital Medical University (BSH) were reviewed. Data from these sites included both prospectively captured data housed in a database and retrospective data. RESULTS: 283 cT1N0M0 OSCC patients were analyzed. The tongue was the most common subsite (56.2%). Ninety-nine (99) patients received END and 184 neck observation. PNI was found in only 8 patients (2.83%) all in the tongue or inferior gingiva. END was performed in 7 of the PNI patients. The mean depth of invasion for tumors with PNI was 3.97 mm, compared to 2.54 mm in tumors without PNI. PNI was statistically correlated with nodal disease (pN+) and extranodal extension (ENE+). After using DOI as a primary indicator for END, no additional PNI patients benefited from END. In addition, only 2 patients had PNI identified preoperatively both with 5 mm DOI, and therefore, already indicated for END on the basis of DOI status. CONCLUSIONS: PNI status is statistically correlated with pN+ and ENE + pathology in cT1N0M0 OSCC. After using DOI as a primary indicator for END, PNI status had no added value in decision making. Even surgeons who do not use DOI will find very limited value in PNI status with less than 1% preoperative prevalence (0.71%) in this patient population.