(A589) Oral Cavity Cancer Depth of Invasion and Rate of Lymph Node Metastasis: A Systematic Review

Introduction: Despite the importance of depth of invasion (DOI) as a prognostic indicator for lymph node metastasis (LNM) in oral squamous cell carcinoma (OSCC), its use in clinical management is discrepant due to inconsistencies in definitions, cut-off values for different treatment methods, and varying use of subsite classification. The aim of this study was to systematically review the relationship between DOI as defined in the 8th edition AJCC guidelines and rate of LNM in OSCC.

Methods: The Preferred Reporting Systems for Systematic Reviews and Meta-Analysis (PRISMA) guidelines were followed for this systematic review. A literature search of DOI and LNM was performed by searching PubMed, SCOPUS, Web of Science, Science Direct, and Embase. Articles presenting data associating DOI per millimeter (mm) with the rate of LNM were included. Review articles, animal studies, non-English-language articles, and abstracts were excluded. The population included adult patients with histologically confirmed OSCC. The primary outcome was reported pathological DOI following surgical excision and rate of LNM on neck dissection. Two reviewers independently performed the review, data extraction for analysis, and a quality assessment. Meta-analysis was performed using a random effects model.

Results: 3,958 articles were identified and 10 studies met inclusion criteria. All studies were retrospective. Sample sizes ranged from 60 to 623 patients. Yaers of publication ranged from 2012 to 2023. All oral cavity subsites were represented, with the tongue accounting for most patients (59.5%). With a high level of heterogeneity present among studies, the random effects model was appropriate. Random effects meta-analysis resulted in estimates for LNM rate at each level of DOI (per mm) as follows: 14.9% at 2mm [95% confidence interval (CI): 10.1%-19.7%], 17.3% at 3mm [95% CI: 12.2%-22.4%], and 31.1% at 4mm [95% CI: 24.5%-37.6%].

Conclusions: DOI is an important predictor of LNM. Based on the results of this study, elective neck dissection could be considered as low as 2mm, as that is the lowest DOI at which there is approximately a 15% prevalence of LNM.