Correlation of preoperative radiological lymph node assessment with pathological nodal and tumor findings in tongue and oral floor squamous cell carcinoma

Summary Purpose Preoperative imaging is central to cervical lymph node staging in oral squamous cell carcinoma (OSCC). However, the pathological and biological determinants underlying radiological lymph node suspicion remain poorly defined. Methods This retrospective cohort study included 438 patients with OSCC of the tongue or floor of mouth treated with curative surgery and neck dissection between 2011 and 2021. Radiological lymph node assessment was classified using a three-level scale (unremarkable, suspicious, metastatic) and a dichotomous scale (unremarkable vs. suspicious/metastatic). Associations with patient characteristics, primary tumor features, and pathological nodal parameters were analyzed using univariable tests and multivariable ordinal logistic regression. A secondary model was performed in patients with available depth of invasion (DOI). Results Radiological lymph node assessment was strongly associated with pathological indicators of nodal tumor burden, including number of positive lymph nodes, lymph node ratio (LNR), extranodal extension (ENE), multilevel nodal involvement, and pathological N stage (all p < 0.001). In multivariate analyses, LNR emerged as the strongest independent predictor of radiological lymph node category, with increasing LNR associated with higher odds of radiological suspicion. Extranodal extension remained independently associated with radiological lymph node status. Depth of invasion showed a positive association that was attenuated after adjustment for nodal tumor burden, suggesting partial mediation. Conclusion Radiological lymph node suspicion in OSCC primarily reflects cumulative nodal tumor burden and aggressive pathological features rather than isolated nodal involvement or patient-related factors. Radiological lymph node assessment may therefore serve as a surrogate marker of biologically aggressive nodal disease in clinical decision-making.