The National Comprehensive Cancer Network guidelines recommend END or sentinel lymph node biopsy for neck management in early tongue cancer cases when the depth of invasion (DOI) of the primary tumor is greater than 3 mm, given the risk of OLNM [4]. Although DOI and the worst pattern of invasion are currently used as predictive models for OLNM in oral cancer, applying those criteria to actual preoperative evaluations is problematic because they are based on postoperative pathological results. Preoperative DOI measurements using imaging tests such as magnetic resonance imaging or intraoperative ultrasonography have been reported to have relevance to the postoperative pathological results; however, there are practical limitations in accurately assessing the DOI of small lesions using radiological methods [5]. A recently introduced deep learning-based modeling method can increase the prediction accuracy of OLNM, but it is difficult to apply this model in the real clinical field considering time and cost [6]. Therefore, in addition to the above indicators, it is necessary to discover new markers for preoperative OLNM prediction.

Lee et al. [7] recently reported the significance of defensin-associated genes as predictors of OLNM in early tongue cancer. The authors performed RNA sequencing on primary tumor samples from 35 tongue cancer patients and analyzed differentially expressed genes (DEGs) according to OLNM status in samples after END. Their analyses confirmed that lower expression of defensin-associated genes (DEFB4A, DEFB103B, and DEFB4B) in the primary tumor could be associated with an increased risk of OLNM in early tongue cancer, suggesting that those genes could be used as biomarkers [7]. Another study published by the same author found that the expression of actin-associated genes such as ACTA1 had a significant relationship with regional metastasis in early tongue cancer [8]. Yang et al. [9] reported that the expression of TNFRSF10C, a gene involved in immune-related pathways, had a significant relationship with OLNM and DFS, and that the log2 ratio of copy number variation of the gene could suggest a favorable prognosis in patients with node-negative early tongue cancer.

In early tongue cancer, preoperative biomarker-based OLNM prediction for the primary tumor may provide more precise information regarding neck management. Nonetheless, it is questionable whether specific genetic information can be accurately confirmed from a small specimen obtained from a punch biopsy before surgery, and another limitation is that an accurate cut-off value for whether to perform END based on the level of marker expression has not yet been determined. Consequently, it is thought that the practical application of genetic biomarkers in the clinical setting will be possible in the future when technological advances are made to reduce the time and cost required to confirm DEG results before surgery in early tongue cancer patients.