The association between preoperative serum transferrin and occult metastasis in early T classification oral squamous cell carcinoma

BACKGROUND: This study aimed to investigate the potential of preoperative serum transferrin levels as a biomarker for predicting occult metastasis in patients with clinically node-negative (cN0), early T-classification (T1-T2) oral squamous cell carcinoma (OSCC). METHODS: This retrospective cohort study included 386 eligible patients with cN0 early-stage OSCC who underwent primary tumor resection and elective neck dissection at the Second Xiangya Hospital, Changsha, China. Preoperative serum transferrin levels were measured, and their association with occult metastasis (determined by postoperative pathology) was analyzed. Statistical analyses included univariate and multivariable logistic regression, restricted cubic spline (RCS) analysis for nonlinear relationships, subgroup analyses, and predictive performance metrics (AUC, calibration, decision curve analysis). RESULTS: Occult metastasis was identified in 104 patients (26.9%). Serum transferrin levels did not differ significantly between patients with and without metastasis (2.16 ± 0.37 vs. 2.14 ± 0.35 g/L, P = 0.693). On multivariable analysis, transferrin was not associated with occult metastasis (P > 0.05), whereas T classification and poorly differentiated histology emerged as significant predictors. RCS analysis confirmed a linear, non-significant relationship. None of the 27 subgroups showed significant differences. Adding transferrin to the baseline model did not improve AUC (unchanged at 0.637), net benefit, or calibration. CONCLUSIONS: Preoperative serum transferrin level was not associated with occult metastasis in early-stage, cN0 OSCC. Tumor-related factors, specifically differentiation grade and T classification, remained the primary predictors of metastatic risk. Serum transferrin does not appear to be a useful independent biomarker for preoperative metastasis risk stratification in this patient population.