Development and internal validation of a prognostic nomogram incorporating F-NLR and HAGR scores in patients with oral squamous cell carcinoma: a single-center retrospective cohort study.

Background and objective: Traditionally, the survival prognosis for patients with oral squamous cell carcinoma (OSCC) has predominantly relied on the TNM staging system, which rarely accounts for the biological heterogeneity of individual patients. This study aimed to construct a novel individualized prognostic model for patients with OSCC by integrating traditional clinical parameters with a systemic inflammatory marker fibrinogen-to-neutrophil-lymphocyte ratio (F-NLR) and a nutritional-metabolic indicator the hemoglobin-albumin-globulin ratio (HAGR). Methods: We retrospectively analyzed the clinical data of 292 patients with OSCC who underwent radical surgical resection at a single center. The optimal cut-off values for continuous variables, including F-NLR and HAGR, were determined using the Youden index derived from receiver operating characteristic (ROC) curves. Univariate and multivariate Cox proportional hazards regression analyses were performed to identify independent prognostic factors, which were subsequently used to construct a nomogram predicting 1-, 3-, and 5-year cancer-specific survival (CSS) rates. The model underwent internal validation only; external validation was not performed. Results: During a median follow-up of 41 months, 119 cancer-related deaths were observed. Multivariate analysis identified age, history of precancerous lesions, N classification, postoperative adjuvant therapy, and F-NLR and HAGR scores as independent prognostic factors for OSCC. The nomogram demonstrated a C-index of 0.73, with areas under the curve (AUC) for predicting 1-, 3-, and 5-year CSS of 0.798, 0.754, and 0.836, respectively, indicating acceptable model discrimination. Calibration plots revealed high consistency between the nomogram-predicted probabilities and actual survival observations. Furthermore, DCA suggested a potential net benefit when utilizing this nomogram to guide clinical interventions across a broad range of threshold probabilities. Conclusion: The proposed nomogram, incorporating F-NLR and HAGR scores alongside traditional clinical parameters, demonstrates acceptable predictive accuracy and promising potential for individualized risk stratification in patients with OSCC. However, further external validation in multi-center cohorts is required before it can be routinely recommended for clinical decision-making.