Abstract
Background: Tonsillar squamous cell carcinoma (TSCC) is characterized by a high tendency to metastasize to lymph nodes, significantly impacting the treatment modality and recurrence rates in head and neck cancer patients. Therefore, the development of accurate predictive models, such as nomograms, is imperative for the early identification of risk factors associated with lymph node involvement. Various lymph node classification systems, including the number of positive lymph nodes (NPLNs), the ratio of positive lymph nodes (pLNRs), and the logarithm of the odds of positive lymph nodes (LODDS), have been proposed to provide prognostic information. However, the optimal system for classifying lymph nodes remains uncertain, necessitating further investigation to determine which system offers the most accurate prediction of patient outcomes. Thus, our objective was to identify the most effective prognostic nomogram for predicting outcomes in TSCC patients. Material and Methods: In this study, we retrospectively analyzed data from 1,775 TSCC patients extracted from the Surveillance, Epidemiology, and End Results (SEER) database, following predefined criteria for inclusion. We evaluated the performance of prognostic models using Harrell's concordance index (C-index) and Akaike information criterion (AIC). Subsequently, variables were utilized to construct nomograms for predicting cancer-specific survival and overall survival. Nomograms' predictive capabilities were assessed using Integrated Discrimination Improvement (IDI) and Net Reclassification Improvement (NRI). Results: The nomogram comprising pLNR, LODDS, and NPLN showed superior efficacy in predicting the survival outcome of patients with laryngectomy for TSCC. Conclusion: The nomograms developed in this study have the potential to serve as valuable tools for forecasting patient survival following surgical interventions for TSCC.