# Evaluating the predictive value of log odds of positive lymph nodes on postoperative survival in patients with laryngeal cancer: a SEER population-based study

## TL;DR

This study shows that using the log odds of positive lymph nodes improves survival predictions for laryngeal cancer patients compared to traditional staging methods.

## Contribution

The study introduces a multivariate model combining LODDS and N staging for better postoperative prognosis prediction in laryngeal cancer.

## Key findings

- LODDS, age, and other factors were consistent independent prognostic indicators for survival.
- The multivariate model outperformed individual N staging and LODDS models in predictive accuracy.
- Combining LODDS with N staging improves postoperative prognosis prediction for laryngeal cancer patients.

## Abstract

Traditionally, the AJCC TNM staging system has been the primary tool for assessing the severity and prognosis of laryngeal cancer. Although several studies have demonstrated that the log odds of positive lymph nodes (LODDS) offers superior predictive accuracy compared to the TNM staging for other cancers, there is limited research for laryngeal cancer. This study analyzed data from SEER database (2000–2019). Independent risk factors for survival were identified using univariate and multivariate Cox regression analyses, and different prognostic models were constructed based on the multivariate analysis results. The predictive performance of these models was evaluated using receiver operating characteristic (ROC) curves and area under the curve (AUC) values. The results indicated that LODDS subgroup, age, marital status, histologic grade, T-stage, and N-stage were consistent independent prognostic factors for overall survival (OS) and cancer-specific survival (CSS). Assessment metrics showed that the multivariate model, which incorporated both LODDS and N staging, outperformed the individual N staging and LODDS models in predicting postoperative prognosis in laryngeal cancer patients. Overall, the multivariate model constructed in this study is a superior tool for predicting the postoperative status of laryngeal cancer.