Comparison of LNR- and LODDS-based predictive models for prognosis in non-elderly patients with locally advanced rectal cancer undergoing neoadjuvant therapy.

purposeThe purpose of this study was to comprehensively assess the efficacy of lymph node ratio (LNR) and log odds of positive lymph nodes (LODDS) in predicting survival outcomes in non-elderly locally advanced rectal cancer (LARC) patients treated with neoadjuvant chemoradiotherapy (NCRT).

methodsThe 1643 non-elderly LARC patients undergoing NCRT between 2010 and 2015 were extracted from the Surveillance, Epidemiology, and End Results (SEER) database and were randomly assigned at a ratio of 7:3. We used Cox regression models to identify independent prognostic factors, then constructed nomogram models to predict cancer-specific survival (CSS) and overall survival (OS). The relative weight in nomogram models, receiver operating characteristic (ROC), area under the curve (AUC), concordance index (C-index), calibration curve, and decision curve analysis (DCA) were performed to evaluate and compare the predictive performance between LNR and LODDS.

resultsThe distribution of LNR and LODDS showed that LODDS exhibited a more detailed stratification when LNR was equal to the extreme value. In predicting CSS, the LNR (hazard ratio (HR), 1.987; 95% confidence interval (CI), 1.375-2.872) and LODDS (HR, 1.568; 95% CI, 1.154-2.131) were independent risk factors in corresponding nomogram models. Regarding OS, the LODDS (HR, 1.387; 95% CI, 1.060-1.816) showed an independent predictive value. All evaluation methods confirmed the reliability of both models. Although the LNR-based model showed better performance for short-term CSS and the LODDS-based model demonstrated slightly better prediction for long-term CSS and OS, these two models showed largely comparable predictive ability.

conclusionsThe LNR and LODDS can provide complementary prognostic value in survival prediction, offering clinicians a comprehensive basis for clinical decision-making.