LODDS is superior to lymph node ratio for the prognosis of node-positive rectal cancer patients treated with preoperative radiotherapy

Purpose: Yielding pathologic-lymph node ratio (yp-LNR) was considered to be a better staging system than yp-N stage in rectal cancer patients treated with preoperative radiotherapy (pre-RT). We aimed to compare the predictive ability of yielding pathologic log odds of positive lymph nodes (yp-LODDS) with that of yp-LNR for cancer-specific survival (CSS) in stage III rectal cancer patients treated with pre-RT. Methods: We analyzed stage III rectal cancer patients treated with pre-RT in the Surveillance, Epidemiology and End Results (SEER) database. Patients were classified into 4 groups, yp-LNR1 to 4, based on the LNR cutoff points 0.25, 0.50, and 0.75. Subjects were categorized into 5 groups, yp-LODDS1 to yp-LODDS5, based on the LODDS cutoff points -1, 0, 1, and 2. Univariate and multivariate Cox proportional hazards models were performed to analyze the risk factors for survival outcome. Results: A total of 4,612 patients were included from the SEER database. Patients in the yp-LNR4 group could be further divided into yp-LODDS4 and yp-LODDS5 groups with 5-year CSS of 47.6% and 31.5%, respectively (p<0.001). In the multivariate analysis without yp-LODDS, yp-LNR was an independent prognostic factor (hazard ratio [HR] 2.006, 95% confidence interval [CI] 1.619-2.484, p<0.001). However, after adjusting for yp-LODDS, yp-LNR was no longer associated with CSS (p = 0.393), and yp-LODDS was identified as an independent prognostic factor (HR 1.274, 95% Cl 1.069-1.520, p = 0.007). Conclusions: The prognostic value of yp-LNR can be confounded by yp-LODDS. In stage III rectal cancer patients treated with pre-RT, yp-LODDS has superior discrimination power over yp-LNR and can more accurately evaluate CSS.