Open AccessThe prognostic performance of the log odds of positive lymph nodes in patients with esophageal squamous cell carcinoma: A population study of the US SEER database and a Chinese single‐institution cohort
Author(s) -
Zhang Hongdian,
Xiao Wanyi,
Ren Peng,
Zhu Kai,
Jia Ran,
Yang Yueyang,
Gong Lei,
Yu Zhentao,
Tang Peng
Publication year - 2021
Publication title -
cancer medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.403
H-Index - 53
ISSN - 2045-7634
DOI - 10.1002/cam4.4120
Subject(s) - medicine , oncology , cohort , multivariate analysis , ajcc staging system , proportional hazards model , stage (stratigraphy) , lymph node , cancer , staging system , biology , paleontology
Background The purpose of this study was to assess the prognostic performance of the log odds of positive lymph nodes (LODDS) value compared with the pathological N stage and lymph node ratio (LNR) in patients with esophageal squamous cell carcinoma (ESCC). Method In total 1144 patients diagnosed with ESCC from the Surveillance, Epidemiology, and End Results (SEER) database and 930 patients from our validation cohort were eligible. Kaplan–Meier plotter and multivariate Cox proportional hazards models were conducted to investigate the prognostic value of the N stage, LNR stage, and LODDS stage. The homogeneity, discriminatory ability, and monotonicity of these variables were evaluated using the linear trend χ 2 test, likelihood ratio χ 2 test, Akaike information criterion (AIC), and consistency index (C‐index) to determine the potential superiorities. Results The prognostic LODDS cutoff values were determined to be −1.49 and −0.55 ( p < 0.001). Univariate analyses showed significant association among the N, LNR, and LODDS stages and overall survival of the patients (all p < 0.001). Multivariate analyses confirmed that the LODDS stage remained an independent prognostic indicator in both the SEER database and our validation cohort. Subgroup analyses identified the ability of LODDS stage to distinguish heterogeneous patients within various groups in both independent databases. Furthermore, the model with the highest C‐index and smallest AIC value was the one incorporating the LODDS stage among the three investigated nodal classifications of both cohorts. Conclusion The novel LODDS stage demonstrated better prognostic performance than the traditional N or LNR stages in ESCC patients. It can serve as an auxiliary factor to improve prognostic performance and can be applied to evaluate the lymph node status to increase the precision of staging and evaluation of survival.