Construction and verification of prognostic nomogram for early-onset esophageal cancer

• DOI: 10.17305/bjbms.2021.5533
• Publication Date: 2021-03-23T16:23:19Z
• Authors: Xiaoxiao Liu; Wei Guo; Xiaobo Shi; Yue Ke; Yuxing Li; Shupei Pan; Yingying Jin; Yuchen Wang; Qinli Ruan; Hongbing Ma
• Keywords: Nomogram; Univariate; T-stage; Concordance; Univariate analysis

This study aimed to build up nomogram models evaluate overall survival (OS) and cancer-specific (CSS) in early-onset esophageal cancer (EOEC). Patients diagnosed with (EC) from 2004 2015 were extracted the Surveillance Epidemiology End Results (SEER) database. Clinicopathological characteristics of younger versus older patients compared, analysis was performed both groups. Independent related factors influencing prognosis EOEC identified by univariate multivariate Cox analysis, which incorporated construct a nomogram. The predictive capability estimated concordance index (C-index), calibration plot, receiver operating characteristic (ROC) curve, decision curve (DCA). A total 534 17,243 available SEER Younger randomly segmented into training set (n=266) validation (n=268). In terms set, C-index OS 0.740 (95% CI: 0.707-0.773), that CSS 0.752 0.719-0.785). view 0.706 0.671-0.741) 0.723 (95%CI: 0.690-0.756), respectively. Calibration curves demonstrated consistent degree fit between actual predicted values models. From perspective DCA, more beneficial than tumor-node-metastasis (TNM) stage for EOEC. brief, model can be considered as an individualized quantitative tool predict assist clinicians making treatment decisions.