Prognostic nomogram for overall survival of elderly esophageal cancer patients receiving neoadjuvant therapy: a population-based analysis.

Background: As the population of elderly patients with esophageal cancer (EC) increases, it becomes more important to understand the prognostic factors. The aim of the present study is to identify prognostic factors among elderly (>60 years) patients with EC receiving neoadjuvant therapy.

Methods: Patients with EC (>60 years) receiving neoadjuvant chemotherapy (nCT) or chemoradiotherapy (nCRT) diagnosed between 2004 and 2015 in the Surveillance, Epidemiology, and End Results (SEER) database were included and divided into a training group and a validation group. Nomograms were constructed based on the Cox proportional risk model. Receiver operating characteristic (ROC) curves, calibration curves, decision curve analysis (DCA), and integrated discrimination improvement (IDI) were used to evaluate the nomogram model. We determined the optimal cutoff value for the scores in terms of overall survival (OS) by X-tile software and divided patients into three different risk groups.

Results: A total of 1,392 patients were included [training group (n=976) and a validation group (n=416)]. Male, T stage, M stage, and N stage were revealed as independent risk factors for poor prognosis (P<0.05). There was no significant difference between nCT and nCRT in prognosis. A novel nomogram model was established based on the above factors. The ROC curve indicated a moderate discriminative power of the nomogram. The DCA demonstrated the clinical value of the nomogram. The nomogram model was superior to the tumor-node-metastasis (TNM) staging system, with an IDI value of 0.006 (P=0.02). Patients classified as low-risk had a better OS, with P values of <0.001 and <0.001 in the validation cohort and training cohort, respectively.

Conclusions: The established nomogram and risk-stratification system were able to improve the precision of prognosis prediction for elderly EC patients.