Development and validation of a nomogram for predicting overall survival of node-negative ampullary carcinoma

J Surg Oncol. 2020 Mar;121(3):518-523. doi: 10.1002/jso.25816. Epub 2019 Dec 26.

Abstract

Background: The accuracy of the current staging system for predicting the overall survival (OS) of patients with ampullary carcinoma (AC) is still unsatisfactory, especially in node-negative (N0) patients. We aimed at establishing a nomogram to accurately predict OS in N0 AC.

Methods: This study enrolled 697 N0 AC patients from the Surveillance, Epidemiology, and End Results database (design cohort [DC], n = 697) and the First Affiliated Hospital of Sun Yat-sen University (validation cohort [VC], n = 112), who underwent surgical resection. The nomogram was established by using prognostic factors determined by univariate and multivariate regression analyses.

Results: The nomogram for OS was developed by using four independent prognostic factors, including age, grade, T stage, and a number of examined lymph nodes. The C-index of a nomogram for OS in DC and VC was 0.665 and 0.731, respectively. Calibration curves showed good consistency of the nomogram. The nomogram had a better accuracy in predicting OS compared with conventional staging system (P < .05). On the basis of nomogram-predicted scores, the patients were stratified into groups with different risk. The OS of low-risk patients was significantly longer than high-risk ones (P ≤ .010).

Conclusions: The nomogram could be used to predict the OS of N0 AC. It could help guide further treatment in clinical practice.

Keywords: ampullary carcinoma; lymph node; nomogram; prediction; prognosis.

MeSH terms

  • Aged
  • Ampulla of Vater / pathology*
  • China / epidemiology
  • Common Bile Duct Neoplasms / mortality*
  • Common Bile Duct Neoplasms / pathology
  • Common Bile Duct Neoplasms / surgery*
  • Female
  • Humans
  • Kaplan-Meier Estimate
  • Lymph Nodes / pathology
  • Male
  • Middle Aged
  • Neoplasm Staging
  • Nomograms*
  • Prognosis
  • Reproducibility of Results
  • Survival Analysis
  • United States / epidemiology