Oops, you're using an old version of your browser so some of the features on this page may not be displaying properly.

MINIMAL Requirements: Google Chrome 24+Mozilla Firefox 20+Internet Explorer 11Opera 15–18Apple Safari 7SeaMonkey 2.15-2.23

Poster Display session 2

1843 - Multicenter Validation of the Postoperative Carcinoembryonic Antigen Combined Prognostic Model for Stage Ⅲ Colon Cancer


29 Sep 2019


Poster Display session 2


Tumour Site

Colon and Rectal Cancer


Ji Zhu


Annals of Oncology (2019) 30 (suppl_5): v198-v252. 10.1093/annonc/mdz246


J. Zhu1, Y. Liu2, R. Zhang3, J. Fan1

Author affiliations

  • 1 Department Of Radiation Oncology, Fudan Univerisity Shanghai Cancer Center, 200032 - Shanghai/CN
  • 2 Department Of Colorectal Surgery, Harbin medical university cancer hospital, Harbin/CN
  • 3 Department Of Colorectal surgery, Liaoning Cancer Hospital & Institute, Shenyang/CN


Login to access the resources on OncologyPRO.

If you do not have an ESMO account, please create one for free.

Abstract 1843


There are limitations in the prognosis stratification of stage III colon cancer patients using only the T and N stages, while other factors are unsure to be equally important. This study aimed to prove the prognostic importance of postoperative carcinoembryonic antigen (CEA) and stratify patients with stage III colon cancer more accurately by building a prognostic model combining CEA.


This was a retrospective cohort study. Patients with colon cancer who had CEA recorded within 100 days after resection at the centers from 2006 to 2017 were identified. The data included three different cancer centers in China: one training set and two validation sets, which were collected and analyzed in 2018. Postoperative CEA, T stage and N stage were combined to build the prognostic model, and patients with stage III colon cancer were divided into the high-risk group and the low-risk group. Harrell’s C-statistics, and net reclassification improvement (NRI) were used to assess the model performance. The disease-free survival (DFS) was compared between the high-risk group and the low-risk groups by the log-rank test.


Among patients with elevated postoperative CEA, we found that CEA tested within 20 days (early test group) after resection was not reliable, since DFS of the early test group was higher than that of the delayed test group (p = 0.006). So, 834 patients with stage III disease which had CEA data for 21-100 days after resection were analyzed. In the training set, C-index for the model (combining T stage, N stage and postoperative CEA) was 0.74. Patients with elevated postoperative CEA, T4b, N2b or T4aN2a were classified as the high-risk group, with 3-year DFS of 55.8%, which was lower than the value of 86.0% in the low-risk group [hazard ratio (HR), 4.03; 95% CI, 2.49-6.54; p < 0.001]. When compared to the TNM model, this new model achieved a 10% NRI. These results were confirmed in the validation sets.


Patients with stage III colon cancer could be stratified much more accurately with the prognostic model combining T stage, N stage and postoperative CEA.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.


Harbin Medical University Cancer Hospital Preeminence Youth Fund (JCQN2019-04).


All authors have declared no conflicts of interest.

This site uses cookies. Some of these cookies are essential, while others help us improve your experience by providing insights into how the site is being used.

For more detailed information on the cookies we use, please check our Privacy Policy.

Customise settings
  • Necessary cookies enable core functionality. The website cannot function properly without these cookies, and you can only disable them by changing your browser preferences.