The novel prognosis prediction model based on preoperative serum tumor markers in gastric cancer:A multicenter,retrospective study
WANG Lin-jun
SHEN Yi-kai
YANG Kun
ZHANG Zi-zhen
LIN Jian-xian
YAN Chao
CHEN Hao
LIN Jie
HUANG Hong-xin
ZHU Cheng-jun
XIA Yi-wen
JIANG Tian-lu
SHEN Xu-sheng
LI Qing-ya
LIU Hong-da
ZHANG Dian-cai
XU Hao
YANG Li
LI Guo-xin
ZHU Zheng-gang
HUANG Chang-ming
CAO Hui
HU Jian-kun
XU Ze-kuan
Abstract：Objective To explore a novel prediction model based on the combined use of multiple common tumor markers to improve the accuracy of prognostic prediction for gastric cancer(GC).Methods This multi-center retrospective analysis collected 5119 GC patients treated with radical resection from January 2015 to December 2022 in 6 hospitals.Based on the inclusion and exclusion criteria,2947 patients were finally enrolled.Basic information about patients,including sex,age,tumor location,tumor size,differentiation,vascular invasion,neural invasion,depth of invasion(T),lymph node metastasis(N),pathological stage and preoperative serum tumor markers were assessed for analysis.The calibration curves,consistency index(C-index)and decision curve analysis(DCA)were employed to evaluate the precision and credibility of nomogram.Results R software randomly selected 4/5 patients(2357/2947)to be included in the training cohort and the remaining 1/5(590/2947)for the validation cohort.CEA,CA72-4,CA19-9 and CA125 were incorporated into risk-score.Patients were divided into high or low risk-score group based on the optimal cut-off value.Age,tumor location,T,N,and risk-score group were enrolled in the nomogram.The C-index of nomogram was 0.782(95％CI:0.744-0.820)in training cohort and 0.802(95％CI:0.733-0.871)in the validation cohort.The DCA curves showed the proposed predictive model had good clinical applicability.Calibration curves indicated the OS values predicted by nomogram fit well with the actual OS observed in the patients.Conclusions The novel prognostic prediction model based on preoperative serum tumor markers has potential clinical application value in predicting OS probability of gastric cancer patients undergoing radical surgery.Thus,it can guide the follow-up treatment of GC patients after surgical intervention.
Keywords：gastric cancerserum tumor markersLASSO-Cox regressionprognostic modelnomogram
Publication Date：2025-01-31
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：14( 95-108 )