Related Experiment Video
Updated: Jul 25, 2025
Establishing a Competing Risk Regression Nomogram Model for Survival Data
Published on: October 23, 2020
[Establishment and Clinical Significance of Prognostic Nomogram Model for Diffuse Large B-Cell Lymphoma Based on
Ya-Wen Xu1, Yu-Lan Zhou2, Fan-Cong Kong3
1Center of Hematology, The First Affiliated Hospital of Nanchang University; Institute of Hematology, Academy of Clinical Medicine of Jiangxi Province; Center of Clinical Trials and Research, Jiangxi Provincial People's Hospital, Nanchang 330006, Jiangxi Province, China.
Objective:
To retrospectively analyze clinical characteristics and survival time of patients with diffuse large B-cell lymphoma (DLBCL), detect prognosis-related markers, and establish a nomogram prognostic model of clinical factors combined with biomarkers.
Methods:
One hundred and thirty-seven patients with DLBCL were included in this study from January 2014 to March 2019 in the First Affiliated Hospital of Nanchang University. The expression of GCET1, LMO2, BCL-6, BCL-2 and MYC protein were detected by immunohistochemistry (IHC), then the influences of these proteins on the survival and prognosis of the patients were analyzed. Univariate and multivariate Cox regression analysis were used to gradually screen the prognostic factors in nomogram model. Finally, nomogram model was established according to the result of multivariate analysis.
Results:
The positive expression of GCET1 protein was more common in patients with Ann Arbor staging I/II (P =0.011). Compared with negative patients, patients with positive expression of LMO2 protein did not often show B symptoms (P =0.042), and could achieve better short-term curative effect (P =0.005). The overall survival (OS) time of patients with positive expression of LMO2 protein was significantly longer than those with negative expression of LMO2 protein (P =0.018), though the expression of LMO2 protein did not correlate with progression-free survival (PFS) (P >0.05). However, the expression of GCET1 protein had no significant correlation with OS and PFS. Multivariate Cox regression analysis showed that nomogram model consisted of 5 prognostic factors, including international prognostic index (IPI), LMO2 protein, BCL-2 protein, MYC protein and rituximab. The C-index applied to the nomogram model for predicting 4-year OS rate was 0.847. Moreover, the calibrated curve of 4-year OS showed that nomogram prediction had good agreement with actual prognosis.
Conclusion:
The nomogram model incorporating clinical characteristics and IHC biomarkers has good discrimination and calibration, which provides a useful tool for the risk stratification of DLBCL.
