Related Experiment Video
Updated: May 8, 2025
04:57
Establishing a Competing Risk Regression Nomogram Model for Survival Data
Published on: October 23, 2020
10.0K
Factors affecting refractoriness or recurrence in diffuse large B-cell lymphoma: development and validation of a
Yiwei Guo1, Jie Lian2, Yao Chen1
1Hematology Department, Harbin Medical University Cancer Hospital, Harbin, People's Republic of China.
Hematology (Amsterdam, Netherlands)
|December 26, 2024
Summary
A new nomogram accurately predicts refractoriness or recurrence in diffuse large B-cell lymphoma (DLBCL) patients. This tool aids in identifying high-risk individuals for better treatment strategies.
Area of Science:
- Hematology
- Oncology
- Medical Statistics
Background:
- Relapsed/Refractory (R/R) diffuse large B-cell lymphoma (DLBCL) has a poor prognosis.
- Current models for predicting R/R DLBCL are insufficient.
- Identifying patients at high risk for refractoriness/recurrence is crucial.
Purpose of the Study:
- To develop a novel nomogram for predicting refractoriness/recurrence in DLBCL.
- To identify key determinants for inclusion in the predictive model.
- To enhance the identification of DLBCL patients at elevated risk.
Main Methods:
- Retrospective study of 293 newly-diagnosed DLBCL patients.
- Patients divided into training (n=206) and validation (n=87) cohorts.
- Univariable analysis, random forest, and binary logistic regression used to develop the nomogram.
- Nomogram evaluated using ROC curves, calibration curves, and DCA.
Main Results:
- Pathological subtype, LDH, treatment, and ALC identified as key predictors.
- The developed nomogram demonstrated excellent calibration and accuracy.
- Nomogram outperformed traditional indexes (IPI, R-IPI, NCCN-IPI) in predictive ability.
Conclusions:
- The novel nomogram is a valuable tool for predicting DLBCL refractoriness/recurrence.
- It aids in risk stratification for personalized treatment approaches.
- The nomogram offers superior clinical utility compared to existing prognostic indices.
