Development and Validation of a Novel Prognostic Nomogram Model and a Web-Based Survival Rate Calculator for Diffuse Large B-Cell Lymphoma in the Rituximab Era
Qiaofeng Zhong,Haizhu Chen,Yan Qin,Xiaohui He,Sheng Yang,Shengyu Zhou,Peng Liu,Yuankai Shi,Hongmei Jing
DOI: https://doi.org/10.1182/blood-2021-154210
IF: 20.3
2021-11-05
Blood
Abstract:Abstract Objective This study aimed to develop a novel prognostic nomogram and a web-based survival rate calculator to improve discriminative and predictive accuracy for diffuse large B cell lymphoma (DLBCL) in rituximab era. Methods We retrospectively collected data about 906 DLBCL patients who received R-CHOP or R-CHOP-like chemotherapy from January 2005 to December 2018 in National Cancer Center / Cancer Hospital, Chinese Academy of Medical Sciences. COX proportional hazard regression analysis was performed to assess the independent prognostic significance of clinical and histopathological variables on overall survival, which were used to develop the nomogram. We performed internal and external validation of the nomogram, and compared it's discriminative and predictive accuracy with the traditional IPI, R-IPI, and NCCN-IPI. ResultsThe patients were divided into the training (n = 636) and validation (n = 270) cohorts at a ratio of 7:3. We identified six independent predictors for survival including age, ECOG, number of extranodal sites, bulky, and β2-MG, which were used to construct the nomogram and the web-based survival rate calculator. The C-index of the nomogram was 0.767 (95% CI, 0.735-0.799) in the training cohort and 0.732 (95% CI, 0.678-0.786) in the validation cohort. The AUC values of the nomogram for predicting the 1-, 5-, and 10-year OS rates for DLBCL were 0.794, 0.799, 0.805 in the training cohort and 0.828, 0.773, 0.771 in the validation cohort. All calibration curves revealed optimal consistency between predicted and actual survival both in the training and validation cohort. The nomogram divided DLBCL patients into three risk groups, the score of low risk, moderate risk, and high risk were 0-45,46-228,and higher than 228, respectively. The risk stratification model generated based on the nomogram showed a favorable level of predictive accuracy compared with the IPI, R-IPI, and NCCN-IPI in both cohorts according to the K-M survival curves. The AUC values of the nomogram model compared with the IPI, R-IPI, and NCCN-IPI for predicting the 10-year OS rates for DLBCL patients were 0.756 vs 0.748, 0.732, 0.713 in the training cohort and 0.742 vs 0.712, 0.713, 0.757 in the validation cohort. ConclusionsIn conclusion, we have established and validated a novel prognostic nomogram model and a web-based survival rate calculator, which revealed more discriminative and predictive accuracy than IPI, R-IPI, and NCCN-IPI in the rituximab era. Disclosures No relevant conflicts of interest to declare.
hematology