KEY TAKEAWAYS
- The study aimed to develop a sophisticated competing risk model for predicting specific mortality in patients with DLBCL.
- The study found age, stage, and chemotherapy as key predictors of DLBCL mortality, aiding high-risk patient identification.
Diffuse large B-cell lymphoma (DLBCL) is a common cancer, but current methods may not fully capture all patient risks. Competing risk models offer a promising approach to improve predictions. These models consider not only death from DLBCL but also the possibility of death from other causes. By incorporating this information, doctors could better identify patients with high-risk DLBCL.
Hui Xu and the team aimed to bridge this void by crafting an advanced competing risk model tailored for predicting specific mortality in patients with DLBCL.
The study obtained data on patients with DLBCL from the SEER (Surveillance, Epidemiology, and End Results) database. Relevant variables were identified through a two-step screening process involving univariate and multivariate Fine and Gray regression analyses.
Based on these results, a nomogram was then developed. The model’s performance was evaluated using the consistency index (C-index), and its effectiveness was validated through calibration curves and receiver operator characteristic (ROC) curves.
They enrolled 24,402 patients and identified 13 statistically significant variables for inclusion in the model through feature selection analysis. Model validation revealed an area under the ROC curve AUC values of 0.748, 0.718, and 0.698 for predicting 6-month, 1-year, and 3-year DLBCL-specific mortality in the training cohort. In the validation cohort, AUC values were 0.747, 0.721, and 0.697. Calibration curves demonstrated good consistency between the training and validation cohorts.
The study concluded that the patient’s age is the most significant predictor of DLBCL-specific mortality, followed by the Ann Arbor stage and chemotherapy administration. This predictive model offers the potential to help clinicians identify patients with DLBCL, thereby improving prognosis.
No funding was provided.
Source: https://pubmed.ncbi.nlm.nih.gov/38643217/
Xu H, Yan R, Ye C, et al. (2024) “Specific mortality in patients with diffuse large B-cell lymphoma: a retrospective analysis based on the surveillance, epidemiology, and end results database.” Eur J Med Res. 2024 Apr 20;29(1):241. doi: 10.1186/s40001-024-01833-4. PMID: 38643217; PMCID: PMC11031870.
