Introduction

Low-grade B-cell lymphomas (LGBCL) account for about 40% of non-Hodgkin lymphomas, characterized by slow progression. In South Korea, marginal zone lymphoma (17.98%), follicular lymphoma (4.9%), and mantle cell lymphoma (1.6%) predominate. While anti-CD20 monoclonal antibodies have improved progression-free survival, non-lymphoma-related mortality (NLM) remains a challenge. Current prognostic indices focus on disease-specific factors, potentially overlooking non-lymphoma-related factors. This study aims to analyze prognostic factors considering both lymphoma-related and non-lymphoma-related causes of death, focusing on secondary malignancies (SM), Richter's syndrome (RS), and COVID-19, using competing risk analysis and machine learning algorithms.

Materials and Methods

This multicenter retrospective study analyzed 1,047 adult LGBCL patients diagnosed between January 2011 and December 2022 across seven South Korean hospitals. Patients with follicular lymphoma (n=689), marginal zone lymphoma (n=312), and mantle cell lymphoma (n=46) were included. Data collection encompassed demographic, clinical, and pathological information based on established prognostic indices. Statistical analysis included Kaplan-Meier survival curves, Cox proportional hazard models, competing risk analysis, Aalen's additive regression model, and Random Survival Forests. The study assessed overall survival (OS), cumulative incidence of progression-related death (CIP), and non-lymphoma-related mortality (NLM).

Results

The median follow-up was 4 years. The 4-year OS rates were 95.4% for follicular lymphoma, 93.1% for marginal zone lymphoma, and 62% for mantle cell lymphoma. In the SM (3.8% of patients) and RS (2.6% of patients) groups, the 4-year OS rates were 77.9% and 87.4%, respectively.

Multivariate analysis identified SM, age over 60 years, male sex, pleural effusion, and elevated LDH levels as factors associated with poorer OS. Competing risk analysis revealed that these factors, except male sex, were also associated with worse NLM. Age over 60 years, mantle cell lymphoma subtype, and anemia were linked to poorer outcomes in CIP.

In the SM group, NLM was more substantial than CIP. The incidence of NLM at 2, 4, and 8 years was 10%, 18.9%, and 26.7%, respectively, while CIP was 0%, 3.4%, and 11.2% at the corresponding time points.

Aalen's additive regression model identified SM, age over 60 years, male sex, and mantle cell lymphoma subtype as factors consistently affecting mortality over time. The random forest survival model demonstrated high concordance (c-index for NLM: 0.79, CIP: 0.77) and identified age over 60 years, SM, pleural effusion, and COVID-19 as key variables for NLM.

Conclusion

This comprehensive analysis reveals the significant impact of secondary malignancies on non-lymphoma-related mortality in LGBCL patients. The study highlights the need for a nuanced approach to risk assessment, considering both lymphoma-specific and non-lymphoma-related factors. The findings suggest that improving survival outcomes in LGBCL patients requires not only effective lymphoma-targeted therapies but also strategies to prevent and manage secondary malignancies and infections. The study emphasizes the importance of early detection of secondary malignancies and vigilant infection risk management. Future prospective studies are needed to confirm these findings and evaluate the long-term impact of emerging therapies on survival outcomes in LGBCL patients.

Disclosures

No relevant conflicts of interest to declare.

This content is only available as a PDF.
2024
Sign in via your Institution