1090 - BALT-NRI: A Novel Prognostic Model for Bronchus-Associated Lymphoid Tissue Lymphoma with Disease-Specific Survival Implications

05:35pm - 05:40pm ET

Room 162

Presenter(s)

Yue Liu, MD - National Cancer Center CAMS & PUMC, Beijing, Beijing

Y. Wang¹, X. Zhang², Y. Wu¹, L. Xin¹, X. Feng³, R. Zheng⁴, Y. W. Song¹, C. Xia⁵, S. Wang¹, Y. X. Li^1,6, S. Qi¹, and Y. Liu¹; ¹Department of Radiation Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China, ²Department of Radiotherapy, Tianjin Medical University Cancer Institute and Hospital, Tianjin, China, ³Department of Pathology, National Cancer Center/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China, ⁴Department of Nuclear Medicine, National Cancer Center/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China, ⁵Office of Cancer Screening, Chinese Academy of Medical Sciences Cancer Institute and Hospital: Cancer Hospital Chinese Academy of Medical Sciences, Beijing, China, ⁶The Affiliated Cancer Hospital of Zhengzhou University, Henan Cancer Hospital, Cancer Hospital & Henan Hospital, Chinese Academy of Medical Sciences, Zhengzhou, China

Purpose/Objective(s):

Mucosa-associated lymphoid tissue (MALT) lymphoma of bronchus-associated lymphoid tissue (BALT) is an indolent non-Hodgkin disease with heterogeneous outcomes. Existing prognostic models derived from MALT lymphoma with diverse primary sites, including MALT-IPI and revised MALT-IPI, may not be optimally applicable to this specific patient population. We aimed to develop a novel model tailored for BALT lymphoma, enabling refined risk stratification and clinical decision-making.

Materials/Methods:

In this multicenter retrospective cohort study, patients with primary BALT lymphoma were enrolled and assigned to the discovery or validation cohorts. Clinicopathologic features, first-line management, and survival outcomes were analyzed. A BALT lymphoma-specific nomogram risk index (NRI) was constructed and validated.

Results:

A total of 552 patients were analyzed. First-line management included surgery (54.3%, n=300), radiotherapy (6.9%, n=38), active surveillance (17.2%, n=95), and systemic therapy (21.6%, n=119). Ten-year progression-free survival (PFS), overall survival (OS), and cancer-specific survival (CSS) were 48.4%, 85.5%, and 93.8%, respectively. Patients with surgery or radiotherapy showed no significant differences in PFS, OS, or CSS (all P > 0.05). Patients managed with active surveillance had significantly shorter PFS (P < 0.001) and OS (P = 0.019) than those with surgery, although CSS did not differ significantly between the two groups (P = 0.182). Locoregional failure represented the most predominant pattern (10.2%), followed by distant failure (4.5%) and non-lymphoma-related death (3.7%). With non-lymphoma-related death as a competing risk, local treatment was associated with a significantly lower risk of progression compared to active surveillance (P = 0.003) or systemic therapy (P = 0.005). However, no significant differences in CSS were observed across different management strategies in patients with progression (P = 0.185). Multifocal pulmonary lesions, nodal involvement, and multiple mucosal site (MMS) involvement were identified as CSS-associated factors and incorporated into BALT-NRI. In the discovery cohort (n=352), BALT-NRI achieved an AUC of 0.958 and C-index of 0.920 for predicting 5-year CSS. In the validation cohort (n=200), BALT-NRI score exhibited superior prediction (AUC 0.922, C-index 0.796, Youden index 0.743), outperforming MALT-IPI (AUC 0.614), revised MALT-IPI (AUC 0.851), and Ann Arbor staging (AUC 0.822). BALT-NRI stratified patients into low-risk (0 point), intermediate-risk (1–2 points), and high-risk (3 points) groups with significantly distinct PFS and CSS (both P < 0.001), while MALT-IPI and revised MALT-IPI demonstrated limited discrimination for CSS.

Conclusion:

BALT-NRI is a robust, clinically applicable model that improves long-term disease-specific survival risk stratification for BALT lymphoma, with enhanced predictive effects and the potential to inform personalized management.