Background:

Splenic marginal zone lymphoma (SMZL) is an uncommon, typically indolent type of B-cell lymphoma originating in the spleen. While generally associated with a favorable prognosis compared to other lymphomas, patient outcomes can vary significantly. Understanding which factors – including demographic characteristics, clinical presentation, and treatment choices influence survival in the real-world setting remains crucial for optimizing patient management and counseling. This study aimed to identify key predictors of survival in a contemporary cohort of SMZL patients across the United States.

Methods

We conducted a retrospective cohort study of patients diagnosed with histologically confirmed primary splenic SMZL (ICD-O-3: 9689; Site: C42.2). All cases were included from the Surveillance, Epidemiology, and End Results database (November 2024 release, covering diagnoses 2000-2022). Individuals with missing critical survival or demographic information were excluded. We analyzed demographics (age, sex, race, ethnicity, marital status, residence location, income), clinical factors (year of diagnosis, Ann Arbor stage), and treatment modalities (chemotherapy, radiation therapy, splenic surgery, lymph node surgery). Statistical analyses were done with descriptive statistics, univariate Cox proportional hazards regression to assess individual variable impact on overall survival (OS) and Lymphoma Specific Survival (LSS), and multivariate Cox regression including variables showing a potential association (p<0.1) in univariate analysis. Survival trends were visualized using Kaplan-Meier curves.

Results

The study cohort comprised 3,548 SMZL patients. The average age at diagnosis was 68.2 years (±11.6), with a slight majority being female (53.6%). The cohort was predominantly White (89.8%) and non-Hispanic (92.1%), and over half were married (57.9%). Stage IV disease was the most frequently recorded stage (38.9%). Treatments included chemotherapy (26.4%), beam radiation (0.7%), primary site (spleen) surgery (21.4%), and lymph node surgery.

  • Univariate Analysis: Significant associations with improved overall survival included: female sex (14.2% lower mortality risk vs males; HR=0.858, p=0.003), non-Hispanic ethnicity (17.7% lower risk vs Hispanic; HR=0.823, p=0.030). Marital status was highly significant (p<0.001), with married patients showing 53.3% lower risk versus widowed. Age increased risk by 7.0% per year (HR=1.070, p<0.001). Lymph node surgery (no surgery: 15.5% lower risk, HR=0.845, p=0.003) and no/unknown chemotherapy (12.7% lower risk vs chemotherapy; HR=0.873, p=0.013) also conferred benefits. Year of diagnosis, income, rural-urban status, primary site surgery, and lymphoma stage were non-significant.

  • Multivariate Analysis: After adjustment, female sex retained a 26.3% lower risk (HR=0.737, p<0.001), non-Hispanic ethnicity a 17.6% lower risk (HR=0.824, p=0.032). Marital status remained highly protective (p<0.001), led by married patients (58.5% risk reduction). Lymph node surgery significance persisted: 1–3 nodes removed (37.1% lower risk; HR=0.629, p=0.034) and no surgery (14.6% lower risk; HR=0.854, p=0.006). No/unknown chemotherapy maintained a 13.0% lower risk (HR=0.870, p=0.012). Age, income, and primary surgery lost significance in the adjusted model.

Conclusion

This study reveals that specific demographic characteristics and treatment patterns are strongly linked to overall survival outcomes. Female sex, non-Hispanic ethnicity, and marital status—particularly being married—emerge as powerful independent predictors of improved survival. The pronounced protective effect associated with non-widowed status, especially marriage, highlights a substantial survival advantage. Notably, the absence of chemotherapy and minimal or no lymph node surgery also demonstrate significant independent survival benefits, warranting further investigation into potential confounding factors. These findings underscore the influence of both patient demographics and selected treatment approaches on SMZL prognosis.

This content is only available as a PDF.
2025
Sign in via your Institution