Survival outcomes in the elderly with limited stage diffuse large B-cell lymphoma treated with chemotherapy: SEER propensity-matched analysis.

e19046 Background: Current data shows a median age of diffuse large b-cell lymphoma (DLBCL) diagnosis of 70 years. More than 50% of elderly DLBCL patients can be expected to be cured by modern immunochemotherapy. Survival outcomes in elderly patients with limited stage DLBCL treated with chemotherapy outside of clinical trials are poorly characterized. Our objective is to determine survival pattern and factors that influence survival in elderly patients with limited stage DLBCL treated with chemotherapy through analysis of data from a large nationwide cancer registry in modern treatment era. Methods: The Surveillance Epidemiology and End Results (SEER 18) treatment database (2001-2013) was used to detect limited stage (Ann Arbor lymphoma stage I and II) DLBCL (ICD-O-3 code: 9680/3) adult cases with ages between 60-80 years treated with chemotherapy. Patients were divided into two cohorts. Cohort 1 included patients aged ≥60 to 70 yrs and Cohort 2 included patients aged > 70 to 80 yrs. The variable "First Malignant Primary Indicator" was used to differentiate between primary DLBCL and secondary DLBCL cases. Overall survival (OS) was calculated using the Kaplan-Meier methods and multivariate cox regression model to determine the impact of race, gender, radiation use and primary malignant indicator on survival using 1:1 propensity score matching. Chi square test was applied to delineate for any significant difference between two cohorts. Results: Overall, 11138 patients were included with 5569 patients in each cohort. The relative incidence of limited stage secondary DLBCL was greater for cohort 2 when compared to cohort 1 (26% vs. 15%, p < 0.01). The odds of receiving radiation was 27% higher in cohort 1 compared to cohort 2 (p < 0.0001). Compared to cohort 1 (median OS-130 months), cohort 2 had worse median OS of 72 months (p < 0.0001). Female gender, not receiving radiation and secondary status have shown strong relationship with increased risk of death on multivariate cox regression analysis in the elderly limited stage DLBCL patients. Conclusions: Overall survival was worse in age group 70-80 yrs when compared to age group 60-70 yrs in limited stage DLBCL patients treated with chemotherapy. This may be due to differences in treatment approaches. Female gender, not receiving radiation and secondary status are poor prognostic factors of survival in above group of patients. Race has no impact on survival.