Role of radiation (RT) in primary mediastinal large B-cell lymphoma (PMBCL): An analysis of the Surveillance, Epidemiology, and End Results (SEER) database.

Lymphoma and Plasma Cell Disorders
Session Type and Session Title: 
Poster Session, Lymphoma and Plasma Cell Disorders
Abstract Number: 


Poster Board Number: 
Board #380
J Clin Oncol 33, 2015 (suppl; abstr 8562)
Smith Giri, Vijaya Raj Bhatt, Ranjan Pathak, Gregory Bociek, Julie Vose, James O. Armitage; The University of Tennessee Health Science Center, Memphis, TN; University of Nebraska Medical Center, Omaha, NE; Reading Health System, Wyomissing, PA

Abstract Disclosures


Background: In the rituximab era, patients with PMBCL demonstrate high complete remission rate and a plateau in survival curve beyond 2-3 years. The use of RT in these young patients may predispose to the risk of cardiopulmonary toxicities and secondary malignancies. Methods: We used SEER 18 database and histology code 9679/3 to identify adult patients with PMBCL between 1973 and 2011. FDA approved rituximab for diffuse large B cell lymphoma in 2006. Hence, using the year 2006 as a cutoff, we compared the survival differences among patients treated with and without radiotherapy. Kaplan Meier survival curves with log rank test were plotted to compare survival statistics. Chemotherapy regimen could not be ascertained. Cox proportional hazard regression model was done to adjust for other covariates including age, year of diagnosis, race, Ann Arbor stage and gender. All p-values were two-sided and level of significance was chosen at 0.05. Results: Of 358 PMBCL patients, 50.5% patients (n = 181) received RT. Those who received radiation were more likely to be younger than 50 years (90% vs. 81%, p = 0.01) but were similar in terms of gender, race and stage at diagnosis. The unadjusted five-year overall survival (OS) was higher for RT vs. non-RT group (84% vs. 74%, p = 0.02). When stratified by year of diagnosis, OS was higher for RT vs. non-RT group in pre-rituximab era (before 2006) but not in post-rituximab era (after 2006). In multivariate analysis, RT remained an independent predictor of improved OS in pre-rituximab era (HR 0.36; 95% CI 0.17-0.79; p = 0.01) but not in post-rituximab era (HR 0.96; 95% CI 0.45-2.05; p = 0.93) after adjusting for age, year of diagnosis, gender, race and stage at diagnosis. Conclusions: Our study suggests that the effect of rituximab may reduce the benefit of RT in select patients such as patients treated with R-EPOCH or those with important risk factors for breast cancer/coronary artery disease. However, select patients such as those with large masses, positive PET scan after chemotherapy or those treated with R-CHOP may still benefit from RT. Resolving these questions is an important topic for future research.