Background: Few elderly patients with RR-DLBCL are offered HSCT and little is known about their overall survival (OS) rates or economic burden. This study evaluated OS rates and healthcare resource utilization (HRU) among Medicare patients who received HSCT. Methods: Using the Surveillance, Epidemiology, and End Results (SEER)-Medicare database, patients who relapsed after receiving R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone) as first-line therapy (1L) for DLBCL diagnosed between 2000 and 2011 were selected for our cohort if they received a subsequent HSCT at 2L or beyond. Subgroups were further divided by time to early relapse (≤12 months) or late relapse ( > 12 months) based on time from end of R-CHOP to 2L therapy. Median (95%) OS were estimated by Kaplan-Meier method from date of HSCT. Descriptive healthcare utilization rates were reported for inpatient, outpatient, and emergency department visits. Results: Among 6361 patients with DLBCL who received 1L R-CHOP, 1869 (29%) received 2L therapy; outcomes for the 117 (6.3%) receiving subsequent HSCT are presented here. At time of diagnosis, patients had a mean age of 69 (SD = 2.99 years) and were predominantly males (59% (n = 69)), with stage III/IV disease (68%, n = 80), and a mean NCI comorbidity index (CI) of < 1. The majority of patients receiving HSCT were early relapsers (n = 86). With a median follow-up of 10.3 months (0.00 -115.92) from the date of transplant, the median OS was 12.38 months (95% CI: 8.06, 29.43), with no statistically differences in OS between pts with early and late relapse. As for HRU, patients had frequent inpatient admissions (100% utilizied inpatient services), with 5.82 mean inpatient visits and an average length of stay of 8.53 days (SD = 5.15 days), outpatient visits (94%, with 25.48 mean outpatient visits), and emergency department visits (60%, with 1.21 mean ER visits). Conclusions: In this elderly population with RR-DLBCL, HSCT with curative intent was infrequently utilized, associated with short OS, and substantial HRU. Poor real world outcomes demonstrate an unmet need for novel therapies in RR-DLBCL.