A Single Center Retrospective Study of Real World Relapsed/Refractory DLBCL Patients Eligible for CAR T-Cell Therapy: Patient Characteristics and Outcomes

Document Type


Publication Date


Publication Title



washington; swedish; swedish cancer


Introduction: Patients with Diffuse large B-cell lymphoma (DLBCL) may experience excellent long-term outcomes after initial anthracycline containing therapy. However, patients with relapsed or refractory (R/R) DLBCL often have poor outcomes. Select R/R DLBCL patients may be treated with additional chemoimmunotherapy (CIT) followed by hematopoietic stem cell transplant (HSCT). However, as many as 50% of R/R DLBCL patients are unable to undergo HSCT due to lack of response to CIT or comorbidities (Gisselbrecht C, et al JCO 2010). Recent data reported in the SCHOLAR-1 study suggest a median overall survival (OS) of 6.3 months for these patients, with only 20% of patients alive at 2 years (Crump M, et al Blood 2017). Chimeric antigen receptor T-cell (CAR-T) therapy, a novel form of immunotherapy, offers improved outcomes for such patients with complete response rates of approximately 40% and 50% OS at 12 months or greater (Neelapu SS et al. NEJM 2017; Borchmann P et al, EHA 2018). Delivery of CAR-T therapy is specialized and remains limited to a small number of centers at present. The broad applicability of CAR-T therapy in a real world population of R/R DLBCL patients remains unknown. This retrospective study aimed to identify the characteristics and outcomes of a cohort of R/R DLBCL patients who would have been eligible for CAR-T cell therapy if available over a 4 year period at Swedish Cancer Institute (SCI). The SCI is a specialty cancer research center, based in a non-profit, non-university affiliated medical center.

Methods: All patients with recorded diagnosis of DLBCL (ICD9/10) seen for an outpatient encounter in a SCI facility between the dates of 01/01/2014 and 01/01/2018 were identified from an electronic medical record database. Patients who had received anthracycline-based chemotherapy only at SCI as initial therapy and then subsequently received 2nd line or beyond therapy at SCI between January 2014 and January 2018 were included. This population was defined as the study cohort of R/R DLBCL patients and was then evaluated for would-be eligibility for CAR-T therapy by application of the defined Kite Zuma-1 clinical trial Inclusion/Exclusion (I/E) criteria. Patient characteristics for the CAR-T eligible population were obtained by retrospective medical record review. Overall survival of the potential CAR-T eligible population was assessed including stratification by receipt of HSCT at any time during the study period.

Results: 486 patients with a diagnosis of DLBCL were seen during outpatient SCI encounters between Jan 2014 and Jan 2018. Of these, 60 patients received prior 1st line anthracycline therapy exclusively at SCI and then received 2nd line or beyond therapy at SCI between 2014 and 2018 for R/R DLBCL. The majority of patients, 82% (n=49), met all Zuma-1 I/E criteria for CAR-T therapy. Characteristics of these patients are identified in Table 1. Among all CAR-T eligible patients, OS was 37.1% at 24 months (Figure 1). Patients received a variety of 2nd line or beyond therapies, including 47% (n=23) who received HSCT. OS at 24 months for CAR-T eligible patients was significantly better for those receiving HSCT in 2nd line or beyond versus those who did not receive HSCT (61.6% vs 12.0%, respectively; p<0.001; Figure 2).

Conclusion: In a retrospective cohort of real-world R/R DLBCL patients treated between 2014 and 2018 at SCI, a non-university based specialty cancer research center, 49 of 60 patients (82%) would have been eligible for CAR-T therapy based on Zuma 1 I/E criteria. This suggests that the majority of the patients with R/R DLBCL in the real-world may have an opportunity to receive CAR-T. Moreover, while those who underwent a successful HSCT as part of 2nd line or beyond therapy had greatly improved outcomes, those patients who did not undergo HSCT had poor outcomes. For such patients not receiving HSCT, the availability of CAR -T may lead to significantly improved outcomes.

Clinical Institute