Real-World Outcomes of CD19CAR T Cell Therapy in Adult Patients with Relapsed Refractory Transformed Indolent Lymphoma

Introduction

Transformation of indolent lymphomas to diffuse large B-cell lymphoma (DLBCL) is associated with poor outcomes, particularly for patients (pts) with relapsed or refractory (r/r) disease. While chimeric antigen receptor (CAR) T cell therapy has revolutionized the management of r/r DLBCL and is approved for transformed indolent lymphoma (tiNHL), pts with tiNHL were largely underrepresented in the pivotal trials leading to the approval of CD19CAR. In this study, we evaluated the safety and efficacy of standard of care CD19CAR T in adults with r/r tiNHL relative to de novo DLBCL (dDLBCL).

Methods

We conducted a multicenter retrospective study in adults treated at 6 centers. Eligibility criteria included age ≥ 18 years old, diagnosis of DLBCL/high-grade B-cell lymphoma (HGBCL) (both de novo and transformed), r/r disease, and receipt of SOC CD19CAR T. Pts who received CAR T on clinical trials were excluded. We defined tiNHL as DLBCL/HGBCL transformed from follicular lymphoma, marginal zone lymphoma, or Waldenstrom’s macroglobulinemia. Pts with Richter’s syndrome were excluded. Safety endpoints included incidence and severity of cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) per American Society for Transplantation and Cellular Therapy (ASTCT) consensus grading. Response was assessed using 2014 Lugano Criteria. Efficacy endpoints include best overall response rate (ORR) (proportion of pts achieving a complete or partial response), complete response rate (CRR), progression/relapse-free survival (PFS) and overall survival (OS). Pts were followed post-CAR T through 6/2023, lost to follow-up, or death. The study was approved by the Institutional Review Board at each site.

Results

A total of 1182 pts were included (338 (29%) tiNHL, 884 (71%) dDLBCL) with the CAR T infusion date ranging from 12/2017 to 10/2022. In the tiNHL cohort, 284 (84%) pts transformed from FL, 41 (12%) MZL, 13 (4%) WM. Most baseline characteristics were comparable in the tiNHL and dDLBCL cohorts. Among all pts, the median age at CAR T cell therapy was 64 (range: 18-89) years and 36% were women. At the time of CAR T cell therapy, 934 (79%) pts had advanced stage disease (III-IV), 649 (55%) had an elevated LDH, 479 (41%) had more than one site of extranodal disease, and 135 (11%) had bulky disease (defined as ≥ 10 cm). Most pts (77%) received axicabtagene ciloleucel. Compared to dDLBCL, tiNHLpts were more heavily pre-treated (67% vs 51% had received ≥3 prior lines of therapy before CAR), more likely to have HGBCL (13% vs 9%), and more likely to have received bendamustine within 12 months of CAR (17% vs 6%), but less likely to have central nervous system involvement (6% vs 11%).

Rates of CRS were similar with any grade / grade ≥ 3 CRS observed in 79%/7% tiNHL pts and 84%/8% of dDLBCL pts, respectively (p=0.6). Any grade ICANS occurred in 42% of tiNHL and 52% of dDLBCL, with less frequent grade ≥ 3 ICANS in the tiNHL cohort (21% vs 27%, p=0.024). Tocilizumab use for CRS (50% vs 61%, p <0.001) and glucocorticoid use for ICANS (36% vs 44%, p=0.01) were lower in tiNHL. Rate of CAR T toxicity related ICU stay within first 30 days (13%) was comparable in both cohorts.

ORR was similar in both cohorts (83% vs 81%, p=0.3), while CRR was higher in tiNHL (67% vs 59%, p=0.017). With a median follow-up of 22.3 months, PFS and OS were similar between the tiNHL and dDLBCL cohorts (24-month PFS 41% vs 38%, p=0.16; 24-month OS 58% vs 52%, p=0.15). The cumulative incidence of non-relapse mortality at 24 months was 10% (95%CI: 6.9–14%) in the tiNHL and 11% (95%CI: 8.7–13%) in the dDLBCL cohort. After adjusting for age, stage, ECOG, LDH, extranodal site, number of prior therapy, prior bendamustine, CNS involvement, and bridging therapy, the hazard of disease progression, relapse or death post-CAR appeared to be 16% lower for tiNHL pts compared to dDLBCL pts (HR: 0.84 [95%CI: 0.69-1.0], p=0.07). Multivariable analysis suggested that elevated LDH, advanced stage, ≥3 prior lines of therapy, prior bendamustine within 12 months of CAR, CNS involvement prior to CAR, and receipt of bridging therapy were each associated with inferior PFS.

Conclusion

This study represents the largest evaluation to date of outcomes of CD19CAR in tiNHL, and demonstrates that CAR is highly effective with an acceptable toxicity profile in tiNHL pts. Thus, CAR may be able to overcome the historically poor prognosis associated with conventional therapies for tiNHL.