Real-World Outcomes of Early Relapsed/Refractory Large B Cell Lymphoma Patients Treated with 2nd Line CAR T Therapy

Background

Axicabtagene ciloleucel (axi-cel) and lisocabtagene maraleucel (liso-cel) are autologous CD19-directed chimeric antigen receptor (CAR) T cells that have been approved for second line (2L) use for patients (pts) with early relapse or refractory (early R/R) large B cell lymphoma (LBCL). However, there are limited data regarding outcomes of pts treated with commercial CAR T cell therapies in 2L.

Methods

This is a single center retrospective study conducted at the University of Pennsylvania of pts diagnosed with early R/R (detected ≤12 months [mo] following completion of first-line therapy) LBCL who received 2L CAR T cell therapy between May 2022 and March 2024. Bridging therapy (BT), lymphodepleting chemotherapy (LDC) and CAR T product were prescribed at the discretion of the treating physician. Cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANs) were graded as per ASTCT consensus criteria. Response assessments were made as per the Lugano Classification. Baseline characteristics that demonstrated statistically significant (P<0.05) association with outcomes on univariate analysis were included in multivariate analysis. Data were censored on 7/1/24.

Results

51 pts were identified with 2 excluded due to being lost to follow-up <3 mo in remission and 1 due to receipt of maintenance therapy while in remission. For the 48 included pts, pt/disease characteristics included 60% male sex, 58% primary refractory, at the time of relapse 44% age >60 and 54% elevated LDH, 33% history of indolent lymphoma, cell of origin by Hans algorithm 60% germinal center B (GCB)/29% non-GCB/11% not classified, double expressor positive 27%/negative 50%/unknown 23%, MYC-BCL2 double hit positive 15%/negative 67%/unknown 18% and TP53 loss of function mutation positive 7%/negative 17%/unknown 76%. At the time of apheresis, CD3 count was ≤293 (lowest quartile) in 27%/>293 69%/unknown 4%. Any/systemic/radiation BT was received by 94%/73%/40%. LDC received was bendamustine 83%/fludarabine-cyclophosphamide 15%/other 2%. Last imaging was performed with a median of 0.45 (interquartile range [IQR] 0.37-0.68) (mo prior to infusion and after completion of BT for 91% of pts receiving BT. Progressive disease (PD)/elevated LDH at (the time of receipt of) LDC was detected in 23%/40% with no significant association between these characteristics (P=0.21). Product received was axi-cel 67%/liso-cel 33%.

The overall response rate (ORR) at 3 mo was 65% (95% confidence interval [CI] 49-78%) and complete response rate (CRR) at 3 mo 48% (95% CI 33-63%). With a median length of follow-up of 13.0 mo, the estimated 12 mo progression free survival (PFS) was 46% (95% CI 31-61%) and 12 mo overall survival (OS) 83% (66-92%). Non-relapse mortality was 0%. Rates of any grade/grade ≥3 CRS were 65% (95% CI 49-78%)/4% (95% CI 1-14%) and any grade ICANS/grade ≥3 ICANS 17% (95% CI 7-30%)/4% (95% CI 1-16%).

On univariate analysis for OR, only PD at LDC (OR 0.21, 95% CI 0.06-0.79, P=0.02) was predictive (no multivariate analysis performed). On multivariate analysis for CR, primary refractory disease (OR 0.22, 95% CI 0.06-0.89, P=0.03) and PD at LDC (OR 0.09, 95% CI 0.02-0.53, P<0.01) were predictive. On multivariate analysis for progression at 12 mo, age >60 (hazard ratio [HR] 2.65, 95% CI 1.17-6.01, P=0.02) and PD at LDC (HR 2.63, 95% CI 1.17-5.91, P=0.02) were predictive. On univariate analysis for death at 12 mo, only PD at LDC (HR 11.63, 95% CI 1.32-97.40, P=0.03) was predictive (no multivariate analysis performed).

For pts with vs without PD at LDC, ORR was 40% vs 76%, CRR 13% vs 64%, estimated 12 mo PFS 27% vs 56% and estimated 12 mo OS 57% vs 97%.

On multivariate analysis for development of CRS, only receipt of liso-cel product (OR 0.12, 95% CI 0.03-0.50, P=<0.01) was predictive. On multivariate analysis for development of ICANS, receipt of radiation bridging therapy and elevated LDH at LDC (OR 9.54, 95% CI 1.36-66.55, P=0.02 for both) were predictive.

Conclusion

Analysis of early R/R LBCL pts treated with 2L CAR T at our center demonstrates lower ORR and CRR but similar 12 mo estimates of PFS and OS to those reported in ZUMA-7 and TRANSFORM. Outcomes did not differ by product received with the exception of lower odds of development of CRS in liso-cel pts. PD at the time of receipt of LDC predicted for inferior ORR, CRR, PFS and OS, a finding which may inform management of R/R LBCL pts who are candidates for CAR T therapy.