Efficacy of CD19 Directed Therapies in Large B-Cell Lymphoma (LBCL) Relapsing after Chimeric Antigen Receptor (CAR) T-Cell Therapy

Introduction

Durable remission after CD19 directed CAR T-cell therapy (CD19-CAR-T) is achieved in around 40% of patients with LBCL. For patients with relapsed and/or refractory (R/R) disease after CAR-T, there is no standard of care and outcomes are generally poor with an anticipated median overall survival (OS) of only 5 months.

Tafasitamab is a CD19 directed monoclonal antibody, approved in combination with lenalidomide (tafa-len) for patients with R/R LBCL. Preclinical models and limited clinical data suggest that prior treatment with tafa does not limit efficacy of CD19-CAR-T but there is currently no available data on tafa-len efficacy for patients who are R/R post CAR-T. One study suggested that len can provide a high response rate in patients relapsing after CAR-T, particularly in those who receive len early after relapse (Thieblemont, C., et al).

Loncastuximab tesirine (loncaT) is an antibody drug conjugate that also targets CD19 and is currently approved in R/R LBCL. The LOTIS-2 trial of loncaT in R/R LBCL included 13 patients with post-CAR-T relapse, with an overall response rate (ORR) of 46.2% (Caimi, P., et al). There remains limited understanding of the efficacy of CD19-directed therapies when utilized in patients with R/R disease post-CD19-CAR-T.

Methods

This is a multi-center observational analysis of patients with R/R LBCL that received either tafa-len or loncaT at any point post-CAR-T. For patients who received both regimens, data was only captured for the first administered treatment. Kaplan-Meier method was used to estimate median survival times for OS and progression free survival (PFS) from the time of tafa or loncaT initiation. Cox proportional hazards regression models were used to evaluate statistically significant univariate predictors of OS and PFS.

Results

Forty-one patients were included in this study. The median patient age was 65 (38-79 years), 25 (61%) were male and 37 (90%) had an international prognostic score of >3. The median number of systemic therapies pre–CAR T was 3 (1-6) with 12 (29%) patients receiving more than 3 prior lines of therapy. Axicabtagene ciloleucel was the most utilized CAR-T product (n=26 ,63%). Seventeen (41%) patients were refractory to CAR-T. CD19 status at post CAR-T relapse was assessed in 31 patients and was positive in 21 (68%). Median time from CAR-T to tafa-len or loncaT was 6.4 (1.4-38.2 months) with median number of lines of therapy between CAR-T and these regimens of 1 (0-5). In 16 (39%) patients, tafa-len or loncaT were the first treatment after CAR T-cell therapy. At time of treatment, median absolute neutrophil count was 3.3 (0.4-12.5 x 10⁹/L), median platelet count was 107.5 x 10⁹/L (13-281 x 10⁹/L), and 29 (83%) patients had elevated serum lactate dehydrogenase (LDH).

Twenty-eight (68%) received tafa-len and 13 (32%) received loncaT. Median number of tafa and loncaT cycles was 2 (1-13). The combined ORR and complete response (CR) rates were 27% and 12%, respectively. For tafa-Len and loncaT, the ORR and CR were 17% and 5% and 10 and 5%, respectively. All patients received len in combination with tafa with median number of lenalidomide cycles at 2.5 (1-9) and median time from initiation of tafa to len being 0 (minus 261-29 days). One patient initiated len before the approval of tafa and tafa was combined later. The median time to response from initiation of tafa or loncaT was 47(6-158 days) with a median duration of response at 13.3 (2.1-56.7 weeks). Thirty-eight (95%) patients discontinued treatment: 28 (70%) due to progressive disease, 5(13%) due to treatment intolerance, 2 (5%) infection, 1 (3%) secondary malignancy and 2 (5%) transitioned to hospice presumably due to progressive disease.

At a median follow-up of 51 (9.1 -199) weeks from CAR T infusion, 30 (73%) progressed a with median PFS of 8 (95% CI 5-13) weeks. Twenty-four (59%) died with a median OS of 18.6 weeks (95% CI 14.1- NA). Causes of death included progressive disease in 18 (75%), infection in 5 (21%) and secondary malignancy in 1 (4%). On univariate analysis, a shorter median PFS was observed in patients with a high LDH at the time of treatment initiation.

Conclusions

In this retrospective study, the use of currently approved CD19-directed therapies for relapse post-CAR T showed limited clinical activity and short-lived responses. There remains an important unmet need for developing effective therapies for patients relapsing post CD19-CAR T.