Relapsed or Refractory CLL after CD19-Specific CAR-T Therapy: Treatment Patterns and Clinical Outcomes

Background: Treatment of high-risk CLL remains challenging despite the introduction of novel therapeutic agents. Chimeric antigen receptor T-cell (CAR-T) therapy has shown promising efficacy in these patients (pts) (Turtle, JCO, 2017; Gauthier, ASH, 2019), but progressive disease after CAR-T is not uncommon. Understanding the outcomes of pts with relapsed or refractory CLL after CAR-T is important for establishing a benchmark for trials in this setting and to study optimal treatment sequencing in high-risk CLL.

Methods: CLL pts treated with CD19-specific CAR-T on a clinical trial (NCT01865617) at Fred Hutch/University of Washington were reviewed. Using the 2018 IWCLL definitions, we identified pts who had stable or progressive disease (SD/PD) on initial assessment (4 weeks) or had a relapse after a complete or partial response (CR/PR) at any time. Overall survival (OS) was calculated using the Kaplan-Maier estimates. For univariate models, we included del17p, complex karyotype (CK), ibrutinib (IB) and venetoclax (Ven) failure before CAR-T, number of prior treatment lines, Richter's transformation (RT), use of bridging treatment, bulky disease (>5cm) before CAR-T, cytokine release syndrome (CRS; ≥ grade 3), neurotoxicity (NT; ≥ grade 3), refractory vs. relapsed (after CAR-T), use of IB, Ven, idelalisib (Idela), repeat CAR-T or allogeneic transplant (allo-SCT) after CAR-T progression. Only factors with significant association in the univariable models (p < 0.05) were included in the multivariable analysis.

Results: 28 pts were identified with refractory (n=16; 57%) or relapsed disease [n=12; 43% - median time to relapse after CAR-T was 11 months (1.8-23)] after CAR-T. Median age was 60 (41-70) and 25% were female. Nine pts (32%) had a history of RT. Cytogenetic changes were: del17p (71%), CK (78%), del11q (37%), del13q (43%) and +12 (11%). Details of pre-CAR-T treatments are outlined in table-1. Before CAR-T, 24 pts (86%) had progressed on IB, 10 (36%) on Ven and 6 (21%) on Idela. Nine pts (32%) had progressed on both IB and Ven before CAR-T and 5 pts (18%) had prior allo-SCT. Five pts (18%) received bridging therapy after leukapheresis. Lymphodepletion (LD) was with cyclophosphamide and fludarabine (CyFlu) in 25 pts (89%). CAR-T adverse events included CRS in 22 pts (grade 3 in 2 pts, no grade 4) and NT in 7 (grade 3 in 5 and no grade 4).

Details of Post CAR-T failure treatments are summarized in table-2 Both IB and Ven were used after CAR-T even if pts had progressed on them before. For Ven, 5 of 11 patients who received Ven after CAR-T had progressed on it before but had a median duration of response (DOR) of 5 months (2-8) which was not different from pts who were Ven responsive before CAR-T [5 months (1-10); p = 0.95). This was not true for IB with DOR of 2 months (0.5-7) in pre-CAR-T IB failed vs. 12.25 months (6-18.5) in pre CAR-T IB responsive pts (p = 0.001). Fourteen pts (51%) had repeat CAR-T therapy on the same protocol and 6 had CR (3) or PR (3). Four of these patients relapsed later with a DOR after second CAR-T of 5.5 months (1-33). Six pts (23%) received an allo-SCT median 6.5 months after CAR-T progression (2-14) and 3 achieved a CR (2) or PR (1), 2 did not respond (PD=1; SD=1) and one died before assessment. Two of these pts relapsed 9 and 39 months after allo-SCT. Chemotherapy was used in 7 pts with no responses.

From variables tested in univariable models, only 4 were entered to the multivariate analysis (p < 0.05) and all remained significantly associated with OS: history of progression on both IB and Ven before CAR-T was strongly associated with poor OS [hazard ratio (HR) 11.2 (95% CI: 2.5-50.5) ; p=0.002]. Receiving an allo-SCT after CAR-T progression was associated with an improved OS [HR 0.14 (95% CI: 0.02-0.97); p= 0.04]. Other factors were post CAR-T progression treatment with IB [HR 0.16 (95% CI: 0.04-0.68); p = 0.01] or Ven [HR 0.17 (95% CI: 0.03-0.8); p = 0.02]. Median OS for the entire cohort was 10.4 months (0.23-59). [Fig-1]

Conclusion: This data sets a benchmark for clinical trials that intend to improve outcome of CLL pts with progression after CAR-T. OS after CAR-T progression was significantly shorter in pts who received CAR-T after failing both IB and Ven compared to others. This finding supports referring high-risk CLL pts for CAR-T treatment while still responsive to either IB (BTKis) or Ven. For eligible pts, allo-SCT seems to provide a higher chance of survival in pts who progress after CAR-T.

View largeDownload slide

View largeDownload slide

Close modal