Five-year follow-up update of defined-composition CD19 CAR T-cell therapy for relapsed/refractory CLL.

7511 Background: We have described high response rates in patients (pts) with relapsed/refractory (R/R) chronic lymphocytic leukemia (CLL) treated with CD19-targeted CAR T cells. We now report comprehensive analyses of factors associated with duration of response (DOR) with a 68-month median follow-up. Methods: We included 49 pts with R/R CLL and/or Richter’s transformation (n = 10) treated on a phase I/II clinical trial of defined-composition CD19 CAR T-cell therapy (1:1 ratio of CD8+:CD4+). Response rates were investigator-assessed by 2018 iwCLL criteria. Measurable residual disease (MRD) in bone marrow (BM) was evaluated by multiparameter flow cytometry (MFC) and IGH next-generation sequencing (NGS; clonoSEQ assay). CAR T-cell enumeration was assessed by MFC. Serum cytokine concentrations (20 analytes) were measured by Luminex assay. Univariate associations between duration of response (DOR) and patient, disease, CAR T-cell manufacturing, and treatment-related variables were estimated using Cox regression (50+ variables). Association between CAR T-cell in vivo persistence and DOR was modeled with a time-dependent Cox model including blood CAR transgene levels (copies/µg genomic DNA). Results: Median age was 61 years (IQR, 55-67). Forty-six pts (94%) had complex karyotype and/or 17p deletion; 46 (94%) received cyclophosphamide/fludarabine lymphodepletion. In the response-evaluable cohort (n = 47), median follow-up was 68.3 months (IQR, 54.3-81.6). Overall and complete response rates at day+28 were 70% and 17%, respectively, and the median DOR was 22.2 months (95% CI, 10.7-not reached [NR]). MRD-negativity at day+28 by MFC occurred in 33/47 (70%). In pts with MRD-negativity by MFC and available data, MRD negativity by NGS was seen in 18/29 (62%). In NGS MRD-negative responders, the median DOR was 56.8 months (95% CI, 39.5-NR). In all evaluable patients, the 5-year OS and PFS were 35.5% (95% CI, 24.0%-52.5%) and 21.5% (95% CI, 12.0%-38.5%), respectively. In univariate Cox regression, day+28 MRD-negativity by MFC (HR = 0.03, 95% CI, 0.01-0.18, p < 0.001), day+28 MRD-negativity by NGS (HR = 0.21, 95% CI, 0.07-0.61, p = 0.004), peak CD8+ CAR T-cell expansion (HR = 0.49, 95% CI, 0.24-1.00, p = 0.05), CAR T-cell persistence (HR = 0.71, 95% CI, 0.54-0.91, p = 0.01), and lower peak MIP-1β (HR = 3.12, 95% CI, 1.13-8.65, p = 0.03) were associated with longer DOR. Peak MIP-1β remained associated with DOR (HR = 3.83, 95% CI, 1.35-10.9, p = 0.012) in a multivariable model including pre-LD CLL cells in BM, tumor cross-sectional area, peak CD8+ CAR T-cell expansion, and day+28 MRD by MFC. Conclusions: CD19 CAR T-cell therapy achieved durable remissions in R/R CLL pts. Day+28 MRD, CAR T-cell peak expansion, and CAR T-cell persistence were strongly associated with DOR. Lower peak serum levels of MIP-1β, a cytokine known to be produced by CLL cells upon BCR activation, was strongly and independently associated with DOR. Clinical trial information: NCT01865617 .