CD19 Antigen Density Down-Regulation at Time of Progression in Large B-Cell Lymphoma Patients Treated with Axicabtagene Ciloleucel

Introduction Axicabtagene ciloleucel (axi-cel) is a CD19-targeting chimeric antigen receptor T (CAR-T) cell therapy approved in aggressive large B cell lymphomas. Despite impressive response rates, durable remissions occur in 30-40% of patients. We have shown CD19 downregulation or loss occurs in approximately 30% -50% of patients at time of axi-cel relapse and that lower pre-treatment CD19 antigen density associated with disease progression (Spiegel, Nature Medicine 2021). Here we report CD19 expression using quantitative flow cytometry in a larger cohort. before and after axi-cel therapy. Results CAR-T treated patients at Stanford University were enrolled on an IRB-approved bio-repository protocol. Fifty-five unique patients with available quantitative flow cytometry data were identified and were treated between January 2018 and June 2022. Fifty-four of these patients received axi-cel in the 3rd line or later, with 3 median prior lines of therapy. Twenty-six (47.3%) patients received bridging therapy between leukapheresis and axi-cel infusion. The best response to axi-cel was CR in 56.4%, PR in 5.5% and PD in 38.2%. As of June 30, 2023, 24 patients (43.6%) remain in remission and 35 (63.6%) remain alive. With a median follow-up of 36.6 months, median progression free survival was 7.1 months (95% CI 4.20 - 26.7 months) and median overall survival was 31.8 months (95% CI 17 - NE). There were 4 NRM events, with a 3 year cumulative risk of relapse of 56% and 7% risk of NRM.We assessed antigen expression both at pre-treatment (n = 36) and at the time of relapse (n = 25). Median CD19 antigen density prior to axi-cel was 5810 CD19 molecules/cell (IQR 3442 - 8725) and was significantly decreased (2021 CD19 molecules/cell, IQR 439 - 5208) at the time of progression (p = 0.001). Six patients had paired assessment pre (4916 CD19 molecules/cell, IQR 2828 - 7679) and post axi-cel (4496 CD19 molecules/cell, IQR 2450 - 5390) which was not statistically different (p = 0.7). CD22 antigen density pre (n = 22) and post axi-cel (n = 26) was 6365 CD22 molecules/cell (IQR 3319 - 8705) and 2566 CD22 molecules/cell (IQR 1132 - 7288), respectively (Wilcoxon p = 0.05). CD20 antigen density post axi-cel (n = 26) (7090 CD20 molecules/cell, IQR 1232 - 21636) was not significantly decreased relative to pre axi-cel (n = 22) levels (5390 CD20 molecules/cell, IQR 2870 - 54213)(p = 0.45). Linear regression did not find an association between pre-treatment CD19 antigen density and disease progression (b coefficient for log10(CD19) = -0.5, p =0.8). Conclusion This study is part of an ongoing analysis of quantitative antigen density in patients treated with axi-cel. CD19was significantly downregulated at time of axi-cel relapse, however, there was no association between pre-treatment CD19 antigen density and disease progression, consistent with the multiple mechanisms of CAR resistance.