Evasion of αCD19 CAR T-Mediated Killing By Rituximab-Chemotherapy Resistant Lymphoma Cells

Introduction

While αCD19 chimeric antigen receptor T cell (CAR T) therapy for the treatment of relapse/refractory (R/R) diffuse large B cell lymphoma (DLBCL) in the second- or third-line setting is associated with a sustained remission in 40% of patients, a significant number of patients progress. The median survival after CAR T failure is 1 month - stressing the need to understand mechanisms of resistance to CAR T therapy. To this end, we studied phenotypic and molecular changes in rituximab-chemotherapy resistant lymphoma cell lines surviving CAR T therapy in laboratory models.

Methods

CAR T Transductionwas performed on healthy donor T cells with lentiviral vectors encoding for 2nd generation α-CD19 CAR constructs containing CD28 (28ζ) or 4-1BB (BBζ) costimulatory domain, 24 hours after activation with αCD3/CD28 TransACT (Miltenyi). Following 10-14 days of expansion, cells were harvested and used in co-culture with Raji4RH at an effector-to-target ratio of 1:1. After 24-hour culture, CAR T cells and Raji4RH were separated via Pan T cell Isolation Kit (Miltenyi), and Raji4RH cultures were submitted for flow cytometry and RNA sequencing (RNA-seq).

Flow cytometry was performed on separated Raji4RH cell fractions from co-culture, and antibody markers anti-PD-L1, anti-PD-L2, anti-IL-10, anti-BTLA, anti-CD19, and Live/Dead Dye were used. For intracellular antibody staining, cells were given Golgi inhibitor for 3 hours prior to harvest, then fixed and permeated using eBioscience Foxp3 Staining Buffer Set (ThermoFisher).

RNA-seq data was analyzed using Spliced Transcripts Alignment to a Reference (STAR) algorithm, and a profile of differentially expressed genes (DEG) was created for genes of interest.

Results

We investigated surviving Raji4RH fractions from either 28ζ (28ζ + Raji4RH) or BBζ (BBζ + Raji4RH) co-culture, relative to control (Raji4RH alone). Flow cytometry revealed significant increase in suppressive cytokine production by surviving Raji4RH cells. PD-L1 was significantly increased by 19% (p=0.0026) within BBζ+Raji4RH cells, in addition to a 33% increase in IL-10 (p=0.0353). In 28ζ+Raji4RH cells, PD-L1 expression increased by 39% (p=0.0062) and IL-10 expression increased by 35% (p=0.0032). Significant increase of BTLA expression within BBζ+Raji4RH cells by 17% (p=0.0014) was noted in addition to 29% (p=0.0289) within 28ζ+Raji4RH cells. Upregulation of CD57 was observed in CAR T-cell surviving Raji4RH cells. We next investigated DEGs within the surviving Raji4RH fractions from either 28ζ (28ζ+Raji4RH) or BBζ (BBζ+Raji4RH) co-culture, relative to control.

Within DEGs, a modest CD19 downregulation was observed in both 28ζ+Raji4RH and BBζ+Raji4RH cells.In 28ζ+Raji4RHasignificant upregulation of IL-10 (log2FC= 1) and CD274/PD-L1 (log2FC= 1.64) was noted. PAX5, responsible for B-cell development and proliferation, was significantly increased (log2FC= 0.13) in addition to proliferative marker MKI67 (log2FC= 0.33), and NFκB signaling gene NKFB2 (log2FC= 0.24). Significant downregulation of B cell receptor kinase BTK was observed (log2FC= -0.35). Indicator of anaerobic glycolysis, LDHA (log2FC= 0.18) saw increased expression, as well as glycolytic enzymes HKI (log2FC= 0.29) and HKII (log2FC= 0.49) within 28ζ+Raji4RH cells.

In contrast, BBζ + Raji4RH exhibitedsignificant upregulation of IL-10 (log2FC= 1.32); however, CD274/PD-L1 transcript remained unchanged. PAX5 (log2FC= 0.031), MKI67 (log2FC= 0.35), and NFKB2 (log2FC= 0.19) transcripts were significantly upregulated. BTK (log2FC= -0.30) was significantly decreased in BBζ+Raji4RH cells. In addition, LDHA (log2FC= 0.17) and HKI (log2FC= 0.23) but not HKII were significantly upregulated within BBζ+Raji4RH cells.

Conclusion

Our data suggest the existence of additional mechanisms for CAR T therapy resistance beyond CD19 downregulation. Lymphoma cells surviving CAR T therapy appear to enter a proliferative state and upregulate immunosuppressive signaling. Additionally, CAR T resistant lymphoma cells preferentially upregulate glycolytic pathways, contributing to an immunosuppressive microenvironment impacting CAR T cell function. Our findings stress the need to further characterize the mechanism(s) by which regulation of B cell proliferation and the Warburg Effect impact CAR T fitness and clinical outcomes in patients with R/R DLBCL treated with CAR T therapies.

Cortese, MD, MPH:Abbvie: Consultancy, Speakers Bureau; AstraZeneca: Consultancy, Honoraria; Synthekine: Consultancy, Membership on an entity's Board of Directors or advisory committees; ADC Therapeutics: Consultancy; SecuraBio: Consultancy; Targeted Oncology: Honoraria; Cellectar Biosciences: Consultancy; Bristol Myers Squibb: Consultancy; Binaytara Foundation: Honoraria; OncLive: Honoraria; Curio Science: Honoraria. Hernandez-Ilizaliturri:Cellectar Biosciences: Consultancy, Research Funding; AbbVie: Consultancy, Research Funding; ADC Therapeutics: Consultancy; Amgen: Consultancy; Bristol-Myers Squibb: Consultancy; Celgene: Consultancy; Dava Oncology: Consultancy; Epizyme: Consultancy; Gilead: Consultancy; Incyte: Consultancy, Honoraria; Ipsen: Honoraria; Morphosys: Consultancy; Kite Pharmaceuticals: Consultancy; Novartis: Consultancy; Pharmacyclics: Consultancy; BioGene: Consultancy; AbbVie: Consultancy.