Preclinical Development of Lymphotoxin Beta Receptor-Overexpressing CAR T Cells Demonstrates Potentiation of CAR T Cell Antitumor Activity

Diffuse large B cell lymphoma (DLBCL) is one of the most common yet aggressive B cell lymphomas, with poor prognosis for patients who do not respond to or relapse after first-line chemo-immunotherapy. Recent advances in cellular immunotherapy, specifically engineering autologous T cells with chimeric antigen receptors (CARs) targeting B cell lineage antigens, have resulted in improved outcomes for patients with relapsed/refractory (R/R) DLBCL. However, progression of disease even after initial complete response (CR) to CAR therapy is common and only 1/3 of patients are cured. To address the shortcomings of the current CAR therapies, we designed an unbiased, genome-scale gain-of-function approach to identify new genes that can potentiate T cell functions (Legut et al, Nature 2022). The top candidate identified in the screen was lymphotoxin beta receptor (LTBR), a member of the tumor necrosis factor receptor superfamily that is not endogenously expressed in lymphocytes. Here we provide preclinical characterization of LTBR-overexpressing CAR T cells in terms of safety and efficacy.

We have previously shown that LTBR not only boosts T cell effector function in vitro but increases T cell proliferation and mitigates activation-induced cell death and apoptosis. To confirm that LTBR overexpression does not result in leukemic CAR T cell transformation, we demonstrated that LTBR CAR T cells proliferate only in response to antigenic stimulation and do not survive in absence of interleukin-2. We also tested whether LTBR-driven T cell phenotypes are dependent on specific media composition used for ex vivo expansion and observed similar phenotypic and functional results across five media compositions tested.

Using a tricistronic lentiviral vector, we delivered a CD19-targeting CAR (utilizing either a CD28- or 4-1BB-based costimulatory domain) together with LTBR and a selection marker to human CD4+ and CD8+ T cells. Across all six healthy donors tested, we observed a striking enhancement of cytotoxicity in LTBR-overexpressing CAR T cells, as compared to CAR T cells overexpressing an irrelevant gene. One of the mechanisms supporting this enhanced cytotoxicity is an up to 70% increase in synapse formation between CAR T cells and target cancer cells, as measured by super-resolution fluorescence imaging. Then, we further optimized the lentiviral vector encoding LTBR and CAR for high LTBR expression and, consequently, anticancer activity.

Finally, we demonstrated that LTBR drives potentiation of T cell effector function in DLBCL patient-derived CAR T cells responding to autologous, CD19+ target cells (Fig. 1A). LTBR CAR T cells similarly showed superior efficacy in a disseminated B cell leukemia model in vivo without overt indications of toxicity (Fig. 1B). These data highlight that ectopic expression of LTBR is a promising next-generation CAR T cell product that warrants further translation; this novel CAR construct may provide substantial benefit to patients with relapsed or refractory DLBLC, or other B cell malignancies.

Figure 1. LTBR co-delivery potentiates CAR T cell antitumor activity ex vivo in DLBCL patient samples and in vivo. A) T cells from a treatment-naïve DLBCL patient were engineered with an anti-CD19 CAR (utilizing the 4-1BB costimulatory domain, based on tisagenlecleucel) co-delivered with LTBR or an irrelevant gene, truncated NGFR (tNGFR). Engineered T cells were then co-incubated with a CD19-negative leukemia cell line or CD19+ autologous target cells for the measurement of secreted interferon (IFN)-g. n = 3 B) Female NSG mice (n = 4 mice per group) were inoculated with 5x10⁵ CD19⁺ Nalm6 cells. Four days later, mice were treated with 5x10⁶ CAR T cells (CD4:CD8 mix at 1:1 ratio), either untransduced or engineered with an anti-CD19 CAR (utilizing the 4-1BB costimulatory domain, based on tisagenlecleucel) co-delivered with LTBR or tNGFR. Survival was measured over the period of 32 days. The study was blinded. Log-rank Mantel-Cox p value is shown.

Diefenbach:MorphoSys: Consultancy, Research Funding; FATE Therapeutics: Research Funding; Genentech/Roche: Consultancy, Research Funding; Genmab: Consultancy; Gilead: Current equity holder in publicly-traded company; IMAB: Consultancy; Incyte: Research Funding; Kite: Consultancy; MEI Pharma: Research Funding; Merck: Consultancy, Research Funding; BMS: Consultancy, Research Funding; Celgene: Consultancy; Seattle Genetics: Research Funding. Sanjana:Qiagen: Consultancy; Vertex: Consultancy.