Targeting PD-L1 in non-small cell lung cancer using CAR T cells

Abstract

Antibodies against programmed cell death protein 1 (PD-1) and its ligand (PD-L1) have dramatically changed the landscape of therapies for non-small cell lung carcinoma (NSCLC); however, the majority of patients do not respond to these agents. In addition, hyperprogressive disease (HPD) develops in a larger portion of NSCLC patients treated with PD-1/PD-L1 inhibitors than in patients treated with standard chemotherapy. The use of chimeric antigen receptor (CAR) T cells has been successful to treat blood cancers but not for solid tumors like NSCLC. In this work, we constructed CAR T cells that target PD-L1 and evaluated their efficacy in NSCLC with either high or low PD-L1 expression. PD-L1-CAR T cells exhibited antigen-specific activation, cytokine production, and cytotoxic activity against PD-L1^high NSCLC cells and xenograft tumors. Furthermore, the addition of a subtherapeutic dose of local radiotherapy improved the efficacy of PD-L1-CAR T cells against PD-L1^low NSCLC cells and tumors. Our findings indicate that PD-L1-CAR T cells represent a novel therapeutic strategy for patients with PD-L1-positive NSCLC, particularly for those who are susceptible to HPD.

Fig. 1. Characteristics of PD-L1-CAR T cells.

a Schematic diagram of PD-L1-CAR and CD19-CAR constructs. scFv for PD-L1 is derived from atezolizumab (Roche). b Surface expression of PD-L1-CAR on transduced T cells as measured by flow cytometry using Biotin-Protein L and APC-streptavidin on day 5 post transduction. c Viability of PD-L1-CAR T cells on day 7 and 14 post-transduction. d Expansion of PD-L1-CAR and CD19-CAR T cells in vitro for 14 days. e Percentage of CAR T cells that were positive for CD3, CD4, CD8, PD-1, PD-L1, and TIM3 on day 7 and 14. f Percentage of CAR T cells that were positive for memory cell markers on day 7 and 14. Data represented technical triplicates using T cells from one donor and were displayed as mean ± SEM. *p ≤ 0.05, ns not significant.

Fig. 2. PD-L1-CAR T cells kill NSCLC cells in a PD-L1-dependent manner.

a Flow cytometry histogram of the surface antigen expression of PD-L1 in human NSCLC cell lines and BEAS-2B. b Cytotoxic activity of PD-L1-CAR T cells after 4 and 20 h of co-culture with human NSCLC cell lines and BEAS-2B. PD-L1-CAR T cells and CD19-CAR T cells were used as effector cells at various ratios of effector (E): target (T). c Secretion of cytokines analyzed by ELISA in supernatants obtained after a 20-h co-culture of effector and target cells at a 2:1 E:T ratio. Data represented technical triplicates using T cells from one donor and were shown as mean ± SEM. ****p ≤ 0.0001.

Fig. 3. PD-L1-CAR T cells inhibit the growth of human PD-L1high NSCLC in a xenograft model.

a Experimental design of the tumor xenograft model infused with PD-L1-CAR or CD19-CAR T cells. b.NSG mice were inoculated with 1.0 × 10⁶ H1975-Fluc cells and infused intravenously with 5 × 10⁶ PD-L1-CAR T cells or CD19-CAR T cells twice on day 7 and 0 (n = 5 mice per group). Bioluminescence imaging was used to assess tumor growth on day 7, 14, and 28 post tumor cell inoculation. c Bioluminescence kinetics of H1975-Fluc (n = 5 mice per group). d The percentage of PD-L1-positive cells within tumors. All cells were extracted from tumors of each treatment group on day 28 after inoculation and the expression of PD-L1 was evaluated by flow cytometry. e Representative IHC images of CD19-CAR or PD-L1-CAR T cell-treated NSCLC tumors for PD-L1 and Ki67. Scale bars, 100 µm. f Hematoxylin and eosin staining of tumors or organs on day 28. Scale bars, 200 µm. ****p ≤ 0.0001.

Fig. 4. Persistence of PD-L1-CAR T cells in mice.

a Experimental design of HCC827-Fluc tumors with PD-L1-CAR T-cell therapy and re-challenge. b Serial bioluminescence imaging of tumor progression and regression. c, d Bioluminescence kinetics of HCC827-Fluc tumors (n = 5 mice per group). For the CD19-CAR T cell control group, a separated cohort of NSG mice was challenged with HCC-827-Fluc cells at day 70.

Fig. 5. Enhanced tumor PD-L1 expression after irradiation treatment.

a Signal intensities of PD-L1 expression in cell lines treated with 5 Gy radiation as analyzed by flow cytometry. b Percentage of PD-L1-positive cells and cell viability in A549 cells treated with different doses of radiation for 24 or 48 h. c The effect of radiation treatment on anti-tumor efficacy of PD-L1-CAR T cells at different effector (E): target (T) ratios. Data represented technical triplicates using T cells from one donor and were shown as mean ± SEM. *p ≤ 0.05, **p ≤ 0.01, ****p ≤ 0.0001, ns not significant.

Fig. 6. Synergistic efficacy of irradiation and PD-L1-CAR T-cell therapy in a PD-L1 low NSCLC xenograft model.

a Experimental design of tumor cell xenograft model treated with CAR T cells and/or irradiation. b Serial bioluminescence imaging of tumor progression and regression in each group (n = 3 mice per group). c Bioluminescence kinetics of A549-Fluc (n = 3 mice per group) in each treatment group. d Representative IHC of PD-L1 in irradiation-treated NSCLC tumors. Scale bars, 100 µm. e Representative images of CD3 IHC in PD-L1-CAR T cell-treated and irradiation-treated NSCLC tumors. Scale bars, 100 µm. f Hematoxylin and eosin staining of tumors. Scale bars, 50 µm. *p ≤ 0.05.