Adaptive single-KIR+NKG2C+ NK cells expanded from select superdonors show potent missing-self reactivity and efficiently control HLA-mismatched acute myeloid leukemia

Abstract

Background: Natural killer (NK) cells hold great promise as a source for allogeneic cell therapy against hematological malignancies, including acute myeloid leukemia (AML). Current treatments are hampered by variability in NK cell subset responses, a limitation which could be circumvented by specific expansion of highly potent single killer immunoglobulin-like receptor (KIR)⁺NKG2C⁺ adaptive NK cells to maximize missing-self reactivity.

Methods: We developed a GMP-compliant protocol to expand adaptive NK cells from cryopreserved cells derived from select third-party superdonors, that is, donors harboring large adaptive NK cell subsets with desired KIR specificities at baseline. We studied the adaptive state of the cell product (ADAPT-NK) by flow cytometry and mass cytometry as well as cellular indexing of transcriptomes and epitopes by sequencing (CITE-Seq). We investigated the functional responses of ADAPT-NK cells against a wide range of tumor target cell lines and primary AML samples using flow cytometry and IncuCyte as well as in a mouse model of AML.

Results: ADAPT-NK cells were >90% pure with a homogeneous expression of a single self-HLA specific KIR and expanded a median of 470-fold. The ADAPT-NK cells largely retained their adaptive transcriptional signature with activation of effector programs without signs of exhaustion. ADAPT-NK cells showed high degranulation capacity and efficient killing of HLA-C/KIR mismatched tumor cell lines as well as primary leukemic blasts from AML patients. Finally, the expanded adaptive NK cells had preserved robust antibody-dependent cellular cytotoxicity potential and combination of ADAPT-NK cells with an anti-CD16/IL-15/anti-CD33 tri-specific engager led to near-complete killing of resistant CD45^dim blast subtypes.

Conclusions: These preclinical data demonstrate the feasibility of off-the-shelf therapy with a non-engineered, yet highly specific, NK cell population with full missing-self recognition capability.

Keywords: immunity, innate; immunotherapy, adoptive; killer cells, natural.

Figure 1

Selective expansion of single-self KIR⁺NKG2C⁺ adaptive NK cells. (A) NKG2C⁺ out of CD56^dim NK cells in a cohort of 202 healthy blood donors and in (B) KIR2DL1⁺KIR2DL3^- or KIR2DL1^-KIR2DL3⁺ out of NKG2C⁺CD56^dim NK cells in the 48 healthy blood donors with over >20% NKG2C⁺ NK cell subsets. (C) ADAPT-NK protocol design. Flow cytometry analysis of (D, E) EpCam⁺ K562-HLA-E feeders, CD14⁺ monocytes and CD56⁺ NK cells on days 0–11 and (F, G) NKG2C⁺, KIR2DL1⁺ KIR2DL3^- or KIR2DL1^- KIR2DL3⁺ (‘single self-KIR’), CD57⁺ and NKG2A⁺ among total CD56⁺ NK cells on day 0 and day 11. (H) NKG2C⁺ frequencies on day 11 where each dot represents a separate expansion of the same donor material. (I) Fold expansion in cell numbers of total and NKG2C⁺ ADAPT-NK cells on day 11. (A), n=202, (B), n= 48,) (E), n=6, (G-I), n=27–31, in G and I the median of each donor from 34 independent experiments is shown. (H) n=19 from 2–10 independent experiments. Statistical differences were tested using paired t-tests, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. KIR, killer immunoglobulin-like receptor; NK, natural killer; PBMC, peripheral blood monunuclar cells.

Figure 2

ADAPT-NK cells show a retained adaptive transcriptional signature with activation of effector programs. Extended phenotyping by flow cytometry and mass cytometry as well as CITE-Seq analysis. (A) Relative expression for assessed markers by mass cytometry shown on a t-SNE clustering of sampled events from all individuals (day 0 n=4, day 11 n=6). (B) Frequencies of marker expression or mean metal intensity depicted as median log2 fold change as compared with NKG2C^- day 0 from flow- and mass cytometry data. Significances in heatmaps are given as compared with NKG2C^- day 0. (C) UMAP embedding of the CITE-seq data colored based on CD56-biotin labeling for day 0 and day 11, cluster assignment (0–2) and expression of KLRC2 in those clusters. Dotplot showing expression of KLRC2 by cluster. (D) 177 downregulated (blue) and 37 upregulated (red) DE genes between KLRC2⁺ day 0 and KLRC2^- day 0 and (E) 176 downregulated (blue) and 221 upregulated (red) DE genes between KLRC2⁺ day 0 and KLRC2⁺ day 11 after filtering for log2FC>1.2. (F) Relative gene expression of selected adaptive genes. ‘Maintained’ indicates significance of KLRC2⁺ day 11 to KLRC2^- day 0 while non-significant to KLRC2⁺ day 0. (G) Topology and content of the protein-protein interaction (PPI) network driven by upregulated DE genes for day 11 KLRC2 ⁺ as compared with day 0 KLRC2 ⁺ NK cells. (H) Gene set enrichment analysis of the Steiner forest PPI network with genes that when identified in GO (BP) processes were highlighted (orange). (I) Topology and content of the PPI Steiner forest network driven by downregulated DE genes for day 11 KLRC2 ⁺ as compared with day 0 KLRC2 ⁺ NK cells. (J) GSEA of the PPI network with genes that when identified in GO (BP) processes were highlighted (teal). For flow cytometry and mass cytometry n=4–7, CITE-seq n=1. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001, in one-way analysis of variance (ANOVA) with Tukey’s correction. BP, biological process; GSEA, gene set enrichment analysis; DE; differentially expressed; GO; gene ontology; GSEA, gene set enrichment analysis; PPI, protein–protein interaction; t-SNE, t-distributed stochastic neighbor embedding; UMAP, uniform manifold approximation and projection.

Figure 3

ADAPT-NK cells are highly functional with predictable alloreactivity. (A) Functional analysis of NK cells by flow cytometry in terms of degranulation (CD107a) and IFN-γ production. (B) ADAPT-NK responses to 721.221 target cells with or without anti-CD20 (MabThera) addition, and to IL-12+IL-18 or PMA+Ionomycin stimulation. (C, D) ADAPT-NK responses to K562 or NALM-6 tumor target cells with varying expression levels of HLA-E. (E) Determination of HLA-C expression by means of KIR-Fc staining for K562 engineered with single chain β2m HLA-C1 or HLA-C2 dimers. (F) Responses of KIR2DL1⁺ KIR2DL3^- KIR2DL2^- NKG2A^- or KIR2DL1^- KIR2DL3⁺ KIR2DL2^- NKG2A^- ADAPT-NK cells to HLA-C1 or HLA-C2 K562 in the HLA-C/KIR matched and mismatched setting. (G) Target cell death judged by Dead Cell Marker and caspase-3 staining to determine freq. of remaining live target cells. (H) Frequency of live target cells after mixed target cell assays with ADAPT-NK cells in the HLA-C/KIR matched and mismatched setting. (I–K) The cytotoxic potential of individual NK cells in the HLA-C/KIR matched and mismatched setting was assessed in a microwell screening assay. (J) Average number of kills performed by the cytotoxic NK cells. (K) Corresponding fraction of cytotoxic NK cells killing >1 target. For (B) n=8 in two independent experiments, (D) n=5 in one independent experiment, (F–H) n=11 in three independent experiments with KIR2DL1⁺ and KIR2DL3⁺ subsets analyzed in all donors, (I–K) n=7 in four independent experiments. Statistical differences were calculated using paired t-tests (in B, D, J–K) and a one-way ANOVA with Sidak’s correction (F–H). *p<0.05, **p<0.01, ***p<0.001, p<0.0001. ANOVA, analysis of variance; KIR, killer immunoglobulin-like receptor; NK, natural killer; PMA, phorbol myristate acetate.

Figure 4

Efficient killing of HLA-C/KIR mismatched tumor cell lines and in vivo efficacy of expanded ADAPT-NK cells. (A–C) Determination of killing ability of ADAPT-NK cells over 48 hours against tumor cell lines of various HLA-C genotypes in the IncuCyte platform, in bold the expected KIR-Fc binding. n=4–9 donors, in at least two independent experiments for each target cell line. Data are displayed as mean (±SD) and significance is given between HLA-C/KIR mismatched and targets only. Scalebar represents 300 µM. (D) a representative example from two independent experiments of bioluminescence in NSG mice injected with luciferase tagged HL-60 4 days prior to injection with a flat-dose of HLA-C/KIR matched or mismatched ADAPT-NK cells and subsequent evaluation for 35 days, compiled data displayed as means in (E). Fifteen mice were used with n=2 ADAPT-NK cells. Four control mice were used. Statistical differences were tested using a one-way ANOVA with Sidak’s correction (B, C) or a two-way ANOVA followed by Tukey’s multiple comparison correction (E). *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. ANOVA, analysis of variance; KIR, killer immunoglobulin-like receptor; NK, natural killer.

Figure 5

ADAPT-NK cells show potent and specific alloreactivity against mismatched primary AML blasts and combination with CD33/IL-15/CD16 TriKE overcomes resistant blast subtypes. (A) Gating strategy for phenotyping of primary AML blast cells (CD45^dim) by flow cytometry, as well as specific target cell killing calculated by the change in Dead Cell Marker⁺ Caspase-3⁺ target cells in co-cultures with ADAPT-NK cells. (B) Specific target cell cytotoxicity of AML blasts in HLA-C/KIR -matched and -mismatched conditions at different E:T ratios. (C) Spearman correlation between the mean specific cytotoxicity against each blast sample at 5:1 E:T ratio and the HLA-E expression on the corresponding CD45^dim AML blasts. (D) Specific target cell killing of leukemic stem cells (LSC) (NKG2DL^-) and non-LSCs (NKG2DL⁺) in HLA-C/KIR mismatched conditions at multiple E:T ratios. (E) t-SNE plot of ADAPT-NK sensitive and resistant AML samples co-cultured at 1:1 with HLA-C/KIR mismatched ADAPT-NK cells with and without the addition of CD33/IL-15/CD16 TriKE (TriKE). Heatmap overlays expression of CD33 (top row) and NKG2DL (bottom row). (F) Mean specific cytotoxicity of different blast subsets at 1:1 E:T ratio with or without the addition of TriKE. (A–F) Primary AML samples n=7 and ADAPT-NK products n=8 (overall 14 HLA-C/KIR-matched and 14 HLA-C/KIR-mismatched interactions) in three independent experiments. Statistical differences were tested using a two-way ANOVA followed by Sidak’s multiple comparison correction (B, D), using Spearman’s correlation (C) or using one-way ANOVA followed by Sidak’s multiple comparison correction (F). *p<0.05, **p<0.01, ***p<0.001. AML, acute myeloid leukemia; ANOVA, analysis of variance; E:T, effector to target; KIR, killer immunoglobulin-like receptor; NK, natural killer; t-SNE, t-distributed stochastic neighbor embedding.