Memory-like natural killer cell and CD19-antibody based immunotherapy in combination with tyrosine-kinase inhibition of Ph(-like) acute lymphoblastic leukemia

doi:10.21203/rs.3.rs-4437299/v1

Download PDF

Article

Memory-like natural killer cell and CD19-antibody based immunotherapy in combination with tyrosine-kinase inhibition of Ph(-like) acute lymphoblastic leukemia

https://doi.org/10.21203/rs.3.rs-4437299/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Philadelphia-like acute lymphoblastic leukemia (Ph-like ALL) is a molecularly distinct tyrosine kinase-driven entity burdened with a high risk of relapsing disease and poor response toward combinatorial chemotherapy. Tyrosine kinase inhibitors (TKI) have been introduced into recent treatment protocols to improve the survival of patients with Ph-like ALL, yet preliminary data indicate subpar outcome. To advance treatment concepts for this genetic subtype of ALL, engineered antibody and CAR T-cell based immunotherapy has been proposed. Alternatively, the clinical use of natural killer cells endowed with a CAR or combined with leukemia epitope-directed antibodies is a conceivable strategy, which allows for an adoptive transfer of immune cells in an allogeneic setting with a low risk of graft-versus-host reaction. Here, we explored memory-like NK cells and Fc-enhanced CD19 antibody-dependent cell-mediated cytotoxicity (ADCC) in combination with TKI directed against in vitro models of kinase-driven leukemia. We demonstrate that the memory-like state of NK cells is determined by interleukin-mediated epigenetic reprogramming resulting in enhanced antileukemic effector functions. TKI differentially interfere with NK cell function and receptor repertoire, but CD19 antibody-mediated cytotoxicity operates ABL and JAK-independent allowing for simultaneous administration of memory-like NK cells, CD19 antibody, and specific TKI with high treatment efficacy in vitro.

Biological sciences/Cancer/Tumour immunology

Health sciences/Medical research/Preclinical research

Philadelphia chromosome-like acute lymphoblastic leukemia (Ph-like or BCR-ABL-like ALL) is a distinct high-risk subtype of ALL that was initially identified by a gene expression signature similar to Philadelphia chromosome positive (Ph⁺) ALL in the absence of a BCR-ABL rearrangement (1). In lieu of BCR-ABL, other fusion genes involving various cytokine receptors or tyrosine kinases have been identified which can be broadly subdivided into JAK and ABL-class alterations (2, 3).The prevalence of Ph-like ALL is age-dependent and varies from 10% in childhood ALL to almost 30% in adults (3, 4). Importantly, Ph-like NCI high-risk ALL is associated with a poor prognosis requiring intensified chemotherapy and/or hematopoietic stem cell transplantation (HSCT) (1, 3–5).

Kinase-activating alterations in Ph-like ALL are amenable to a targeted treatment approach using ABL1/2 inhibitors in ABL-class fusions and JAK1/2 inhibitors in JAK-STAT-activating lesions. Several ongoing trials initiated by the Children’s Oncology Group (COG) (NCT01406756, NCT02723994), St. Jude Children’s Research Hospital (SJCRH) (NCT03117751, NCT03117751) in the United States and the ALL-Together Trial Consortium in Europe (EudraCT-No. 2018-001795-38) evaluate the impact of these targeted interventions in children and young adults at primary diagnosis of Ph-like ALL. Overall, tyrosine kinase inhibitors (TKI) appear to improve remission rates and outcome of various ABL-class ALL, whereas JAK-class mutated ALL likely require a more complex intervention directed against multiple signaling pathways (6, 7). Notwithstanding, high treatment intensity of combination chemotherapy and HSCT as well as the emergence of resistance to kinase inhibition impose limits to concepts of precision-medicine integrated into a chemotherapy backbone.

Hence, it is plausible to explore mutation-directed TKI also in the context of cellular and antibody-based immunotherapy. Beyond established cytotoxic T-cells that express B-cell epitope-directed chimeric antigen receptors (CAR T-cells), an adoptive natural killer cell-based immunotherapy in combination with TKI could be a complementary approach toward Ph-like ALL, since NK cells have proven tumor cell killing activity (8, 9). The adoptive transfer of allogeneic NK cells from various sources (peripheral or umbilical cord blood, induced pluripotent stem cells) is associated with a low risk of graft-versus-host reaction due to a lack of TCR expression and is capable of inducing remissions in patients with leukemia (9–11). To increase functionality, NK cells can be endowed with a CAR in analogy to T-cells or combined with a therapeutic B-cell epitope-targeted antibody (12–14). Generation of an adaptive or memory-like state of NK cells by interleukin-dependent pre-activation also results in enhanced cytotoxicity, expansion and long-term persistence in vivo (15–17). Adaptive memory formation in NK cells has been linked to epigenetic re-programming, but its mechanistic underpinnings have not been clearly defined yet. In particular, the effect of a memory-like cellular state on antibody-dependent cell-mediated cytotoxicity (ADCC) and the potential interference by TKI have not been elucidated. TKI has been reported to impact hematopoiesis including NK cell development, repertoire and effector function depending on target selectivity potentially affecting SYK/Pi3K-mediated ADCC (18, 19). Notably, the multi-target TKI dasatinib switches off CARs containing either CD28-CD3ζ or 4-1BB-CD3ζ activation modules by interference with lymphocyte-specific protein tyrosine kinase (LCK) (20). Hence, a preemptive investigation of potential interferences of selected kinase inhibitors with cellular and antibody-based immunotherapy is mandatory to optimize treatment efficacy.

Herein, we set out to explore the interplay between memory-like NK cells, CD19-mediated ADCC and TKI as a novel combinatorial treatment approach toward Ph-like ALL.

Cell lines

Human BCR-ABL⁺ K562 CML (subclone ACC86: MHC class I negative) cells were obtained from the Leibniz Institute DSMZ (Braunschweig, Germany) and gene-modified human K562 cells expressing membrane-bound IL-21/4-1BB ligand (mbIL21/4-1BBL) were kindly provided by Dr. Dean Lee (Center for Childhood Cancer Research Institute, Nationwide Children's Hospital, Columbus, OH, USA). Both cell lines were cultured at 0.5 x 10⁶ cells/mL in RPMI media supplemented with 10% fetal calf serum (FCS) and incubated at 37°C and 5% CO₂. Human CRLF2/JAK2_mut MHH-CALL-4 and BCR_p190-ABL SUP-B15 cells were also obtained from DSMZ and cultured in RPMI media supplemented with 20% FCS at 37°C and 5% CO₂.

Peripheral blood and bone marrow-derived natural killer cells

Peripheral blood mononuclear cells (PBMCs) were isolated from blood samples of anonymous CMV-negative healthy donors collected at the blood bank of the University Medical Center Hamburg-Eppendorf. Peripheral blood-derived natural killer cells (PBNKC) were purified using an NK cell isolation kit (STEMCELL Technologies, Cologne, Germany) or enriched in co-culture with mbIL-21/4-1BBL K562 cells (see Supplementary Information). Anonymized bone marrow samples from a healthy individual and patients with B-precursor ALL were obtained from the biobank of the Dept. of Pediatric Oncology at the University Medical Center Hamburg-Eppendorf. The collection and use of bio-banked human specimens for research purposes have been approved by competent authorities (Ethical Board Medical Association Hamburg; PV5777, 2018).

Reagents

Tyrosine kinase inhibitors imatinib (Cell Signaling Technology, Leiden, the Netherlands), ruxolitinib (MCE MedChemTronica, Sollentuna, Sweden), and dasatinib (STEMCELL Technologies, Cologne, Germany) were purchased and EC50 and LC50 of target cells were tested prior to use. In K562 co-culture assays ruxolitinib and imatinib were used at a concentration of 1 µM vs. 0.1% dimethyl sulfoxide (DMSO). MHH-CALL-4 cells were treated with 1 µM ruxolitinib. SUP-B15 cells were exposed to 0.5 or 1 µM imatinib or 100 nM dasatinib. Recombinant interleukins (IL-2, IL-15, IL-18, IL-12) were obtained from PEPROTECH (Hamburg,

Deutschland), R&D system (Minneapolis, USA) and InvivoGEN (Toulouse, France) and used as indicated. Fc-engineered CD19-antibody (designated CD19-DE) has been generated in-house (21, 22).

NK cell cytotoxicity assay

Serial immune cell killing assays were performed on an IncuCyte SX5 live cell imaging platform using CellEvent caspase 3/7 green detection reagent and CellTrace™ far red cell proliferation kit (both obtained from ThermoFisher, Waltham, USA) according to manufacturer’s instructions. For details refer to Supplementary Information.

Flow cytometry analysis and single cell sorting

Cell surface receptor and cytotoxic marker expression of expanded PBNK cells were examined by multiparametric flow cytometry using BD Fortessa and Cytek Aurora FACS instruments. FACS single cell sorting was performed using a BD Aria Fusion device. Cells isolated from bone marrow were sorted based on markers co-expressed on B-precursor ALL cells. For antibody panels, source and technical details refer to Supplementary Information.

DNA extraction and genome-wide methylation analysis

A pure population of NK cells was obtained by sorting the co-cultured cells based on NK cell markers as described above (Supplementary Fig. 1). DNA was extracted from the sorted NK cells using a NucleoSpin kit (Macherey-Nagel, Düren, Germany). DNA was bisulfite-treated and converted with the NucleoSpin Tissue kit according to manufacturer’s instructions. Genome-wide methylation was performed using Infinium Methylation arrays (850k) (Illumina) as described in Supplementary Information.

RNA sequencing

RNA sequencing was done using a SMART-Seq mRNA kit (Takara, Saint-Germain-en-Laye, France) for cDNA synthesis and Nextera XT kit (Illumina, Berlin, Germany) for library preparation. Quality controlled libraries were sequenced on an Illumina NextSeq2000 in a paired end mode (2x61bp). For further details refer to Supplementary Information.

Single cell RNA sequencing

Gene expression libraries were generated using the 10X Chromium Single Cell 3' Reagent Kit v3 (10x GENOMICS, CA, USA), with 3,500 cells per sample. Quality control was achieved using Agilent Bioanalyzer High Sensitivity DNA Assay. Sequencing was performed on an Illumina NextSeq 500 machine, and demultiplexing and initial downstream analysis were performed using the 10X cellranger analysis pipeline.

Statistical analysis

All data were statistically analyzed by GraphPad Prism 10.2.0. Student’s t test or Mann-Whitney test was used to compare two groups. One-way analysis of variance (ANOVA) was used to compare multiple groups. Differences were considered significant at p < 0.05. Data are presented as mean ± standard error of the mean (SEM). For further details refer to Supplementary Information.

Cytokine-induced pre-activation of NK cells induces epigenetic re-programming

by differential DNA-methylation

Epigenetic re-programming has been proposed to prime NK cells for an adaptive immune response as shown upon CD2 and NKp46 receptor engagement (23). Moreover, a discontinuous pre-activation of NK cells with IL-12, IL-15, and IL-18 evokes a memory-like state characterized by enhanced immune effector functions upon subsequent target cell exposure (11, 17, 24, 25). To gain a more comprehensive insight into the cytokine-induced memory-like state of NK cells, we investigated the effect of activating cytokines on genome-wide DNA methylation in peripheral blood-derived NK cells (PBNKC) after antibody-bead isolation or upon expansion in a 7-day co-culture with irradiated K562 cells expressing membrane-bound IL-21 and co-stimulatory 4-1BB ligand (mbIL-21/4-1BB). To induce a memory-like state, either NK cell batch was stimulated with IL-12, IL-15, and IL-18 for 16 hours and compared with non-preactivated magnetically isolated or mbIL-21/4-1BB K562 co-cultured and expanded NK cells (Supplementary Fig. 1) (11). Pre-activation was confirmed by an increased expression of interferon-γ (IFN-γ) and the degranulation marker CD107a in IL-12/-15/-18-treated NK cells upon short-term exposition toward unmodified K562 cells (Supplementary Fig. 1). FACS sorted NK cells underwent genome-wide DNA methylation analyses, which revealed differential methylation patterns as demonstrated by unsupervised hierarchical clustering. Magnetic bead-isolated NK cells clearly separated into preactivated and non-preactivated clusters in contrast to mbIL21/4-1BB K562 expanded NK cells which exhibited methylation patterns largely mimicking epigenetic features of preactivated magnetic bead-isolated PBNK cells (Fig. 1A, B; Supplementary Fig. 2). Thus, IL-12, IL-15, and IL-18 induced pre-activation as well as mbIL21/4-1BB K562 co-culture resulted in hypomethylation of CpG regions at multiple common gene loci indicative of a robust epigenetic response generating a transcriptionally permissive chromatin state at specific sites. Overall, we observed n = 307 differentially methylated genes (DMG) (applying following filter criteria: |deltaBeta| > 0.2, p-value < 0.05, feature != “IGR”). DMGs comprised various functional categories such as immune effectors (CD274/PDL1, HIVEP2, CBLB, ICOS, NFAT2C), inflammatory chemokine receptor and stress-response signaling (CCR5, ZAK, MAP3K7IP2/TGFB-activated kinase, TNFRSF8, AOAH, PDE4B, TP53BP2), protein biogenesis, trafficking and folding (BCAT1, ATE1, RAB29, SYTL3, SGMS1, FCHSD2, PDIA6, SGMS1), glycosylation-dependent immune cell homeostasis (ST3GAL1, FUT8, B4GALT5), transcription and splicing (JAZF1, MED1, TTF2, PSPC1, TRERF1, PRPF33, RUNX3, ZEB2), peroxisome and energy metabolism (PEX2; ATP5S, CMC1), cytoskeleton/migration/adhesion (DOCK10, COTL1, SPECC1L), and lncRNAs with miscellaneous functions (FLJ22447, TPRG1-AS1, DLEU1) among others. As a distinctive feature of magnetic bead-isolated PBNKC we observed a pronounced differential hypomethylation upon IL12/15/18-preactivation at the ATP receptor ion channel P2RX7 gene, the BRISC and BRCA1-A complex member 2 (BRE) gene, GTPase activator gene ARHGAP26, and lncRNA FLJ22447 in contrast to baseline hypomethylation of the cell-cell adhesion gene TENM2 under experimental control conditions. In aggregate, a transcriptionally permissive state at multiple gene loci coding for distinct but convergent functional effector proteins warrants IL-induced memory-like features of NK cells accounting for enhanced immune effector activity upon later exposure toward target cells.

Epigenetic reprogramming determines transcriptional output of preactivated NK cells

To decipher the transcriptional activities of preactivated vs. canonical PBNKC upon K562 target cell exposure we performed RNA-sequencing of single cell FACS-sorted NKC. IL12/15/18-induced pre-activation of bead-isolated PBNKC (marked as PBNKC_i vs. non-preactivated PBNKC) resulted in n = 1643 differentially expressed genes (DEGs) (n = 835 upregulated; n = 808 downregulated DEGs) upon short-term exposure toward K562 target cells (FDR ≤ 0.1; log₂ f.c.≥1.0) (Fig. 2A). Gene set enrichment analysis (GSEA) of DEGs showed a positive enrichment of the following Gene Ontology (GO) terms among others: Ribosome and mitochondrial function, immune system process, cell adhesion and motility (Fig. 2B). Under identical experimental conditions, IL12/15/18-dependent pre-activation of mbIL21/4-1BB K562-expanded NK cells (PBNKC_ei) yielded only n = 229 DEGs (n = 138 upregulated; n = 91 downregulated DEGs) compared to mbIL21/4-1BB K562 co-culture expanded NK cells (PBNKC_e), which had not undergone prior pre-activation with IL12/15/18 (Fig. 2A). Despite the smaller number of DEGs, enrichment analysis identified very similar GO terms in comparison to bead-isolated NK cells yielding the highest normalized enrichment scores for ribosome and ribosome biogenesis (Fig. 2B). By contrast, single cell RNA-seq of leukemia-infiltrating NK (LINK) cells at primary diagnosis of B-cell precursor ALL revealed a marked downregulation of multiple ribosomal biogenesis transcripts suggesting that IL-mediated pre-activation of NK cells could potentially compensate for the observed translational deficiency of LINK cells (Fig. 2C; Supplementary Fig. 3). Overall, IL12/15/18-induced pre-activation of bead-isolated PBNKC_i or mbIL-21/4-1BB K562 co-culture of PBNKC_e lead to robust epigenetic reprogramming and differential transcriptional output. Hereby, PBNKC_i share many epigenetic and transcriptional features with PBNKC_e indicating that a memory-like state of NK cells can also be instilled under sole mbIL21/4-1BB K562 co-culture conditions (Fig. 1A; Supplementary Fig. 2). An integrated analysis of differential gene methylation and expression of NK cells under the different experimental conditions outlined above is depicted in Fig. 2D, E. Upon pre-activation, the number of differentially methylated and expressed genes (DMEGs) in bead-isolated PBNKC is significantly greater than the one in mbIL21/4-1BB K562-expanded PBNKC (Fig. 2D, E). Overall, DEGs clearly outnumbered DMGs in bead-isolated and mbIL21/4-1BB K562-co-cultured PBNKC, although only a fraction of DMGs displayed a robust transcriptional change compatible with transcriptionally permissive, but inactive chromatin sites. The greater number of DEGs compared to DMGs might be explained by a multilayered transcriptional response of a rewiring gene regulatory network driven by translated DMEGs such as DNA binding factors (ZEB2, RUNX3, TRERF1) and signaling effectors (PD-L1/CD274, CCR5, ZAK, TNFRSF8, ICOS) as shown for select cell surface and cytoplasmic proteins by flow cytometry (Fig. 2F, G; Supplementary Fig. 4). Beyond DMEGs, analysis of transcription factor target genes revealed ZNF263, MYC, NFKB, STATs and other DNA binding factors (Supplementary Fig. 5). Intriguingly, PR domain zinc finger protein 14 gene (PRDM14) involved in re-acquisition of pluripotency is consistently downregulated upon IL-induced pre-activation (Supplementary Fig. 5) (26).

Using bead-isolated canonical PBNKC as a reference, we performed a crosswise comparison of differential gene regulation in PBNKC_i, PBNKC_ei, and PBNKC_e which demonstrated a substantial overlap of gene activities reflecting PD-1-signaling, IFN-signaling, MHC class II antigen presentation, allograft rejection, MTORC1-induced growth and TNF-NFKB mediated survival signaling, among other functional categories which are in accordance with earlier observations (27–29) (Fig. 3A-D). Among PD-1-signaling enriched genes, PD-L1 (CD274) as well as PD-L2 (PDC1LG2) were upregulated in IL-preactivated PBNKCs (Fig. 3E, F). MHC-class II antigen expression is associated with an upregulation of X-box binding RFX5 which binds at MHC class II gene promoters in cooperation with NF-Y (Supplementary Fig. 5). As a non-overlapping exclusive feature, PBNKC_ei exhibited an enhanced IFN- α, β, and γ-signaling which might be associated with further augmented levels of effector function linked to downregulation of the lymphotactin XCL1 and the IFN-signaling regulator NLRC5 (Fig. 3D, G, H). In summary, largely overlapping epigenomic and transcriptional patterns of IL12/15/18-preactivated bead-isolated and IL21/4-1BB K562 co-culture expanded NK cells indicate similar memory-like states and improved functional outcomes prompting us to proceed with mbIL21/4-1BB K562 co-culture expanded NK cells (hereafter denoted as memory-like (ML-)PBNKC_e or (ML-)PBNKC_ei upon additional IL12/15/18-preactivation). Notably, gene activities of ML-PBNKC_e analyzed by RNA-seq are at least partially congruent with previously mass cytometry-defined memory-like features of IL12/15/18-preactivated PBNKC reflecting the multilayered regulation of memory formation and adaptive response as described here (Supplementary Fig. 6) (11).

Tyrosine kinase inhibition interferes with memory-like NK cell effector function

To explore the feasibility of a combinatorial phenotype and genotype-directed approach toward Ph-like ALL, we first investigated NK cell cytotoxicity of BCR-ABL rearranged K562 target cells in the presence vs. absence of ruxolitinib or imatinib which inhibit JAK or ABL-dependent signaling pathways, respectively. To this end, we set up serial killing experiments including ML-PBNKC_e and ML-PBNKC_ei effector cells and K562 target cells under treatment with TKI compared to DMSO-treated control cells (Supplementary Fig. 7). Cell death was monitored in real-time over a period of 72 hours by live cell imaging using caspase 3/7 labeling. During the initial 24-hour period, ML-PBNKC_ei cells demonstrated a more rapid and greater cytotoxicity compared to ML-PBNKC_e clearly discernible at an E:T ratio of 5:1 (60.2%, SD ± 8.5% vs. 44.9%, SD ± 6.1%; p < 0.05), but less evident at an E:T ratio of 1:1 without a significant effect of TKI on NKC-mediated killing efficacy except for a slight initial delay under ruxolitinib treatment (Fig. 4A, B; Supplementary Fig. 7). After 48 hours of K562-replenished co-culture, solvent-treated ML-PBNKC_ei and ML-PBNKC_e showed an incremental increase in killing efficacy likely due to a training effect achieving 68.7% (SD ± 11.1%) and 70.1% (SD ± 9.5%) at the 60-hour time point in comparison to 50.4% (SD ± 15.2%) and 55.4% (SD ± 14.8%) cell death rate during the initial 24 hours of the time course experiment (Fig. 4C). However, prolonged TKI treatment exerted detrimental effects on NKC function: Imatinib treatment abrogated the observed enhancement of cytotoxicity over time attaining 55.6% (SD ± 18.4%) and 57.2% (SD ± 19.3%) cell death rate after 60 hours with ML-PBNKC_ei compared to ML-PBNKC_e. Ruxolitinib treatment was associated with a more pronounced reduction in K562-directed cytotoxicity of NKC exhibiting a death rate of 21.8% (SD ± 7.9%) and 24% (SD ± 10.3%) at the 60-hour time point using ML-PBNKC_ei vs. ML-PBNKC_e (Fig. 4B). Imatinib treatment alone resulted in a maximum of 11.2% K562 cell death over background during a period of 24 hours, whereas ruxolitinib did not yield any discernable cytotoxicity in the absence of NKC in previous K562-targeted serial killing experiment (Fig. 4A).

To elucidate the activity profile of NKC under the experimental conditions of serial K562 cell killing, we performed multiparametric flow cytometry analyses of various NKC receptors and cytosolic effector proteins (Fig. 4D-G; Supplementary Fig. 8). In accordance with enhanced transcriptional activities of IFN-signaling in PBNKC_ei, we initially observed a greater IFN-γ protein expression level in ML-PBNKC_ei in comparison to ML-PBNKC_e. The increased expression of IFN-γ in ML-PBNKC_ei decreased after 48-hour K562 co-culture in the absence of TKI. Under treatment with ruxolitinib, the expression of IFN-γ was completely abrogated independent of IL12/15/18 pre-activation, whereas imatinib treatment was associated with a significantly lower but still discernible IFN-γ activity in ML-PBNKC_ei (Fig. 4D). Among multiple other NKC effectors, NKp30 which is activated by BAG6 and NCR3LG1 binding showed a marked downregulation under ruxolitinib in 48-hour K562 cell co-culture (30, 31). NKp30 expression was not rescued by IL12/15/18 pre-activation (Fig. 4E). The activating and co-stimulatory NKG2D receptor showed a similar trend towards reduced expression in the presence of ruxolitinib (Supplementary Fig. 8). Unexpectedly, the cytotoxicity activating receptors NKG2C and NKp46 exhibited an inverse behavior with increased expression in the presence of ruxolitinib in ML-PBNKC_e and ML-PBNKC_ei alike (Fig. 4F, G). The inhibitory receptor NKG2A revealed high expression levels under any experimental condition applied herein (Supplementary Fig. 8). In contrast to ruxolitinib, imatinib did not discernibly affect the expression of NK cell receptors (Fig. 4D-G; Supplementary Fig. 8). Importantly, the FcγRIIIa receptor (CD16a) mediating ADCC maintained robust expression levels which were only slightly reduced upon K562 cell co-culture independent of TKI and IL-coactivation (Supplementary Fig. 8). Together, the observed effects of JAK and to a lesser extent ABL-class-directed TKI on NKC effector protein expression result in impaired cytotoxicity of K562 target cells, which might pose a significant obstacle to the concomitant adoptive transfer of NKC and administration of TKI in a clinical setting.

CD19 antibody-dependent cellular cytotoxicity rescues the TKI-associated impairment of NK cell effector functions

To potentially overcome the inhibitory effects of TKI on NKC cytotoxicity we investigated the combinatorial use of a humanized Fc-enhanced CD19-targeting mAb, designated CD19-DE, an in-house generated analogue of tafasitamab featuring ADCC and antibody-dependent cellular phagocytosis (ADCP) (14, 21, 22). We reasoned that a sustained CD16 expression under JAK or ABL-directed TKI treatment could potentially be exploited for ADCC mediated by non-targeted spleen tyrosine kinase (SYK) and Pi3K upon FcγRIIIa receptor phosphorylation leading to granule release and target cell killing (32). To test this hypothesis, we assessed the cytotoxicity of CRLF2/JAK_mut MHH-CALL-4 and BCR-ABL⁺ SUP-B15 cell lines, the latter of which served as a model system in the absence of an established ABL-class Ph-like ALL cell line. We applied various treatment conditions including ML-PBNKC_e or ML-PBNKC_ei (E:T ratio of 5:1), TKI, and CD19-DE compared to DMSO-treated samples (Fig. 5A-C; Supplementary Fig. 9). Strikingly, ML-PBNKC_ei directed against MHH-CALL-4 cells combined with the CD19-DE exhibited the highest and most rapid killing rate (mean = 88.3%, SD ± 4.0%) vs. ML-PBNKCe (mean = 78.6%, SD ± 6.5%; p < 0.05), which was not discernibly affected by ruxolitinib during the first 24 hours of co-culture indicative of JAK-pathway independent ADCC (Fig. 5A; Supplementary Fig. 9). By contrast, in the absence of CD19-DE, ML-PBNKC_ei and ML-PBNKC_e cytotoxicity of MHH-CALL-4 decreased to 56.4% (SD ± 11.6%) and 39.4% (SD ± 12.9%), respectively. Under those experimental conditions, co-administration of ruxolitinib resulted in an incremental reduction of cytotoxicity over time. Under a treatment regimen with ruxolitinib and/or CD19-DE devoid of ML-PBNKC, MHH-CALL-4 cell death did not significantly differ from background cytotoxicity (Fig. 5A; Supplementary Fig. 9).

The eminent efficacy and synergy of ML-PBNKC_ei and to a slightly lesser extent ML-PBNKC_e in conjunction with CD19-DE was recapitulated in BCR-ABL⁺ SUP-B15 cells, albeit peak cytotoxicity was lower and delayed attaining a maximum of 65.6% (SD ± 20.4% at 12 hours) cell death with greater variability in comparison with MHH-CALL-4 cells (Fig. 5A, B; Supplementary Fig. 9). As anticipated, a distinctive feature of SUP-B15 cells is the sensitivity toward imatinib which showed incremental additive effects in combination with ML-PBNKC, but those were clearly inferior to the synergistic action of ML-PBNKC in combination with CD19-DE. In stark contrast, co-administration of the multi-target TKI dasatinib resulted in a significantly lower cytotoxicity of ML-PBNKC_ei and ML-PBNKC_e which could not be compensated by CD19-DE indicative of an interference of dasatinib with ADCC-dependent signaling pathways (Fig. 5C). Finally, we analyzed ML-PBNKC_ei in combination with CD19-DE in serial killing experiments comparing cytotoxicities of MHH-CALL-4 and SUP-B15 in dependence on the duration of TKI exposure (Fig. 5D). Prolonged exposure of ML-PBNKC_ei toward ruxolitinib in MHH-CALL-4 co-culture resulted in a temporarily delayed cytotoxicity which was fully compensated after 72 h. By contrast, prolonged treatment with imatinib did not discernibly affect the cytotoxicity of SUP-B15 under combined usage of ML-PBNKC_ei and CD19-DE. In accordance with the above inhibitory effects of dasatinib on the initial ML-PBNKC and CD19-DE mediated cytotoxicity of SUP-B15 cells, prolonged exposure toward this multi-target TKI caused a marked reduction in cytotoxic efficacy of immunotherapy during the entire experimental time course (Fig. 5C, D). Together, these findings suggest a compatibility of memory-like NKC and CD19-DE based immunotherapy with carefully selected TKI intervention tailored to the specific signaling requirements of cellular and antibody-based immunotherapies.

The implementation of novel immunotherapeutic concepts into the backbone of chemotherapy and TKI-based Ph-like ALL trials is supported by anecdotal reports on CAR T-cell efficacy in these patients and encouraging data on blinatumomab plus dasatinib for adult Ph⁺ ALL (33, 34). Herein, we explored the potential of memory-like NK cells in combination with TKI and a humanized Fc-engineered CD19-antibody. An integrative analysis of interleukin-mediated genome-wide DNA methylation and transcriptional regulation identified key effectors of memory formation such as PD-L1/CD274, CCR5, TNFRSF8, ICOS among others. PD-L1 has been previously shown to be transactivated by p65 (NFkB) in NK cells which has been identified in ORA and GSEA of K562-expanded and/or IL12/15/18 pre-activated NK cells (Fig. 3). A PD-1-independent mechanism of antitumor activity of PD-L1⁺ NK cells has recently been identified by usage of an anti-PD-L1 mAb resulting in a cytolytic effector phenotype rather than immune exhaustion (35). As previously reported, TKI had differential inhibitory effects on NK cell function reflected by decreased cytotoxicity which was most pronounced under prolonged exposure to ruxolitinib (18, 19). Importantly, a robust expression of CD16 under TKI treatment is a prerequisite for the sustained efficacy of ADCC which primarily signals via SYK not targeted by specific ABL or JAK inhibitors. However, ADCC signaling is negatively affected by dasatinib likely targeting upstream SRC kinase family members such as LCK (Fig. 5C). In addition, the observed decrease in NKp30 surface expression under ruxolitinib indicates that the integration of bi-specific NK cell engager antibodies in future immunotherapeutic concepts involving TKI will require a detailed analysis of activating and inhibitory NK cell surface receptors to identify robust target epitopes (Fig. 4E).

The in vitro efficacy of memory-like NK cell and antibody-based immunotherapy in combination with TKI for Ph-(like) ALL as demonstrated herein warrants further exploration. To overcome the very limited availability of in vitro models of Ph-like ALL, next generation treatment concepts should be tested in a variety of accessible patient-derived xenografted mouse models of ABL- and JAK class Ph-like ALL.

Competing interests

The authors declare no competing financial interests in relation to the work described herein.

Author contributions

ZE and RH were responsible for experimental design and analyses of NK cell activities. Both contributed equally to writing the report and providing feedback on the final draft of the manuscript. MS performed all bioinformatic analyses of methylome, bulk and single cell RNA-seq data, curated and uploaded data into repositories, and revised methods and figure legends. PN performed ingenuity pathway analyses using scRNA-seq data. DW and KK generated and characterized essential immunotherapeutic tools used herein. HG, CI, SS, ALH, MG, KR, GWE, and JS performed NK cell cytotoxicity, immune effector function, metabolome, and other experiments. US supervised methylome analyses, MP supervised antibody studies, and MAH conceptualized and supervised the research project and wrote the manuscript together with ZE and RH. All authors read, revised, and approved the manuscript.

Acknowledgements

This work has been funded by the German Research Foundation (DFG Ho2176/6 − 1) and J.J. Ganzer Foundation to MAH. RH and DW were supported by a joint grant from University Cancer Centers Hamburg and Schleswig-Holstein, the Christian Albrecht Universität Kiel (Medical Faculty) & German Cancer Aid (No. 70113304). MP and KK are funded by the German Research Foundation (project number 444949889 (KFO 5010/7 Clinical Research Unit “CATCH ALL”). We thank all team members of the Horstmann lab, next generation sequencing and flow cytometry units at the Leibniz Institute of Virology (LIV) Hamburg for fruitful discussions and technical support. We are also very grateful for technical advice from members of the Altfeld lab at the LIV.

Data Availability Statement

Methylome and sequencing data are available through Gene Expression Omnibus under accessions GEO NN.

Den Boer ML, van Slegtenhorst M, De Menezes RX, Cheok MH, Buijs-Gladdines JGCAM, Peters STCJM, et al. A subtype of childhood acute lymphoblastic leukaemia with poor treatment outcome: a genome-wide classification study. Lancet Oncol 2009;10(2):125–34.
Roberts KG, Morin RD, Zhang J, Hirst M, Zhao Y, Su X, et al. Genetic alterations activating kinase and cytokine receptor signaling in high-risk acute lymphoblastic leukemia. Cancer Cell 2012;22(2):153–66.
Roberts KG, Li Y, Payne-Turner D, Harvey RC, Yang YL, Pei D, et al. Targetable kinase-activating lesions in Ph-like acute lymphoblastic leukemia. N Engl J Med 2014;371(11):1005–15.
Roberts KG, Gu Z, Payne-Turner D, McCastlain K, Harvey RC, Chen IM, et al. High Frequency and Poor Outcome of Philadelphia Chromosome-Like Acute Lymphoblastic Leukemia in Adults. J Clin Oncol 2017;35(4):394–401.
van der Veer A, Waanders E, Pieters R, Willemse ME, Van Reijmersdal SV, Russell LJ, et al. Independent prognostic value of BCR-ABL1-like signature and IKZF1 deletion, but not high CRLF2 expression, in children with B-cell precursor ALL. Blood 2013;122(15):2622–9.
Tanasi I, Ba I, Sirvent N, Braun T, Cuccuini W, Ballerini P, et al. Efficacy of tyrosine kinase inhibitors in Ph-like acute lymphoblastic leukemia harboring ABL-class rearrangements. Blood 2019;134(16):1351–5.
Moorman AV, Schwab C, Winterman E, Hancock J, Castleton A, Cummins M, et al. Adjuvant tyrosine kinase inhibitor therapy improves outcome for children and adolescents with acute lymphoblastic leukaemia who have an ABL-class fusion. Br J Haematol 2020;191(5):844–51.
Maude SL, Laetsch TW, Buechner J, Rives S, Boyer M, Bittencourt H, et al. Tisagenlecleucel in Children and Young Adults with B-Cell Lymphoblastic Leukemia. N Engl J Med 2018;378(5):439–48.
Handgretinger R, Lang P, André MC. Exploitation of natural killer cells for the treatment of acute leukemia. Blood 2016;127(26):3341–9.
Miller JS, Soignier Y, Panoskaltsis-Mortari A, McNearney SA, Yun GH, Fautsch SK, et al. Successful adoptive transfer and in vivo expansion of human haploidentical NK cells in patients with cancer. Blood 2005;105(8):3051–7.
Romee R, Rosario M, Berrien-Elliott MM, Wagner JA, Jewell BA, Schappe T, et al. Cytokine-induced memory-like natural killer cells exhibit enhanced responses against myeloid leukemia. Sci Transl Med 2016;8(357):357ra123.
Liu E, Tong Y, Dotti G, Shaim H, Savoldo B, Mukherjee M, et al. Cord blood NK cells engineered to express IL-15 and a CD19-targeted CAR show long-term persistence and potent antitumor activity. Leukemia 2018;32(2):520–31.
Li Y, Hermanson DL, Moriarity BS, Kaufman DS. Human iPSC-Derived Natural Killer Cells Engineered with Chimeric Antigen Receptors Enhance Anti-tumor Activity. Cell Stem Cell 2018;23(2):181–192.e5.
Horton HM, Bernett MJ, Pong E, Peipp M, Karki S, Chu SY, et al. Potent In vitro and In vivo Activity of an Fc-Engineered Anti-CD19 Monoclonal Antibody against Lymphoma and Leukemia. Cancer Res 2008;68(19):8049–57.
O’Leary JG, Goodarzi M, Drayton DL, von Andrian UH. T cell– and B cell–independent adaptive immunity mediated by natural killer cells. Nat Immunol 2006;7(5):507–16.
Paust S, von Andrian UH. Natural killer cell memory. Nat Immunol 2011;12(6):500–8.
Romee R, Schneider SE, Leong JW, Chase JM, Keppel CR, Sullivan RP, et al. Cytokine activation induces human memory-like NK cells. Blood 2012;120(24):4751–60.
Damele L, Montaldo E, Moretta L, Vitale C, Mingari MC. Effect of Tyrosine Kinase Inhibitors on NK Cell and ILC3 Development and Function. Front Immunol 2018 Oct 23:9:2433.
Salih J, Hilpert J, Placke T, Grünebach F, Steinle A, Salih HR, et al. The BCR/ABL-inhibitors imatinib, nilotinib and dasatinib differentially affect NK cell reactivity. Int J Cancer 2010;127(9):2119–28.
Mestermann K, Giavridis T, Weber J, Rydzek J, Frenz S, Nerreter T, et al. The tyrosine kinase inhibitor dasatinib acts as a pharmacologic on/off switch for CAR T cells. Sci Transl Med 2019;11(499):eaau5907.
Schewe DM, Alsadeq A, Sattler C, Lenk L, Vogiatzi F, Cario G, et al. An Fc-engineered CD19 antibody eradicates MRD in patient-derived MLL-rearranged acute lymphoblastic leukemia xenografts. Blood 2017;130(13):1543–52.
Gehlert CL, Rahmati P, Boje AS, Winterberg D, Krohn S, Theocharis T, et al. Dual Fc optimization to increase the cytotoxic activity of a CD19-targeting antibody. Front Immunol 2022;13:957874.
Wiencke JohnK, Butler R, Hsuang G, Eliot M, Kim S, Sepulveda MA, et al. The DNA methylation profile of activated human natural killer cells. Epigenetics 2016;11(5):363–80.
Cooper MA, Elliott JM, Keyel PA, Yang L, Carrero JA, Yokoyama WM. Cytokine-induced memory-like natural killer cells. Proc Natl Acad Sci 2009;106(6):1915–9.
Ni J, Miller M, Stojanovic A, Garbi N, Cerwenka A. Sustained effector function of IL-12/15/18-preactivated NK cells against established tumors. J Exp Med 2012;209(13):2351–65.
Chia NY, Chan YS, Feng B, Lu X, Orlov YL, Moreau D, et al. A genome-wide RNAi screen reveals determinants of human embryonic stem cell identity. Nature 2010;468(7321):316–20.
Erokhina SA, Streltsova MA, Kanevskiy LM, Telford WG, Sapozhnikov AM, Kovalenko EI. HLA-DR + NK cells are mostly characterized by less mature phenotype and high functional activity. Immunol Cell Biol 2018;96(2):212–28.
Evans JH, Horowitz A, Mehrabi M, Wise EL, Pease JE, Riley EM, et al. A distinct subset of human NK cells expressing HLA-DR expand in response to IL-2 and can aid immune responses to BCG. Eur J Immunol 2011;41(7):1924–33.
Niehrs A, Altfeld M. Regulation of NK-Cell Function by HLA Class II. Front Cell Infect Microbiol 2020;10:55.
Reiners KS, Topolar D, Henke A, Simhadri VR, Kessler J, Sauer M, et al. Soluble ligands for NK cell receptors promote evasion of chronic lymphocytic leukemia cells from NK cell anti-tumor activity. Blood 2013;121(18):3658–65.
Brandt CS, Baratin M, Yi EC, Kennedy J, Gao Z, Fox B, et al. The B7 family member B7-H6 is a tumor cell ligand for the activating natural killer cell receptor NKp30 in humans. J Exp Med 2009;206(7):1495–503.
Mócsai A, Ruland J, Tybulewicz VLJ. The SYK tyrosine kinase: a crucial player in diverse biological functions. Nat Rev Immunol 2010;10(6):387–402.
Gardner RA, Finney O, Annesley C, Brakke H, Summers C, Leger K, et al. Intent-to-treat leukemia remission by CD19 CAR T cells of defined formulation and dose in children and young adults. Blood 2017;129(25):3322–31.
Foà R, Bassan R, Vitale A, Elia L, Piciocchi A, Puzzolo MC, et al. Dasatinib–Blinatumomab for Ph-Positive Acute Lymphoblastic Leukemia in Adults. N Engl J Med 2020;383(17):1613–23.
Dong W, Wu X, Ma S, Wang Y, Nalin AP, Zhu Z, et al. The Mechanism of Anti–PD-L1 Antibody Efficacy against PD-L1–Negative Tumors Identifies NK Cells Expressing PD-L1 as a Cytolytic Effector. Cancer Discov 2019;9(10):1422–37.
Aryee MJ, Jaffe AE, Corrada-Bravo H, Ladd-Acosta C, Feinberg AP, Hansen KD, et al. Minfi: a flexible and comprehensive Bioconductor package for the analysis of Infinium DNA methylation microarrays. Bioinforma Oxf Engl 2014;30(10):1363–9.
Peters TJ, Buckley MJ, Chen Y, Smyth GK, Goodnow CC, Clark SJ. Calling differentially methylated regions from whole genome bisulphite sequencing with DMRcate. Nucleic Acids Res 2021;49(19):e109.
Tian Y, Morris TJ, Webster AP, Yang Z, Beck S, Feber A, et al. ChAMP: updated methylation analysis pipeline for Illumina BeadChips. Bioinformatics 2017;33(24):3982–4.
Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z, u. a. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. The Innovation 2021;2(3):100141.
Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, et al. Gene Ontology: tool for the unification of biology. Nat Genet 2000;25(1):25–9.
Durinck S, Spellman PT, Birney E, Huber W. Mapping identifiers for the integration of genomic datasets with the R/Bioconductor package biomaRt. Nat Protoc 2009;4(8):1184–91.
Chen S, Zhou Y, Chen Y, Gu J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics. 2018;34(17):i884–90.
Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 2014;15(12):550.
Introduction to msigdbr [Internet]. https://igordot.github.io/msigdbr/articles/msigdbr-intro.html
Garcia-Alonso L, Iorio F, Matchan A, Fonseca N, Jaaks P, Peat G, et al. Transcription Factor Activities Enhance Markers of Drug Sensitivity in Cancer. Cancer Res 2018;78(3):769–80.

There is NO conflict of interest to disclose.

SupplementaryInformationNKCCD19PhlikeALL20240517.pdf

Download PDF

Version 1

posted

You are reading this latest preprint version

Memory-like natural killer cell and CD19-antibody based immunotherapy in combination with tyrosine-kinase inhibition of Ph(-like) acute lymphoblastic leukemia

Status:

Version 1

Abstract

Figures

Introduction

Materials and Methods

Cell lines

Peripheral blood and bone marrow-derived natural killer cells

Reagents

NK cell cytotoxicity assay

Flow cytometry analysis and single cell sorting

DNA extraction and genome-wide methylation analysis

RNA sequencing

Single cell RNA sequencing

Statistical analysis

Results

Cytokine-induced pre-activation of NK cells induces epigenetic re-programming

by differential DNA-methylation

Epigenetic reprogramming determines transcriptional output of preactivated NK cells

Tyrosine kinase inhibition interferes with memory-like NK cell effector function

CD19 antibody-dependent cellular cytotoxicity rescues the TKI-associated impairment of NK cell effector functions

Discussion

Declarations

Competing interests

Author contributions

Acknowledgements

Data Availability Statement

References

Additional Declarations

Supplementary Files

Status:

Version 1