J Immunother Cancer: first published as 10.1136/jitc-2022-005400 on 26 September 2022. Downloaded from http://jitc.bmj.com/ on January 24, 2023 at Univesity Autonomous of Barcelona. Open access Original research Generation and proof-­of-­concept for allogeneic CD123 CAR-­Delta One T (DOT) cells in acute myeloid leukemia Diego Sánchez Martínez ‍ ‍ ,1,2 Néstor Tirado,1,2 Sofia Mensurado,3 Alba Martínez-­Moreno,1,2 Paola Romecín,1,2 Francisco Gutiérrez Agüera,1,2 Daniel V Correia,3 Bruno Silva-­Santos,3 Pablo Menéndez1,2,4,5 To cite: Sánchez Martínez D, ABSTRACT Tirado N, Mensurado S, et al. Background Chimeric antigen receptor (CAR)-­T cells WHAT IS ALREADY KNOWN ON THIS TOPIC Generation and proof-­of-­concept have emerged as a breakthrough treatment for relapse/ ⇒ Allogeneic cytotoxic cells obtained from healthy for allogeneic CD123 CAR-­Delta donors are a very interesting alternative for next-­ refractory hematological tumors, showing impressive One T (DOT) cells in acute generation ‘off-­ the-­ shelf’ chimeric antigen re- complete remission rates. However, around 50% of the myeloid leukemia. Journal for patients relapse before 1-­year post-­treatment. T-­cell ceptor (CAR)-­T cell therapies. In particular, MHC/ ImmunoTherapy of Cancer 2022;10:e005400. doi:10.1136/ ‘fitness’ is critical to prolong CAR-­T persistence and HLA-­independent γδ T cells have emerged as a jitc-2022-005400 activity. Allogeneic T cells from healthy donors are less promising candidate based on their high cytotoxic dysfunctional or exhausted than autologous patient-­ activity, stimulatory cytokine release and recruit- ► Additional supplemental derived T cells; in this context, Delta One T cells (DOTs), ment of other immune cells to the tumor site. material is published online only. a recently described cellular product based on MHC/ WHAT THIS STUDY ADDS To view, please visit the journal HLA-­independent Vδ1+γδ T cells, represent a promising ⇒ Our data demonstrate the capacity of CAR-­ Delta Protected by copyright. online (http://​dx.d​ oi.​org/​10.​ allogeneic platform. 1136/j​ itc-​2022-0​ 05400). Methods Here we generated and preclinically validated, One T cells (DOTs) to be a disruptive ‘off-­the-­shelf’ for the first time, 4-­1BB-­based CAR-­DOTs directed against allogeneic cellular immunotherapy for acute my- Accepted 26 August 2022 the interleukin-­3α chain receptor (CD123), a target antigen eloid leukemia. widely expressed on acute myeloid leukemia (AML) blasts. HOW THIS STUDY MIGHT AFFECT RESEARCH, Results CD123CAR-­DOTs showed vigorous, superior PRACTICE OR POLICY to control DOTs, cytotoxicity against AML cell lines and ⇒ CAR-­DOTs exhibit a combination of properties that primary samples both in vitro and in vivo, even on tumor may change allogeneic cell immunotherapy of he- rechallenge. matological (and probably solid) tumors in the future. Conclusions Our results provide the proof-­of-­concept for a DOT-­based next-­generation allogeneic CAR-­T therapy for AML. T cells are being actively investigated in order to increase T-­cell persistence and efficacy. INTRODUCTION Allogeneic cytotoxic cells obtained from Immunotherapy has promoted marked healthy donors (HD) are an especially attrac- improvements in cancer treatment over the tive avenue for ‘off-­the-­shelf’ next-­generation last decade. The immune system offers a CAR-­ T cell therapies. In particular, MHC/ © Author(s) (or their wide range of alternatives, including cyto- HLA-­independent γδ T cells have emerged as employer(s)) 2022. Re-­use toxic lymphocytes—native or engineered—to a promising candidate based on their potent permitted under CC BY-­NC. No eliminate treatment-­resistant tumors. In this cytotoxic activity and release of cytokines commercial re-­use. See rights and permissions. Published by context, CD19-­ directed chimeric antigen stimulating and recruiting other immune BMJ. receptor (CAR)-­transduced T cell (CAR-­T) cells to the tumor site.7–9 While the low abun- For numbered affiliations see therapies have shown impressive rates of dance of γδ T cells in the human peripheral end of article. complete remissions (CR) in relapse/refrac- blood (PB) has hindered their clinical appli- tory (r/r) B-­ cell malignancies, especially cation, recent advances in protocol develop- Correspondence to in B-­ cell acute lymphoblastic leukemia ment have enabled their ex vivo expansion Dr Diego Sánchez Martínez; ​dsanchez@​carrerasresearch.​org (ALL).1 2 However, 1 year progression-­ free to large numbers.10–13 In particular, we have survival remains ~50% due to frequent characterized Delta One T (DOT) cells, a Vδ1 Dr Bruno Silva-­Santos; relapses.3 4 Antigen loss and phenotypic T cell-­enriched cellular product expressing ​bssantos@​medicina.u​ lisboa.​pt escape as well as CAR-­T cell disappearance, enhanced levels of natural cytotoxicity recep- Dr Pablo Menéndez; dysfunctionality or exhaustion are commonly tors, and demonstrated that they consti- ​pmenendez@​carrerasresearch.​ responsible for such failures.5 6 Thus, new tute a safe and efficient effector platform to org therapeutic options based on ‘fitter’ effector eliminate cancer cells in in vitro and in vivo Sánchez Martínez D, et al. J Immunother Cancer 2022;10:e005400. doi:10.1136/jitc-2022-005400 1 J Immunother Cancer: first published as 10.1136/jitc-2022-005400 on 26 September 2022. Downloaded from http://jitc.bmj.com/ on January 24, 2023 at Univesity Autonomous of Barcelona. Open access preclinical models of solid and hematological malignan- depleteS protocol (Miltenyi Biotec, Biergisch Gladbach, cies,10 most notably acute myeloid leukemia (AML).14 Germany). DOT cells were generated from αβ-depleted AML, the most common acute leukemia in adults, is PBMCs cultured in either 96-­well plates or G-­REX plat- characterized by the accumulation of differentiation-­ form (Wilson Wolf Manufacturing), based on an adapta- defective immature and proliferative myeloid blasts in tion from our previous protocol.10 14 Briefly, αβ-depleted the bone marrow (BM) and PB.15 Unfortunately, patient PBMCs were resuspended in OpTmizer-­ CTS medium overall survival has not improved significantly over the supplemented with 2.5% heat-­inactivated human plasma last decades, with relapses being frequent and presenting (LifeSciences), 2 mmol/L L-­glutamine (Thermo Fisher) poor outcome, thus requiring hematopoietic stem cell and 50 U/mL/50 µg/mL of penicillin/streptomycin transplantation as rescue therapy.16 Despite its molecular (Thermo Fisher) and cultured for 16–20 days. Animal-­ enormous heterogeneity,15 17 18 CD123, the α chain of the free human cytokines rIL-­4 (100 ng/mL), rIFN-γ (70 ng/ interleukin (IL)-­3 receptor, is consistently expressed in mL), rIL-­21 (7 ng/mL), and rIL-­1β (15 ng/mL; all from >90% of AML blasts, in similar levels to leukemic stem PeproTech), and a soluble mAb anti-­CD3 (clone OKT-­3, cells, and its expression is maintained at relapse.19 20 140 ng/mL; BioLegend), were added to the medium at Based on this rational, we previously showed robust effi- day 0. At day 7, cultures were supplemented with anti-­ cacy of second generation (4-­1BB-­based) CD123-­directed CD3 (clone OKT-­3, 1 µg/mL or 2 µg/mL for 96-­well CAR-­T cells in preclinical AML models.21 plate protocol), rIL-­21 (13 ng/mL) and rIL-­15 (70 ng/ Clinically implemented CAR-­ T cell approaches are mL; also from PrepoTech). On day 11, new medium was based on abundant αβ T cells22–24 which have limited appli- added to cultures, supplemented with anti-­CD3 (1 µg/ cability in the allogeneic setting due to their MHC/HLA mL) and rIL-­15 (100 ng/mL). Cells were incubated at restriction and graft-­versus-­host potential.25 Importantly, 37°C and 5% CO2. DOT cells were harvested at the end immunological BM dysfunction and T-­ cell exhaustion of the culture and either used fresh (for in vitro assays) has been reported in intensively chemotherapy-­treated of cryopreserved (for in vivo experiments) in OpTmizer r/r AMLs.26 Thus, safety allowing, allogeneic T cells media+20% human plasma+10% DMSO, and stored in from HDs would represent an ‘off-­the-­shelf’ cost-­efficient liquid nitrogen. therapy to eliminate r/r AML. Building on these foun- Protected by copyright. dations, we prompted to explore the potential of MHC/ CD123-CAR retroviral production and transduction HLA-­independent DOT cells as vehicle for our second To construct the retroviral transfer vector, the complete generation CD123CAR in preclinical AML models. CD123-­ directed CAR (including CSL362 scFv, CD8 Here we describe, for the first time, the generation transmembrane domain, 4-­ 1BB costimulatory domain, of CAR-­expressing DOT cells (CAR-­DOTs) and provide CD3z endodomain, and a T2A-­eGFP cassette) previously the proof-­of-­concept for their application in AML treat- described21 was cloned into an SFG retroviral backbone. ment. Retrovirally-­transfected CD123CAR-­DOTs potently The SFG vector expressing eGFP alone (mock vector) was eliminated AML cell lines and primary samples both in used as a control. Viral particles pseudotyped with RD114 vitro and in vivo, increasing the efficacy of ‘naked’ (mock-­ were generated using 293 T cells with GeneJuice transfec- transduced) DOTs in different models and conditions, tion reagent (Sigma-­Aldrich, Saint Louis, Missouri, USA) without phenotypic alterations. Moreover, CD123CAR-­ following manufacturer’s instructions with a 1.5:1.5:1 DOTs released type 1 (antitumor) cytokines and chemo- µg ratio of the SFG:Peq-­Pam:RD114 DNA plasmids and kines on exposure to primary AML cells. In in vivo concentrated using Retro-­ X Concentrator (Takara, rechallenge experiments, CD123CAR-­ DOTs persisted Kusatsu, Japan) following manufacturer’s instructions. and remained functional against AML, 9 weeks after their DOT cells were expanded for 7 days prior to viral trans- infusion into mice. IL-­15 significantly improved their in duction. Proper CAR transduction was checked by flow vivo efficacy by sustaining longer action and enabling cytometry by eGFP expression and by using AffiniPure one-­single-­ dose treatment. Our data demonstrate the F(ab’)2 Fragment Goat Anti-­Human IgG (H+L) (Jackson potential of CAR-­DOTs to represent a disruptive ‘off-­the-­ ImmunoResearch Laboratories). Other conditions tested shelf’ allogeneic cellular immunotherapy for AML. during transduction optimization of DOT cells included the use of lentiviral particles as previously reported27 at multiplicities of infection between 10 and 50, as well METHODS as with retroviral particles with RetroNectin (Takara) DOT cell generation following manufacturer’s instructions. All lentiviral trans- PB mononuclear cells (PBMCs) were isolated from buffy duction conditions were tested at days 7 and 11 of the coats from HDs by Ficoll-­Hypaque gradient centrifuga- DOT cell expansion protocol. tion. Buffy coats were obtained from the Barcelona Blood and Tissue Bank on Institutional Review Board-­approval Immunophenotyping of DOT cells, cell lines and primary AML (HCB/2018/0030). PBMCs were incubated with anti- samples αβTCR Biotin mAb (Miltenyi Biotec) followed by incu- Proper differentiation and activation of DOT cells was bation with anti-­Biotin mAb microbeads to deplete αβ confirmed by surface staining with CD3-­PB (UCHT1), T cells by magnetic separation using autoMACS under CD3-­APCCy7 (UCHT1), TCRγδ-PE (B1), TCRαβ-APC 2 Sánchez Martínez D, et al. J Immunother Cancer 2022;10:e005400. doi:10.1136/jitc-2022-005400 J Immunother Cancer: first published as 10.1136/jitc-2022-005400 on 26 September 2022. Downloaded from http://jitc.bmj.com/ on January 24, 2023 at Univesity Autonomous of Barcelona. Open access (IP26), CD4-­ APCCy7 (RPA-­ T4), CD8-­ BV510 (RPA-­ indicated, mice were injected intraperitoneally with 1 T8), CD25-­ APCCy7 (BC96), CD69-­ BV510 (FN50), µg of hrIL-­15 every 3–4 days or daily from day 7 to the NKp30-­BV421 (P30-­15), NKG2D-­BV510 (1D11), end of the experiment, as a strategy to support DOT-­cell CD45RA-­APC (HI100), CD27-­PB (O323), CD62L-­BV510 persistence in vivo. Tumor burden was followed by biolu- (DREG-­ 56), DNAM-­ 1-­BV510 (11A8) from BioLegend minescence (BLI) using the Xenogen IVIS 50 Imaging (San Diego, California, USA) and TCR Vδ1-­PE (REA173), System (PerkinElmer). To measure luminescence, mice TCR Vδ2-­ APC (123R3) and NKp44-­ APC (2.29) from received 150 mg/kg of D-­luciferin intraperitoneally, and Miltenyi Biotec. Briefly, 2.5×105 cells were incubated with tumor burden was monitored at the indicated time points the antibodies for 30 min at 4°C and then washed. For as we previously described.27 30 Living Image software intracellular staining, DOT cells were incubated after (PerkinElmer) was used to visualize and calculate total membrane labeling with anti-­perforin-­BV241 (dG9) and luminescence. Besides, tumor burden was followed-­up at anti-­ granzyme B-­ BV510 (GB11) from BD Biosciences, different time points by bleeding and BM aspirate using and anti-­granzyme A-­APC (CB9) from BioLegend (San fluorescence activated cell sorting (FACS) analysis. Mice Diego, California, USA), using FIX&PERM Sample Kit were sacrificed when control and/or mock-­DOT-­treated (Nordic MUbio) under manufacturer’s instructions. animals were leukemic, and tumor burden (hHLA-­ Non-­reactive, isotype-­matched fluorochrome-­conjugated ABC+hCD45+hCD33+hCD123+ graft) and effector DOT mAbs were systematically used to set the gates. Dead cells persistence (hHLA-­ABC+hCD45+hCD3+Vδ1+Vd2+) were were discarded by 7-­AAD staining. analyzed in BM and PB by FACS. In rechallenge experi- The immunophenotyping of AML cell lines and ments, leukemia-­free animals were reinfused with 2.5×105 primary samples (obtained from Hospital Clínic of Barce- CD123+AML primary cells, and disease reappearance was lona) was done by surface staining with CD45-­PE (HI30), followed-­up by BLI and FACS as above. All procedures CD33-­ BV421 (HIM3-­ 4) and CD123-­ APC (7G3) (BD were performed in compliance with the institutional Biosciences, Franklin Lakes, New Jersey, USA). A FACS- animal care committee of the PRBB (DAAM9624). Canto-­II flow cytometer equipped with FACSDiva soft- ware (BD Biosciences) was used for the analysis.27 Statistical analysis Data from at least three individual donors are shown in Protected by copyright. In vitro cytotoxicity assays and cytokine release all figures. All p values were calculated by unpaired two-­ determination tailed or one-­tailed Student’s t-­test using Prism software The cell lines MOLM13, THP-­1 and Jurkat were purchased (GraphPad). A p value *<0.05/**<0.01/***<0.001/****< from DSMZ (Germany) and expanded according to 0.0001 was considered statistically significant. DSMZ recommendations. Target cells (cell lines and primary AML blasts) were labeled with 3 µM eFluor 670 (eBioscience) and incubated with CAR-­DOTs or mock-­ RESULTS DOTs at different Effector:Target (E:T) ratios for the CD123-directed DOT cells (CD123CAR-DOTs) expand robustly indicated time periods in complete DOT media. CAR-­ while preserving DOT phenotype DOT-­mediated cytotoxicity was determined by analyzing We set out to generate, for the first time, CAR-­DOT cells, the residual alive (7-­AAD–) eFluor 670+ target cells at each and purpose them for AML targeting. DOT cells were time point and E:T ratio. Absolute cell counts were deter- expanded and activated as described,10 14 and transduc- mined using Trucount absolute count beads (BD Biosci- tion was tested at days 7 and 11 using different retroviral ences).28 Cytokine production was measured by G-­Series (PeqPam/RD114) and lentiviral (VSV-­G/psPAX2) condi- Human Cytokine Antibody Array 4000 from RayBiotech tions (online supplemental figure S1). Retro-­X Concen- in supernatants harvested after 48 hours-­incubation CAR-­ trator at day 7 showed the highest rate of anti-­CD123CAR DOT/mock-­DOT (four donors) with two primary AML expression and was chosen for downstream generation samples. of CD123CAR-­DOTs (figure 1A). CD123CAR transduc- tion/expression did not impact γδ TCR+ and Vδ1+ T cell In vivo AML patient-derived xenograft models percentages, which were ~97% and ~70%, respectively, The 7–12 weeks old non-­obese diabetic Cg-­Prkdcscid Il2rgt- as in the original ‘naked’ DOT product.10 14 Likewise, m1Wjl /SzJ (NSG) mice (Jackson Laboratory) were bred total and Vδ1+ T-­cell numbers at the end of the process and housed under pathogen-­ free conditions in the were similar for mock-­DOTs and CAR-­DOTs (figure 1C). animal facility of the Barcelona Biomedical Research Retroviral particles infected the Vδ1+ and Vδ1–γδ T-­cell Park (PRBB). Mice were intravenously transplanted populations in similar proportions, and the CD123CAR in the tail with 2.5×105 Luc-­GFP-­expressing primary expression as determined by GFP positivity correlated CD123+ AML blasts (primograft-­ expanded) (579AML-­ well with anti-­ScFv labeling (figure 1B,C). CAR+ DOTs patient-­derived xenograft (PDX)LUC).29 CAR-­ DOTs or (n=14) presented a similar T-­ cell differentiation stage mock-­DOTs were thawed and resuspended in complete to CAR– DOTs, as demonstrated by CD62L/CD45RA media and mice were intravenously infused in the tail staining by FACS. The majority of DOTs were effector T vein with 10×106 cells once, two times or three times 7, 14 cells (TEFF), with an expected degree of variability between and 21 days post-­tumor inoculation, respectively. Where donors, followed by effector memory (TEM) DOTs, Sánchez Martínez D, et al. J Immunother Cancer 2022;10:e005400. doi:10.1136/jitc-2022-005400 3 J Immunother Cancer: first published as 10.1136/jitc-2022-005400 on 26 September 2022. Downloaded from http://jitc.bmj.com/ on January 24, 2023 at Univesity Autonomous of Barcelona. Open access Protected by copyright. Figure 1 CD123CAR-­DOT cell generation and phenotypic characterization. (A) CD123CAR-­DOT cell production scheme. αβ-depleted PBMCs were cultured under OKT3 antibody and cytokine stimulation for 3 weeks and transduced with γ-retroviruses at day 7. (B) Upper panels, γδ-TCR+ and Vδ1+T cell percentages evolution from day 0 to end-­product for mock DOTs and CAR-­DOTs. Lower panels, representative FACS dot plots depicting the expression of Vδ1/Vδ2 and Vδ1/CD3 in end-­ products. (C) Upper panels, cell number for total cells and Vδ1+T cells along the CAR-­DOT generation process. Lower panels, representative FACS dot plot for Vδ1/GFP and anti-­ScFv/GFP expression. (D) Left-­middle panels, scheme and representative FACS plot depicting TCM, TN/SCM, TEM, TEFF T-­cell subsets as defined by CD62L versus CD45RA staining. Right panel, percentage of TCM, TN/SCM, TEM, TEFF T-­cell subsets for CAR+DOT and CAR−DOT subpopulations (n=14). (E) NKG2D, NKp30, NKp44, DNAM-­1 and PD-­1 expression (n=10), and (F) perforin (Prf), granzyme A (GzmA) and granzyme B (GzmB) expression (n=6) in CAR−DOT versus CAR+DOT cells. CAR, chimeric antigen receptor; DOT, Delta One T cells; PBMCs, peripheral blood mononuclear cells; TCM, central memory; TEFF, effector T cells; TEM, effector memory; TN/SCM, naïve/stem central memory. 4 Sánchez Martínez D, et al. J Immunother Cancer 2022;10:e005400. doi:10.1136/jitc-2022-005400 J Immunother Cancer: first published as 10.1136/jitc-2022-005400 on 26 September 2022. Downloaded from http://jitc.bmj.com/ on January 24, 2023 at Univesity Autonomous of Barcelona. Open access and lower contributions of central memory (TCM) and These data demonstrate an enhanced effector poten- naïve/stem central memory (TN/SCM) DOTs (figure 1D). tial of CD123CAR-­DOT cells in direct comparison with Furthermore, both CAR+ and CAR– DOTs expressed their naked/mock-­DOT cells. natural killer (NK) cell receptor signature according to the published10 hierarchy: NKG2D (>90%) > DNAM-­1 Serial infusions of CD123CAR-DOTs exhibit robust anti- (~60%) > NKp30 and NKp44. By contrast, the immune leukemic effect in vivo checkpoint programmed cell death protein-1 (PD-­1) was CD123CAR-­ DOT function was evaluated in vivo by strikingly absent from both DOT populations (figure 1E). employing Luc-­expressing AML-­PDXLUC. NSG mice The expression of the key components of cytotoxic gran- were transplanted with 2.5×105 AML-­PDXLUC cells and, ules, perforin, granzyme (Gzm) A and GzmB, was also based on our previous extensive experience with DOTs not affected by CAR transduction (figure 1F). Collec- in AML xenograft models,14 infused with 10×106 (CAR-­ tively, these data demonstrate the feasibility of CAR-­DOT DOTs or mock-­ DOTs) DOTs in three serial infusions generation, and further show that retroviral infection (one injection per week (day 7, 14, 20), starting 1 week and CD123CAR expression on the cell surface does not post-­ tumor injection), and leukemia progression was modify DOT expansion properties or phenotype, thus followed by BLI (figure 3A). CD123CAR-­DOTs controlled generating NK-­like cytotoxic effector γδ T cells highly AML tumor burden between days 20 and 47 significantly enriched in Vδ1+ T cells. better than mock-­DOTs which only managed to delay tumor growth and could not prevent an eventual loga- CD123CAR-DOTs specifically augment cytotoxicity against rithmic tumor growth from day 47 onwards (figure 3B). AML cell lines and primary blasts in vitro At this point, we decided to provide an additional fourth In order to test the anti-­AML activity of CAR-­DOTs, infusion at day 53, which resulted in complete control we performed in vitro cytotoxicity assays against the of leukemia in CD123CAR-­DOT mice, in stark contrast CD123+ AML cell lines MOLM13 and THP-­1, and the with continued tumor growth in mock DOT-­treated mice CD123– T-­ cell ALL line Jurkat, as negative control. (figure 3B). Furthermore, FACS analyses of PB and BM CD123CAR-­DOTs specifically eliminated CD123+ AML samples collected at days 47 and 67 confirmed the BLI cells in an E:T ratio-­dependent manner, with substan- data. Leukemic burden, as determined by HLA-­ ABC/ Protected by copyright. tially increased cytotoxic activity compared with mock-­ CD45/CD123/CD33 expression, showed high, interme- DOTs. Even at low E:T ratios (1:16, 1:8, 1:4) we achieved diate and absent blast cells in untreated, mock-­DOT and 20–60% more AML cell lysis than with mock-­DOTs in CAR-­DOT-­treated mice, respectively (figure 3C). These 48 hours and 24 hours assays (figure 2, online supple- data highlight the therapeutic advantage over 12 weeks of mental figure S2). Importantly, CD123CAR-­ DOTs CD123CAR-­DOTs over mock-­DOTs in in vivo AML-­PDX showed similar potency to conventional CD123CAR-­Ts models. (online supplemental figure S3), thus allowing us to concentrate on the allogeneic advantages of Provision of IL-15 supports single-dose CD123CAR-DOT CD123CAR-­DOTs. Next, to evaluate their capacity to activity in vivo kill primary tumors, CD123CAR-­ DOTs were co-­ cul- Having achieved the initial proof-­ of-­ concept for tured with primary CD123+ AML samples, with different CD123CAR-­ DOT activity in AML xenografts, we next percentages of CD123+CD33+ blasts (figure 2B). aimed to maximize it through provision of key DOT CD123CAR-­DOTs showed enhanced cytotoxicity over survival factors, particularly since the murine milieu lacks mock-­DOTs against CD123+ AML primary blasts in 48 human cytokines. Given that the DOT protocol relies hours assays (figure 2C). Moreover, compared with on IL-­15 as the critical cytokine in the second stage of mock-­DOTs, CD123CAR-­DOTs produced significantly expansion/differentation,10 14 we combined CD123CAR-­ higher levels of master antitumor mediators, namely DOTs with intraperitoneal IL-­15 administration every 3–4 the cytokines tumor necrosis factor-α and interferon-γ, days. In this set of experiments, NSG mice were trans- as well as IL-­13, recently implicated in orchestrating planted with 2.5×105 AML-­PDXLUC cells, and starting tumor surveillance in mice31; the T-­cell costimulator 1 week later, they received one, two or three doses of 1BB, a major determinant of γδ T-­cell-­mediated 4-­ 10×106 CD123CAR-­ DOTs, with or without IL-­ 15 injec- tissue surveillance32; the signal transducer Axl, shown tion (figure 4A). Leukemia engraftment was followed to maximize IL-­ 15R signaling in human NK cell weekly by BLI and by flow cytometry analysis of BM and differentiation33; the myeloid differentiation factor, PB samples. In the absence of exogenous IL-­15, three granulocyte-­macrophage colony-­stimulating factor doses of CD123CAR-­ DOT (CAR3) were clearly more (GM-­CSF); and the chemokines MIP-­1a (CCL3), CCL5, efficacious than one (CAR1) or two (CAR2) doses, since CXCL13 and lymphotactin (XCL1), all important in these only managed to delay tumor burden (compared mobilizing multiple leukocyte subsets in cancer immu- with untreated control mice) (figure 4B,C). Strikingly, nity.34 While not significantly augmented, IL-­ 2 and the provision of IL-­15 every 3–4 days overrode all such FasL, key mediators of T-­cell proliferation and cytotox- differences, thus allowing complete leukemia control icity, respectively, also showed a tendency for higher 60 days after even with a single CD123CAR-­DOT infu- expression in CD123CAR-­ DOT cells (figure 2D). sion (figure 4B,C). These results demonstrate a critical Sánchez Martínez D, et al. J Immunother Cancer 2022;10:e005400. doi:10.1136/jitc-2022-005400 5 J Immunother Cancer: first published as 10.1136/jitc-2022-005400 on 26 September 2022. Downloaded from http://jitc.bmj.com/ on January 24, 2023 at Univesity Autonomous of Barcelona. Open access Protected by copyright. Figure 2 CD123CAR-­DOT cells specifically target and eliminate CD123+AML cell lines and primary samples in vitro. (A) Cytotoxicity of CAR-­DOTs and mock-­DOTs against CD123+ AML (MOLM13 and THP-­1) and CD123− T-­ALL (Jurkat, negative control) cell lines at the indicated E:T ratios in 48 hours assays (n=5). Small insets, CD123 expression in each cell line. (B) AML primary sample (Pt, patient) phenotype showing CD33/CD123 analysis by FACS. AML blasts are highlighted in blue. (C) Absolute counts of alive eFluor+CD123+ AML blasts measured by FACS in 48-­hour cytotoxicity assays at 1:1/2:1 E:T ratios. (D) Cytokine array determination in supernatants obtained from four donors of CAR-­DOTs/mock-­DOTs exposed to two different AML primary samples for 48 hours. Relative expression was normalized to mock-­DOT levels. *p<0.05, **p<0.01, ***p<0.001. ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; CAR, chimeric antigen receptor; DOTs, Delta One T cells; E:T, Effector:Target. 6 Sánchez Martínez D, et al. J Immunother Cancer 2022;10:e005400. doi:10.1136/jitc-2022-005400 J Immunother Cancer: first published as 10.1136/jitc-2022-005400 on 26 September 2022. Downloaded from http://jitc.bmj.com/ on January 24, 2023 at Univesity Autonomous of Barcelona. Open access Protected by copyright. Figure 3 CD123CAR-­DOT cells fully control growth of primary CD123+AML blasts in a PDX model. (A) Schematic of the AML-­PDX model. NSG mice (n=4/group) were intravenously injected with 2.5×105 AML-­PDXLUC cells and then received three intravenous injections (one per week, starting 7 days after tumor injection) of 10×106 CAR-­DOT or mock-­DOT cells. Tumor burden was monitored weekly by BLI using IVIS imaging. At day 53, mock-­treated and DOT-­treated mice received an extra infusion of 10×106 mock-­DOTs or CAR-­DOTs, respectively. (B) IVIS imaging of tumor burden monitored by BLI at the indicated time points(escale: 5x104-­1x106). Right panel shows the total radiance quantification (p/sec/cm2/sr) at the indicated time points. †: sacrifice. (C) Tumor burden analyzed by FACS in PB and BM at days 47 and 67. Primary AML blasts are shown in blue. Mouse cells are shown in black. *p<0.05, **p<0.01, ***p<0.001. AML, acute myeloid leukemia; BLI, bioluminescence; BM, bone marrow; CAR, chimeric antigen receptor; DOTs, Delta One T cells; PB, peripheral blood; PDX, patient-­derived xenograft. Sánchez Martínez D, et al. J Immunother Cancer 2022;10:e005400. doi:10.1136/jitc-2022-005400 7 J Immunother Cancer: first published as 10.1136/jitc-2022-005400 on 26 September 2022. Downloaded from http://jitc.bmj.com/ on January 24, 2023 at Univesity Autonomous of Barcelona. Open access Protected by copyright. Figure 4 A single-­dose of CD123CAR-­DOT treatment combined with administration of IL-­15 suffices to fully abolish AML progression in a PDX model. (A) Schematic of the AML-­PDX model. NSG mice (n=4/group) were intravenously injected with 2.5×105 AML-­PDXLUC cells followed 7 days after by one, two or three intravenous injections (one per week) of 10×106 CAR-­ DOT cells, with or without administration of IL-­15 intraperitoneally. (B) IVIS imaging of tumor burden monitored by BLI at the indicated time points(scale: 5x104-­1x106). Right panel shows the total radiance quantification (p/sec/cm2/sr) at the indicated time points. (C) Tumor burden analyzed by FACS in PB and BM at the end of the experiment. AML blasts are highlighted in blue. AML, acute myeloid leukemia; BLI, bioluminescence; BM, bone marrow; CAR, chimeric antigen receptor; DOTs, Delta One T cells; IL, interleukin; PB, peripheral blood; PDX, patient-­derived xenograft. adjuvant effect of exogenous IL-­15 on CD123CAR-­DOT single-­ dose CD123CAR-­ DOT versus mock-­ DOT activities therapy in AML-­PDX in vivo. in vivo. In this experimental design, CD123CAR-­DOTs or mock-­DOTs were infused (in a single-­one dose) 1 week Daily provision of IL-15 maximizes DOT-cell therapeutic after intravenous injection of 2.5×105 AML-­PDXLUC cells, efficacy in vivo followed by three different IL-­15 administration regimens: Building on the observed major impact of IL-­ 15 on daily, every 3–4 days, or none at all (figure 5A). Strikingly, CD123CAR-­DOT activity, we decided to evaluate side-­by-­ mock-­ DOTs plus daily IL-­ 15 showed similar leukemia side how IL-­15 provision, in various regiments, would boost 8 Sánchez Martínez D, et al. J Immunother Cancer 2022;10:e005400. doi:10.1136/jitc-2022-005400 J Immunother Cancer: first published as 10.1136/jitc-2022-005400 on 26 September 2022. Downloaded from http://jitc.bmj.com/ on January 24, 2023 at Univesity Autonomous of Barcelona. Open access Protected by copyright. Figure 5 Daily-­infused IL-­15 increases DOT cell persistence and enables elimination of primary CD123+AML blasts in a PDX model. (A) Schematic of the AML-­PDX model. NSG mice (n=4/group) were intravenously injected with 2.5×105 AML-­PDXLUC cells followed 7 days after by a one-­single dose intravenous injection of 10×106 CAR-­DOTs or mock-­DOTs. Mock-­DOT and CAR-­DOT-­treated mice were followed in the absence or presence of IL-­15, intraperitoneal administered either daily or every 3–4 days. (B) IVIS imaging of tumor burden monitored weekly by BLI at the indicated time points (scale: 5x104-­1x106). (C) Total radiance quantification (p/sec/cm2/sr) at the indicated time points. (D) Tumor burden was monitored by FACS in BM at day 30. (E) Dot plots of representative Vδ1+ and Vδ2+T cells determined in CAR-­DOT/ mock-­DOT-­treated mice in BM at day 30. (F) Total γδ T cells were monitored by FACS in BM at day 30. AML, acute myeloid leukemia; BLI, bioluminescence; BM, bone marrow; CAR, chimeric antigen receptor; DOTs, Delta One T cells; IL, interleukin; PB, peripheral blood; PDX, patient-­derived xenograft. Sánchez Martínez D, et al. J Immunother Cancer 2022;10:e005400. doi:10.1136/jitc-2022-005400 9 J Immunother Cancer: first published as 10.1136/jitc-2022-005400 on 26 September 2022. Downloaded from http://jitc.bmj.com/ on January 24, 2023 at Univesity Autonomous of Barcelona. Open access control to all groups of CD123CAR-­ DOT-­ treated mice, The BM of patients with leukemia is exposed to several as evaluated by BLI up to day 42 (figure 5B,C). The daily cycles of chemotherapy and immunosuppressants, which provision of IL-­15 was critical to enhance mock-­DOT cell alter the status of normal T cells.5 39 40 Particularly in activity, since the 3–4 day IL-­15 regimen barely reduced patients with AML, T cells have been characterized with tumor burden compared with control groups (figure 5B,C). an exhausted phenotype at relapse.26 In this context, allo- These data were consolidated by flow cytometry analysis of geneic T cells represent an attractive alternative, since BM aspirates, which showed that AML cell engraftment circulating T cells from HDs have not suffered disease-­ correlated with IVIS imaging (figure 5D). Importantly, the induced or therapy-­induced dysfunction. While various infused (CAR- or mock-) DOTs were only detected in the groups have reported the generation of allogeneic αβ BM at the endpoint of the experiment when daily IL-­15 T cells, these cells must undergo genetic manipulation, was administered, thus highlighting the importance of in generally CRISPR/Cas9 editing, to eliminate the endog- vivo persistence of the TEFF (figure 5E,F). These data unveil enous T-­ cell receptor (TCR) since it would otherwise the importance of in vivo IL-­15 administration regimens on drive a (potentially severe) graft-­ versus-­ host reaction. DOT-­based therapies in AML. Elegant studies have reported the feasibility of these TCRKO CAR-­T cells in preclinical models and even in Optimized CD123CAR-DOT treatment controls AML growth on pediatric patients.41–43 However, a relevant recent paper in vivo rechallenge has reported that TCR elimination in CAR-­T cells leads to Persistence is a key goal and a major challenge of adop- a decrease in persistence.44 Furthermore, from a clinical tive cellular immunotherapy. Aiming to determine the point-­of-­view, CRISPR/Cas9 handling would still have to sustained functionality and potency of (CAR- or mock-) solve complex regulatory requirements. DOTs after >40 days of controlling AML progression, Since the vast majority of γδ T cells are HLA-­ we then performed a tumor rechallenge experiment. independent,45 they do not require genetic manipula- Disease-­free mice that eliminated the previous AML graft tion to be used as allogeneic therapies. It was with this (figure 5), either under all CD123CAR-­DOT treatments purpose that we previously developed DOT cells, a Vδ1 or under mock-­ DOTs+daily IL-­ 15, were rechallenged T-­cell-­enriched cell product characterized by the upreg- with an additional AML-­PDXLUC intravenous infusion at ulation of multiple activating/cytotoxicity-­ associated Protected by copyright. day 47 (figure 6A). In stark contrast to the other groups, NK-­cell receptors (NKG2D, DNAM-­1, NKp30, NKp44), CD123CAR-­DOTs plus daily IL-­15 was the only condition and showed their anti-­leukemic potential.10 14 However, in which tumor growth kept controlled; in fact, AML our previous data in AML xenograft models left a signif- cells were barely detectable by BLI or FACS analysis of icant margin for improvement in terms of control of PB and BM samples (figure 6B–E). These results firmly tumor burden.15 Here we investigated whether DOT demonstrate a vigorous and persistent anti-­ leukemia transduction with a CD123-­directed CAR would improve effect of CD123CAR-­DOTs plus daily IL-­15, sustained over efficacy and lead to complete and durable leukemia 70 days and even on tumor rechallenge, thus providing remissions. We show that, compared (side-­by-­side) with seminal proof-­of-­concept for their application for adop- control (mock-­ )DOTs, CAR123-­ DOTs are substantially tive cellular immunotherapy in AML. more potent at eliminating AML blasts (both cell lines and primary samples) in vitro; release increased levels of key cytokines involved in immune orchestration and amplification of the antitumor response; and are more DISCUSSION efficient at controlling tumor growth in vivo, in AML-­PDX In the last few years, treatments for hematological experiments conducted over 70 days. neoplasias have undergone a revolution due to immu- We also found that on administration of IL-­15 in our notherapy. In B-­ cell malignancies, autologous CAR-­ T AML-­PDX models, CD123CAR-­DOTs (and also mock-­ cell therapies have rescued r/r patients on the path to DOTs) persisted better, thus allowing a single dose to palliative care, obtaining very impressive CR rates.1 2 The maintain mice free of leukemia. Critically, daily-­infused number of CAR-­T clinical trials keeps increasing, aiming IL-­15 enabled CD123CAR-­DOTs to sustain their anti-­ to implement alternatives for bad prognosis tumors leukemic activity for more than 70 days, even after AML and expanding their horizons to other types of diseases rechallenge. These results are in agreement with recent (HIV, COVID-­19, autoimmune diseases, cardiac damage, elegant publications showing CAR-­Vδ1 T-­cell efficacy etc).35–37 However, CAR-­T cells are still at the beginning and the adjuvant effect of IL-­15 in B-­cell lymphoma of their clinical implementation, and a significant frac- and hepatocellular carcinoma models.12 46 Of note, our tion (around 50%) of the patients relapse after treatment study is the first to compare CAR-­transduced with mock-­ in a relative short period of time.3 4 Factors like the low transduced Vδ1 T cells, thus being able to ascertain the number of T cells retrieved from the patient’s leukapher- increased potency provided by CAR targeting. Only the esis, tumor blast contamination in the starting material, clinical setting can establish the real benefit of IL-­15 or T-­ cell exhaustion in multitreated patients, can all co-­administration (potentially with an inducible ‘switch explain autologous CAR-­T failures stemming from the off’ system) with DOTs or CD123CAR-­DOTs. Obviously, manufacturing process.38 our (and other) preclinical mouse xenograft models lack 10 Sánchez Martínez D, et al. J Immunother Cancer 2022;10:e005400. doi:10.1136/jitc-2022-005400 J Immunother Cancer: first published as 10.1136/jitc-2022-005400 on 26 September 2022. Downloaded from http://jitc.bmj.com/ on January 24, 2023 at Univesity Autonomous of Barcelona. Open access Figure 6 A single-­dose of CD123CAR-­DOT cells plus daily IL-­15 sustain the ability to control AML progression on rechallenge Protected by copyright. in a PDX model. (A) Schematic of the tumor rechallenge AML-­PDXLUC experiments. Mock-­DOTs plus daily IL-­15-­treated mice and CAR-­DOTs-­treated mice (regardless of IL-­15 regiment) were rechallenged with 2.5×105 primary CD123+AML, 47 days after initial (CAR- or mock-) DOT cell infusion. (B) IVIS imaging of tumor burden monitored by BLI at the indicated time points (scale: 5x104-­1x106). (C) Total radiance quantification (p/sec/cm2/sr) over time in mice rechallenged with AML-­PDXLUC. Tumor burden was monitored at endpoint analysis (day 70) by FACS in BM (D) and PB (E). *p<0.05, **p<0.01, ***p<0.001 with one-­tailed Student’s t-­test. AML, acute myeloid leukemia; BLI, bioluminescence; BM, bone marrow; CAR, chimeric antigen receptor; DOTs, Delta One T cells; IL, interleukin; PB, peripheral blood; PDX, patient-­derived xenograft. endogenous human IL-­15; thus, basal circulating IL-­15 the need for exogenous (adjuvant) IL-­ 15 should be levels in patients could suffice to sustain (CD123CAR-) carefully examined since it caused some toxicities in DOT persistence and activity/efficacy in the clinical a recent clinical trial in patients with cancer.49 This setting. Importantly, lymphodepletion regimens that notwithstanding, this clinical trial employed a contin- are used prior to CAR T-­cell infusions were shown to uous bolus of IL-­15 as single therapy, and thus, lower increase the levels of IL-­ 15 in patients.47 Therefore, and/or intermittent IL-­15 dosing may be a promising although described as safe in non-­human primates,48 solutions.50 51 Sánchez Martínez D, et al. J Immunother Cancer 2022;10:e005400. doi:10.1136/jitc-2022-005400 11 J Immunother Cancer: first published as 10.1136/jitc-2022-005400 on 26 September 2022. Downloaded from http://jitc.bmj.com/ on January 24, 2023 at Univesity Autonomous of Barcelona. Open access An important final aspect to discuss is the potential provided, Dr Clara Bueno and Dr Virginia Rodríguez for laboratory maintenance, Dr advantage of using DOTs as effector cells for CAR trans- Pedro Roda for technical support, and Natacha Gonçalves-­Sousa for administrative help. duction. Compared with conventional αβ-dominated Contributors Conception and design of the study: DSM, BS-­S, PM. Acquisition of T-­cell products, DOTs offer several advantages: first data: DSM, NT, SM, AM-­M, PR, FGA, DVC. Analysis and interpretation of data: DSM, and foremost, the previously mentioned suitability for NT, SM, AM-­M, PR, FGA, DVC. Writing original draft: DSM, BS-­S, PM. Review and allogeneic use, without any need for additional genetic editing: DSM, NT, SM, BS-­S, PM. Guarantors: PM, BS-­S. engineering,52 and without any loss in potency (online Funding “la Caixa” Foundation (ID 100010434) under the agreement LCF/PR/ supplemental figure S3). Additionally, the immune HR19/52160011 (BS-­S, PM) ISCIII-­RICORS within the Next Generation EU program checkpoint PD-­1 is absent in DOTs10 and CD123CAR-­ (plan de recuperación, transformación y resilencia) (PM). CERCA/Generalitat de Catalunya (PM) Fundació Josep Carreras-­Obra Social la Caixa (PM). Sara Borrell DOTs (figure 1E), whereas the upregulated NK-­associated fellowship from the Instituto de Salud Carlos III (DS-­M). FPU PhD Scholarship form cytotoxicity machinery (including DNAM-­1, NKp30 and MINECO (NT). NKp44)10 14 may be critical on CAR antigen loss as observed Competing interests BS-­S and DVC were co-­founders and shareholders with CD19 in B-­ALL.6 Importantly, this enhanced NK-­like of Lymphact/Gamma Delta Therapeutics. PM is co-­founder of OneChain cytotoxicity is also an advantage over other γδ T cell-­based ImmunoTherapeutics, a spin-­off from the Josep Carreras Leukemia Research Institute. products being developed for adoptive immunotherapy of cancer.7–9 12 46 On the other hand, in comparison with Patient consent for publication Not applicable. NK cells, which obviously share such activating NK-­cell Ethics approval Not applicable. receptors, DOTs offer as key advantages7 8 53 the previ- Provenance and peer review Not commissioned; externally peer reviewed. ously documented10 absence of inhibitory KIRs, namely Data availability statement All data relevant to the study are included in the KIR2DL1, KIR2DL2, KIR2DL4, KIR2DL5A and KIR3DL1, article or uploaded as supplementary information. and the additional input of TCR-­dependent activation and Supplemental material This content has been supplied by the author(s). It has expansion (up to very high yields as reported).10 14 In fact, not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-­reviewed. Any opinions or recommendations discussed are solely those the presence of a stochastically recombined and highly of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and polyclonal TCR repertoire,14 allowing HLA-­unrestricted responsibility arising from any reliance placed on the content. Where the content recognition of stress-­induced molecules,7 may also coun- includes any translated material, BMJ does not warrant the accuracy and reliability teracts tumor immune evasion (via CAR antigen loss of the translations (including but not limited to local regulations, clinical guidelines, Protected by copyright. terminology, drug names and drug dosages), and is not responsible for any error or HLA downregulation, eg). Moreover, the ‘trophic’ and/or omissions arising from translation and adaptation or otherwise. (homeostatic) signals received through γδ TCR binding Open access This is an open access article distributed in accordance with the to butyrophilins constitutively expressed in multiple Creative Commons Attribution Non Commercial (CC BY-­NC 4.0) license, which tissues54 may be beneficial for in vivo persistence, and may permits others to distribute, remix, adapt, build upon this work non-­commercially, underlie the recent description of a striking Vδ1+ T cell-­ and license their derivative works on different terms, provided the original work is associated CD19CAR-­T cell expansion (from a very low properly cited, appropriate credit is given, any changes made indicated, and the use is non-­commercial. See http://creativecommons.org/licenses/by-nc/4.0/. initial frequency), over 10 years in a patient with disease-­ free Chronic Lymphocytic Leukemia (CLL).55 Further- ORCID iD more, although allogeneic CD19-­directed CAR-­NK cells Diego Sánchez Martínez http://orcid.org/0000-0003-4605-5325 have recently shown promising clinical results,56 their the-­ ‘off-­ shelf’ implementation may be difficult due to the reported substantial decrease in viability after the freezing/thawing process,57 58 which contrasts with the REFERENCES 1 Maude SL, Frey N, Shaw PA, et al. Chimeric antigen receptor robustness we observed with CD123CAR-­DOTs. We there- T cells for sustained remissions in leukemia. N Engl J Med fore strongly believe that CAR-­DOTs combine a set of 2014;371:1507–17. 2 Ortíz-­Maldonado V, Rives S, Castellà M, et al. CART19-­BE-­01: A properties that may revolutionize allogeneic cell immu- Multicenter Trial of ARI-­0001 Cell Therapy in Patients with CD19+ notherapy of hematological (and possibly also solid) Relapsed/Refractory Malignancies. Mol Ther 2021;29:636–44. 3 Gardner RA, Finney O, Annesley C, et al. Intent-­to-­treat leukemia cancers in the near future. remission by CD19 CAR T cells of defined formulation and dose in children and young adults. Blood 2017;129:3322–31. Author affiliations 4 Maude SL, Laetsch TW, Buechner J, et al. Tisagenlecleucel in 1 children and young adults with B-­cell lymphoblastic leukemia. N Engl Josep Carreras Leukaemia Research Institute, Department of Biomedicine, School J Med 2018;378:439–48. of Medicine, University of Barcelona, Barcelona, Spain 5 Finney OC, Brakke HM, Rawlings-­Rhea S, et al. CD19 CAR T cell 2 Red Española de Terapias Avanzadas (TERAV) - Instituto de Salud Carlos III (ISCII) product and disease attributes predict leukemia remission durability. (RICORS, RD21/0017/0029) J Clin Invest 2019;129:2123–32. 3 Instituto de Medicina Molecular João Lobo Antunes, Faculdade de Medicina, 6 Xu X, Sun Q, Liang X, et al. Mechanisms of relapse after CD19 CAR Universidade de Lisboa, Portugal T-­cell therapy for acute lymphoblastic leukemia and its prevention 4 and treatment strategies. Front Immunol 2019;10:1–15. Centro de Investigación Biomédica en Red-­Oncología (CIBERONC), Instituto de 7 Sebestyen Z, Prinz I, Déchanet-­Merville J, et al. Translating Salud Carlos III, Barcelona, Spain gammadelta (γδ) T cells and their receptors into cancer cell therapies. 5 Institució Catalana de Recerca i Estudis Avançats (ICREA), Barcelona, Spain Nat Rev Drug Discov 2020;19:169–84. 8 Silva-­Santos B, Mensurado S, Coffelt SB. γδ T cells: pleiotropic immune effectors with therapeutic potential in cancer. Nat Rev Acknowledgements We are indebted with Dr Talía Velasco-­Hernández for Cancer 2019;19:392–404. technical support with interleukin-­15 infusions, Dr Matteo Libero Baroni for 9 Kabelitz D, Serrano R, Kouakanou L, et al. Cancer immunotherapy retroviral cloning, Professor Irmela Jeremias for the Luc+ acute myeloid leukemia with γδ T cells: many paths ahead of US. Cell Mol Immunol primografts, Dr Maksim Mamonkin for retroviral generation protocol kindly 2020;17:925–39. 12 Sánchez Martínez D, et al. J Immunother Cancer 2022;10:e005400. doi:10.1136/jitc-2022-005400 J Immunother Cancer: first published as 10.1136/jitc-2022-005400 on 26 September 2022. Downloaded from http://jitc.bmj.com/ on January 24, 2023 at Univesity Autonomous of Barcelona. Open access 10 Almeida AR, Correia DV, Fernandes-­Platzgummer A, et al. Delta one 34 Nagarsheth N, Wicha MS, Zou W. Chemokines in the cancer T cells for immunotherapy of chronic lymphocytic leukemia: clinical-­ microenvironment and their relevance in cancer immunotherapy. Nat grade expansion/differentiation and preclinical proof of concept. Clin Rev Immunol 2017;17:559–72. Cancer Res 2016;22:5795–804. 35 Aghajanian H, Kimura T, Rurik JG, et al. Targeting cardiac fibrosis 11 Wang H, Chen H, Liu S, et al. Costimulation of γδTCR and TLR7/8 with engineered T cells. Nature 2019;573:430–3. promotes Vδ2 T-­cell antitumor activity by modulating mTOR pathway 36 Rurik JG, Tombácz I, Yadegari A, et al. CAR T cells produced in vivo and APC function. J Immunother Cancer 2021;9:e003339–15. to treat cardiac injury. Science 2022;375:91–6. 12 Nishimoto KP, Barca T, Azameera A, et al. Allogeneic CD20‐ 37 Maldini CR, Ellis GI, Riley JL. CAR T cells for infection, autoimmunity targeted γδ T cells exhibit innate and adaptive antitumor activities and allotransplantation. Nat Rev Immunol 2018;18:605–16. in preclinical B‐cell lymphoma models. Clin Transl Immunology 38 Abou-­El-­Enein M, Elsallab M, Feldman SA, et al. Scalable 2022;11:1–23. manufacturing of CAR T cells for cancer immunotherapy. Blood 13 Weng RR, Lu H-­H, Lin C-­T, et al. Epigenetic modulation of immune Cancer Discov 2021;2:408–22. synaptic-­cytoskeletal networks potentiates γδ T cell-­mediated 39 Das RK, Vernau L, Grupp SA, et al. Naïve T-­cell deficits at diagnosis cytotoxicity in lung cancer. Nat Commun 2021;12:1–18. and after chemotherapy impair cell therapy potential in pediatric 14 Di Lorenzo B, Simões AE, Caiado F, et al. Broad cytotoxic targeting cancers. Cancer Discov 2019;9:492–9. of acute myeloid leukemia by polyclonal delta one T cells. Cancer 40 Das RK, O'Connor RS, Grupp SA, et al. Lingering effects of Immunol Res 2019;7:552–8. chemotherapy on mature T cells impair proliferation. Blood Adv 15 Döhner H, Estey E, Grimwade D, et al. Diagnosis and management 2020;4:4653–64. of AML in adults: 2017 ELN recommendations from an international 41 Qasim W, Zhan H, Samarasinghe S, et al. Molecular remission of expert panel. Blood 2017;129:424–47. infant B-­ALL after infusion of universal TALEN gene-­edited CAR T 16 Kantarjian H, Kadia T, DiNardo C, et al. Acute myeloid leukemia: cells. Sci Transl Med 2017;9:1–9. current progress and future directions. Blood Cancer J 2021;11:41. 42 Rasaiyaah J, Georgiadis C, Preece R, et al. TCRαβ/CD3 disruption 17 DiNardo CD, Wei AH. How I treat acute myeloid leukemia in the era enables CD3-­specific antileukemic T cell immunotherapy. JCI Insight of new drugs. Blood 2020;135:85–96. 2018;3. doi:10.1172/jci.insight.99442. [Epub ahead of print: 12 07 18 Johansson B, Harrison CJ. Acute myeloid leukemia. Cancer 2018]. Cytogenet 2010:45–139. 43 Eyquem J, Mansilla-­Soto J, Giavridis T, et al. Targeting a CAR to the 19 Bras AE, de Haas V, van Stigt A, et al. CD123 expression TRAC locus with CRISPR/Cas9 enhances tumour rejection. Nature levels in 846 acute leukemia patients based on standardized 2017;543:113–7. immunophenotyping. Cytometry B Clin Cytom 2019;96:134–42. 44 Stenger D, Stief TA, Kaeuferle T, et al. Endogenous TCR promotes in 20 Haubner S, Perna F, Köhnke T, et al. Coexpression profile of leukemic vivo persistence of CD19-­CAR-­T cells compared to a CRISPR/Cas9-­ stem cell markers for combinatorial targeted therapy in AML. mediated TCR knockout CAR. Blood 2020;136:1407–18. Leukemia 2019;33:64–74. 45 Vantourout P, Hayday A. Six-­of-­the-­best: unique contributions of γδ T 21 Baroni ML, Sanchez Martinez D, Gutierrez Aguera F, et al. cells to immunology. Nat Rev Immunol 2013;13:88–100. 41BB-­based and CD28-­based CD123-­redirected T-­cells ablate 46 Makkouk A, Yang XC, Barca T, et al. Off-­the-­shelf Vδ1 gamma delta human normal hematopoiesis in vivo. J Immunother Cancer T cells engineered with glypican-­3 (GPC-­3)-­specific chimeric antigen 2020;8:e000845. receptor (CAR) and soluble IL-­15 display robust antitumor efficacy Protected by copyright. 22 Mardiana S, Gill S. Car T cells for acute myeloid leukemia: state of against hepatocellular carcinoma. J Immunother Cancer 2021;9. the art and future directions. Front Oncol 2020;10:1–12. 47 Dudley ME, Yang JC, Sherry R, et al. Adoptive cell therapy for 23 Gill S, Tasian SK, Ruella M, et al. Preclinical targeting of human acute patients with metastatic melanoma: evaluation of intensive myeloid leukemia and myeloablation using chimeric antigen receptor-­ myeloablative chemoradiation preparative regimens. J Clin Oncol modified T cells. Blood 2014;123:2343–54. 2008;26:5233–9. 24 Kenderian SS, Ruella M, Shestova O, et al. CD33-­specific chimeric 48 Berger C, Berger M, Hackman RC, et al. Safety and immunologic antigen receptor T cells exhibit potent preclinical activity against effects of IL-­15 administration in nonhuman primates. Blood human acute myeloid leukemia. Leukemia 2015;29:1637–47. 2009;114:2417–26. 25 Liu P, Liu M, Lyu C, et al. Acute graft-­versus-­host disease after 49 Conlon KC, Potter EL, Pittaluga S, et al. Il15 by continuous humanized Anti-­CD19-­CAR T therapy in relapsed B-­ALL patients intravenous infusion to adult patients with solid tumors in a phase after allogeneic hematopoietic stem cell transplant. Front Oncol I trial induced dramatic NK-­cell subset expansion. Clin Cancer Res 2020;10:573822. 2019;25:4945–54. 26 Noviello M, Manfredi F, Ruggiero E, et al. Bone marrow central 50 Waldmann TA, Dubois S, Miljkovic MD, et al. Il-­15 in the combination memory and memory stem T-­cell exhaustion in AML patients immunotherapy of cancer. Front Immunol 2020;11. relapsing after HSCT. Nat Commun 2019;10:1–15. 51 Dubois SP, Miljkovic MD, Fleisher TA, et al. Short-­course IL-­15 given 27 Sánchez-­Martínez D, Baroni ML, Gutierrez-­Agüera F, et al. Fratricide-­ as a continuous infusion led to a massive expansion of effective NK resistant CD1a-­specific CAR T cells for the treatment of cortical cells: implications for combination therapy with antitumor antibodies. T-­cell acute lymphoblastic leukemia. Blood 2019;133:2291–304. J Immunother Cancer 2021;9. 28 Menendez P, Redondo O, Rodriguez A, et al. Comparison between a 52 Handgretinger R, Schilbach K. The potential role of γδ T cells after lyse-­and-­then-­wash method and a lyse-­non-­wash technique for the allogeneic HCT for leukemia. Blood 2018;131:1063–72. enumeration of CD34+ hematopoietic progenitor cells. Cytometry 53 Morandi F, Yazdanifar M, Cocco C, et al. Engineering the bridge 1998;34:264–71. between innate and adaptive immunity for cancer immunotherapy: 29 Ebinger S, Zeller C, Carlet M, et al. Plasticity in growth behavior focus on γδ T and NK cells. Cells 2020;9:1757. of patients' acute myeloid leukemia stem cells growing in mice. 54 Hayday AC, Vantourout P. The innate biologies of adaptive antigen Haematologica 2020;105:2855–60. receptors. Annu Rev Immunol 2020;38:487–510. 30 Sánchez-­Martínez D, Gutiérrez-­Agüera F, Romecin P, et al. Enforced 55 Melenhorst JJ, Chen GM, Wang M, et al. Decade-­long leukaemia sialyl-­Lewis-­X (SLex) display in E-­selectin ligands by exofucosylation remissions with persistence of CD4 + CAR T cells. Nature is dispensable for CD19-­CAR T-­cell activity and bone marrow 2022;602:503–9. homing. Clin Transl Med 2021;11:e280. 56 Liu E, Marin D, Banerjee P, et al. Use of CAR-­Transduced natural 31 Dalessandri T, Crawford G, Hayes M, et al. Il-­13 from intraepithelial killer cells in CD19-­Positive lymphoid tumors. N Engl J Med lymphocytes regulates tissue homeostasis and protects against 2020;382:545–53. carcinogenesis in the skin. Nat Commun 2016;7:12080. 57 Mark C, Czerwinski T, Roessner S, et al. Cryopreservation impairs 32 McKenzie DR, Hart R, Bah N, et al. Normality sensing licenses local T 3-­D migration and cytotoxicity of natural killer cells. Nat Commun cells for innate-­like tissue surveillance. Nat Immunol 2022;23:411–22. 2020;11:1–8. 33 Park I-­K, Giovenzana C, Hughes TL, et al. The Axl/Gas6 pathway is 58 Liu S, Galat V, Galat Y, et al. Nk cell-­based cancer immunotherapy: required for optimal cytokine signaling during human natural killer cell from basic biology to clinical development. J Hematol Oncol development. Blood 2009;113:2470–7. 2021;14:7. Sánchez Martínez D, et al. J Immunother Cancer 2022;10:e005400. doi:10.1136/jitc-2022-005400 13