Flt-3L Expansion of Recipient Cd8a Dendritic Cells Deletes Alloreactive

Published OnlineFirst January 24, 2018; DOI: 10.1158/1078-0432.CCR-17-2148

Cancer Therapy: Preclinical Clinical Cancer Research Flt-3L Expansion of Recipient CD8aþ Dendritic Cells Deletes Alloreactive Donor T Cells and Represents an Alternative to Posttransplant Cyclophosphamide for the Prevention of GVHD Kate A. Markey1,2,3, Rachel D. Kuns1, Daniel J. Browne1, Kate H. Gartlan1,3, Renee J. Robb1, J. Paulo Martins1, Andrea S. Henden1,2,3, Simone A. Minnie1, Melody Cheong1, Motoko Koyama1, Mark J. Smyth1,3, Raymond J. Steptoe4, Gabrielle T. Belz5, Thomas Brocker6, Mariapia A. Degli-Esposti7,8, Steven W. Lane1,2,3, and Geoffrey R. Hill1,2

Abstract

Purpose: Allogeneic bone marrow transplantation (BMT) pro- kinase-3 ligand (Flt-3L) treatment to the current clinical strat- vides curative therapy for leukemia via immunologic graft-versus- egy of posttransplant cyclophosphamide (PT-Cy) therapy. Our leukemia (GVL) effects. In practice, this must be balanced against results demonstrate superior protection from GVHD with the life threatening pathology induced by graft-versus-host disease immunomodulatory Flt-3L approach, and similar attenuation (GVHD). Recipient dendritic cells (DC) are thought to be impor- of GVL responses with both strategies. Strikingly, Flt-3L treat- tant in the induction of GVL and GVHD. ment permitted maintenance of the donor polyclonal T-cell Experimental Design: We have utilized preclinical models pool, where PT-Cy did not. of allogeneic BMT to dissect the role and modulation of Conclusions: These data highlight pre-transplant Flt-3L recipient DCs in controlling donor T-cell–mediated GVHD therapy as a potent new therapeutic strategy to delete allo- and GVL. reactive T cells and prevent GVHD, which appears particularly þ Results: We demonstrate that recipient CD8a DCs promote well suited to haploidentical BMT where the control of infec- activation-induced clonal deletion of allospeciﬁc donor tion and the prevention of GVHD are paramount. Clin Cancer T cells after BMT. We compared pretransplant fms-like tyrosine Res; 24(7); 1604–16. 2018 AACR.

Introduction these two immunologic phenomena remains the ultimate goal in the field. Furthermore, retaining functionality within the Understanding the modifiable determinants of alloreactivity nonalloreactive T-cell pool is paramount for the prevention of is critical for the design of rational strategies to prevent and severe infection that remains responsible for both short- and treat graft-versus-host disease (GVHD) after allogeneic bone long-term mortality after BMT. marrow transplantation (BMT). The elimination of GVHD Outside of selecting optimally matched donors, the preven- must be balanced against maintenance of protective graft- tion of pathogenic alloreactive T-cell responses (i.e., GVHD) versus-leukemia (GVL) effects and meaningful separation of relies on T-cell depletion or pharmacologic immune suppression, usually with calcineurin inhibitors (i.e., cyclosporin or tacrolimus). Posttransplant cyclophosphamide (PT-Cy), which 1 2 QIMR Berghofer Medical Research Institute, Brisbane, Australia. Royal Brisbane is thought to eliminate activated, proliferating alloreactive and Women's Hospital, Brisbane, Australia. 3School of Medicine, University of Queensland, Herston, Queensland, Australia. 4University of Queensland Dia- donor T cells (while sparing regulatory T-cell populations) is mantina Institute, Translational Research Institute, Brisbane, Australia. 5Walter also highly effective in reducing the incidence of chronic GVHD and Eliza Hall Institute of Medical Research, Melbourne, Victoria, Australia. after haploidentical BMT (1–5). 6Institute for Immunology, Ludwig-Maximilians Universitat, Munich, Germany. Retrospective data suggest that PT-Cy may be superior to 7 Centre for Experimental Immunology, Lions Eye Institute, Nedlands, Western standard immune suppression (with or without anti-thymocyte 8 Australia. Immunology and Virology Program, Centre for Ophthalmology and globulin) with regard to the prevention of chronic GVHD after Visual Science, University of Western Australia, Crawley, Western Australia. unrelated donor transplantation (6) although this requires con- Note: Supplementary data for this article are available at Clinical Cancer firmation in prospective randomized studies. Interestingly, the Research Online (http://clincancerres.aacrjournals.org/). effect of PT-Cy on leukemia relapse is unclear, and importantly, Corresponding Author: Kate A. Markey, Bone Marrow Transplantation Labo- no prospective studies to date have been sufficiently powered ratory, QIMR Berghofer Medical Research Institute, Brisbane, Australia. Phone: to analyze relapse after BMT using PT-Cy compared with stan- 617-3362-0290; Fax: 617-3845-3509; E-mail: [email protected] dard immune suppression. With the expanding role of alternate doi: 10.1158/1078-0432.CCR-17-2148 donor transplantation, it is critically important to understand 2018 American Association for Cancer Research. the drivers of early T-cell responses that characterize GVL effects

1604 Clin Cancer Res; 24(7) April 1, 2018

Recipient DCs Drive Deletion of Donor Ag-speciﬁc T Cells

þ / bm1 (lacking in CD8a DC, (17)], b2m , Bm1 (H-2K ), and Translational Relevance bm1 Bm1.ActmOVA (H-2K and ubiquitous ovalbumin expression Graft-versus-host disease (GVHD) is a major barrier to driven off the b-actin promoter). Female mice at 8 to 12 weeks of successful allogeneic bone marrow transplantation, a proce- age were used throughout the study. dure offering curative potential to patients with hematologic malignancies and marrow failure syndromes. In our preclin- Bone marrow chimeras ical models, pretreatment with Flt-3L (pre-T Flt-3L) expands For IRF8 / chimeras, bone marrow was harvested from þ recipient CD8a DCs that subsequently activate and delete IRF8 / mice and 5 106 bone marrow cells/mouse transferred antigen-specific donor T cells, an effect similar to that seen with by intravenous injection into lethally irradiated (1,000 cGy) B6. posttransplant cyclophosphamide (PT-Cy). Pre-T Flt-3L offers WT or B6.CD45.1 (PTprca) recipients and allowed to reconstitute protection from GVHD while facilitating the maintenance of for 3 months prior to second transplantation. third-party reactive donor T cells. Pre-T Flt-3L, like PT-Cy, resulted in marked reductions in graft-versus-leukemia effects. Bone marrow transplantation, leukemia induction, and As Flt-3L has shown acceptable safety profiles in phase I/II treatment schedules clinical trials, this approach to prevent GVHD is readily B6 recipients of C3HSw grafts received 1 106 FACS purified þ þ testable. (CD90.2 /CD4 )CD8 Tcellsand5 106 bone marrow cells. Where T cell depleted (TCD) bone marrow controls were included, cells were prepared using an antibody incubation followed by complement depletion as described previously and infectious immunity, such that they may be targeted for (18). In the C3H.Sw model, donor mice were immunized via therapeutic benefit. intraperitoneal injection with B6 splenocytes 2 weeks prior to 6 It is now clear that recipient antigen-presenting cells (APC), transplantation. B6D2F1 recipients received 5 10 TCD bone þ – fi þ both hematopoietic and nonhematopoietic, initiate alloreac- marrow cells either magnetic bead puri ed CD8 T cells (MACS-purified according to the manufacturer's instructions, tive donor T-cell responses and GVHD (7), while reconstituting þ Miltenyi Biotec), CD3 T cells (as previously described; ref. 18), donor DCs determine final GVHD severity (8, 9). As yet, no þ preventive or therapeutic approaches have been developed to or OT-I Tg CD8 T cells in doses as stated. OT-I were MACS- fi specifically target the antigen presentation component of T-cell puri ed from spleen and lymph nodes. Where OT-I T cells were fi activation. With regard to GVL responses, recipient APC (puta- transferred to read out antigen-speci c responses, MHC class – fi b / tively dendritic cells, but this is as yet unproven) are thought to I de cient ( 2m ) bone marrow was used to exclude any determine the magnitude of responses (10), with donor APC contribution of indirect antigen presentation by donor APC. playing a limited role (11). Total body irradiation (TBI) doses were as follows: B6 back- þ Given the important effects of CD8 TcellsinHLA-matched ground, 900 cGy in the presence of DT treatment and 1,000 cGy þ clinical transplantation (12), we have focused on CD8 T cell- otherwise; B6D2F1 mice, 1,100 cGy; Balb/B mice 400 cGy (in fl driven mouse models of allogeneic BMT. Here, we have defined conjunction with 2 mg udarabine d-4 to d-2). the contribution of dendritic cells (DC) to GVL effects, and For in vivo depletion of DTR-expressing DCs, diphtheria explored the influence of recipient DCs on donor T-cell acti- toxin from Corynebacterium diphtheriae (Sigma Aldrich) was vation following allogeneic BMT in models that recapitulate administered intraperitoneally at 160 ng per dose on day both HLA matched and haploidentical transplantation in the 2, 1, 0, 1, and 2 for day 3 analyses and continued on day þ clinic. Surprisingly, we find that recipient CD8a DCs potently 5, 7, 9, for day 10 analyses. Primary leukemia cells were induce antigen-specific deletion of donor cytotoxic T cells generated using the expression of the human oncogenes þ (CTL). We demonstrate that this effect can be exploited to MLL-AF9 or BCR-ABL NUP98-HOXA9 to model human AML prevent GVHD by further expanding recipient DCs with recom- and myeloid blast-crisis leukemia, as described previously (19) binant fms-like tyrosine kinase-3 ligand (Flt-3L); notably, this and cryopreserved at disease onset, for subsequent transplan- strategy appears superior to current approaches for in vivo T-cell tation. Leukemia cells were thawed on the day of injection and – 6 depletion using PT-Cy, as the GVHD outcome is superior and included in grafts at 0.5 1 10 /mouse. Mice were scored fi the polyclonal T-cell pool appears maintained. according to standard protocols and sacri ced if clinical score reached 6, in accordance with animal ethics guidelines and a previously established scoring system (20). For a death to be Methods attributed to leukemia, leukemia burden in peripheral blood Mice at either terminal or last routine bleed had to meet the fol- Female C57BL/6 (B6.WT, H-2b), Ptprca (B6.Ptprca, H-2b, lowing criteria to avoid overstatement of leukemic deaths for CD45.1; also described as B6.CD45.1 for clarity), and B6D2F1 therarecaseswhenbothGVHDandleukemiawerepresent: þ (H-2b/d) mice were purchased from the Animal Resources Centre greater than 10% GFP cells in peripheral blood (with any total (WA, Australia). The following strains were bred and housed at white cell count) or present in the peripheral blood at any level QIMRB: C3H.Sw (H-2b, Ly9.1), B6.CD11c.DOG (H-2b) and B6. but with a total WCC 1 107/mL. CD11c.DOGxDBA/2F1 (H-2Db/d; diphtheria toxin (DT) receptor, For in vivo expansion of DCs, mice received 10 mg of Flt-3 ovalbumin (OVA), and enhanced GFP (eGFP) driven off the ligand, daily via subcutaneous injection (Flt-3L; Celldex CD11c promoter), B6.CD11cGCDL (eGFP, Cre, the DTR, lucif- Therapeutics) from d-10 to -1 (21, 22). Posttransplant cyclo- erase driven off the CD11c promoter), B6.IL12p40 eYFP (13), phosphamide (PT-Cy) was administered on dþ3anddþ4at OT-I Tg, CD11c.Rac Tg (14, 15), IRF8 / mice [>10 x backcrosses 100 mg/kg i.p. Cyclosporin was administered at 5 mg/kg i.p. þ to B6 background, lacking in CD8a DC; ref. 16), Batf3 / from d0 to dþ14.

www.aacrjournals.org Clin Cancer Res; 24(7) April 1, 2018 1605

Markey et al.

For poly I:C experiments, mice received 100 mgofpolyI:C Flow cytometry (Invivogen) in saline via intraperitoneal injection 1 hour fol- A full list of mAbs utilized is given in Supplementary lowing the second dose of TBI, and prior to the injection of OT-I Table S2. Annexin V was assessed as previously described T cells. (18). Intracellular staining with activated caspase-3 was performed according to the manufacturer's instructions (BD Phar- CMV studies mingen). Flow cytometry analysis was performed using an Female C57BL/6 mice were infected intraperitoneally LSR Fortessa II (BD Biosciences) using FACSDiva software 4 with 10 PFU of salivary gland propagated MCMV-K181-Perth (Version 8.0.1). Offline analysis was performed using FlowJo for >90 days to establish a latent MCMV infection, as deter- (Version 10, Treestar). mined by the absence of replicating virus. Splenic T cells were isolated from latently infected mice and transplanted using Statistical analysis the same methods described below. MCMV-infected mice Survival curves were plotted using Kaplan–Meier estimates were housed at the University of Western Australia prior to and compared by log-rank analysis. Unpaired two-tailed fi þ being used as BMT donors. CMV-speci cCD8 T cells were Mann–Whitney tests were used throughout. Data are mean fi b identi ed using PE-conjugated tetramer for H-2K -SSPPMFRV SEM and P < 0.05 considered significant. GraphPad Prism MCMV-m38 (ImmunoID Tetramers, University of Melbourne, (Version 6.00 for Windows, GraphPad Software, www.graph Australia). pad.com) was used for the generation of graphs and for statistical analysis. Researchers were not blind to the groups Gene expression atthetimeofanalysis., P 0.05; , P 0.01; , P 0.001; Total RNA was extracted with the RNeasy Mini Plus kit (Qia- þ , P 0.0001. Sample sizes for mouse experiments gen) from sort-purified (>95% purity) CD8 donor-type T cells were estimated on the basis of our expected effect size and 2 and gene expression measured using the Qiagen RT Profiler PCR previous experience with these models. Where all recipient Array Mouse Apoptosis kit (Qiagen), and validated using TaqMan mice were WT, formal randomization was not conducted, GE assays (Applied Biosystems). but groups were matched for weight at the commencement of experiments. Xenogen imaging Bioluminescent imaging was performed to demonstrate the depletion and expansion of DCs (using B6.CD11c.GCDL mice, Results and chimeras in which the hematopoietic compartment alone Recipient DCs regulate GVL effects was of B6.CD11cGCDL origin). Recipients were injected subcu- Immune-mediated antitumor effects are increasingly recog- taneously (SC) with D-Luciferin (0.5 mg, PerkinElmer) and then nized as central to long-term disease control following clin- anesthetized with isoflurane 5 minutes before imaging using the ical transplantation. The contribution of recipient DC to þ þ Xenogen imaging system (Xenogen IVIS 100; Caliper Life CD8 T-cell–mediated GVL effects was examined in a CD8 Sciences; ref. 23). Bioluminescence (BLI) shown as photons per T-cell–dependent, miHA mismatched C3H.Sw ! B6 mouse second (ph/s). model of BMT (25). Recipient B6.CD11c.DOG mice (where diphtheria toxin receptor, OVA, and GFP are driven off the CD11c In vivo and in vitro cytotoxicity assays promoter) with either DCs intact (saline treated) or DCs depleted In vivo cytotoxicity assays were performed as follows: 10 (diphtheria toxin treated) were transplanted with C3H.Sw bone days after BMT, recipient mice received 2 107 congenic marrow and T cells, as well as B6-derived (recipient-type) MLL- þ donor-type (PTprca, CD45.1 ) unlabeled splenocytes and AF9–transduced leukemia, to mimic the clinical potential for þ 2 107 host-type B6D2F1 CD45.2 CFSE-labeled splenocytes relapse in parallel with the immunologic GVL effect exerted by IV (24). Eighteen hours later, peripheral blood and spleen the donor graft (19, 26). Unexpectedly, mice developed higher þ were analyzed for remaining donor (CD45.1 PE-stained) and leukemia burdens and died more rapidly when recipient DCs host-type (CFSE-labeled) cells by FACS analysis. The ratio of were present compared with when they had been depleted, adoptively transferred donor to recipient cells in spleen after suggesting that DCs somehow function to attenuate GVL effects 18 hours reflects degree of cytotoxicity of the resident donor (Fig. 1A and B). This model does not induce GVHD mortality T-cell population and is reported as: n donor-type/n host-type under control conditions, and as such, survival data reflects solely cells recovered. leukemia-related death. Of note, when we performed these For in vitro cytotoxicity, allogeneic (B6D2F1 BCR-ABL þ experiments using a lymphoma cell line (EL4, which generates NUP98-HOXA9) and syngeneic (B6 MLL-AF9) primary leukemia hepatosplenic masses rather than a leukemic phase) to model target cells were labeled with 51Cr, prior to culture with donor GVL there were no differences between DC-depleted and -replete þ CD8 effector T cells (FACS purified from recipient spleens on day recipients (data not shown). þ þ þ 10 following BMT, CD45.1 /CD8 ) for 5 hours at 37 C, 5% CO2. As CD8a DCs have been implicated in tolerance induction 51Cr release into culture supernatant was determined via gamma (27, 28), we next performed transplants using recipients lacking counter (TopCount microplate scintillation counter, Packard IFN regulatory factor 8 (Irf8) gene expression in the hematopoietic þ Instruments). Spontaneous release was determined using wells compartment (Irf8 / ! B6 chimeras), and thus missing CD8a containing labeled target only, and maximum release from wells DC (Fig. 1C). The improvement in survival in the isolated absence þ containing targets þ 1% Triton X100. Percentage cytotoxicity of CD8a DC mimicked that observed in the setting of pan-DC þ was determined as follows: percentage cytotoxicity ¼ (experimen- depletion (Fig. 1D and E), suggesting CD8a DCs are the key tal release spontaneous release)/(maximum release sponta- population required for control of leukemia relapse. Recipient þ neous release) 100. DCs as a whole, and the CD8a DC subset specifically are lost

1606 Clin Cancer Res; 24(7) April 1, 2018 Clinical Cancer Research

Recipient DCs Drive Deletion of Donor Ag-speciﬁc T Cells

A B **** **** 100 100

C3H.Sw TCD BM → B6.WT + DT (DC intact) 50 50 C3H.Sw BM + CD8 T → B6.WT + DT (DC intact) *** C3H.Sw BM + CD8 T → B6.CD11c.DOG + DT (DC depleted) Percent death

% GFP (peripheral blood) 0 0 7142128 0 204060 C Time (days) Time (days) B6.CD45.1 D E 100 100

*** C3H.Sw TCD BM → [B6.CD45.1 → B6] -/- C3H.Sw TCD BM → [IRF8 → B6] 50 50 C3H.Sw BM + CD8 T → [B6.CD45.1 → B6] -/- IRF8 → -/- → Percent death C3H.Sw BM + CD8 T [IRF8 B6]

% GFP (peripheral blood) 0 0 0 204060 0 204060 Time (days) Time (days) α CD8 CD11b F G H 2.5 50 100 Control TBI + )

-6 2.0 40 C3H.Sw TCD BM → B6.WT 1.5 30 C3H.Sw TCD BM → B6.CD11c Rac Tg 50 C3H.Sw BM + CD8 T → B6.WT 1.0 20 C3H.Sw BM + CD8 T → B6.CD11c Rac Tg

0.5 10 Percent death Splenic DC # ( ¥ 10 % Splenic DC IL-12 0.0 0 0 Control TBI + + + + 0 20406080 a DN DN CD8 CD8a Time cDC CD11b CD11b CD8α+DC subset

Figure 1. IL12-producing recipient CD8aþ DCs accelerate leukemia relapse. A, B6.CD11c.DOG or B6.WT mice were irradiated (900 cGy, d0) and transplanted with 1 106 sort-purified CD8þ T cells and 5 106 bone marrow cells, or TCD bone marrow alone, from C3H.Sw donors, with 1 106 GFPþ MLL-AF9 primary mouse leukemia cells. Mice were treated with saline (DC intact) or DT (DC depleted) from day 2toþ7. Mice were bled weekly and samples were counted and lysed prior to analysis via FACS for presence of GFPþ cells. Data represent three replicate experiments, d7 n ¼ 5/group; d14 and 21, n ¼ 15/group TCD mice, n ¼ 17–18/group BM þ Tmice;d28 n ¼ 2 TCD mice (only living mice at this time point from a large number of replicate mice), n ¼ 13–16/group BM þ Tmice.Mean SEM shown. d28 DCs depleted versus DCs intact; P ¼ 0.0007. B, Kaplan–Meier curve showing time to death from leukemia for two replicate experiments as described in A.Totaln ¼ 10 (TCD group); 15 (DC intact BM þT); 16 (DC depleted BMþT). TCD versus BMþTwithDCintact,P < 0.0001; BMþTDCintactversusDCdepleted,P < 0.0001. C, Splenic DC subsets in WT [B6.CD45.1 ! B6] chimeras and [IRF8/ ! B6] chimeras. D, Leukemia engraftment of IRF8/ chimeric mice receiving C3H.Sw grafts as described. E, Kaplan–Meier curve showing time to death from leukemia for two replicate experiments. n ¼ 7–9 in TCD groups, n ¼ 17 in each BMþTgroup.F, Splenic DC were enumerated in from WT B6 mice at rest, and 18 hours following TBI (1,100 cGy). n ¼ 9 for control mice, 16 for TBI group. Data shown from three replicate experiments. G, IL12 production was measured using IL12 eYFP reporter mice, as described in F. n ¼ 9 for control animals, 8 for post-TBI. H, CD11c.Rac1 Tg recipient mice were transplanted as described in A.Kaplan–Meiercurveforleukemiadeathfrom two replicate experiments, n ¼ 10/TCD group, n ¼ 16/BMþTgroups. rapidly after total body irradiation (TBI), and as expected, the Recipient DCs induce apoptosis in donor T cells, which results þ CD8a DCs preferentially produce IL12, both homeostatically in contraction of the polyclonal T-cell compartment þ and following TBI (Fig. 1F and G). To understand the profound modulation of CD8 Tcell- þ CD8a DCs are known to be highly efficientattheuptake mediated GVL effects by recipient DCs, we examined the donor þ of cell-associated antigen from dying cells (29, 30); we there- T-cell compartment after BMT. Donor CD8 T-cell numbers fore investigated the contribution of this pathway (i.e., cross- were equivalent following transplantation into DC-depleted or presentation) to leukemia control after allogeneic BMT. We DC-intact mice at d3, but were significantly reduced by d7 in utilized Rac1 transgenic recipients, where DCs specifically lack recipients with DCs intact (Fig. 2A). This was associated with a the capacity to acquire apoptotic antigen due to deficiency of marked increase in donor cells undergoing apoptosis at both þ theRhoGTP-aseRac1withinCD11c cells (14). Surprisingly, time points examined (Fig. 2B). The same effect was observed in there was no impairment in GVL (Fig. 1H), suggesting that both IFR8 / ! B6 chimeras and Batf3 / recipients (17) that þ recipient DCs regulate GVL effects via the presentation of lack the CD8a DC subset in isolation (Fig. 2C and D). þ endogenous alloantigen, rather than cross-presentation of We next analyzed donor CD8 Tcellssort-purified from DC exogenous alloantigen. depleted or intact recipients on d7 using a targeted PCR array.

www.aacrjournals.org Clin Cancer Res; 24(7) April 1, 2018 1607

Markey et al.

Figure 2. Recipient DCs induce apoptosis in polyclonal donor CD8þ T cells. A, Recipient B6.CD11c.DOG mice were treated with DT or saline and transplanted with bone marrow and sort-purified CD8þ T cells from C3H.Sw donors. Splenic donor CD8þ T cells were analyzed at day 3 and 7, with enumeration from two replicate experiments shown. n ¼ 9/group at d3 and 10/group at d7. B, Proportion of cells undergoing apoptosis, defined by Annexin Vþ/7AAD shown in left. Representative FACS plots in right panel. Data shown from two replicate experiments. n ¼ 5/group at d3, 10/group at d7. IRF8/ chimeras (C)andBatf3/ mice were used as recipients (D). Cells were phenotyped using flow cytometry. Enumeration was performed on the basis of whole-organ WBC (established by Coulter count) and phenotype and data from day 3 and 7 is shown. Data represent two replicate experiments for each. n ¼ 5/group at d3, 10/group at d7 for the [IRF8/ ! B6] and n ¼ 8/group/time point in the Batf3/ experiments. E, Apoptosis gene array on sort-purified donor T cells (CD8þ/Ly9.1þ from the C3H.Sw ! B6.CD11c.DOG DT transplant model. Differentially regulated genes (Igfr1, Trp63, Trp73,andDapK1)shown in red. Array performed on individual sorted T cells from day 7 posttransplant, n ¼ 4/group. F, TaqMan real-time qPCR results for d10 ex vivo T cells from the B6 ! B6.CD11c.DOG DBA/2F1 DT system. qPCR-sorted splenic T cells from individual mice, n ¼ 4/group.

1608 Clin Cancer Res; 24(7) April 1, 2018 Clinical Cancer Research

Recipient DCs Drive Deletion of Donor Ag-speciﬁc T Cells

We identified four genes that were expressed at higher levels in transfer when compared with saline-treated controls (Fig. 4C the T cells from mice that had not been exposed to recipient and D). Strikingly, by day 3, the reverse was true, with OT-I DCs, all of which are involved in regulation of cellular growth undetectable in Flt3-L–treated animals, consistent with and apoptosis: Insulin-like growth factor 1 receptor (Igfr1;4.44 the induction of widespread deletion following initial stimu- fold), transformation-related protein 63 and 73 (Trp63;2.29 lation and early expansion (Fig. 4E). By day 10, splenic OT-I fold and Trp73;3.00fold),anddeathassociatedproteinkinase numbers were equivalent in the Flt-3L and saline groups, 1(DapK1; 2.22 fold; Fig. 2E). This likely reflects the loss of, suggesting a reexpansion of surviving T cells in the Flt-3L or lack of requirement for, regulation among the nonalloreac- pretreated mice. These were, however, markedly different tive pool of T cells that remain post-DC-mediated deletion. to the OT-I present in saline pretreated mice, skewed toward These changes in gene expression were validated by qPCR in the aTEM/TEFF phenotype and a high proportion were undergoing haploidentical model at d10, with consistent results obtained active apoptosis, as measured by activated caspase-3 intracel- (Fig. 2F). The full panel of genes analyzed is listed in Supple- lular staining (Fig. 4F). mentary Table S1. Flt-3L–expanded DCs specifically delete alloreactive T cells þ Recipient DC:CD8 T-cell interactions result in while maintaining the polyclonal T-cell pool antigen-specific deletion and loss of cytolytic function To examine the functional consequences of Flt-3L expansion We next measured antigen-specificT-cellfunction on alloreactivity, we first enumerated polyclonal B6 T cells þ þ þ using CD8 OT-I transgenic donor T cells and B6.CD11c. (both CD4 and CD8 ) following transplantation into saline DOGxDBA/2 F1 recipients, with or without DT treatment to or Flt-3L–treated B6D2F1 recipients, a model of haploidentical deplete DCs. Ovalbumin (OVA), which is expressed by recip- transplantation in the clinic. Interestingly, polyclonal T cells ient DC (driven by the CD11c promoter), serves as a model were preserved in the setting of Flt3L treatment (Fig. 5A). To alloantigen in this system. OT-I T cells exclusively recognize the explore the role of antigen-specific T-cell deletion in leukemia OVA-derived peptide SIINFEKL in the context of MHC class I, relapse, Pre-T Flt-3L recipient mice were transplanted with and therefore OT-I responses quantify DC-specific, endogenous BCR-ABL þ NUP98-HOXA9 cotransduced primary leukemia presentation of OVA-derived peptide. Once again, we observed with bone marrow polyclonal CD3 T cells (31, 32). Median DC-associated compartment size contraction and enhanced leukemia survival was just 11 days in the Pre-T Flt-3L BM þ T apoptosis (Fig. 3A and B), as well as hyperactivation of OT-I group, equivalent to recipients of T-cell–depleted (TCD) con- transgenic T cells (CD25 and CD69 expression, Fig. 3C). OT-I trol grafts. The saline pretreated mice receiving bone marrow þ acquired an effector phenotype (CD62Lneg/CD44high)more T grafts survived long-term (Fig. 5B and C). rapidly in the presence of DCs (78.0 0.9% in DC replete We confirmed that this was a DC-specific effect by using Flt-3L vs. 33.2 1.8% in DC deplete at 3 days after BMT; Fig. 3D). pretreatment in combination with the DT-depletion approach Strikingly, recipient DCs induced high levels of exhaustion previously described (Fig. 5D and E). The mice treated with Flt-3L markers PD-1 and Lag3 on T cells early after BMT, and Tim3 and DT (such that DCs were expanded and subsequently deleted) was present at high levels, but was not differentially expressed had equivalent leukemia control to that seen in the WT bone (Fig. 3E). Exhaustion marker expression was markedly marrow þ T recipients pretreated with saline (Fig. 5E), confirming decreased in both T-cell groups by d10, with enhanced expres- the primacy of recipient DCs in this effect. sion seen in the T cells from DC-depleted recipients likely To assess whether this deletion was due to the potentially þ reflecting the delayed, more measured activation that occurs tolerogenic nature of CD8a DCs, we treated mice with Flt-3L when initial interaction with DC is denied. or saline as described, and then activated recipient DCs with To assess the impact of DCs on antigen-specific T cells within toll-like receptor 3 (TLR3) ligand polyinosinic:polycytidylic þ þ a polyclonal pool (containing both CD4 and CD8 Tcells), acid (poly I:C) prior to the infusion of antigen-specificOT-IT we used the parent-into-F1 (B6 ! B6D2F1) model of haploi- cells. As expected, DCs were highly activated following poly I: dentical transplantation and waited until donor T-cell numbers C treatment (Fig. 5F; ref. 33). Importantly, complete deletion were equal between DC-depleted and -replete recipients (d10, of antigen-specific T cells in Flt-3L pretreated recipients was Fig. 3F). We noted reduced cytotoxic function in vivo (Fig. 3G), maintained in the setting of this DC activation (Fig. 5G and a striking absence of in vitro cytotoxic function against and H). Interestingly, while deletion was most effective in recipient-type leukemia (Fig. 3H), consistent with the specific the Flt-3L pretreated mice, poly I:C activation of residual deletion of alloreactive CTL, even when a full complement of DCsalsoleadtomarkeddonorT-celldepletioninsaline polyclonal T cells are present in the initial graft. pretreated recipients. Thus, the ability of recipient DCs to attenuate GVHD reflects direct alloantigen presentation and Expansion of recipient DCs with Flt-3L results in an early their capacity as highly potent APC rather than any intrinsic expansion of antigen-specific T cells followed by their complete regulatory property. deletion in lymphoid organs We next assessed the impact of posttransplant cyclophos- þ Having established recipient CD8a DCsaskeydriversof phamide (PT-Cy) on antigen-specific T-cell depletion, main- alloantigen-specific donor T-cell deletion, we sought to further tenance of the polyclonal T-cell pool and ultimately, GVL harness this effect by administering Flt-3L pretransplantation effects. As expected, the high-dose chemotherapy effectively þ (Pre-T Flt-3L). Expansion of CD8a DC was confirmed by eliminated the alloreactive OT-I T cells (Fig. 5I), but also had bioluminescence and flow cytometry (Fig. 4A and B). When significant impact on the polyclonal T-cell pool when WT B6 OT-I T cells were transferred into Pre-T Flt-3L conditioned grafts were transplanted (Fig. 5J). PT-Cy had a significant B6.CD11c.DOGxDBA/2F1 recipients, OT-I were present in impact on leukemia relapse. While the PT-Cy TCD group increased numbers in spleen and lymph nodes 12 hours after experienced delayed relapse compared with saline-treated

www.aacrjournals.org Clin Cancer Res; 24(7) April 1, 2018 1609

Markey et al.

A B C

30,000 1,500

+ 20,000 1,000

10,000 500 % Annexin V

0 0 Day 3 Day 10 Day 10 Day 10 Day 10

D E

15,000 4,000 8,000 Day 3 3,000 6,000 10,000

2,000 4,000

5,000 1,000 2,000

0 0 0 Day 10 Day 3 Day 10 Day 3 Day 10 Day 3 Day 10 Day 3 Day 10 Day 3 Day 10

F G H

Day 10

Figure 3. Donor T-cell deletion is antigen-specific. A, Recipient B6.CD11c.DOG DBA/2 F1 mice were treated with DT or saline from d2toþ9, received TBI on d0 6 6 / and were transplanted with 1 10 OT-I T cells and 5 10 b2m BM. d3 and d10 enumeration. Four replicate experiments, n ¼ 7/group at d3; 8, 9, 17 respectively at d10. B, Proportion of d10 OT-I T cells undergoing apoptosis, n ¼ 7/group from two replicate experiments. C, Plots show splenic activation marker expression by OT-I T cells on d10. n ¼ 4/group, representative data from one of two replicate experiments. D, Effector/memory phenotype; representative flow cytometry plots and enumeration. Data shown are representative of three replicate experiments. n ¼ 5/group at d3, n ¼ 10/group at d10. E, Exhaustion markers at day 3 and 10; MFI shown for Lag3, PD-1, Tim3. Data representative of three replicate experiments. Data shown, n ¼ 5/group for PD-1 and 10/group for Lag3 and Tim3. F, Polyclonal B6 CD8þ T cells were transplanted into B6.CD11c.DOG DBA/2F1 recipients. T cells were enumerated at day 10. Data shown from four replicate experiments, n ¼ 17/group. G, In vivo cytotoxicity assays were performed as described, cytotoxicity index is reported, with data shown combined from three experiments, n ¼ 14–15/group. H, Polyclonal donor CD8þ T cells were sort-purified on day 10 post-BMT and plated with allogeneic (B6D2F1 BCR-ABL þ NUP98-HOXA9) and syngeneic (B6 MLL-AF9) primary leukemias in a chromium release assay. Data shown is one of two experiments performed with equivalent results. Each replicate experiment, n ¼ 5/group (T cells purified from individual mice).

animals (median survival 22 days compared with 14.5), likely CsA-treated mice relapsed in an equivalent fashion to the saline due to direct effects of the cyclophosphamide on the leukemia treated controls, and the PT-Cy mice once again relapsed at cells. PT-Cy resulted in rapid relapse in the bone marrow þ T similar rates whether T cells were present in the graft, or grafts recipients (median survival 26 days, compared with unreach- were TCD marrow alone. ed in the saline-treated controls, as no mice died of leukemia; Fig. 5K). Flt-3L pretreatment is a superior strategy for prevention of Following on from this striking and rapid relapse in the GVHD when compared with PT-Cy setting of PT-Cy, we next performed experiments using clini- Until there are clear therapeutic strategies which separate cally relevant calcineurin inhibition (CsA, 5 mg/kg, day 0–14, GVHD and GVL, it is likely that any strategy chosen to prevent achieving trough levels between 200 and 300 mg/dL; Fig. 5L). and treat GVHD will have some inﬂuence on relapse risk. In

1610 Clin Cancer Res; 24(7) April 1, 2018 Clinical Cancer Research

Recipient DCs Drive Deletion of Donor Ag-speciﬁc T Cells

A B

Day 0 Day 0

C D

E F

Day 3 Day 10 Day 10 Day 10 Day 10

Figure 4. Flt-3L pretreatment results in early antigen-specific T-cell expansion followed by near-complete deletion. A, B6.CD11c.GCDL mice (luciferase driven off the CD11c promoter) were pretreated with Flt-3L for 10 days and imaged using the Xenogen IVIS. Representative images are shown of 5 mice/group. Quantification is demonstrated for whole body, n ¼ 5/group. B, Flow cytometry plots demonstrating DC identification and subset, n ¼ 5/group. C, OT-I T cells were transferred into irradiated B6.CD11c.DOG DBA/2 F1 recipients pretreated with either Flt-3L or saline for 10 days. Mesenteric lymph node (mLN), peripheral lymph node (pLN) and spleen shown, plots are concatenated and representative of 4 individual animals harvested at 12 hours post-transfer. Enumeration shown in D. E, Representative flow plots of spleen at day 3. Data pooled from two replicate experiments (n ¼ 10/group). OT-I T cells were enumerated in spleen, at day 3 and day 10. F, T-cell phenotype at day 10 from experiment as described in C. Donor OT-I T cells were identified on the basis of their TCR expression (Va2/Vb5.1)andphenotypedonthebasisofCD62LandCD44expression. Apoptosis was assessed using intracellular staining for activated caspase-3. Data pooled from two replicate experiments n ¼ 6–7/group. that setting, we sought to assess the relative benefitfroma superior to PT-Cy (median survival 45 vs. 36 days) for GVHD point of view from PT-Cy and Flt-3L pretreatment, GVHD prevention (Fig. 6A) in a haploidentical model. Given given that both have a detrimental (and equivalent) impact that Flt-3L is a myeloid growth factor, efforts to translate on leukemia relapse. We found that Flt-3L pretreatment was this strategy for GVHD prevention would need to begin in

www.aacrjournals.org Clin Cancer Res; 24(7) April 1, 2018 1611

Markey et al.

A B C D

E F G H

I J K L

Figure 5. Flt-3L pretreatment results in functional deletion of the alloreactive T-cell pool with maintenance of the polyclonal T-cell pool. A, CD3þ T cells from B6 donors were transplanted into B6D2F1 mice pretreated with Flt-3L or saline as described. Enumeration of polyclonal splenic T cells on d3 is shown. Data are from two replicate experiments, n ¼ 8/group. B, B6D2F1 mice were treated with either Flt-3L or saline for 10 days prior to TBI and transplantation of B6. WT BM 0.5 106 CD3þ TwithBCR-ABLþ NUP98-HOXA9 leukemia cells on day 0. Kaplan–Meier survival curve shown. Data shown are combined from two replicate experiments. n ¼ 9/TCD BM group, n ¼ 14/BMþTgroup.C, Leukemia burden at d10 is shown, from the two combined experiments described. BM þ T grafts, saline versus Flt-3L, P < 0.0001. D, B6D2F1 mice were treated with Flt-3L and either saline or DT. Representative FACS plots pre- and post-TBI shown. Plots shown are gated on FSC/SSC, 7AAD-negative, single cells. Axes for each plot are as shown. E, Mice were transplanted following treatment described in D. Data shown is from two combined replicate experiments. n ¼ 9 in TCD group; 12 in saline controls, 16 in Flt-3L DT depletion. F, Recipient dendritic cell activation status 18 hours following Poly I:C treatment (100 mg, i.p.). Gated on CD11cþ/MHC class IIþ cDC. G, Mice were pretreated with Flt-3L as described, and additionally treated with either Poly I:C or saline post-TBI and prior to the injection of 1 106 OT-I T cells. Representative FACS plots shown in G are gated on FSC/SSC, 7AAD-negative, single cells. Enumeration of splenic OT-I cells in spleen on day 3 is shown in H with data pooled from two replicate experiments, n ¼ 8/group. I, Antigen-specific OT-I CD8 T cells were transplanted as previously described into irradiated B6.CD11c.DOGxDBA/2F1 recipients. Mice were treated with saline or cyclophosphamide on day 3 and 4 posttransplant, and spleens analyzed on day 7. Data shown from two replicate experiments, n ¼ 13 (saline) and n ¼ 5(PT-Cy).Representativeflow cytometry plots are shown, gated on FSC/SSC, 7AAD-negative, single cells. OT-I T cells are defined on the basis of TCR expression (Va2/Vb5.1). J, CD3þ T cells from B6 donors were transplanted into B6D2F1 mice and treated with PT-Cy as described. Enumeration of polyclonal splenic T cells on d7 is shown. Data shown is representative of two replicate experiments, n ¼ 4/group. K, B6D2F1 mice were transplanted as described above, and treated with saline or 100 mg/kg cyclophosphamide on d3 and d4. Kaplan–Meier curve for leukemia death. Data shown is pooled from three replicate experiments, n ¼ 14 in saline and Cy treated TCD; n ¼ 22 in the BMþT. L, B6D2F1 mice were transplanted as described, with grafts containing BCR-ABL þ NUP98-HOXA9 primary leukemia, but treated with either PT-Cy or CsA (5 mg/kg/day for day 1–14). Kaplan–Meier curve shown for leukemia death. Data are pooled from two replicate experiments, n ¼ 7ineachTCDgroupand11ineachBMþTgroup.

the lymphoid malignancies, in which setting patients are often this setting, with no GVHD deaths in the Flt-3L pretreated conditioning with reduced intensity regimens. We therefore recipients of bone marrow þ T grafts (Fig. 6B), and equivalent performed transplants in an additional miHA mismatch mod- engraftment. el (B6 to Balb/B) with ﬂudarabine and low-dose TBI condi- Having observed profound effects on broad alloreactivity (i.e., tioning. The beneﬁt of Flt3L pretreatment was maintained in both GVL and GVHD) in response to both Pre-T Flt-3L and PT-Cy,

1612 Clin Cancer Res; 24(7) April 1, 2018 Clinical Cancer Research

Recipient DCs Drive Deletion of Donor Ag-speciﬁc T Cells

A C

Day 3

B D

Day 7

Figure 6. Flt-3L treatment results in superior GVHD prevention when compared with PT-Cy. A, B6D2F1 mice were pretreated with either Flt-3L or saline, transplanted with B6 BM 5 106 CD3þ T cells, and administered posttransplant cyclophosphamide or saline on dþ3, dþ4. Kaplan–Meier survival curve shown. Pooled data from three replicate experiments, n ¼ 15/group. B, Balb/B mice were conditioning with fludarabine (2 mg/dose d3, 2, 1) and sublethal irradiation (400 cGy, d1), followed by transplantation with B6 grafts (1 107 BM CD3þ T cells). Kaplan–Meier survival curve shown, pooled from three replicate experiments. n ¼ 10 in the TCD arm and 18 in each of the BM þ Tarms.C, 5 106 CD3þ T cells from MCMV immune donors were transplanted into B6D2F1 mice pre-treated with Flt-3L or saline as described. Splenic T cells, including MCMV-specificm38þ CD8þ T cells, were examined 3 days later; enumeration is shown in the right panels. FACS data shown is representative of two replicate experiments. Enumeration data is pooled n ¼ 8/group. D, Mice were transplanted as described, but treated with cyclophosphamide on dþ3anddþ4 and splenic T cells, including MCMV-specificm38þ CD8þ T cells examined 3 days later at dþ7(n ¼ 4/group). we sought to examine maintenance of virus-specific immunity but in the era of posttransplantation cellular therapies (e.g., using murine cytomegalovirus (MCMV) immune B6 donor mice with engineered CAR T cells or suicide-gene transfected poly- in the haploidentical B6 ! B6D2F1 model. We hypothesized that clonal donor T cells) it is likely that this can be overcome. in the absence of CMV-specific antigen in the peritransplant Importantly, pretransplant Flt-3L therapy appears to spare the period, the CMV-specific memory T-cell pool would be intact polyclonal T-cell pool and in light of the importance of oppor- following transplant into Flt-3L-pretreated recipients. At the time tunistic infection after transplant, this represents a significant of maximal clonal deletion of alloreactive T cells, CMV-specific advantage over nonselective chemotherapy-based approaches þ (m38 tetramer cells) were indeed present in equivalent numbers for the deletion of alloreactive T-cell populations that are in Flt-3L pretreated mice compared with saline pretreated controls currently used. In addition to the polyclonal T-cell pool, Flt- (Fig. 6C). In contrast, there was profound reduction in all T 3L appears to spare CMV-specific immunity where PT-Cy does lymphocytes in the PT-Cy mice, with almost complete loss of not. This is likely explained by the relative absence of viral CMV-specific CD8 T cells (Fig. 6D). Thus unlike PT-Cy, pre-T Flt- antigen in the immediate post-transplant period. 3L preserves virus-specific T cells. Mechanistically, we demonstrate that alloantigen presenta- þ tion within MHC class I by recipient CD8a DC results in activation induced cell death and exhaustion of allospecific þ Discussion CD8 cytolytic T cells. We thus demonstrate the ability of DCs þ In this study, we have examined the role of recipient DCs in to delete MHC class I–dependent GVL effects, as well as CD8 þ controlling CD8 T-cell–mediated alloreactivity and propose T-cell–mediated GVHD. A previous study using Batf3-deficient that pretransplant Flt-3L therapy may provide a new strategy recipients and a T-cell lymphoma cell line to model relapse þ for the prevention of GVHD in the clinic. It is clear that this after BMT reported that CD8a DCs were required for optimal comes at the cost of decreasing T-cell–mediated GVL effects, GVL effects, which is in contrast to our findings using primary

www.aacrjournals.org Clin Cancer Res; 24(7) April 1, 2018 1613

Markey et al.

myeloid leukemias (34). Indeed, we could see no effect of clinical studies can analyze the relative impact of these com- recipient DC depletion on GVL against the EL4 cell line (data parative GVHD prophylaxis strategies on GVL effects and not shown). This likely reflects the high mutational burdens in relapse after BMT. Given that Flt-3L is a myeloid growth factor, the multiply passaged cell lines used in those studies relative its receptor (Flt3) is ubiquitously expressed by AML blasts (44), to primary leukemia (35) and the fact that these models are and that Flt3 mutations portray an adverse prognosis in AML poor discriminators of quantitative defects in GVL (12). The (45), we would not advocate its use in patients undergoing results presented here also serve to confirmandextenddata BMT for myeloid malignancies. Instead, we propose that initial from ourselves and others (7, 36, 37) demonstrating that trials be undertaken in patients with lymphoid malignancies. In recipient DCs are not required for the induction of GVHD. addition, we hypothesize that Pre-T Flt-3L, like PT-Cy, may be Indeed, we have previously shown that recipient DCs also most appropriate for use in haploidentical BMT, where HLA potently delete antigen-specific donor CD4 T cells to attenuate mismatches are permissive of strong GVL effects mediated by GVHD (7). low numbers of donor T cells escaping deletion. These strate- Our work demonstrates that DCs act to control alloreactivity gieswouldalsoseemparticularlysuitedtononmalignant via their profound stimulatory capacity, which results in clonal conditions, indolent hematopoietic malignancies, or settings T-cell deletion. This effect can be augmented by expanding where an engineered graft is employed (e.g., CAR or suicide- þ recipient CD8a DCs (e.g., with Flt-3L), or enhancing residual gene T cells; ref. 46) such that there is little reliance on donor T DC activation (e.g., with poly I:C). Interestingly, the adminis- cells within the graft for antitumor effects. Our data demon- tration of exogenous IL12 within 12 hours of BMT has been strating maintenance of CMV-specificmemoryTcellswiththe shown to prevent GVHD by inducing donor T-cell apoptosis Flt-3L approach may mean that this strategy results in lower þ and is again consistent with the CD8a DC–mediated effect infection risk in the clinic, when compared with other T-cell demonstrated here (38–40). While the administration of IL12 depletion approaches that inevitably target both allo-specific has significant potential for toxicity in clinical BMT and has not and bystander T cells. progressed into the clinic, Flt-3L has been widely administered to patients (41, 42) and does represent a feasible clinical Disclosure of Potential Conflicts of Interest approach. No potential conflicts of interest were disclosed. When Flt-3L therapy was previously studied in GVHD, it appeared to have mixed effects, dependent on the timing of administration (21, 43). Flt-3L administration post-BMT expan- Authors' Contributions ded donor DCs and resulted in the expected acceleration of Conception and design: K.A. Markey, M.A. Degli-Esposti, G.R. Hill GVHD (43). Conversely, treatment of recipients prior to BMT Development of methodology: K.A. Markey, S.W. Lane, G.R. Hill appeared to reduce GVHD, and the authors observed lower Acquisition of data (provided animals, acquired and managed patients, numbers of donor T cells early after transplantation. At that provided facilities, etc.): K.A. Markey, R.D. Kuns, D.J. Browne, K.H. Gartlan, – R.J. Robb, J.P. Martins, S.A. Minnie, M. Cheong, M. Koyama, R.J. Steptoe, G. Belz, time, this was interpreted as a failure of Flt-3L expanded DCs to M.A. Degli-Esposti stimulate T-cell expansion, as DCs were thought to have an Analysis and interpretation of data (e.g., statistical analysis, biostatistics, exclusively stimulatory function (21). computational analysis): K.A. Markey, K.H. Gartlan, A.S. Henden, M. Koyama, þ Expansion of CD8a DC using Pre-T Flt-3L represents a G. Belz, M.A. Degli-Esposti potent immunomodulatory strategy that appears to be at least Writing, review, and/or revision of the manuscript: K.A.Markey,R.D.Kuns, equivalent to high-dose PT-Cy for the deletion of alloreactive T K.H. Gartlan, R.J. Robb, M.J. Smyth, G. Belz, M.A. Degli-Esposti, S.W. Lane, G.R. Hill cells using our haploidentical transplant model. While PT-Cy Administrative, technical, or material support (i.e., reporting or organizing has been shown to spare regulatory T cells (4), it is still highly data, constructing databases): D.J. Browne, M.J. Smyth þ active in systems where CD4 T cells are absent (5), suggesting Study supervision: G.R. Hill that the augmentation of recipient DC-induced clonal deletion Other (provision of mice): T. Brocker is highly relevant to the protection seen with this strategy. DCs fi are specialized cells that are highly ef cient at presenting Acknowledgments limiting quantities of antigen and their paradoxical ability to TheauthorswouldliketoacknowledgetheassistanceofGrace decrease alloreactive T-cell responses reflects the nonphysiolo- Chojnowski, Paula Hall, and Michael Rist. T. Brocker is supported by DFG gic nature of transplantation. Importantly, the stimulatory SFB 1054 B03. The project was supported by grants from the NHMRC, capability of recipient DCs, including their secretion of IL12, Australia. is markedly augmented by TBI (12). Furthermore, unlike path- ogen-derived antigen, alloantigen is ubiquitous, present in The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked excess, and persists indefinitely. advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate It is clear from our data that both PT-Cy and Pre-T Flt-3L have this fact. important implications for GVL effects, and their use for GVHD prevention must be balanced against the inevitable impact on Received July 25, 2017; revised November 1, 2017; accepted January 8, 2018; GVL activity. Thus, only well-designed and large prospective published OnlineFirst January 24, 2018.

References 1. Ciurea SO, Zhang MJ, Bacigalupo AA, Bashey A, Appelbaum FR, Aljitawi mide vs. matched unrelated donor transplant for acute myeloid leukemia. OS, et al. Haploidentical transplant with posttransplant cyclophospha- Blood 2015;126:1033–40.

1614 Clin Cancer Res; 24(7) April 1, 2018 Clinical Cancer Research

Recipient DCs Drive Deletion of Donor Ag-speciﬁc T Cells

2. Luznik L, O'Donnell PV, Fuchs EJ. Post-transplantation cyclophosphamide transplants expands host CD8 alpha(þ) dendritic cells and re- for tolerance induction in HLA-haploidentical bone marrow transplanta- duces experimental acute graft-versus-host disease. Blood 2002;99: tion. Semin Oncol 2012;39:683–93. 1825–32. 3. McCurdy SR, Kanakry JA, Showel MM, Tsai HL, Bolanos-Meade~ J, Rosner 22. Markey KA, Koyama M, Gartlan KH, Leveque L, Kuns RD, Lineburg KE, et al. GL, et al. Risk-stratified outcomes of nonmyeloablative HLA-haploiden- Cross-dressing by donor dendritic cells after allogeneic bone marrow tical BMT with high-dose posttransplantation cyclophosphamide. Blood transplantation contributes to formation of the immunological synapse 2015;125:3024–31. and maximizes responses to indirectly presented antigen. J Immunol 4. Kanakry CG, Ganguly S, Zahurak M, Bolanos-Meade~ J, Thoburn C, Perkins 2014;192:5426–33. B, et al. Aldehyde dehydrogenase expression drives human regulatory T cell 23. Morris ES, MacDonald KP, Rowe V, Banovic T, Kuns RD, Don AL, et al. NKT resistance to posttransplantation cyclophosphamide. Sci Transl Med cell-dependent leukemia eradication following stem cell mobilization 2013;5:211ra157. with potent G-CSF analogs. J Clin Invest 2005;115:3093–103. 5. Ganguly S, Ross DB, Panoskaltsis-Mortari A, Kanakry CG, Blazar BR, 24. Banovic T, MacDonald KP, Morris ES, Rowe V, Kuns R, Don A, et al. TGF- Levy RB, et al. Donor CD4þ Foxp3þ regulatory T cells are necessary beta in allogeneic stem cell transplantation: friend or foe? Blood for posttransplantation cyclophosphamide-mediated protection 2005;106:2206–14. against GVHD in mice. Blood 2014;124:2131–41. 25. Robb RJ, Lineburg KE, Kuns RD, Wilson YA, Raffelt NC, Olver SD, et al. 6. Bashey A, Zhang X, Jackson K, Brown S, Ridgeway M, Solh M, et al. Identification and expansion of highly suppressive CD8(þ)FoxP3(þ) Comparison of outcomes of hematopoietic cell transplants from T- regulatory T cells after experimental allogeneic bone marrow transplanta- replete haploidentical donors using post-transplantation cyclophospha- tion. Blood 2012;119:5898–908. mide with 10 of 10 HLA-A, -B, -C, -DRB1, and -DQB1 allele-matched 26. Zhang P, Tey SK, Koyama M, Kuns RD, Olver SD, Lineburg KE, et al. unrelated donors and HLA-identical sibling donors: a multivariable Induced regulatory T cells promote tolerance when stabilized by rapamycin analysis including disease risk index. Biol Blood Marrow Transplant and IL-2 in vivo. J Immunol 2013;191:5291–303. 2016;22:125–33. 27. Hawiger D, Inaba K, Dorsett Y, Guo M, Mahnke K, Rivera M, et al. Dendritic 7. KoyamaM,KunsRD,OlverSD,RaffeltNC,WilsonYA,DonAL,etal. cells induce peripheral T cell unresponsiveness under steady state condi- Recipient nonhematopoietic antigen-presenting cells are sufficient tions in vivo. J Exp Med 2001;194:769–79. to induce lethal acute graft-versus-host disease. Nat Med 2012;18: 28. Bonifaz L, Bonnyay D, Mahnke K, Rivera M, Nussenzweig MC, Steinman 135–42. RM. Efficient targeting of protein antigen to the dendritic cell receptor DEC- 8.MacDonaldKP,KunsRD,RoweV,MorrisES,BanovicT,Bofinger H, 205 in the steady state leads to antigen presentation on major histocom- et al. Effector and regulatory T-cell function is differentially regulated by patibility complex class I products and peripheral CD8þ T cell tolerance. RelB within antigen-presenting cells during GVHD. Blood 2007;109: J Exp Med 2002;196:1627–38. 5049–57. 29. Schulz O, Reis e Sousa C. Cross-presentation of cell-associated antigens by 9. Markey KA, Banovic T, Kuns RD, Olver SD, Don AL, Raffelt NC, et al. CD8alphaþ dendritic cells is attributable to their ability to internalize dead Conventional dendritic cells are the critical donor APC presenting alloan- cells. Immunology 2002;107:183–9. tigen after experimental bone marrow transplantation. Blood 2009;113: 30. Iyoda T, Shimoyama S, Liu K, Omatsu Y, Akiyama Y, Maeda Y, et al. The 5644–9. CD8þ dendritic cell subset selectively endocytoses dying cells in culture 10. Reddy P, Maeda Y, Liu C, Krijanovski OI, Korngold R, Ferrara JL. A crucial and in vivo. J Exp Med 2002;195:1289–302. role for antigen-presenting cells and alloantigen expression in graft-versus- 31. Gartlan KH, Markey KA, Varelias A, Bunting MD, Koyama M, Kuns RD, et al. leukemia responses. Nat Med 2005;11:1244–9. Tc17 cells are a proinflammatory, plastic lineage of pathogenic CD8þ T 11. Matte CC, Liu J, Cormier J, Anderson BE, Athanasiadis I, Jain D, et al. Donor cells that induce GVHD without antileukemic effects. Blood 2015;126: APCs are required for maximal GVHD but not for GVL. Nat Med 1609–20. 2004;10:987–92. 32. Dash AB, Williams IR, Kutok JL, Tomasson MH, Anastasiadou E, 12. Markey KA, MacDonald KP, Hill GR. The biology of graft-versus-host Lindahl K, et al. A murine model of CML blast crisis induced by disease: experimental systems instructing clinical practice. Blood 2014; cooperation between BCR/ABL and NUP98/HOXA9. Proc Natl Acad 124:354–62. Sci U S A 2002;99:7622–7. 13. Reinhardt RL, Hong S, Kang SJ, Wang ZE, Locksley RM. Visualization of IL- 33. Alexopoulou L, Holt AC, Medzhitov R, Flavell RA. Recognition of double- 12/23p40 in vivo reveals immunostimulatory dendritic cell migrants that stranded RNA and activation of NF-kappaB by Toll-like receptor 3. Nature promote Th1 differentiation. J Immunol 2006;177:1618–27. 2001;413:732–8. 14. Luckashenak N, Schroeder S, Endt K, Schmidt D, Mahnke K, Bachmann MF, 34. Toubai T, Sun Y, Luker G, Liu J, Luker KE, Tawara I, et al. Host-derived et al. Constitutive crosspresentation of tissue antigens by dendritic cells CD8þ dendritic cells are required for induction of optimal graft-versus- controls CD8þ T cell tolerance in vivo. Immunity 2008;28:521–32. tumor responses after experimental allogeneic bone marrow transplanta- 15. Kerksiek KM, Niedergang F, Chavrier P, Busch DH, Brocker T. Selective Rac1 tion. Blood 2013;121:4231–41. inhibition in dendritic cells diminishes apoptotic cell uptake and cross- 35. Domcke S, Sinha R, Levine DA, Sander C, Schultz N. Evaluating cell lines as presentation in vivo. Blood 2005;105:742–9. tumour models by comparison of genomic profiles. Nat Commun 16. Seillet C, Jackson JT, Markey KA, Brady HJ, Hill GR, Macdonald KP, et al. 2013;4:2126. CD8alphaþ DCs can be induced in the absence of transcription factors Id2, 36. Koyama M, Hill GR. Alloantigen presentation and graft-versus-host dis- Nfil3, and Batf3. Blood 2013;121:1574–83. ease: fuel for the fire. Blood 2016;127:2963–70. 17. Hildner K, Edelson BT, Purtha WE, Diamond M, Matsushita H, Kohyama 37. Li H, Demetris AJ, McNiff J, Matte-Martone C, Tan HS, Rothstein DM, et al. M, et al. Batf3 deficiency reveals a critical role for CD8alphaþ dendritic cells Profound depletion of host conventional dendritic cells, plasmacytoid in cytotoxic T cell immunity. Science 2008;322:1097–100. dendritic cells, and B cells does not prevent graft-versus-host disease 18. Markey KA, Burman AC, Banovic T, Kuns RD, Raffelt NC, Rowe V, et al. induction. J Immunol 2012;188:3804–11. Soluble lymphotoxin is an important effector molecule in GVHD and GVL. 38. Yang YG, Dey BR, Sergio JJ, Pearson DA, Sykes M. Donor-derived interferon Blood 2010;115:122–32. gamma is required for inhibition of acute graft-versus-host disease by 19. Bruedigam C, Bagger FO, Heidel FH, Paine Kuhn C, Guignes S, Song A, et al. interleukin 12. J Clin Invest 1998;102:2126–35. Telomerase inhibition effectively targets mouse and human AML stem 39. Sykes M, Pearson DA, Taylor PA, Szot GL, Goldman SJ, Blazar BR. Dose and cells and delays relapse following chemotherapy. Cell Stem Cell 2014; timing of interleukin (IL)-12 and timing and type of total-body irradiation: 15:775–90. effects on graft-vs.-host disease inhibition and toxicity of exogenous IL-12 20. Cooke KR, Kobzik L, Martin TR, Brewer J, Delmonte J Jr, Crawford JM, et al. in murine bone marrow transplant recipients. Biol Blood Marrow Trans- An experimental model of idiopathic pneumonia syndrome after bone plant 1999;5:277–84. marrow transplantation: I. The roles of minor H antigens and endotoxin. 40. Dey BR, Yang YG, Szot GL, Pearson DA, Sykes M. Interleukin-12 inhibits Blood 1996;88:3230–9. graft-versus-host disease through an Fas-mediated mechanism associated 21. Teshima T, Reddy P, Lowler KP, KuKuruga MA, Liu C, Cooke KR, with alterations in donor T-cell activation and expansion. Blood 1998;91: et al. Flt3 ligand therapy for recipients of allogeneic bone marrow 3315–22.

www.aacrjournals.org Clin Cancer Res; 24(7) April 1, 2018 1615

Markey et al.

41. Morse MA, Nair S, Fernandez-Casal M, Deng Y, St Peter M, Williams R, et al. transplantation accelerates GVHD lethality. Biol Blood Marrow Trans- Preoperative mobilization of circulating dendritic cells by Flt3 ligand plant 2001;7:197–207. administration to patients with metastatic colon cancer. J Clin Oncol 44. Zheng R, Levis M, Piloto O, Brown P, Baldwin BR, Gorin NC, et al. FLT3 2000;18:3883–93. ligand causes autocrine signaling in acute myeloid leukemia cells. Blood 42.AnandasabapathyN,BretonG,HurleyA,CaskeyM,TrumpfhellerC, 2004;103:267–74. Sarma P, et al. Efﬁcacy and safety of CDX-301, recombinant human Flt3L, 45. Papaemmanuil E, Gerstung M, Bullinger L, Gaidzik VI, Paschka P, Roberts at expanding dendritic cells and hematopoietic stem cells in healthy ND, et al. Genomic classiﬁcation and prognosis in acute myeloid leukemia. human volunteers. Bone Marrow Transplant 2015;50:924–30. N Engl J Med 2016;374:2209–21. 43. Blazar BR, McKenna HJ, Panoskaltsis-Mortari A, Taylor PA. Flt3 ligand 46. Di Stasi A, Tey SK, Dotti G, Fujita Y, Kennedy-Nasser A, Martinez C, et al. (FL) treatment of murine donors does not modify graft-versus-host Inducible apoptosis as a safety switch for adoptive cell therapy. N Engl J disease (GVHD) but FL treatment of recipients post-bone marrow Med 2011;365:1673–83.

1616 Clin Cancer Res; 24(7) April 1, 2018 Clinical Cancer Research

Flt-3L Expansion of Recipient CD8α+ Dendritic Cells Deletes Alloreactive Donor T Cells and Represents an Alternative to Posttransplant Cyclophosphamide for the Prevention of GVHD

Kate A. Markey, Rachel D. Kuns, Daniel J. Browne, et al.

Clin Cancer Res 2018;24:1604-1616. Published OnlineFirst January 24, 2018.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-17-2148

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2018/01/20/1078-0432.CCR-17-2148.DC1

Cited articles This article cites 46 articles, 29 of which you can access for free at: http://clincancerres.aacrjournals.org/content/24/7/1604.full#ref-list-1

Citing articles This article has been cited by 1 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/24/7/1604.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/24/7/1604. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.