Peptide-Loaded Langerhans Cells, Despite Increased...

Cancer Therapy: Clinical

Peptide-Loaded Langerhans Cells, Despite Increased IL15 Secretionand T-Cell Activation In Vitro, Elicit Antitumor T-Cell ResponsesComparable to Peptide-Loaded Monocyte-Derived Dendritic Cells In Vivo

Emanuela Romano1,7, Marco Rossi1,7, Gudrun Ratzinger1,7, Maria-Angeles de Cos1,7, David J. Chung1,7,8,Katherine S. Panageas5,7, Jedd D. Wolchock3,4,6,7,8, Alan N. Houghton3,6,7,8, Paul B. Chapman6,7,8,Glenn Heller5,7, Jianda Yuan1,4,7, and James W. Young1,2,6,7,8

AbstractPurpose:We compared the efficacy of human Langerhans cells (LC) as tumor immunogens in vivo with

monocyte-derived dendritic cells (moDC) and investigated how interleukin 15 (IL15) supports optimal

DC-stimulated antitumor immunity.

Experimental Design: American Joint Committee on Cancer stage III/IV melanoma patients partici-

pated in this first clinical trial comparing melanoma peptide-pulsed LC with moDC vaccines

(NCT00700167, www.ClinicalTrials.gov). Correlative studies evaluated mechanisms mediating IL15

support of DC-stimulated antitumor immunity.

Results: Both DC vaccines were safe and immunogenic for melanoma antigens. LC-based vaccines

stimulated significantly greater tyrosinase–HLA-A*0201 tetramer reactivity than themoDC-based vaccines.

The two DC subtypes were otherwise statistically comparable, in contrast to extensive prior data in vitro

showing LC superiority. LCs synthesize much more IL15 than moDCs and stimulate significantly more

antigen-specific lymphocytes with a cytolytic IFN-g profile even without exogenous IL15. When supple-

mented by low-dose IL15, instead of IL2, moDCs stimulate 5 to 6 logsmore tumor antigen–specific effector

memory T cells (TEMRA) over 3 to 4 weeks in vitro. IL2 and IL15 can be synergistic in moDC stimulation of

cytolytic T cells. IL15 promotes T-cell expression of the antiapoptotic bcl-2 and inhibits candidate

regulatory T-cell (Treg) expansion after DC stimulation, countering two effects of IL2 that do not foster

tumor immunity.

Conclusions: MoDC-based vaccines will require exogenous IL15 to achieve clinical efficacy. Alterna-

tively, LCs can couple the endogenous production of IL15 with potent T-cell stimulatory activity.

Optimization of full-length tumor antigen expression for processing into multiple immunogenic peptides

for presentation by both class I and II MHC therefore merits emphasis to support more effective antitumor

immunity stimulated by LCs. Clin Cancer Res; 17(7); 1984–97. �2011 AACR.

Introduction

Antigen-specific expansion of effector and memoryCD8þ T cells is a central goal of immunotherapy againsttumors. For stimulation of MHC-restricted, antigen-speci-

fic, CD8þ, CTL in vitro (1), human Langerhans cells (LC)derived fromCD34þ hematopoietic progenitor cells (HPC)have shown superiority over other known conventional ormyeloid human dendritic cell (DC) subtypes, for example,monocyte-derived DCs (moDC) and dermal-interstitial

Authors' Affiliations: 1Laboratory ofCellular Immunobiology, ImmunologyProgram, Sloan-Kettering Institute for Cancer Research; 2Adult BoneMarrow Transplantation Service, Division of Hematologic Oncology,Department of Medicine, Memorial Hospital; 3Swim Across AmericaLaboratory, Immunology Program, Sloan-Kettering Institute for CancerResearch; 4Immune Monitoring Facility, Ludwig Center for Cancer Immu-notherapy; 5Biostatistics Service, Department of Biostatistics and Epide-miology; 6Melanoma and Sarcoma Service, Division of Solid TumorOncology, Department of Medicine, Memorial Hospital; 7MemorialSloan-Kettering Cancer Center; 8Weill Cornell Medical College, New York,New York

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

Current address for M. Rossi: Department of Clinical and Experi-mental Medicine, Oncology and Onco-Hematology Unit, Magna

Graecia University and T. Campanella Cancer Center, Catanzaro,Italy.

Current address for G. Ratzinger: Department of Dermatology and Vener-eology, Innsbruck Medical University, Anichstrasse 35, AT-6020Innsbruck, Austria.

Current address for M-A. de Cos: Clinical Pharmacology Service,University Hospital "Marques de Valdecilla," ES-39008, Santander,Spain

Corresponding Author: James W. Young, Memorial Sloan-KetteringCancer Center, 1275 York Avenue, New York, NY 10065. Phone: 646-888-2064; Fax: 646-422-0298; E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-10-3421

�2011 American Association for Cancer Research.

ClinicalCancer

Research

Clin Cancer Res; 17(7) April 1, 20111984

on June 21, 2018. © 2011 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst February 25, 2011; DOI: 10.1158/1078-0432.CCR-10-3421





http://clincancerres.aacrjournals.org/



DCs. This has held true for LCs either presenting peptide forrecall responses against viral antigens or cross-presentingdying tumor cells to elicit tumor antigen–specific CTLs (1).Detailed comparisons between resident DC populationsisolated from human skin have shown similar potency ofLCs (2). Clinical trial data have also suggested greaterefficacy of DC vaccines that contain LCs (3); but therehave been no direct comparisons in vivo in humansbetween defined DC subtypes as vaccines. We thereforeundertook a phase I clinical trial to determine safety andtoxicity of melanoma peptide-pulsed LCs compared withthe more commonly used moDCs. Laboratory studies thencompared their immunologic efficacy based on T-cell tetra-mer reactivity and enzyme-linked immunosorbent spot(ELISpot) assay of IFN-g secretion.LCs achieve robust stimulation of CTLs without produ-

cing any IL-12p70 (1), yet they secrete more interleukin 15(IL15) than any other conventional DC subtype (1, 4). IL15is therefore of particular interest because of its role inlymphocyte homeostasis; the expansion of memory T cells,especially CD8þ CTLs; and its autocrine protection of DCsfrom apoptosis (2, 5–11). IL15 also has a contrasting rolewith IL2 in that it counters tolerance and supports auto-immunity (5), rendering IL15 an attractive cytokine tosupport targeted immune responses against self-differen-tiation antigens expressed by tumors like melanoma. MostIL15 data derive from mouse rather than human studies,although a recent non-human primate study providesimportant insight into the safety and cytotoxicity of theIL15 administered in vivo (12).Monocyte precursors of moDCs have proven malleable

in their differentiation, depending on cytokine exposure

(13). Provision of IL15 during development leads tomoDCs with Langerhans-like DC properties (14, 15).The moDCs and LCs used in the vaccine trial, however,were generated according to standard protocols withoutIL15 (1); and we wished to investigate the role of thiscytokine on T-cell responses in the context of restimulationby the same DC subtype to which the T cells had beenexposed in vivo during vaccination. Our findings establishan essential role for IL15 in the generation of cytolytic, IFN-g–secreting T cells, even when stimulated by potent DCs.MoDCs depend on an exogenous source of IL15, yet LCsremain effective even with limiting or no exogenous IL15.These data have important implications for the design ofDC-based immunotherapy trials, which going forward,must also include optimized approaches to provide full-length antigens for processing into multiple immunogenicpeptides for presentation on both class I and II MHC.

Materials and Methods

Human cells, media, and cytokinesHuman cell collection and use adhered to protocols

approved by the Institutional Review and Privacy Boardof Memorial Hospital, Memorial Sloan-Kettering CancerCenter (MSKCC). Healthy volunteers or patients providedperipheral blood mononuclear cells (PBMC) or granulo-cyte colony stimulating factor–elicited CD34þHPCs for theisolation of T cells and the generation of moDCs and LCsexactly as published [(1), for correct FLT-3-ligand dose, seeref. 16]. CD34þ HPCs have greater expansion and differ-entiation potential than monocytes, so the progeny con-taining LCs were more heterogeneous, primarily includingimmature eosinophils (17, 18). Viable LCs and moDCswith large forward scatter by flow cytometry were dosedaccording to phenotypic expression of HLA-DRbrightCD86brightCD83þCD14� epitopes. All cells wereused either fresh or thawed after cryopreservation withoutcompromising phenotype or activity (19). Viable recoverywas consistently 85% or more and usually more than 90%.

Phenotypic analyses by flow cytometryFluorescein isothiocyanate (FITC), phycoerythrin (PE),

PE-cyanine-5 (PE-Cy5), PE-Cy7, peridinin chlorophyll pro-tein-cynine-5.5 (PerCP-Cy5.5), allophycocyanin (APC),APC-Alexa Fluor750, PE-Texas Red (ECD)-conjugatedmouse anti-human monoclonal antibodies (mAb)included anti-CD3, anti-CD8, anti-CD25, anti-CD27,anti-CD28, and anti-CD69 (BD Biosciences); anti-CD4,anti-CD45RA, and anti-CD45RO (Beckman Coulter);anti-CD127, anti-human Foxp3, anti-CD8, and anti-CD62L (eBioscience); and anti-CCR7 (R&D Systems). Iso-type controls included the appropriate fluorochrome-con-jugated mouse IgG1 or rat IgG2a (Dako; eBioscience).Foxp3 detection required intracellular staining, using cellfixation and permeabilization buffers provided with theFoxp3 kit (eBioscience).

HLA-A*0201–restricted tyrosinase, gp100, and flumatrix peptide (fluMP)-streptavidin-PE–labeled tetramers

Translational Relevance

Dendritic cell (DC) vaccination protocols most oftenuse monocyte-derived DCs (moDC), although Langer-hans-type DCs (LCs) have repeatedly proven superior tomoDCs and other conventional DCs for stimulatingimmunity against viral and tumor antigens in vitro. Wereport a phase I comparison trial between melanomapeptide-pulsed LCs and moDCs, where LCs provedmore potent than moDCs for one parameter but wereotherwise comparable in eliciting immunity againstmelanoma antigens. We nevertheless established a cri-tical role for IL15 in countering IL2-induced apoptosisand phenotypic Treg expansion during CTL generationby DC stimulation. MoDCs require exogenous IL15,whereas LCs do not. LCs providing endogenous IL15therefore merit optimization as cancer immunogens invivo, whereas continued use of moDCs will requireexogenous IL15. Circumventing the inherent limita-tions of single peptides will require improved methodsfor loading full-length antigen for processing and pre-sentation of multiple class I and II MHC-restrictedepitopes.

Peptide-Loaded LC versus moDC Anti-Melanoma Vaccines

www.aacrjournals.org Clin Cancer Res; 17(7) April 1, 2011 1985




and negative tetramer controls (Beckman Coulter) detectedantigen-specific T cells. Positive tetramer reactivity requireda distinct population at least 1 log above the mean fluor-escent intensity of the negative control.

Flow cytometry analyses used a Cytomics FC500 (Beck-man Coulter) or a Cyan-ADP flow cytometer (DAKO).Gates were set for collection and analysis of 200,000or more live events based on propidium iodide (PI)exclusion.

Immune responses to tumor antigenic peptide-pulseddendritic cells in vivo

Clinical trial design. We conducted a phase I clinicaltrial to test safety and toxicity and to compare immuneresponses stimulated by tumor peptide-pulsed moDCsversus LCs in HLA-A*0201þ patients with AmericanJoint Committee on Cancer (AJCC) stage III or IVmelanoma (registered as no. NCT00700167 at www.ClinicalTrials.gov; Supplementary Fig. 1). There wereno significant differences between assignment to eithertype DC vaccine with regard to demographics, Kar-nofsky performance score, or disease stage (Table 1).Phase Ia accrued cohorts of 3 patients to each of 3vaccine doses (3 � 106, 10 � 106, or 30 � 106), usingeither peptide-loaded (see below). moDCs or LCs for apriming dose, followed by 2 boosters of the samepeptide-loaded DC subtype at approximately 4-weekintervals. Patients received a total of 3 vaccines withno crossover between moDCs or LCs. MoDCs or LCswere freshly made and used for the initial vaccine.Subsequent vaccines used thawed cells from the cryo-preserved initial product (19). Table 2 lists releasecriteria required for each vaccine, and Table 3 reportsthe viability and phenotype of all administered DCvaccines. Purity of the CD34þ-derived LC progenywas lower than that of the moDCs, reflecting the more

variable differentiation potential of the starting popula-tions of CD34þ HPCs compared with blood monocytes(1). The major contaminants of the LCs were immaturemyeloid cells, mostly eosinophils (1). Vaccines weredosed according to the absolute number ofCD83þCD86brightHLA-DRbrightCD14� moDCs or LCsby flow cytometry, so the DC numbers were equivalent.

Each moDC or LC vaccine, regardless of total dose, wasadministered as ten 0.1 mL deep intradermal injectionsdivided equally between 2 sites in the proximal arms orthighs, excluding areas adjacent to prior lymph node resec-tions. Surrounding erythema and induration were mea-sured for delayed type hypersensitivity (DTH) reactions 48hours after the second and third vaccines.

The dose that yielded a positive response by tetramerreactivity or ELISpot (see below), defined as greater than2 SD above the prevaccine baseline response in at least 2 ofthe 3 initial phase Ia patients at each dose level, determinedthe optimal biologic dose. Hence, the 10 � 106 dose level,which proved most feasible in terms of starting cell num-bers, cytokine usage, costs, and labor, was also the optimalbiologic dose for eachDC subtype. Phase Ib of the trial thenaccrued 9 additional patients to each of the moDC or LCarm, resulting in a total of 12 patients who received eithermoDC or LC vaccines at the 10� 106 dose level in phases Iaand Ib combined.

Antigen-loading of moDCs or LCs. The clinical trial used2 synthetic, HLA-A*0201–restricted, heteroclitic mela-noma peptides: tyrosinase (TYR368–376) YMDGTMSQVand gp100 (gp100209–217) IMDQVPFSV (Research Genet-ics, Invitrogen). Synthetic influenza matrix peptide(fluMP58–66 GILGFVFTL; Research Genetics, Invitrogen)served as a positive control for HLA-A*0201–restrictedresponses. Keyhole limpet hemocyanin (KLH; high purity,endotoxin-free; Calbiochem) served as a positive controlneoantigen for class II MHC-restricted responses.

Table 1. Patient demographics and disease characteristics at protocol entry

All patients, all dose levels 10 � 106 dose level only

Variable LC (n ¼ 18) moDC (n ¼ 18) LC (n ¼ 12) moDC (n ¼ 12)

Gender, n (%)Male 10 (56) 11 (61) 8 (66) 7 (58)Female 8 (44) 7 (39) 4 (34) 5 (42)

Age, yMean 61.4 57.2 63.2 57.4Range 43–81 31–83 46–81 39–77

Karnofsky performance scoreMedian 100 100 100 100Range 90–100 90–100 90–100 90–100

Disease stage (AJCC), n (%)III, NED 15 (83) 16 (89) 9 (75) 10 (83)IV, NED 2 (11) 1 (6) 2 (17) 1 (8)IV, slow progressive disease 1 (6) 1 (6) 1 (8) 1 (8)

Romano et al.

Clin Cancer Res; 17(7) April 1, 2011 Clinical Cancer Research1986




Terminally maturing moDCs or LCs for the clinicaltrial were cultured at 1 � 106 cells/3 mL in 6-well plates,to which peptides (1 mmol/L) or KLH (10 mg/mL) wereadded overnight at 37�C, but in separate cultures toavoid MHC binding competition. Cells were washedand evaluated for release criteria (Table 2) before com-bining and resuspending to the appropriate concentra-tion for vaccination.Response assessments in vitro and correlative laboratory

studies. Response assessments for the clinical trial usedPBMC collected at prevaccine baseline and then approxi-mately 3 weeks after each vaccine. We compared CD8þ T-cell reactivity with HLA-A*0201–peptide tetramers overtime and ELISpot assay of IFN-g secretion by total T cellsover time as fold increase over baseline. These followed asingle 6- to 7-day restimulation of the PBMC responders invitro, using the same peptide-loaded autologous moDCs orLCs used for a patient’s vaccines, but without exogenous

cytokines. Ten to 12 patients, out of total 12 in each DCsubtype vaccine group at the optimal biologic dose of 10�106 cells/vaccine, were evaluable. Zero to 2 patients in eachgroup became inevaluable due to inadequate cell yields orinsufficient replicates (Table 4). For assessment of KLH-specific responses, PBMCs from moDC and LC vaccineeswere harvested at baseline prevaccine and approximately3 weeks after each of 3 vaccines and cultured with KLH(1 mg/mL). KLH-specific immunity was determined by theincorporation of 3HTdR (1 mCi/well) by proliferatingresponder T cells in the last 8 hours of a 5-day culture.

All patients consented to the use of any residual cryo-preserved cells for subsequent thaw and assessment inadditional correlative laboratory studies. These usedresponder PBMCs, total T cells, or T-cell subsets selectedby using immunomagnetic columns (Miltenyi Biotec) asindicated. Responder PBMCs or T cells were resuspended in10% autologous serum (v/v) in complete RPMI-1640 and

Table 2. Criteria for release of dendritic cell vaccines for patient administration

Variable Test Result required for release

Bacteria and fungus Culture in thioglycolate brothand soybean casein digest medium

No growth after 5 days of in-processculture; no growth confirmation offinal product after administration

Gram stain Negative on final productEndotoxin Limulus amebocyte lysate

(BioWhittaker; CBER/FDAbiologic license number 709)

<5 endotoxin units

Mycoplasma PCR Negative result, in process 48 hoursbefore end of culture

Viability PI staining of large FSCcells on flow cytometry

<30% PIþ (or �70% viable)

Phenotype (flow cytometry) Flow cytometry: gated populationof large FSC, CD14�, class IIMHC bright cells

�50% CD83þ, � 50% CD86þ

Abbreviation: FSC, forward scatter.

Table 3. Viability and phenotype of dendritic cell vaccines administered to patients

% viability oftotal

(mean � SEM)

% viable HLA-DRbrightCD14�

of total(mean � SEM)

%CD83 of viableHLA-DRbrightCD14�

(mean � SEM)

%CD86 ofviable HLA-

DRbrightCD14�

(mean � SEM)

Blood monocyte-derived DCs 89.6 � 1.7 81.3 � 2.1 89.6 � 2.1 98.0 � 0.5CD34þ-derived LCs 93.1 � 0.7 56.8 � 1.9a 65.6 � 2.1a 98.2 � 0.4

aPurity of the CD34þ-derived LC progenywas lower than that of themoDCs, reflecting themore variable differentiation potential of thestarting populations of CD34þHPCs comparedwith bloodmonocytes (1). Themajor contaminants of the LCswere immaturemyeloidcells, mostly eosinophils (1). Regardless, vaccines were dosed according to the absolute number of CD83þCD86þ HLA-DRbrightCD14� moDCs or LCs.






plated at 2� 106 cells/mL final volume per well of a 24-wellplate (Costar) or at 1 � 105/100 mL final volume per wellof a 96-well round bottomed plate (Costar). Matureautologous moDCs or LCs were resuspended at 106/mLin cytokine and serum-free RPMI-1640, then loaded with10 mmol/L of a single 9-mer peptide for 1h at RT before

restimulation. Target cells for ELISpot or CTL assays weresimilarly loaded with single peptides. Peptides werealways separately loaded onto the respective DCs to avoidMHC binding competition. Peptide-loaded moDCs or LCswere added in graded doses to fixed numbers of respon-ders. Where experiments required repeated stimulation,

Table 4. Patient accrual to and completion of vaccine trial, including numbers evaluable for responseassessments

Totalaccrued

(n)

Off study(n)

Reason for withdrawalfrom study

Completed all 3vaccines (n)

Evaluable for response assessments(adequate cell yields)

Tetramer ELISpot

gp100 tyr fluMP gp100 tyr fluMP

Phase IaMoDCs

3 � 106 moDCs/vaccine

3 0 3 3 3 3 3 3 3


3 0 3 3 3 3 3 3 3


4 1 Myocardial infarctionunrelated to study

3 3 3 3 3 3 3

LCs3 � 106 LCs/vaccine

4 0 4a 3 3 3 3 3 3

10 � 106 LCs/vaccine

3 0 3 3 3 3 3 3 3

30 � 106 LCs/vaccine

3 0 3 3 3 3 3 3 3

Phase Ibb

MoDCs10 � 106 moDCs/vaccine

10 1 Noncompliance withscheduled responseassessments

9 9 9 7 7 7 7

LCs10 � 106 LCs/vaccine

13 4 Patient concern aboutvitiligo (grade I)possibly related to firstvaccine (n ¼ 1); diseaseprogression withalternative treatment(n ¼ 2); inadequate cellyield for secondvaccine (n ¼ 1).

9 8 8 8 7 8 8

Phase Ia þ Ib, allat 10 � 106 cells/vaccine

MoDCs 12 12 12 10 10 10 10LCs 12 11 11 11 10 11 11

aThree evaluable patients were required at each dose level in phase Ia to select the optimal biologic dose (defined in Materials andMethods) to use in phase Ib. Hence, 1 additional patient was accrued to replace 1 who had inadequate cell yields for responseassessments despite a normal complete blood count.bResponse assessments for the trial overall were based on the evaluable patients at the 10 � 106 moDC or LC dose per vaccine inphase Ia, plus all evaluable patients in phase Ib.

Romano et al.





responder cells were harvested after 6- to 7-day initialstimulation and restimulated by the same DC subtypeand exogenous cytokine combination each round.The cultures described above were supplemented or not

with either recombinant human IL2 (10 IU/mL; Chiron)and/or IL15 (10 ng/mL; R&D Systems; see also Supple-mentary Fig. 2). Although IL2 10 IU/mL is less than whatDC-stimulated allogeneic T cells produce (20), we avoidedsupraphysiologic doses of IL2 in these autologous combi-nations that could activate natural killer (NK) cells amongbulk PBMCs or T cells via their IL15Rs, which sharecommon b- and g-chains with the IL2R (21).

Inhibition of IL15 binding to IL15R-a on dendriticcellsLCs or moDCs were opsonized for 2 hours with a

polyclonal goat anti-human IL15R-a (5ug/mL; R&D Sys-tems) or its isotype control (5ug/mL; R&D Systems). Thesewere then separately cocultured with responder PBMCs infunctional assays at the indicated ratios. Anti-humanIL15R-a or its isotype control was replenished at the sameconcentration in culture every 48 hours. When cells under-went a second round of stimulation after 1 week, respon-ders were harvested, washed, and restimulated for 7 days byanti-human IL15R-a–opsonized gp100-loaded moDCs orLCs from the same population.

IFN-g ELISpot assayAfter antigen-specific stimulation, responder lymphocytes

were tested for IFN-g production by ELISpot assay (1-DIKELISpot for human IFN-g , Mabtech, DiaPharma Group;Vector ABC kit, Vector Laboratories; Automated ELISpotReader System and KS 4.3 software, Carl Zeiss Vision)according to the manufacturer’s instructions. Targets forlymphocyte restimulation during overnight rechallenge intheELISpot assayswere eithermaturepeptide-pulsedmoDCsor LCs, whichever had been used for stimulations in vitro,plated in triplicates at 30:1 effector:target (E:T) ratio. Controlwells contained effectors and unloaded targets. ELISpotreactivity against both moDC and LC targets without anti-genic peptide was subtracted from reactivity against thepeptide-loaded moDC and LC targets, because unloadedmoDCs caused such high background.

Cytolytic T-cell assaysCytolytic activity exerted by T lymphocytes responding to

antigen-specific DC stimulation was assessed in standard51Cr release assays against the same target cells used inELISpot assays. A total of 5�103 51Cr-labeled target cellswere added at the indicated T-cell E:T ratios. Supernatantswere collected after 4 to 6 hours from replicate microwellsfor calculating specific 51Cr release.

Assessment of early and late CD3þCD8þ T-cellapoptosisEarly and late apoptosis were evaluated by a FITC-

conjugated Annexin V probe (BD Biosciences), togetherwith PI (Sigma-Aldrich). Fluorochrome-conjugated mAbs

costained the CD3þCD8þ T-cell subset of interest. Earlyapoptotic cells were Annexin VþPI�, whereas late apoptoticand necrotic cells were Annexin VþPIþ.

Protein extraction and Western blottingPositive immunomagnetic selection (Miltenyi Biotec)

yielded CD3þCD8þ T cells with 97% to 99% purity.Protein extraction from 2 � 106 to 5 � 106 cells used RIPAlysis buffer (Pierce) with protease inhibitors (RocheApplied Science) for 20minutes on ice. Twentymicrogramsof cell lysate were applied to a 12%Bis-Tris gel (Invitrogen).Proteins were blotted onto a polyvinylidene difluoridemembrane (Biorad), blocked with 5% nonfat dry milkin PBS/0.1% Tween20, and then probed with mouseanti-human Bcl-2 mAb (BD Biosciences). Mouse anti-human glyceraldehyde 3 phosphate dehydrogenase(GAPDH; Ambion) served as the loading control. Second-ary antibody was horseradish peroxidase–conjugated goatanti-mouse IgG (PerkinElmer Life Sciences). Immunoreac-tive protein bands were identified by an ECL detection kit(Amersham).

StatisticsResponses to the moDC or LC vaccines in the clinical

trial were tested by a stratified permutation (Wilcoxon)rank sum statistic. This compared moDCs with LCs wherethe outcomes were reactivity against gp100, tyrosinase, orcontrol fluMP. The test was stratified by vaccine numberover time.

For the correlative laboratory studies in vitro, replicatemeans from 3 or more independent experiments wereaveraged and SEM calculated as the measure of variability.Otherwise, SD was calculated for the replicate mean of 1 of3 representative experiments. A stratified t test (stratifiedby the E:T or responder:stimulator ratio) was used forthe functional assays. The t test was used for each compar-ison within a given time point or condition for all otheranalyses.

Online supplementary materialSupplementary Figure 1 shows the protocol schema.

Supplementary Figure 2 is a dose-finding pilot study forIL2 and IL15 by using fluMP-pulsedmoDC stimulators andfluMP–HLA-A*0201 tetramer reactive T-cell readouts. Sup-plementary Figure 3 illustrates the proliferation of totalcells and the absolute number of CD3þCD8þ T cellsreactive with tyrosinase, gp100, and fluMP–HLA-A*0201tetramers over 3 to 4 rounds of weekly stimulation withand without IL2, IL15, or both in combination.

Results

Peptide-loaded moDCs and LCs are safe in patientswith advanced melanoma and generate measurabletumor antigen-specific immune responses

Two groups of 12 patients each received 3 vaccines ofeither 10� 106 peptide-pulsed moDCs or CD34þ-derivedLCs, which was the optimal biologic dose as explained in






Materials and Methods. No patient experienced morethan grade 2 toxicity according to Common TerminologyCriteria for Adverse Events, version 3.0, possibly or prob-ably related to these vaccines. All patients developedsome degree of erythema (Fig. 1A), induration(Fig. 1B), and mild pruritus at the injection sites. Twopatients experienced grade I vitiligo, possibly related tothe vaccine. Grade 2 toxicities were because of combina-tions of grade 1 cutaneous toxicities. Prevaccine baselineand postvaccine responses were assessed simultaneouslyin vitro to avoid interassay variability. Proliferativeresponses to rechallenge with KLH (Fig. 1C) were greaterin the moDC than LC vaccinees (post 2, P¼ 0.002; post 3,P < 0.001; for comparisons between moDC vs LC vac-cines). Figure 2A depicts the absolute tetramer reactivityover time, including the prevaccine baseline. Figure 2Bshows IFN-g secretion measured by ELISpot as the foldincrease over 2 SD above the prevaccine baseline mean.Background reactivity against targets without peptide wassubtracted from each ELISpot condition, and resultingnegative or zero values were normalized to 1 for calcula-tion of fold increase.

Data comparing moDCs with LCs in vitro (1) led to thehypothesis that LCs would be superior immunogens. OnlyCD8þ T-cell reactivity with tyrosinase peptide–HLA-A*0201 tetramers reached statistical significance, however,slightly favoring LCs over moDCs (P ¼ 0.04). There was anonsignificant trend supporting moDCs for stimulatingCD8þ T-cell reactivity with gp100 peptide–HLA-A*0201tetramers (P ¼ 0.11), but LCs and moDCs were equivalentin stimulating fluMP tetramer reactivity (P¼ 0.93). Limitedcell numbers precluded isolation of CD8þ T cells for ELI-Spot assays. Total T-cell production of IFN-g showed atrend without reaching statistical significance in favor ofresponses stimulated by tyrosinase peptide-pulsed moDCs(P ¼ 0.08) and by fluMP-loaded LCs (P ¼ 0.14), whereasthe 2 subtypes showed no difference in stimulating gp100-specific responses (P ¼ 0.30).

LCs maintain superior potency over moDCs ininducing IFN-g–secreting lymphocytes at limitingdoses of exogenous IL15

Pertinent to the generation of antigen-specific CTLs wasour previous finding that LCs secreted significantly more

0

10

20

30

40

50

0

5

10

15

20

25

Ery

them

a

Ind

ura

tio

n

A B

post 2 post 3 post 2 post 3post 2 post 3 post 2 post 3moDC vaccine moDC vaccine LC vaccineLC vaccine

C

post 2

post 3

3 H-T

dR

inco

rpo

rati

on

(cp

m lo

g2)

moDC vaccine LC vaccine

20

2 5

210

215

220

Figure 1. Peptide-pulsed moDCand LC vaccines stimulate DTHreactions and KLH-specificresponses after the second andthird vaccines. Erythema (A) andinduration (B) were measured atthe greatest diameter around eachof 10 injection sites to calculatethe mean for each parameter foreach patient after vaccine 2 andagain after vaccine 3. Shown arethe averages � SD of theindividual patient means (n ¼ 12patients) for each subtype DCvaccine. P ¼ NS comparingmoDCs with LCs. C, PBMCs frommoDC and LC vaccinees wereharvested at baseline prevaccineand approximately 3 weeks aftereach of 3 vaccines and culturedwith KLH (1 mg/mL). KLH-specificimmunity was based on theincorporation of 3H-TdR byproliferating responder T cells (y-axis, log2 scale) in the last 8 hoursof a 5-day culture. Post 2ndvaccine (post 2), P ¼ 0.002,and post 3rd vaccine (post 3),P < 0.001, for moDC vs LCvaccines.

Romano et al.





IL15 than moDCs (1, 4). By titrating doses of exogenousIL15 down to zero, we evaluated the biologic dependenceof LCs compared with moDC on exogenous IL15. T cellsremaining after completion of the clinical trial responseassessments underwent 3 weekly rounds of stimulation invitro by the same gp100 peptide- or fluMP-pulsed DCsubtype used for primary immunization in vivo. LCsstimulated significantly greater reactivity than themoDCs, especially at limiting doses of exogenous IL15

(Fig. 3A and B). Even in the absence of any exogenousIL15, LCs stimulated significantly greater T-cell responsesthan did comparable numbers of moDCs.

We then assessed the effect of blocking IL15R-a(Fig. 3C). After 2 weekly rounds of restimulation in vitrowithout addition of exogenous cytokine, ELISpot assays ofIFN-g secretion by responder PBMCs showed that LCs wereagain more potent APCs than moDCs. Blocking anti–IL15R-a significantly reduced the stimulatory capacity of

0

0.5

1.0

1.5

2.0

0

1

2

3

0

5

10

15

20

25

30

0

10

20

30

40

0

10

20

30

40

0

2

4

6

8

10

HL

A-A

*020

1 te

tram

er-r

eact

ive

CD

3+C

D8+

T c

ells

Fo

ld in

crea

se in

IF

Nγ–

secr

etin

g T

cel

ls

tyrosinase tyrosinase

gp100gp100

fluMPfluMP

pre post 1 post 2 post 3

LC vaccine

pre post 1 post 2 post 3

moDC vaccine

post 1 post 2 post 3post 1 post 2 post 3

A B

LC vaccine moDC vaccine

Figure 2. Peptide-loaded moDCs and LCs generate immune responses in patients with advanced melanoma, but neither is clearly superior inthis setting. Responses against peptide-loaded moDCs or LCs were measured at baseline prevaccine and again approximately 3 weeks after each of 3vaccines given at nearly 4 week intervals. Responses were based on a single 6- to 7-day restimulation in vitro, using the same moDCs or LCs used tovaccinate a given patient, pulsed with the indicated peptides. No exogenous cytokines were added. Box plots show the medians and interquartile ranges(25th to 75th percentiles) with whiskers approximating � 2 SD or 95% of the data (Tukey method). A, absolute numbers of CD3þ CD8þ T cells reactive withtyrosinase–, gp100–, or fluMP–HLA-A*0201 tetramers are shown over time, where positive events exhibited at least 1 log higher fluorescent intensitythan the negative controls. B, the fold increases in IFN-g secretion over time by total T cells in ELISpot assays, relative to the prevaccine baseline averageþ 2SD, are depicted. Empty moDC targets without peptides were associated with high background IFN-g secretion in the ELISpot assay, so the backgroundvalues were subtracted from all conditions at all timepoints to calculate fold increases. LCs were compared with moDCs where the outcomes were reactivityagainst tyrosinase, gp100, or control fluMP. A and B, the test was stratified by vaccine number over time, and a permutation test generated P values.Only tyrosinase–HLA-A*0201 tetramer reactivity achieved significance in favor of LCs over moDCs (P ¼ 0.04). There was a trend in favor of moDCs in theELISpot assays for tyrosinase-specific responses (P ¼ 0.08). P ¼ NS for all other comparisons between LCs and moDCs.






each DC subtype (Fig. 3C). Near complete inhibition ofmoDC activity indicated that the blocking antibody waseffective. Although the inhibition of LC activity by anti–IL15R-a achieved statistical significance (P < 0.03), it wasless complete than with moDCs (P < 0.0001).

IL15 and IL2 synergize in supporting development ofantigen-specific CD3þCD8þ T cells with an IFN-g–secreting cytolytic profile and effector memoryphenotype

We used moDCs to distinguish the effects of IL15 fromIL2, the receptors for which share common b- andg-chains. MoDCs depend on exogenous IL15, unlikeLCs, so endogenous IL15 would not confound the results.Pilot studies established that IL2 at 10 IU/mL was onlyminimally additive to IL15 at 10 ng/mL in supportingproliferation of antigen-specific T cells (SupplementaryFigs. 2 and 3). Representative dot plots from the com-bined IL2 þ IL15 condition in a single set of experimentsillustrate the expansion of tetramer-reactive T cells in vitroagainst moDC-presented tyrosinase, gp100, and influ-enza matrix peptide antigens (Fig. 4A). These CD3þCD8þ

tetramer-reactive T cells attained a cytotoxic IFN-g–secreting profile (Fig. 4B). In contrast to cell prolifera-tion (Supplementary Fig. 3), IL2 and IL15 synergized tosupport CTL development (P < 0.05 for IL15 or IL2þ IL15

vs. IL2 or no cytokine; and P < 0.005 by the third round ofstimulation for the combination of IL2 þ IL15 vs. IL15alone).

Because limited cell numbers precluded isolation ofCD8þ T cells for most ELISpot assays, we confirmed cyto-lytic activity in standard 51Cr release assays (Fig. 4C). Bulkresponder T cells included CTLs capable of dose-depen-dent, MHC-restricted, tumor antigen–specific killing(Fig. 4C; P < 0.001). Although IL15 was more effectivethan IL2 alone in promoting robust expansion of tetramer-reactive CD3þCD8þ T cells, IFN-g secretion and cytolyticfunction benefited from IL15 and IL2 synergy when sti-mulated by peptide-loaded moDCs.

After three 7-day rounds of stimulation in vitro in thepresence of IL2 and IL15, we assessed activation and lym-phoid homing markers on T cells from the previouslyvaccinatedmelanoma patients (Fig. 4D and E). CD3þCD8þ

T-cell responders expressed neither of the lymphoid homingreceptors, CCR7 or L-selectin (CD62L); nor did these T cellsexpress CD28, all of which are usually found on centralmemory (TCM) or naive/precursor memory T cells (22).These tetramer-reactive T cells instead expressed the activa-tion markers CD45RO, CD69, and variable amounts ofCD27, as well as CD45RA (Fig. 4D). Data from 3 indepen-dent experiments gatedon tyrosinase-reactive tetramerswerepooled to illustrate the aggregate phenotype, as well as

10 2 0.1 0.01 0.00

10

20

30

40

50

60

70

80

*

** *

LCmoDC

fluMP

SF

C p

er 1

× 1

05 inp

ut

cells

10 2 0.1 0.01 0.00

10

20

30

40

50

60

70

80

*

*

* *

*

gp100LCmoDC

0

10

20

30

40

50

60

70

80

*

* *

LC moDCIL15 ng/mL IL15 ng/mL

anti-IL15R-αisotype control

A B C

Figure 3. LCs are more potent than moDCs in inducing IFN-g–secreting CTL at limiting doses of exogenous IL15. PBMCs from melanoma patientspreviously vaccinated with peptide-pulsed, mature, autologous DCs (LCs or moDCs) were restimulated in vitro by the same type of autologous DC pulsed with(A) fluMP or (B) gp100 peptide. Following 3 weekly restimulations, ELISpot assays measured IFN-g secretion after overnight exposure to the respectivepeptide-pulsed DC targets. Limited cell numbers precluded isolation of CD3þCD8þ T cells for the ELISpot assays. IL15 was supplemented at the indicatedconcentrations during the restimulations in vitro. Shown are the averaged triplicate means � SEM (n ¼ 3 independent experiments) for the number of IFN-gspot forming cells (SFC) per triplicate of 105 input cells. By Student's t test, P < 0.05 LCs versus moDCs at each IL15 concentration tested. Only LCswere still stimulatory in the absence of exogenous IL15. (See also Fig. 4C, which validates the concordance between IFN-g secretion and actual target lysis byCTLs.) C, PBMCs frommelanoma patients, previously vaccinated with either peptide-pulsed, mature, autologous LCs or moDCs, were restimulated in vitro bythe same type of autologous DCs pulsed with gp100 peptide over each of 2 weeks (stimulator:responder ¼ 1:30). No exogenous cytokines, including IL15,were added to any condition. The peptide-pulsed LCs and moDCs were opsonized at the outset and on each restimulation with anti–IL15R-a orisotype control. IFN-g secretion by PBMCs was analyzed by ELISpot after overnight exposure to the same peptide-pulsed LC or moDC targets. Shownare the averaged triplicate means � SEM (n¼ 3 independent experiments) for the number of IFN-g SFC per triplicate of 105 input cells. *, P < 0.03 for LCs and**, P < 0.0001 for moDCs (by Student's t test), with versus without blockade of IL15R-a.

Romano et al.





variability, characteristic of theseCD45RAþ effectormemoryT cells (TEMRA: CD62L�, CCR7�, CD28�, CD27variable,CD45ROþ, CD69þ, and CD45RAþ; Fig. 5D; refs. 22–24).Figure 4E shows the dot plots from the 1 experiment of 3with thehighest numberof tetramer-reactive cells among thegated CD3þCD8þ T cells.

IL15 prevents contraction of the human immuneresponse by countering apoptosis and expansion ofcandidate Tregs

We then asked whether IL15 rescued T cells from IL2activation–induced cell death or apoptosis. We againfocused on moDCs in lieu of LCs, because we could better

100 101 102 103

100

101

102

103

100 101 102 103

100

101

102

103

100 101 102 103

100

101

102

103

100 101 102 103

100

101

102

103

100 101 102 103

100

101

102

103

100 101 102 103

100

101

102

103

100 101 102 103

100

101

102

103

0.12

99.9100 101 102 103

100

101

102

103

0.5

99.5100 101 102 103

100

101

102

103

2.3

99.7

0.15

99.8

0.3

99.7

5

95

2.11

97.9

13.3

86.7

42.2

57.8

1st 2nd 3rd

CD8

flu

MP

gp

100

Tyro

sin

ase

A

Tyro

sin

ase

tetr

amer

100 101 102 103 104

CCR7100

101

102

103

104

99.1 0.91

100 101 102 103 104

CD45RA100

101

102

103

104

10.4 89.6

100 101 102 103 104

CD28100

101

102

103

104

99.7 0.35

100 101 102 103 104

CD27100

101

102

103

104

5.3 94.7

100 101 102 103 104

CD127100

101

102

103

104

98.3 1.72

100 101 102 103 104

CD45RO100

101

102

103

104

17.1 82.9

100 101 102 103 104

CD69100

101

102

103

104

2.86 97.1

100 101 102 103 104

CD62L100

101

102

103

104

98.7 1.33

40:1 20:1 10:1 5:10

10

20

30

40

50

% s

pec

ific

lysi

s

gp100 K562 A*0201tyr K562 A*0201

E: T ratio

0

20

40

60

80

100

120

140

160

180

200

* *

*

**

SF

C p

er

1X

105

in

pu

t cells

1st stimulation 3rd stimulation

no cytokineIL2IL15IL2 + IL15

B C

E

D

0

20

40

60

80

100

CD62L CCR7 CD45RO CD45RA CD69 CD27 CD28 CD127

% m

arke

r/C

D3+

CD

8+ t

etra

mer

+

Figure 4. IL15 and IL2 synergize in supporting development of moDC-stimulated, antigen-specific CD3þCD8þ T cells with an IFN-g–secreting,cytolytic profile and effector memory phenotype. A, CD3þCD8þ T cells were gated from PBMCs obtained frommelanoma patients previously vaccinated withtyrosinase, gp100, and fluMP-pulsed moDCs in vivo. CD3þCD8þ T cells (x-axis) that reacted with HLA-A*0201 tetramers bearing the respectivepeptides (y-axis) were measured at the outset of each restimulation by autologous peptide-pulsed moDCs in vitro. Dot plots show only the combined IL2(10 IU/mL) þ IL15 (10 ng/mL) condition. Numbers in the top right quadrants represent the percentages of tetramer-reactive cells in the CD3þCD8þ gate.Quadrants are the same for each round of stimulation in a single column but different across rows, because staining and analyses occurred at different timepoints after each round of stimulation. These data represent 1 of 3 independent experiments. B, ELISpot assays measuring IFN-g secretion by responderPBMCs tested as effectors against gp100 peptide, tyrosinase peptide, and fluMP-pulsed moDC targets were performed after the first and third stimulations.Shown are the results for the gp100 target antigen, but the other 2 antigens stimulated similar response patterns (not shown). The averaged triplicatemeans� SEM of IFN-g spot forming cells (SFC) per 105 input T cells combined from 3 independent experiments are shown. **, P < 0.005 for the combination ofIL2 þ IL15 relative to all other conditions after the third round of stimulation. *, P < 0.05 relative to the IL2 or no added exogenous cytokine conditions forIL15 either alone at both time points or in combination with IL2 after the first stimulation. C, gp100-specific CD8þ T cells that developed in the presenceof IL2 (10 IU/mL) and IL15 (10 ng/mL) killed HLA-A*0201–transfected K562 targets (gift of Dr. Thomas Wolfel, University of Mainz, Germany; ref. 37) pulsedwith gp100 but not with tyrosinase peptide in a 51Cr release assay. The triplicate means for specific lysis [({sample release – spontaneous release}/{totalrelease � spontaneous release}) � 100%] were averaged from 3 independent experiments and plotted � SEM, P < 0.001. D, the percentages oftyrosinase tetramer-reactive cells that also expressed each of the epitopes listed along the x-axis were pooled from 3 independent experiments (mean � SD).E, dot plots from 1 of the 3 experiments summarized in D. Numbers in the top quadrants indicate the percentages of tetramer reactive cells that did ordid not express the respective epitope indicated on the x-axis. Although the proportion expressing CD27 was quite variable over the 3 separate experiments,as shown in D, the tetramer-reactive T cells from this experiment displayed a very clear population of CD27þ cells.






control the amount of IL15 in the system in vitro. T cellsfrom the gp100-pulsed moDC-vaccinated melanomapatients underwent a single 7-day restimulation with auto-logous gp100-pulsed moDCs in the presence of IL2 and/orIL15 or neither. CD3þCD8þ-gated responder T cells wereanalyzed cytofluorographically for early (AnnexinVþPI�)and late (AnnexinVþPIþ) apoptosis (Fig. 5A). IL15 signifi-cantly reduced apoptosis, thus prolonging CD3þCD8þ T-cell survival after restimulation (P < 0.05). The controlwithout cytokines was comparable to the condition withexogenous IL2 alone, because such low-dose IL2 did notenhance the higher amounts of endogenous IL2 alreadysecreted by DC-stimulated T cells (20).

CD3þCD8þ responder T cells increased expression of theantiapoptotic protein, Bcl-2, after restimulation by gp100-pulsed moDCs with exogenous IL15 (Fig. 5B). IL15 also

mitigated the reduction in Bcl-2 expression caused by IL2.Levels were nearly undetectable with IL2 alone.

Neither mouse nor human studies have addressed indetail whether IL15 could also attenuate regulatory T-cell(Treg) expansion by IL2. We therefore determinedthe percentage of Foxp3þ Tregs among the gatedCD3þCD4þCD25bright T cells responding to gp100peptide-pulsed (Fig. 5C) or fluMP-pulsed (Fig. 5D) humanmoDCs. IL15 reduced the proportion of candidate Tregssignificantly below the proportions generated by moDCsalone or with only low-dose exogenous IL2 (n ¼ 3 inde-pendent experiments, P < 0.01; Fig. 5C and D). These datashow that IL15, while expanding total and tetramer-reac-tive T cells, reduced the proportion of candidate Tregs thatwould otherwise respond to IL2 produced in culture oradded exogenously. In contrast to the mixed effect of IL15

KDa26

40

pt nl pt nl pt pt nl nl

Bcl-2

GAPDH

- - + + - + - +- - - - + + + +

IL2IL15

B

0

10

20

30

**

A

no cyto IL2 IL15 IL2+15

%

An

nex

in V

o

f C

D3

CD

8 T

cel

ls+

++

0

10

20

30

40

50

* *


%

Fox

p3

of

CD

4 C

D25

T c

ells

bri

gh

t

+

+

Cgp100

0

10

20

30

40

50

* *


%

Fox

p3

of

CD

4 C

D25

T c

ells

bri

gh

t

+

+

DfluMP

43

57

53

47

Figure 5. IL15 helps rescue CD3þCD8þ T cells from IL2-induced apoptotic cell death and reduces the frequency of CD4þCD25bright Foxp3þ Tregafter moDC stimulation. A, responder T cells underwent a single round of stimulation by autologous gp100-pulsedmoDCs and were assessed for cell death byflow cytometry after gating on the CD3þCD8þ population. The percentages of Annexin-Vþ and either PI� (early apoptotic) or PIþ (late apoptotic) were summedto obtain the total percentage of apoptotic cells. Data represent the averages of duplicate means � SEM from 6 independent experiments. P < 0.05for either IL15 condition with or without IL2, compared with IL2 alone or no exogenous cytokine (no cyto) condition. B, Western blot analysis of Bcl-2 wasperformed on total cell lysates from CD3þCD8þ T cells purified by positive immunomagnetic selection and stimulated for 7 days by gp100-pulsed autologousmoDCs in the presence of different cytokine conditions. CD3þCD8þ T cells originated from a previously vaccinated melanoma patient (pt) or from a healthydonor (nl). Results from 1 of 3 independent experiments are shown. GAPDH in CD3þCD8þ T cells served as an internal control. C and D, T cells wererestimulated in vitro twice over 7 days each by either autologous (C) gp100-pulsed or (D) fluMP-pulsed moDCs, using the indicated cytokine conditions.Bar graphs represent the proportion of phenotypic Tregs based on CD4þCD25bright T cells that coexpressed intracellular Foxp3 (mean � SD from 3–4independent experiments). Insets depict the Foxp3þ expression by these gated CD4þCD25bright T cells from a representative condition without cytokine.P < 0.01 for pairwise comparisons between either of the IL15-containing and the IL2 alone or no cytokine (no cyto) conditions.

Romano et al.





and IL2 on Bcl-2 expression, IL15 suppression of thesephenotypic Tregs overrode any effect of IL2.

Discussion

This clinical trial was the first head-to-head comparisonbetween vaccines using peptide-pulsed LCs or moDCs inhumans. Immune responses to tumor peptide and con-trol antigens developed in vivo in patients with advancedmelanoma in response to both DC subtypes, whichproved safe and well tolerated in all patients. Therewas a trend toward greater erythema and induration 48hours after peptide-pulsed moDC injections, possiblyindicating less trafficking away from the local sites todraining nodes; but these reactions were not significantlydifferent from those after LC-based vaccines. Theincreased proliferative response to KLH rechallenge withmoDCs may also account for their trend toward greaterDTH responses. We set a high threshold to define apositive response, and LC-based vaccines stimulatedsignificantly greater tyrosinase–HLA-A*0201 tetramerreactivity than the moDC-based vaccines. Beyond that,the results of this particular trial did not yet corrobo-rate extensive data in vitro that had predicted LC super-iority across the board (1, 2) or confirm the findingsof some investigators suggesting that there is anadvantage to including CD34þ HPC-derived LCs in DCvaccines (3).We conducted additional experiments to establish

mechanisms that could account for LC superiority overmoDCs in vitro (1, 2), using T cells primed in vivo by DCvaccination and focusing on the activity of IL15 with a viewtoward future optimization of tumor antigen presentation.IL15 is a pleiotropic cytokine that affects homeostasis,activation, and homing of lymphocytes in both innateand adaptive immunity (5). LCs secrete more IL15 thanthe moDCs (1, 4), and when exogenous IL15 was eitherlimiting or absent, only LCs could still stimulate a statis-tically significant enhancement of IFN-g production bytumor and viral antigen-specific T cells. Anti–IL15R-a sig-nificantly inhibited both LC- and moDC-stimulated IFN-gresponses, yet blockade was incomplete for LCs. Limitedcell numbers precluded isolation of CD8þ T-cell respon-ders, so CD4þ T cells could have been less sensitive toIL15R-a blockade. Lacking a positive control to guideantibody dosing, complete blockade of LCs may also haverequired a higher mAb concentration. Other unmeasuredfactors could also have played roles downstream of theinitial T-cell activation events, especially over two 7-dayrounds of stimulation.Somewhat paradoxically, we then concentrated on

mature moDCs so that we could control the amount ofIL15 in the system in vitro and not confound our assess-ments by what LCs could produce themselves. Exposure toIL15 during moDC development confers Langerhans-likeproperties on moDCs (14, 15), but does not provide IL15during DC stimulation of tumor-specific T cells, which wasthe focus of our inquiry.

Our data showed that IL15 supported the long-termexpansion by DCs of tumor antigen–specific T cells witheffector memory phenotype (TEMRA) and cytolytic function.There was a modest additive effect on proliferation whenIL2 combinedwith IL15. The 2 cytokines were synergistic inenhancing CTL activity stimulated by moDCs against mel-anoma antigens, however, with IL2 being an essentialcomponent of CD4þ T-cell help for the long-term expan-sion of CD8þ memory T cells (25–30). Low-dose exogen-ous IL2 avoided bystander activation of either IL15R (21)or NK cells in the responder populations, insofar as con-ventional DCs already stimulate T cells to secrete compar-able or higher amounts of IL2 (20). Hence, IL2 did not addmuch to the DC-stimulated controls without exogenouscytokines.

More importantly, IL15 countered the apoptosis inducedby IL2 activation, which otherwise impeded T-cell expan-sion by IL2 beyond the first or second week. IL15 achievedthis by partially abrogating IL2’s reduction of the antiapop-totic protein, Bcl-2. Mouse data predicted these findings,but there have been no comparable studies with humancells. IL15 also trumped all of the IL2 effect on Treg expan-sion, which has not been previously reported. These datawith human cells confirm that IL2 and IL15 have distinctroles in DC-stimulated T-cell apoptosis and survival.

Antigen-experienced T cells comprise both effector mem-ory (TEM) and TCM, which home respectively to inflamedtissues and lymphoid organs (7, 22). Investigators haveascribed TCM development to IL15 and TEM generation toIL2 (6, 7), but IL15 supports TEM in HIV infection (8). Ourdata showed that conventional DCs with IL15 expandedboth tumor and viral antigen-reactive human TEM cellslacking CCR7, CD62L, and CD127, yet coexpressingCD45RO, CD45RA, and CD69. This characterizes aCD45RAþ TEM subpopulation (TEMRA; 22), which hasthe largest amount of perforin and appears late in theimmune response (31–33). Somewhat atypical was thepersistent expression of CD27 on these cells, although asubset of CD27þ TEMRA cells with intermediate effectoractivity exists (34). Resting rather than activated popula-tions like those in this study, however, have mostly estab-lished the phenotypes for TCM and TEM cells. Our data alsoshowed that TEMRA cells developed under antigen-bearingDC-driven conditions with IL15, arguing against theirdevelopment through homeostatic rather than antigen-driven pathways (31, 32).

Despite encouraging findings from the clinical trialestablishing safety and efficacy of LC- and moDC-basedvaccines, a clear winner did not emerge across all antigenstested, although LCs stimulated significantly greater tetra-mer reactivity at least against HLA-A*0201–restricted tyr-osinase peptide. Among the outstanding challenges,foremost is the need to reconcile abundant data in vitrothat LCs are in fact superior to moDCs for stimulation ofvirus and tumor antigen–specific CTLs, with the responsesobserved in vivo. The clinical trial data do not exclude thepossibility that sufficient IL15 may have been presentin vivo, as reported in a mouse model of influenza infection






(35). Given the trend toward greater DTH reactivity aftermoDC vaccination and the significant increase in moDC-stimulated KLH-specific proliferation on rechallenge,we also cannot exclude induction in vivo of IL15 in moDCsby KLH-specific CD4þ T cells.

Additional limitations when translating studies in vitro toa clinical trial included differences in peptide hydrophobi-city that affected solubility and loading onto class I MHC.There was also no reliable way to confirm that sufficientpeptide remained bound toMHC long enough to stimulateT cells in draining lymph nodes. These problems beg forapproaches that achieve durable expression of full-lengthtumor antigens, which could in turn be processed intomultiple epitopes for simultaneous presentation on bothclass I and II MHC. Furthermore, because LCs secrete moreIL15 (1, 2, 4) than the moDCs, and given that LCs canstimulate significantly higher numbers of virus- and tumor-specific T cells in the complete absence of exogenous IL15orIL2, LCsmerit further specific investigation as immunogensfor theoptimal inductionof tumor immunity. Alternatively,continued use of moDCs as vaccines would benefit fromsupplementation with IL15 by transfection (36) or directadministration of the cytokine as a drug (NCT01021059,ClinicalTrials.gov). Available reagents and approaches havenot yet detected IL15R-a by flow cytometry on humanCD34þ HPC-derived LCs (1, 4), which is a prerequisitefor binding and presenting IL15 in trans to responderlymphocytes. Alternative approaches are underway toresolve this conundrum and to ascertain distinctionsbetween the mechanisms used by LCs versus moDCs forearly T-cell activation events that depend on IL15.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Author Contributions

E. Romano, M. Rossi, G. Ratzinger, J. Yuan, and J.W. Young designed,planned, and performed the experiments. J.W. Young, K.S. Panageas, P.B.Chapman, and A.N. Houghton designed the clinical trial. G. Ratzinger, M.Rossi, M-A. de Cos, E. Romano, D.J. Chung, J. Yuan, and J.W. Youngconducted the clinical trial. K.S. Panageas assisted with statistical designof the clinical trial, and G. Heller provided statistical design and analysisof the trial and lab correlative studies. J.W. Young conceived, planned,and reviewed the experiments. E. Romano and J.W. Young wrote themanuscript.

Acknowledgments

We appreciate the stimulating discussion and intellectual inputfrom other members of the Young Lab and the technical assistanceof Jennifer P. Ghith, Nicholas Sakellarios, and Dana E. Pepe. Thesupport and assistance of the MSKCC attending physicians, nurses,and staff of the Melanoma/Sarcoma and Adult Bone Marrow Trans-plant Services, the Blood Bank Donor Room, and the Cytotherapy Labin obtaining patient samples are gratefully acknowledged. We alsothank the clinical trial participants and healthy volunteers who pro-vided samples for research.

Grant Support

This study was supported by grants R01-CA083070 (J.W. Young),R01-CA118974 (J.W. Young), R21-CA119528 (J.W. Young), R21-CA097714 (J.W. Young), P01-CA23766 (G. Heller and J.W. Young),from the National Cancer Institute, NIH; Mr. William H. Goodwin andMrs. Alice Goodwin of the Commonwealth Cancer Foundation forResearch and The Experimental Therapeutics Center of MemorialSloan-Kettering Cancer Center (J.W. Young); Swim Across America(J.D. Wolchock and J.W. Young); and the University of Rome "LaSapienza" (E. Romano).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received December 23, 2010; revised February 14, 2011; acceptedFebruary 18, 2011; published OnlineFirst February 25, 2011.

References1. Ratzinger G, Baggers J, de Cos MA, Yuan J, Dao T, Reagan JL, et al.

Mature human Langerhans cells derived from CD34þ hematopoieticprogenitors stimulate greater cytolytic T lymphocyte activity in theabsence of bioactive IL-12p70, by either single peptide presentationor cross-priming, than do dermal-interstitial or monocyte-deriveddendritic cells. J Immunol 2004;173:2780–91(Erratum in J Immunol2005;174:3818).

2. Klechevsky E, Morita R, Liu M, Cao Y, Coquery S, Thompson-SnipesL, et al. Functional specializations of human epidermal Langerhanscells and CD14þ dermal dendritic cells. Immunity 2008;29:497–510.

3. Banchereau J, Palucka AK, Dhodapkar M, Burkeholder S, Taquet N,Rolland A, et al. Immune and clinical responses in patients withmetastatic melanoma to CD34(þ) progenitor-derived dendritic cellvaccine. Cancer Res 2001;61:6451–8.

4. Munz C, Dao T, Ferlazzo G, de Cos MA, Goodman K, Young JW.Mature myeloid dendritic cell subsets have distinct roles for activationand viability of circulating human natural killer cells. Blood2005;105:266–73.

5. Waldmann TA. The biology of interleukin-2 and interleukin-15: impli-cations for cancer therapy and vaccine design. Nat Rev Immunol2006;6:595–601.

6. Klebanoff CA, Gattinoni L, Torabi-Parizi P, Kerstann K, Cardones AR,Finkelstein SE, et al. Central memory self/tumor-reactive CD8þ T cells

confer superior antitumor immunity compared with effector memory Tcells. Proc Natl Acad Sci U S A 2005;102:9571–6.

7. Weninger W, Crowley MA, Manjunath N, von Andrian UH. Migratoryproperties of naive, effector, and memory CD8(þ) T cells. J Exp Med2001;194:953–66.

8. Mueller YM, Bojczuk PM, Halstead ES, Kim AH, Witek J, Altman JD,et al. IL-15 enhances survival and function of HIV-specific CD8þ Tcells. Blood 2003;101:1024–9.

9. Oh S, Perera LP, Burke DS, Waldmann TA, Berzofsky JA. IL-15/IL-15Ralpha-mediated avidity maturation of memory CD8þ T cells. ProcNatl Acad Sci U S A 2004;101:15154–9.

10. Stonier SW, Ma LJ, Castillo EF, Schluns KS. Dendritic cells drivememory CD8 T cell homeostasis via IL-15 trans-presentation. Blood2008;112:4546–54.

11. Dubois SP, Waldmann TA, M€uller JR. Survival adjustment ofmature dendritic cells by IL-15. Proceedings of the NationalAcademy of Sciences of the United States of America. 2005;102:8662–7.

12. Berger C, Berger M, Hackman RC, Gough M, Elliott C, Jensen MC,et al. Safety and immunologic effects of IL-15 administration innonhuman primates. Blood 2009;114:2417–26.

13. Banchereau J, Palucka AK. Dendritic cells as therapeutic vaccinesagainst cancer. Nat Rev Immunol 2005;5:296–306.

Romano et al.





14. Mohamadzadeh M, Berard F, Essert G, Chalouni C, Pulendran B,Davoust J, et al. Interleukin 15 skews monocyte differentiation intodendritic cells with features of Langerhans cells. J Exp Med2001;194:1013–20.

15. Dubsky P, Saito H, Leogier M, Dantin C, Connolly JE, Banchereau J,et al. IL-15-induced human DC efficiently prime melanoma-specificnaive CD8þ T cells to differentiate into CTL. Eur J Immunol2007;37:1678–90.

16. Ratzinger G, Baggers J, de Cos MA, Yuan J, Dao T, Reagan JL, et al.Mature human Langerhans cells derived from CD34þ hematopoieticprogenitors stimulate greater cytolytic T lymphocyte activity in theabsence of bioactive IL-12p70, by either single peptide presentationor cross-priming, than do dermal-interstitial or monocyte-deriveddendritic cells. J Immunol 2005;174:3818.

17. Szabolcs P,MooreMAS, Young JW. Expansion of immunostimulatorydendritic cells among the myeloid progeny of human CD34þ bonemarrow precursors cultured with c-kit ligand, granulocyte-macro-phage colony-stimulating factor, and TNF-a. J Immunol 1995;154:5851–61.

18. Szabolcs P, Avigan D, Gezelter S, Ciocon DH, Moore MA, SteinmanRM, et al. Dendritic cells and macrophages can mature independentlyfrom a human bone marrow-derived, post-colony-forming unit inter-mediate. Blood 1996;87:4520–30.

19. Maecker HT, Moon J, Bhatia S, Ghanekar SA, Maino VC, Payne JK,et al. Impact of cryopreservation on tetramer, cytokine flow cytometry,and ELISPOT. BMC Immunol 2005;6:17.

20. Young JW, Baggers J, Soergel SA. High dose UV-B radiation altersdendritic cell costimulatory activity but does not tolerize T lympho-cytes to alloantigen in vitro. Blood 1993;81:2987–97.

21. Fehniger TA, Caligiuri MA. Interleukin 15: biology and relevance tohuman disease. Blood 2001;97:14–32.

22. Sallusto F, Geginat J, Lanzavecchia A. Central memory and effectormemory T cell subsets: function, generation, and maintenance. AnnuRev Immunol 2004;22:745–63.

23. Sallusto F, Lenig D, Forster R, Lipp M, Lanzavecchia A. Two subsetsof memory T lymphocytes with distinct homing potentials and effectorfunctions. Nature 1999;401:708–12.

24. Romero P, Zippelius A, Kurth I, Pittet MJ, Touvrey C, Iancu EM, et al.Four functionally distinct populations of human effector-memoryCD8þ T lymphocytes. J Immunol 2007;178:4112–9.

25. Williams MA, Tyznik AJ, Bevan MJ. Interleukin-2 signals duringpriming are required for secondary expansion of CD8þ memory Tcells. Nature 2006;441:890–3.

26. Williams MA, Bevan MJ. Effector and memory CTL differentiation.Annu Rev Immunol 2007;25:171–92.

27. Kamimura D, Bevan MJ. Naive CD8þ T cells differentiate into pro-tectivememory-like cells after IL-2 anti IL-2 complex treatment in vivo.J Exp Med 2007;204:1803–12.

28. Boeckh M, Leisenring W, Riddell SR, Bowden RA, Huang ML, Myer-son D, et al. Late cytomegalovirus disease and mortality in recipientsof allogeneic hematopoietic stem cell transplants: importance of viralload and T-cell immunity. Blood 2003;101:407–14.

29. Bonini C, Lee SP, Riddell SR, Greenberg PD. Targeting antigen inmature dendritic cells for simultaneous stimulation of CD4þ andCD8þ T cells. J Immunol 2001;166:5250–7.

30. Hakki M, Riddell SR, Storek J, Carter RA, Stevens-Ayers T,Sudour P, et al. Immune reconstitution to cytomegalovirus afterallogeneic hematopoietic stem cell transplantation: impact of hostfactors, drug therapy, and subclinical reactivation. Blood 2003;102:3060–7.

31. Alves NL, Hooibrink B, Arosa FA, van Lier RAW. IL-15 inducesantigen-independent expansion and differentiation of human naiveCD8þ T cells in vitro. Blood 2003;102:2541–6.

32. Geginat J, Lanzavecchia A, Sallusto F. Proliferation and differentiationpotential of human CD8þ memory T-cell subsets in response toantigen or homeostatic cytokines. Blood 2003;101:4260–6.

33. Hislop AD, Annels NE, Gudgeon NH, Leese AM, Rickinson AB.Epitope-specific evolution of human CD8(þ) T cell responses fromprimary to persistent phases of Epstein-Barr virus infection. J ExpMed 2002;195:893–905.

34. Rufer N, Zippelius A, Batard P, Pittet MJ, Kurth I, Corthesy P, et al.Ex vivo characterization of human CD8þ T subsets with distinctreplicative history and partial effector functions. Blood2003;102:1779–87.

35. McGill J, Van Rooijen N, Legge KL. IL-15 trans-presentation bypulmonary dendritic cells promotes effector CD8 T cell survival duringinfluenza virus infection. J Exp Med 2010;207:521–34.

36. Steel JC, Ramlogan CA, Yu P, Sakai Y, Forni G, Waldmann TA,et al. Interleukin-15 and its receptor augment dendritic cell vac-cination against the neu oncogene through the induction of anti-bodies partially independent of CD4 help. Cancer Res 2010;70:1072–81.

37. Britten CM, Meyer RG, Kreer T, Drexler I, Wolfel T, Herr W. The use ofHLA-A*0201-transfected K562 as standard antigen-presenting cellsfor CD8(þ) T lymphocytes in IFN-gamma ELISPOT assays. J ImmunolMethods 2002;259:95–110.






Correction

Correction: Peptide-Loaded Langerhans Cells,Despite Increased IL15 Secretion and T-CellActivation In Vitro, Elicit Antitumor T-CellResponses Comparable to Peptide-LoadedMonocyte-Derived Dendritic Cells In Vivo

In this article (Clin Cancer Res 2011;17:1984–97), which was published in theApril 1, 2011 issue of Clinical Cancer Research (1), the name of the seventh authoris incorrect. The correct name is Jedd D. Wolchok.

Reference1. Romano E, Rossi M, Ratzinger G, de Cos M-A, Chung DJ, Panageas KS, et al. Peptide-loaded

Langerhans cells, despite increased IL15 secretion and T-cell activation in vitro, elicit antitumor T-cell responses comparable to peptide-loaded monocyte-derived dendritic cells in vivo. Clin CancerRes 2011;17:1984–97.

Published OnlineFirst May 3, 2011.�2011 American Association for Cancer Research.doi: 10.1158/1078-0432.CCR-11-0931

ClinicalCancer

Research

www.aacrjournals.org 3851

2011;17:1984-1997. Published OnlineFirst February 25, 2011.Clin Cancer Res Emanuela Romano, Marco Rossi, Gudrun Ratzinger, et al.

In VivoDendritic Cells Responses Comparable to Peptide-Loaded Monocyte-Derived

, Elicit Antitumor T-CellIn VitroSecretion and T-Cell Activation Peptide-Loaded Langerhans Cells, Despite Increased IL15

Updated version

10.1158/1078-0432.CCR-10-3421doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2011/03/30/1078-0432.CCR-10-3421.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/17/7/1984.full#ref-list-1

This article cites 36 articles, 25 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/17/7/1984.full#related-urls

This article has been cited by 5 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/17/7/1984To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-10-3421

http://clincancerres.aacrjournals.org/content/suppl/2011/03/30/1078-0432.CCR-10-3421.DC1

http://clincancerres.aacrjournals.org/content/17/7/1984.full#ref-list-1

http://clincancerres.aacrjournals.org/content/17/7/1984.full#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/17/7/1984


Peptide-Loaded Langerhans Cells, Despite Increased...

Documents

Transcript of Peptide-Loaded Langerhans Cells, Despite Increased...