Defining characteristics of classical Hodgkin lymphoma ... Defining characteristics of classical...Regular Article LYMPHOID NEOPLASIA Defining characteristics of classical Hodgkin

doi:10.1182/blood-2013-06-508044Prepublished online September 4, 2013;2013 122: 2856-2863   

 Wilson, Maria Calaminici and John G. GribbenPaul Greaves, Andrew Clear, Andrew Owen, Sameena Iqbal, Abigail Lee, Janet Matthews, Andrew microenvironment T-helper cellsDefining characteristics of classical Hodgkin lymphoma

http://bloodjournal.hematologylibrary.org/content/122/16/2856.full.htmlUpdated information and services can be found at:

(1557 articles)Lymphoid Neoplasia   �Articles on similar topics can be found in the following Blood collections

http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://bloodjournal.hematologylibrary.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

Copyright 2011 by The American Society of Hematology; all rights reserved.Washington DC 20036.by the American Society of Hematology, 2021 L St, NW, Suite 900, Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly    

For personal use only. by RAUL RIBEIRO on November 25, 2013. bloodjournal.hematologylibrary.orgFrom For personal use only. by RAUL RIBEIRO on November 25, 2013. bloodjournal.hematologylibrary.orgFrom

Regular Article

LYMPHOID NEOPLASIA

Defining characteristics of classical Hodgkin lymphomamicroenvironment T-helper cellsPaul Greaves, Andrew Clear, Andrew Owen, Sameena Iqbal, Abigail Lee, Janet Matthews, Andrew Wilson, Maria Calaminici,

and John G. Gribben

Centre for Haemato-oncology, Bart’s Cancer Institute, Barts and the London School of Medicine and Dentistry, Queen Mary University of London,

Charterhouse Square, London, United Kingdom

Key Points

• Hodgkin lymphomamicroenvironment T-helpercells express TH1/activationmarkers and lack TH2/immunosuppression markers.

• These cells are functional,retaining the capacity forcytokine secretion andproliferation in vitro.

CD41 T-helper cells (THs) dominate the classical Hodgkin lymphoma (CHL) microenviron-

ment, but their role is poorly understood. Advances in flow cytometry and immunohisto-

chemistry permit more detailed investigation of this aspect of CHL pathophysiology. To

address the hypothesis that the TH-infiltrate, rather than being TH2-enriched, senescent

and hypofunctional, is TH1 and activationmarker-rich, cytokine-secretory and proliferative,

we applied comprehensive flow cytometric immunophenotyping and functional assays of

cytokine secretion/proliferation to TH cells from 18 CHL-derived single-cell suspensions

(SCSs) compared to reactive lymph nodes (RLNs). CHL-derived TH cells express TH1-

associatedCXCR3/CCR5 andTNFa/IFNg/interleukin-2 (IL-2) and less TH2-associatedCCR3/

CCR4, with no IL-4/IL-13. They lack exhaustion-/suppression-associated PD1, CD57 and

terminally differentiated effector memory cells, with more central memory cells, activation-

associated partners of HodgkinReedSternberg (HRS) cell-expressedCD30/OX40-L/ICOS-L,

and other activation markers. TH cell lines established from CHL and RLN-derived SCSs

remain cytokine-secretory. We confirmed and extended these studies using tissue microarray immunohistochemistry (TMA-IHC) from

a large CHL tissue bank (n = 122) and demonstrate TH1-associated TBET is abundant in CHL, and TH2-associated CMAF/GATA3 and

exhaustion-associated PD1 expressed at significantly lower levels. These molecular insights into the CHL-associated TH offer potential

diagnostic, prognostic and pharmacologically modifiable therapeutic targets and do not support the established view of a TH2-enriched,

senescent/exhausted, hypofunctional, hypoproliferative infiltrate. (Blood. 2013;122(16):2856-2863)

Introduction

The bulk of the cellular immune infiltrate in classical Hodgkinlymphoma (CHL) comprises CD41 T cells (T-helper cells [TH]),along with cytotoxic T cells, macrophages, nonmalignant B cells,and innate immune cells.1 Antitumor immunity is largely T cellmediated,2 but the mechanisms underlying the failure of this richT-cell infiltrate to clear the malignancy are poorly understood. Acell-mediated immune response orchestrated by TH1-polarizedinflammation is likely central to tumor eradication,3,4 perhaps partlyby suppressing a competing protumoral5 TH2-mediated response.6

Previous studies have suggested that in CHL, the TH-cell compart-ment is TH2 and regulatory T cell (Treg) enriched, senescent/exhausted, hypofunctional, and hypoproliferative,1 providing someexplanation for the failed immune response. Systemic T-cell–specific immune defects are well described in CHL,7,8 but thecomplexity and heterogeneity of TH function challenges ourunderstanding of their specific role in the microenvironment.

Previous reports showed that the lymphoid immune infiltrate ofCHL is resistant to proliferation and cytokine secretion in vitro,9

suggesting a predominance of senescent/anergic TH cells. T cellsexpressingPD1 are functionally immunosuppressed through interaction

with ligands PD-L1/210; however, evidence for PD1 expression inthe microenvironment of CHL is limited despite substantial evi-dence that PD-L1 is expressed by the malignant Hodgkin ReedSternberg (HRS) cell.11 Although one study found PD1 expressedon CD41 T cells derived from CHL lymph nodes, it is based on just3 patients,12 and other studies have found only extremely low ex-pression levels.13-16 There are several other well-characterizedmarkers of senescent TH cells, including the terminally differentiatedeffector memory cell (TEMRA), but no detailed investigation ofmemory subsets has been carried out. CD57, associatedwith chronicviral infection and impaired function,17 is well known to be un-derexpressed in CHL.15

In fact, there is already a limited evidence base that an activated,TH1-biased, rather than suppressed, TH2-biased infiltrate, could bea major component of the tumor microenvironment. TH1 cells arecharacterized by TBET expression and interleukin-2 (IL-2)/IFNg/TNFa production, and TH2 by GATA3/CMAF expression and IL-4/IL-13/IL-21 production.18 TH1-associated TBET was found to beexpressed to a greater extent than TH2-associated GATA3 in a smallIHC study,19 although another study20 produced contradictory

Submitted June 11, 2013; accepted August 2, 2013. Prepublished online as

Blood First Edition paper, September 4, 2013; DOI 10.1182/blood-2013-06-

508044.

The online version of this article contains a data supplement.

The publication costs of this article were defrayed in part by page charge

payment. Therefore, and solely to indicate this fact, this article is hereby

marked “advertisement” in accordance with 18 USC section 1734.

© 2013 by The American Society of Hematology

2856 BLOOD, 17 OCTOBER 2013 x VOLUME 122, NUMBER 16

For personal use only. by RAUL RIBEIRO on November 25, 2013. bloodjournal.hematologylibrary.orgFrom

evidence, finding more TH2-associated CMAF expression thanTBET. Several groups have found CHL-infiltrating lymphocytes to becytokine secretory, capable of producing proinflammatory, TH1-biasedcytokines such as IL-2, IFNg, and TNFa as well as TH2 and immu-nosuppressive cytokines.21-24 TH1/TH2 polarization is also suggestedby chemokine receptor (CCR) profiling, with CCR3 and CCR4 pre-ferentially expressed by TH2-polarized cells25 and CXCR3 and CCR5by TH1-polarized cells.26 Although TH2-associated CCR3 and CCR4are receptors for the CHL-associated chemokines Regulated uponActivation Normal T cell Expressed and Secreted/CCL5 (RANTES)and CCL17/thymus- and activation-regulated chemokine, respec-tively,27,28 RANTES may also bind TH1-associated CCR5, andCCR4 is also a marker of Tregs, hence limiting the interpretationof TH polarization data based solely on CCR expression. However,previous publications have shown a significant proportion of in-filtrating TH cells express TH1-associated CXCR3 and CCR5.29,30

TH2-associated CCRs CCR3 and CCR4 were also demonstrated inthe lymphoid microenvironment in these studies, but not to a greaterextent than the TH1-associated CCRs. These studies incorporatedsmall numbers of samples andwere limited to isolated investigations ofonly cytokine, CCR, or transcription factor expression but provide anevidence base supporting an activated TH1-rich microenvironment forfurther investigation in larger, more extensive cohorts.

An activated TH microenvironment may itself contribute todisease pathogenesis. The HRS cell is derived from a germinalcenter B cell with JAK/STAT and nuclear factor-kB pathway over-expression essential to its survival. Having lost of much of its B-cellphenotype, notably surface immunoglobulin, through which prosur-vival signals would normally be delivered, the HRS cell is reliantupon intrinsic genetic defects and signaling through a range ofoverexpressed surface molecules, including LMP1 in Epstein Barrvirus (EBV)-positive cases, and members of the tumor-necrosisfactor and immunoglobulin receptor superfamilies, including CD70,CD80/CD86, CD30, CD40, OX40-L/CD252, and ICOS-L/CD275(for a review, see Kuppers31). The reciprocal molecules, receptor, orligand (CD27, CD28, CD153, CD154, CD134, and CD278, respec-tively) are often present on CHL microenvironment immune cells1

and all are upregulated on activated TH cells. CD27 and CD28 areexpressed at high levels on unstimulated or naıve T cells, loss ofexpression associatedwith activation, and retention ofCD27 expressionassociated with a memory cell phenotype.32 Specific expression ofthese molecules by CHL-infiltrating TH cells has not previouslybeen comprehensively investigated.

New tools in tissue microarray immunohistochemistry (TMA-IHC) and multicolor fluorescence flow cytometric immunopheno-typing (flow) and the discovery of TH-defining transcription factors,cell surface molecules and cytokine secretory profiles haveenabled more extensive investigation of functional subsets in themalignant microenvironment, overcoming the limitations of pre-vious studies. We used diagnostic frozen single-cell suspensions(SCSs) for flow, cytokine-secretion, and proliferation assays and alarge tissue bank of formalin-fixed, paraffin-embedded (FFPE) tissuefor TMA-IHC to characterize the TH infiltrate of CHL. We founda predominance of TH1 over TH2, an absence of markers ofsenescence, an excess of central memory cells (CMs), and retainedcytokine secretory and proliferative capacity for CHL-associatedT cells, with some markers showing associations with clinicaloutcome. These findings challenge the established view of a TH2-enriched, senescent, hypoproliferative infiltrate, provide insightinto molecular interactions between functionally active TH cells andmalignant HRS cells, and offer potential diagnostic, prognostic, andpharmacologically modifiable therapeutic targets.

Methods

Ethics approval was obtained from the local regional ethics committee. Allpatient-derived samples were obtained with informed consent in accordancewith the Declaration of Helsinki and stored under conditions compliant withthe Human Tissue Act 2008.

Patient samples for frozen SCSs

SCSs derived from surplus material available after diagnostic biopsy wereretrieved from nitrogen-frozen archived samples, with patient, histological,and storage characteristics detailed in Table 1 (CHL: n5 18; reactive lymphnode [RLN]: n 5 6). Details of storage/retrieval methods and selection ofcontrol samples are provided in the supplemental Methods. The median ageof patients with CHL (29 years, range 16-81) and those with reactive nodes(24 years, range 18-73) was not significantly different (P 5 .713). Themedian duration of time for which SCSs derived from CHL were held instorage (48 months, range 11-149) was significantly longer (P 5 .024)compared with that from reactive nodes (18 months, range 5-23). However,no correlation between sample age and expression of any single marker byflow could be demonstrated (data not shown).

Flow

Flow was carried out using standard methods for fluorescence immuno-phenotyping and intracellular cytokine stimulation and staining, withdetails of fluorochrome-conjugated antibodies and reagents provided inthe supplemental Material.

In vitro cell culture

SCS-derived cells were cultured based on a technique developed for pro-liferating tumor-infiltrating lymphocytes33 but optimized using conditions

Table 1. Summary of patient characteristics for theimmunophenotype study

Histology Age at biopsy (y) Node origin Sample age (mo) EBV status

CHL (LR) 35 Cervical 81 1

CHL (MC) 16 Cervical 149 1

CHL (MC) 31 Cervical 75 1

CHL (MC) 33 Cervical 49 1

CHL (NK) 23 NK 139 NK

CHL (NS) 38 Cervical 44 2

CHL (NS) 20 Cervical 82 2

CHL (NS) 28 Cervical 11 2

CHL (NS) 45 Axilla 31 2

CHL (NS) 23 Cervical 39 2

CHL (NS) 31 Supraclav 44 2

CHL (NS) 21 Inguinal 35 2

CHL (NS) 81 Cervical 119 2

CHL (NS) 24 Cervical 80 2

CHL (NS) 28 Cervical 149 1

CHL (NS) 40 Cervical 44 1

CHL (NS) 29 Cervical 46 1

CHL (NS) 18 Cervical 18 NK

FH 19 Cervical 23 NA

FH 68 Cervical 18 NA

FH 18 Submental 17 NA

PH 21 Cervical 18 NA

PH 73 Cervical 22 NA

PH 27 Axilla 5 NA

EBV status defined by LMP1 or EBER depending upon era of biopsy.

Median age patient (years)5 29/24/27. Median age SCS sample CHL/Reactive/

PH (mo) 5 48/18/18.

FH, follicular hyperplasia; LR, lymphocyte rich; MC, mixed cellularity; NK, not

known; NS, nodular sclerosis; PH, paracortical hyperplasia.

BLOOD, 17 OCTOBER 2013 x VOLUME 122, NUMBER 16 HELPER T CELLS IN HODGKIN LYMPHOMA 2857

For personal use only. by RAUL RIBEIRO on November 25, 2013. bloodjournal.hematologylibrary.orgFrom

and a cytokine combination found best to support long-term TH growth, asdetailed in the supplemental Methods.

Patient samples for TMA construction

and immunohistochemistry

We identified 122 adult patients of known clinical outcome diagnosed atSt. Bartholomew’s Hospital who had high-quality, FFPE tissue from theoriginal diagnostic biopsy available, the characteristics of which werepreviously published and are summarized in Table 2.12 Median follow-upwas 16.5 years (range, 2-40 years). TMA construction and quality control,immunohistochemistry, EBV status, and automated cell counting werecarried out as described in the supplemental Methods and as previouslypublished.34

Statistical analysis

All pairwise comparisons were carried out using the Mann-Whitney U test.For comparisons of markers detected using flow, we determined percentageof CD31CD41 cells expressing each marker and calculated median ex-pression levels for CHL-derived CD31CD41 cells (CHL-TH; n 5 18),paracortical hyperplasia-derived cells only (PH-TH, n 5 3), and all RLN-derived cells comprising both PH-TH cells and follicular hyperplasia-derived TH cells (RLN-TH; n 5 6). Statistically significant differenceswere determined using pairwise comparisons of medians for CHL-TH vsPH-TH and CHL-TH vs RLN-TH. Full results and statistical analysis arepresented in supplemental Table 1. For TMA-IHC, median cell count permm2 was calculated for each marker derived from all patient samples basedon the automated image analysis cell count and expression levels for eachmarker compared pairwise. Pairwise comparisons were also carried outbased on EBV status (positive or negative) and the major histologicalsubtypes (nodular sclerosis or mixed cellularity).

Survival outcomes were measured from date of diagnosis to occurrenceof event or date of last follow-up. These were: overall survival (OS); deathfrom any cause; disease-specific survival (DSS); death from disease ortreatment; and freedom from first-line treatment failure, the event beingdeath, first relapse, or progression on first-line therapy. Each case was classifiedas high or low expression based on the number of microenvironment cells/mm2

expressing each marker. The cutpoint was generated for each marker byusing the X-Tile statistical package35 (Yale University, New Haven, CT)with the test/validation set methodology, as detailed in the supplementalMethods. Kaplan-Meier curves defined by these cutpoints were generatedand statistical significance of differences arising from differential expressionof each marker determined using the log-rank test. Multivariate modelswere built using a Cox proportional-hazards regression model (forwardstepwise likelihood ratio) incorporating variables selected with a univariateP value , .05. Statistical significance for all tests was set at .05.

Results

Extended immunophenotype of SCS-derived TH cells

Expression levels and statistical comparisons of all markers mea-sured by flow are provided in supplemental Table 1. The relativecomposition of CD4, CD8, and CD20 was similar for SCS derivedfrom CHL, PH, and all RLN (supplemental Figure 4).

TH1 and TH2

Because TH2 polarization is thought to contribute to the failedimmune response in CHL, we first investigated TH1 and TH2polarization in the CHL-TH. Using flow, we assessed TH1- andTH2-associated CCR expression in CHL (CHL-TH) and RLN(RLN-TH), including CCRs recognizing HRS cell-expressed chemo-kines CCL17/thymus- and activation-regulated chemokine andCCL5/RANTES. The median proportion of CHL-TH cells express-ing the TH1-associated CCRs CXCR3 (45%) and CCR5 (22%) wassignificantly greater (P , .005) than expressing TH2-associatedCCR3 (8%) or CCR4 (4%) (Figure 1A-B). Further, TH1-associatedCCRs were more highly expressed in CHL-TH than in RLN-THcells (Figure 1C), with similar proportions expressing TH2-associatedCCRs. Weused acytokine stimulation assay todetermine the functionalcapacity of TH cells to produce TH1- or TH2-defining cytokines.CHL-THproduced TH1 cytokines to a significantly greater (for TNFa)or comparable (for IFNg and IL-2) extent compared with RLN-TH(Figure 1D-E) and although the TH2 cytokines IL-4 and IL-13 weredetected in TH2-polarized cell lines, they were not detectable inSCS-derived TH cells from any primary source (Figure 1F). IL-21(Figure 1E), a cytokine without clear TH subset-defining features, wasexpressed in a significantly greater proportion (P, .005) of CHL-THcells (11%) compared with RLN-TH (1%).

Senescence, immunosuppression, and memory

Because HRS cells overexpress PD-L1 and interaction with itsT cell–expressed ligand PD1 is thought to contribute to the failedimmune response in CHL, we next sought to determine the con-tribution of PD1 expression to immunosuppression in CHL-TH.PD1 was detectable only at low levels in CHL-TH cells and ex-pressed in a higher proportion of RLN-TH cells (Figure 2A). Wedetermined 2 other indicators of functional senescence in TH cells:the expression ofCD57 and the TEMRA phenotype (CCR7-CD45RA1).CD57 was expressed in significantly fewer CHL-TH cells (4%)compared with RLN-TH (19%) (P, .050 (Figure 2A). The TEMRA

phenotype was rarely found inCHL-TH cells, whereas it was detectedin up to 10% of RLN-TH cells (Figure 2B). In fact, the dominantmemory subset of CHL-TH cells was the CM, whereas for RLN-THcells the dominant subset was the effector memory cell (Figure 2C-D).A similar pattern was found using the T-cell–homing selectin CD62Las a marker of central memory (not shown).

Activation, Tregs, and proliferation

We find little evidence to support a TH2-skewed phenotype orevidence of excess senescence or exhaustion in CHL-TH comparedwith RLN-TH cells. We instead demonstrate that CHL-TH cellsexpress high levels of TH1 cytokines and CCRs and CM markers atcomparable or greater levels than in RLN-TH cells, suggesting anactivated phenotype for the CHL-TH. We therefore stained for anextended panel of activation-modulated molecules, including receptor/ligands reciprocal for molecules expressed by the HRS cell (CD30,

Table 2. Summary of patient characteristics for the TMA-IHC study

Total number of patients inoriginal cohort (n 5 122)

Percent of originalcohort (n 5 122)

Male 79 65

Age .45 y 27 22

Advanced (stage IIB-IV) 87 71

Anthracycline-based

chemotherapy

56 46

Alkylator-based

chemotherapy

52 43

Radiotherapy only 14 11

Combined modality 48 39

Histological subtype

Nodular sclerosis 93 78

Mixed cellularity 25 20

Classical lymphocyte

rich

0 —

Lymphocyte deplete 2 2

EBER-ISH 1 38 31

2858 GREAVES et al BLOOD, 17 OCTOBER 2013 x VOLUME 122, NUMBER 16

For personal use only. by RAUL RIBEIRO on November 25, 2013. bloodjournal.hematologylibrary.orgFrom

CD40, CD70, CD80/86, and CD134-L). These markers were allexpressed by a high proportion of CHL-TH cells to a comparableor significantly greater extent than RLH-TH (Figure 3A-B). CD27and CD28, both downregulated with stimulation, were expressedin a large proportion of CHL-TH cells; however, medianCD27 expres-sion levels were significantly lower than in RLN-TH.

Next, we investigated the expression of TH-associated markersof regulatory T-cell function, also nonspecifically upregulated onactivated T cells (FOXP3, CTLA4, CD25/IL-2Ra). There is sub-stantial evidence of a significant infiltrate of FOXP3-expressing cellsin the CHL microenvironment proposed to represent regulatoryT cells. Although we and others have previously demonstrated that

high levels of FOXP3 expression are associated with good prognosisin CHL,34 there was no difference in median expression levels ofFOXP3 compared with RLN-TH. We confirm that FOXP3 wasexpressed in CD41 T cells coexpressing CD25 and CTLA4 (sup-plemental Figure 5). CTLA4 was expressed at significantly greaterlevels (P 5 .011) in CHL-TH (50%) than in RLN-TH (20%).

Given the activated, cytokine secretory, central memory-rich THphenotype demonstrated in the previous experiments, we assessedthe proliferative capacity of the TH cells. We optimized a culturesystem with IL-2/IL-4 supplementation to expand the TH popula-tion without antigen ormitogen. 15/18CHL-derived SCS and 5/6RLN-derived SCSs had adequate cells remaining after immunophenotyping

Figure 1. Expression of TH-polarization

and CHL-associated CCRs as summarized

in supplemental Table 1. (A) Representa-

tive flow plots and (B) summary of expression

of TH1 and TH2 demonstrating a greater

proportion of CHL-TH cells expressing TH1-

related CCRs than TH2-related CCRs. (C)

Greater expression of TH1-related CCRs in

CHL-TH than in RLN-TH. (D) Representative

flow plots and (E) summary of expression of

TH-related cytokines measured by intracel-

lular cytokine assay demonstrates greater or

equivalent expression levels of TH1-related

cytokines TNFa, IFNg, and IL-2, along with

IL-21 in CHL-TH compared with RLN-TH, with

no measurable TH2-related cytokine expres-

sion (F). RLN comprised 6 samples, of which 3

were classifiedPHand 3FH.Results for PHare

additionally presented as a separate category

in these figures for reasons described in the

supplemental Material.

Figure 2. Expression of markers of memory and senes-

cence in TH cells. (A) Representative flow plots and summary

of expression of PD1 and CD57 demonstrates relative absence

of these markers of immunosuppression/senescence in CHL-TH

compared with RLN-TH. (B) Representative flow plots and (C)

summary of T memory subsets demonstrate a relative absence

of TEMRA cells and increased proportion of CM cells in CHL-TH

compared with RLN-TH. (D) The major differences in subset

distribution between CHL-TH and RLN-TH, with bars divided

according to the median proportion of cells defined by each

memory phenotype and error bars omitted for clarity. Those

subsets for which there are significant differences between CHL

and RLN and presented with error bars in panel C.

BLOOD, 17 OCTOBER 2013 x VOLUME 122, NUMBER 16 HELPER T CELLS IN HODGKIN LYMPHOMA 2859

For personal use only. by RAUL RIBEIRO on November 25, 2013. bloodjournal.hematologylibrary.orgFrom

for the cell culture experiment. There was no significant differencein proliferation between CHL- and RLN-derived culture systems(Figure 3C, median doublings in CHL- and RLN-derived systems,respectively, at day 305 2.7 vs 3.0 and day 605 4.2 vs 4.8;P5 .67).Day 30 flow confirmed all culture systems to be predominantlycomprised of TH-cells with ongoing cytokine-secretory capacity(Figure 3D). In contrast, tonsil-derived culture systems failed toproliferate.

Validation by TMA-IHC in a large independent cohort

To corroborate our findings of TH1 predominance over TH2 andlow expression levels of PD1 in situ without the histological ar-chitecture disruption and freeze-/thaw-induced changes of markerexpression, we investigated a large independent cohort of FFPEpatient samples. We used TMA-IHC to examine expression of TH1-associated TBET, TH2-associated CMAF/GATA3, and PD1 for122 cases of CHL with known clinical outcome. Representativeexamples of cases are presented in Figure 4A. There was significantheterogeneity of expression between samples (Figure 4B). TBETwas expressed in significantly greater numbers of microenvironmentcells comparedwith CMAF (median 879 vs 272 cells/mm2,P, .0001)or GATA3 (105 cells/mm2, P , .0001). FOXP3 expression waspreviously described in this cohort,34 but notably was expressed ina higher number of cells than any other TH subset marker, includingTBET (2346 vs 879 cells/mm2, P , .0001). Because a TH1-biasedinfiltrate may be expected to induce a more robust cytotoxic T-cellinfiltrate, we tested for a correlation between TBET expression andnumber of CD81 T cells by using data generated from previouslypublished work.34 Interestingly, we found a significant positivecorrelation between numbers of CD81 cells and numbers ofTBET1 cells (Pearson R 5 0.300; P , .005). There was no cor-relation between numbers of CD81 cells and numbers of CMAF1 orGATA31 cells (supplemental Figure 6).

We observed strikingly little expression of PD1 in the CHLmicro-environment despite high levels seen in the internal positive control of

tonsil (Figure 4A, far right and inset). There was no detectable PD1expression in 42% of patients, and a further 40% showed expression in,0.5% of all nucleated cells.

Clinical associations

In light of the significant heterogeneity of expression of thesemarkersand in order to explore the potential clinical impact of a TH1-polarized, activated TH infiltrate in contrast to a PD1-expressingsuppressive infiltrate, we went on to assess clinical associations. Thiswas only possible with the large number of samples available in TMAand hence only for those markers that were stained by IHC.Univariate survival analysis revealed associations between high levelsof TBET expression and superior DSS (median 5-year DSS for.1500cells/mm2 5 97% vs 77% for ,1500 cells/mm2; test/validation setcorrectedP5 .045) (Figure 4C), andwith EBV status (1750 cells/mm2

for EBV1 vs 669 for EBV2, P 5 .0033; supplemental Figure 7)and between GATA3 and histological subtype, with microenvironmentexpression of GATA3 significantly greater in nodular sclerosis com-pared with mixed cellularity cases (183 vs 24 cells/mm2, P5 .0028;data not shown). No other significant associations between tran-scription factor expression level, clinical outcome, EBV status, andhistological subtype were found. Those patients having detectablePD1 expression (.15 cells/high-power field) had poorer 5-yearDSS (63% vs 86%;P5 .012), but OS (63% vs 84%,P5 .18) was notsignificantly different (Figure 4D). Multivariate analysis incorpo-rating only those markers expressed by the microenvironment thatshowed significant association with survival outcomes in univariateanalysis in this and our previously published study34 (CD20, CD68,FOXP3, TBET, and PD1) was carried out (supplemental Table 8).Only high PD1 expression (P 5 .007) and low FOXP3 expression(P 5 .029) remained associated with adverse OS in this model.However, the limitations of multivariate analysis on incorporation oflarge numbers of variables with limited numbers of survival events areacknowledged, and hence any such analysis must be interpretedwith caution.36 In those cases investigated by flow cytometry where

Figure 3. Expression of activation markers

and proliferative capacity of TH cells. (A)

Summary of all activation markers showing pro-

portions of CHL-TH cells expressing all activation

markers compared with RLN, with those showing

statistically significant differences indicated. (B)

Representative flow plots and summary compar-

ing CHL-TH and RLN-TH expression levels for

selected markers that showed increased median

expression in CHL-TH compared with RLN-TH.

Similar proportions of CHL-TH and RLN-TH ex-

pressed all other markers. (C) Representative

photomicrograph (left,340 objective, Olympus

BX61 microscope) and bar chart (right) summa-

rizing proliferation CHL, RLN, and tonsil-derived

SCSs at 30 and 60 days. (D) Representative flow

plots showing capacity of proliferative cells to

produce TH1 (TNFa and IFNg, top) and TH2

(IL-4 and IL-13, bottom) cytokines at 60 days

(right) compared with only TH1 at baseline

(bottom) having been exposed to IL-2 and IL-4

in the culture medium.

2860 GREAVES et al BLOOD, 17 OCTOBER 2013 x VOLUME 122, NUMBER 16

For personal use only. by RAUL RIBEIRO on November 25, 2013. bloodjournal.hematologylibrary.orgFrom

single cell suspensions were available, we could not find any as-sociations between EBV status and expression levels of markers.However, with only 9 EBV negative cases and 7 EBV positive(2 unknown), the sample size may have been underpowered.

Discussion

Advances in technology and in our understanding of markers ofT-cell function have enabled us to revisit the phenotype of theCHL-infiltrating TH. This study is, to our knowledge, the mostdetailed investigation of the CD41 TH in CHL using flow withcorroboration in TMA-IHC. The TH1-associated CCRs CXCR3 andCCR5 were expressed in a greater proportion of CHL-TH cells thanTH2-associated CCR3 and CCR4, and cytokine expression profilingindicates that the cells are TH1 polarized and proinflammatorywithout expression of TH2-associated cytokines. Compared withRLN-TH cells, the TH cells of the CHL microenvironment expressmore TH1-associated CCRs and produce TH1 cytokines to a similaror greater extent. IL-21 expression was notably restricted to CHL-TH

cells. Although this cytokine has been described in TH2 cells, ithas also been associated with diverse TH function, including TH1,TFH (follicular helper), TH17, memory, and enhancement of cell-mediated antitumor responses.37 There is also evidence that it isimportant for HRS cell survival.38 Although previous studies havesuggested that IL-13 also appears to be a growth factor for HRScells,39 we found no evidence of its production by CHL-TH cells,suggesting its activity only as an autocrine growth factor.

By TMA-IHC, we noted low expression of GATA3 or CMAFcompared with levels of expression of TBET, where the presenceof higher numbers of infiltrating cells was associated with EBV1

status and superior survival outcome. This association has been pre-viously reported in pediatric CHL40 but never to our knowledge in anadult series. Associations between TBET and EBV status may beexpected, with EBV infection inducing a TH1/cytotoxic T-cell re-sponse. A large gene expression profiling study of frozenwhole tumorsamples supports this hypothesis,41 where EBV1 cases showedoverexpression of genes associated with TH1/cytotoxic immuneresponse. However, TBET was not overexpressed in EBV1 cases inthis unsorted whole-tumor gene expression profile. We acknowl-edge that expression of TH polarization-defining transcription factors

Figure 4. Immunohistochemistry to investigate TBET, CMAF, GATA3, and PD1 expression. Photomicrographs (340 objective, Olympus BX61 microscope) of

representative examples (A) showingmedian expression of TH1-associated TBET (extreme left, black arrowhead showing TBET1HRS cell) and TH2-associated CMAF (center

left) and GATA3 (center right, black arrowhead showing GATA31 HRS cell). PD1 expression was undetectable in most cases with the example (extreme right) being one of the

few cases with .15 cells/high-power field associated with adverse outcome. Expression is weak compared with the internal positive control of tonsil (inset). Numbers of cells

expressing each TH-associated marker and PD1 are compared (B) showing TH1-associated TBET is expressed by greater numbers of cells compared with TH2-associated

CMAF/GATA3 and with minimal PD1 expression. DSS curves indicate a positive impact of TBET expression (C) and a negative impact of PD1 expression (D) in univariate analysis.

BLOOD, 17 OCTOBER 2013 x VOLUME 122, NUMBER 16 HELPER T CELLS IN HODGKIN LYMPHOMA 2861

For personal use only. by RAUL RIBEIRO on November 25, 2013. bloodjournal.hematologylibrary.orgFrom

has been described outside the context of a classical TH1/TH2dichotomy. Expression of TBET has been shown in TH1 response-suppressing Tregs and immunoglobulin G2/3 class-switching B cells,42

GATA3 in T-cell precursors and Tregs,43 and CMAF in TFH andTH17 cells.44 This complicates the interpretation of these TMA-IHCfindings in isolation. However, combining functional and flowphenotype data from the SCS work, a predominance of TBET andabsence of CMAF/GATA3 is highly suggestive of a TH1-polarizedimmune infiltrate in CHL with a surprisingly absent TH2 response.Overall, therefore, we find little evidence that TH2 bias can explainthe failure of antitumor immunity in CHL, although we acknowledgethat cells expressing a particular combination of cytokines, CCRs,and polarizing transcription factors in vitro may behave differentlyin the complex multicellular environment in vivo.

The evidence for a TH2 bias in CHL was based largely on earlyimmunophenotyping data,45 which showed that the majority oftumor-infiltrating cells are CD45RO1, CD45RA2, and CD45RBlo.Although these results are consistent with our own immunopheno-typing data, they have been previously interpreted as suggestingexcess TH2-biased cells. This interpretation was in the context ofa more limited understanding of CD45 isoforms.46 It is now under-stood that CD45 isoform expression in T cells discriminates naivefrom antigen-exposed cells and proliferative from secretory/effectorcells and may be an indicator of overall functional plasticity ratherthan TH1/TH2 polarization (for a review, see Penninger et al47).

A TH1-biased infiltrate may be hypothesized to be accompaniedby a more extensive cytotoxic T-cell infiltrate, and indeed we founda positive association between numbers of CD81 T cells and ex-pression of TBET. However, our previous publication34 reported noassociation between numbers of CD81 cells and clinical outcome,although perhaps unexpectedly a number of previous groups haveshown an adverse impact of microenvironment expression ofmarkers of cytotoxic T-cell function, Granzyme B, and TIA1.20,48

Our study used the material available to focus on markers of CD41

function and as such did not investigate these markers.PD-L1 overexpression by HRS cells is well described,11,49,50 but

there is a paucity of evidence for expression of its ligand PD1 in theCHL microenvironment.13,14 and we confirm strikingly low expres-sion levels. This suggests that PD1/PD-L1 immunosuppressive in-teractions cannot readily explain the failure of the immune responseto eradicate tumor. PD1 appears to be important in the small minorityof cases in which it is expressed and in whom it is associated withadverse prognosis, confirming a previous study.16 Based on our pre-vious study of the significant infiltrate of FOXP31 cells in CHL, weinvestigated expression of CTLA4, an immunosuppressive moleculeoverexpressed by FOXP31 Tregs as well as by activated effectorT cells, and found it expressed to a greater extent in CHL than in RLNTH cells, although in the SCS, median FOXP3 expression was similarin CHL and RLN-TH cells.

CTLA4 is also nonspecifically upregulated in activated non-regulatory T cells. We find extensive evidence for T-cell activationwith overexpression of markers of central memory and upregulationof activation-induced molecules, including ligand/receptor pairs forsurface molecules expressed by the HRS cell and a capacity forcytokine-induced proliferation of the TH component of CHL to asimilar extent as that seen in reactive nodes.

Rosetting of the HRS cell by T cells is frequently reported instudies of the histological architecture of CHL. We found no spatial

relationship between expression of TBET, GATA3, or CMAFin this study, with markers being relatively evenly distributedthroughout the lymphoid microenvironment. We acknowledge thata limitation of using SCS is that no comment can be made on spatialrelationships between T cells expressing activation moleculesand HRS cells expressing their receptor/ligand reciprocal, andsuch an investigation in TMA would be an important componentof future work.

Despite the heterogeneity of expression levels of markers betweensamples, reflecting disease heterogeneity, our data do suggest acharacteristic immunophenotype of the CHL-TH cell, namely proin-flammatory, cytokine-secretory, rich in CM cells, lacking TH2-associated CCR3 or IL-4/-13 and senescence markers PD1, CD57,or TEMRA, and overexpressing TH1-associated CXCR3/CCR5and activation-induced CD153/CD30-L, CD134/OX40, CD278/ICOS, CD152/CTLA4, CD95, and CD45RO. This appears to bea pathognomonic “activated” signature of the CHL microenvi-ronment, perhaps suitable for diagnostic application in histologicallyambiguous cases.

The evidence suggests that the malignant tumor is capable ofsequestering substantial numbers of activated, functional T cells, pro-viding one explanation for the profound systemic immune defectencountered in advanced disease. Overexpression of reciprocalreceptor/ligand pairs for HRS cell-expressed molecules capableof transducing survival signals suggests a number of potentiallyimportant therapeutic targets, all of which should continue to beexplored. Overall, we find little to support the hypothesis that theTH infiltrate is TH2 biased or exhausted/senescent, but rather wefind evidence for an activated, proliferative, and proinflamma-tory cytokine-secretory phenotype.

Acknowledgments

This work was supported by The Baker Foundation, Cancer ResearchUK and by aNational Cancer Institute ProgrammeGrant (PO1C81538to J.G.G.).

Authorship

Contribution: P.G. designed and performed the research, performedstatistical analysis and results interpretation, and wrote the paper;A.C. and A.O. constructed the TMA and optimized the IHCreagents; S.I. collected and processed the SCS for flow; J.M. andA.W. managed patient database and performed statistical analysis;A.L. provided expert histopathological review; M.C. provided experthistopathological review and supervised the project; and J.G.G.designed the research, supervised the project and wrote the paper.

Conflict-of-interest disclosure: The authors declare no compet-ing financial interests.

Correspondence: Paul Greaves, Centre for Haemato-oncology,Bart’s Cancer Institute, Barts and the London School of Medicineand Dentistry, Queen Mary University of London, CharterhouseSquare, London EC1M 5BQ, United Kingdom; e-mail: [email protected].

2862 GREAVES et al BLOOD, 17 OCTOBER 2013 x VOLUME 122, NUMBER 16

For personal use only. by RAUL RIBEIRO on November 25, 2013. bloodjournal.hematologylibrary.orgFrom

References

1. Steidl C, Connors JM, Gascoyne RD.Molecular pathogenesis of Hodgkin’s lymphoma:increasing evidence of the importance of themicroenvironment. J Clin Oncol. 2011;29(14):1812-1826.

2. Van Pel A, Vessiere F, Boon T. Protection againsttwo spontaneous mouse leukemias conferred byimmunogenic variants obtained by mutagenesis.J Exp Med. 1983;157(6):1992-2001.

3. Nastala CL, Edington HD, McKinney TG, et al.Recombinant IL-12 administration induces tumorregression in association with IFN-gammaproduction. J Immunol. 1994;153(4):1697-1706.

4. Lowes MA, Bishop GA, Crotty K, Barnetson RS,Halliday GM. T helper 1 cytokine mRNA isincreased in spontaneously regressing primarymelanomas. J Invest Dermatol. 1997;108(6):914-919.

5. Pellegrini P, Berghella AM, Del Beato T, Cicia S,Adorno D, Casciani CU. Disregulation in TH1 andTH2 subsets of CD41 T cells in peripheral bloodof colorectal cancer patients and involvement incancer establishment and progression. CancerImmunol Immunother. 1996;42(1):1-8.

6. Kacha AK, Fallarino F, Markiewicz MA, GajewskiTF. Cutting edge: spontaneous rejection of poorlyimmunogenic P1.HTR tumors by Stat6-deficientmice. J Immunol. 2000;165(11):6024-6028.

7. Sartoris S, Cavallero P, Pegoraro L, Vergnano F,Fazio M. The absence of lymphocyte response invitro to tuberculin challenge in Hodgkin’s disease.Panminerva Med. 1965;7(10):370-372.

8. Han T, Sokal JE. Lymphocyte response tophytohemagglutinin in Hodgkin’s disease. Am JMed. 1970;48(6):728-734.

9. Marshall NA, Christie LE, Munro LR, et al.Immunosuppressive regulatory T cells areabundant in the reactive lymphocytes of Hodgkinlymphoma. Blood. 2004;103(5):1755-1762.

10. Nishimura H, Nose M, Hiai H, Minato N, Honjo T.Development of lupus-like autoimmune diseasesby disruption of the PD-1 gene encoding an ITIMmotif-carrying immunoreceptor. Immunity. 1999;11(2):141-151.

11. Steidl C, Telenius A, Shah SP, et al. Genome-wide copy number analysis of Hodgkin Reed-Sternberg cells identifies recurrent imbalanceswith correlations to treatment outcome. Blood.2010;116(3):418-427.

12. Yamamoto R, Nishikori M, Kitawaki T, et al.PD-1-PD-1 ligand interaction contributes toimmunosuppressive microenvironment ofHodgkin lymphoma. Blood. 2008;111(6):3220-3224.

13. Dorfman DM, Brown JA, Shahsafaei A, FreemanGJ. Programmed death-1 (PD-1) is a markerof germinal center-associated T cells andangioimmunoblastic T-cell lymphoma. Am J SurgPathol. 2006;30(7):802-810.

14. Nam-Cha SH, Roncador G, Sanchez-Verde L,et al. PD-1, a follicular T-cell marker useful forrecognizing nodular lymphocyte-predominantHodgkin lymphoma. Am J Surg Pathol. 2008;32(8):1252-1257.

15. Carbone A, Gloghini A, Cabras A, Elia G.Differentiating germinal center-derivedlymphomas through their cellularmicroenvironment. Am J Hematol. 2009;84(7):435-438.

16. Muenst S, Hoeller S, Dirnhofer S, Tzankov A.Increased programmed death-11 tumor-

infiltrating lymphocytes in classical Hodgkinlymphoma substantiate reduced overall survival.Hum Pathol. 2009;40(12):1715-1722.

17. Andersson E, Dahlenborg K, Ohlin M, BorrebaeckCA, Carlsson R. Immunoglobulin production

induced by CD571 GC-derived helper T cells invitro requires addition of exogenous IL-2. CellImmunol. 1996;169(2):166-173.

18. Zhu JF, Paul WE. CD4 T cells: fates, functions,and faults. Blood. 2008;112(5):1557-1569.

19. Atayar C, van den Berg A, Blokzijl T, et al.Hodgkin’s lymphoma associated T-cells exhibita transcription factor profile consistent withdistinct lymphoid compartments. J Clin Pathol.2007;60(10):1092-1097.

20. Schreck S, Friebel D, Buettner M, et al.Prognostic impact of tumour-infiltrating Th2 andregulatory T cells in classical Hodgkin lymphoma.Hematol Oncol. 2009;27(1):31-39.

21. Gruss HJ, Pinto A, Duyster J, Poppema S,Herrmann F. Hodgkin’s disease: a tumor withdisturbed immunological pathways. ImmunolToday. 1997;18(4):156-163.

22. Frisan T, Sjoberg J, Dolcetti R, et al. Localsuppression of Epstein-Barr virus (EBV)-specificcytotoxicity in biopsies of EBV-positive Hodgkin’sdisease. Blood. 1995;86(4):1493-1501.

23. Serrano D, Ghiotto F, Roncella S, et al. Thepatterns of IL2, IFN-gamma, IL4 and IL5 geneexpression in Hodgkin’s disease and reactivelymph nodes are similar. Haematologica. 1997;82(5):542-549.

24. Poppema S, Potters M, Visser L, van den BergAM. Immune escape mechanisms in Hodgkin’sdisease. Ann Oncol. 1998;9(Suppl 5):S21-S24.

25. Sallusto F, Mackay CR, Lanzavecchia A.Selective expression of the eotaxin receptorCCR3 by human T helper 2 cells. Science. 1997;277(5334):2005-2007.

26. Bonecchi R, Bianchi G, Bordignon PP, et al.Differential expression of chemokine receptorsand chemotactic responsiveness of type 1T helper cells (Th1s) and Th2s. J Exp Med. 1998;187(1):129-134.

27. Hedvat CV, Jaffe ES, Qin J, et al. Macrophage-derived chemokine expression in classicalHodgkin’s lymphoma: application of tissuemicroarrays. Mod Pathol. 2001;14(12):1270-1276.

28. Maggio EM, Van Den Berg A, Visser L, et al.Common and differential chemokine expressionpatterns in rs cells of NLP, EBV positive andnegative classical Hodgkin lymphomas. Int JCancer. 2002;99(5):665-672.

29. Ohshima K, Tutiya T, Yamaguchi T, et al.Infiltration of Th1 and Th2 lymphocytes aroundHodgkin and Reed-Sternberg (H&RS) cells inHodgkin disease: relation with expression of CXCand CC chemokines on H&RS cells. Int J Cancer.2002;98(4):567-572.

30. Buri C, Korner M, Scharli P, et al. CC chemokinesand the receptors CCR3 and CCR5 aredifferentially expressed in the nonneoplasticleukocytic infiltrates of Hodgkin disease. Blood.2001;97(6):1543-1548.

31. Kuppers R. The biology of Hodgkin’s lymphoma.Nat Rev Cancer. 2009;9(1):15-27.

32. Hendriks J, Gravestein LA, Tesselaar K, van LierRA, Schumacher TN, Borst J. CD27 is requiredfor generation and long-term maintenance ofT cell immunity. Nat Immunol. 2000;1(5):433-440.

33. Belldegrun A, Muul LM, Rosenberg SA. Interleukin2 expanded tumor-infiltrating lymphocytes inhuman renal cell cancer: isolation, characterization,and antitumor activity. Cancer Res. 1988;48(1):206-214.

34. Greaves P, Clear A, Coutinho R, et al. Expressionof FOXP3, CD68, and CD20 at diagnosis in themicroenvironment of classical Hodgkin lymphoma

is predictive of outcome. J Clin Oncol. 2013;31(2):256-262.

35. Camp RL, Dolled-Filhart M, Rimm DL. X-tile:a new bio-informatics tool for biomarkerassessment and outcome-based cut-pointoptimization. Clin Cancer Res. 2004;10(21):7252-7259.

36. Concato J, Feinstein AR, Holford TR. The risk ofdetermining risk with multivariable models. AnnIntern Med. 1993;118(3):201-210.

37. Leonard WJ, Spolski R. Interleukin-21:a modulator of lymphoid proliferation, apoptosisand differentiation. Nat Rev Immunol. 2005;5(9):688-698.

38. Scheeren FA, Diehl SA, Smit LA, et al. IL-21 isexpressed in Hodgkin lymphoma and activatesSTAT5: evidence that activated STAT5 isrequired for Hodgkin lymphomagenesis. Blood.2008;111(9):4706-4715.

39. Skinnider BF, Elia AJ, Gascoyne RD, et al.Interleukin 13 and interleukin 13 receptor arefrequently expressed by Hodgkin and Reed-Sternberg cells of Hodgkin lymphoma. Blood.2001;97(1):250-255.

40. Barros MH, Vera-Lozada G, Soares FA,Niedobitek G, Hassan R. Tumormicroenvironment composition in pediatricclassical Hodgkin lymphoma is modulated by ageand Epstein-Barr virus infection. Int J Cancer.2012;131(5):1142-1152.

41. Chetaille B, Bertucci F, Finetti P, et al. Molecularprofiling of classical Hodgkin lymphoma tissuesuncovers variations in the tumormicroenvironment and correlations with EBVinfection and outcome. Blood. 2009;113(12):2765-3775.

42. Lazarevic V, Glimcher LH. T-bet in disease. NatImmunol. 2011;12(7):597-606.

43. Wang Y, Su MA, Wan YY. An essential role of thetranscription factor GATA-3 for the function ofregulatory T cells. Immunity. 2011;35(3):337-348.

44. Bauquet AT, Jin H, Paterson AM, et al. Thecostimulatory molecule ICOS regulates theexpression of c-Maf and IL-21 in the developmentof follicular T helper cells and TH-17 cells. NatImmunol. 2009;10(2):167-175.

45. Poppema S. The nature of the lymphocytessurrounding Reed-Sternberg cells innodular lymphocyte predominance and in othertypes of Hodgkin’s disease. Am J Pathol. 1989;135(2):351-357.

46. Morimoto C, Letvin NL, Distaso JA, Aldrich WR,Schlossman SF. The isolation and characterizationof the human suppressor inducer T cell subset.J Immunol. 1985;134(3):1508-1515.

47. Penninger JM, Irie-Sasaki J, Sasaki T, Oliveira-dos-Santos AJ. CD45: new jobs for an oldacquaintance. Nat Immunol. 2001;2(5):389-396.

48. Alvaro T, Lejeune M, Salvado MT, et al. Outcomein Hodgkin’s lymphoma can be predicted fromthe presence of accompanying cytotoxic andregulatory T cells. Clin Cancer Res. 2005;11(4):1467-1473.

49. Green MR, Monti S, Rodig SJ, et al. Integrativeanalysis reveals selective 9p24.1 amplification,increased PD-1 ligand expression, and furtherinduction via JAK2 in nodular sclerosing Hodgkinlymphoma and primary mediastinal large B-celllymphoma. Blood. 2010;116(17):3268-3277.

50. Steidl C, Shah SP, Woolcock BW, et al. MHCclass II transactivator CIITA is a recurrent genefusion partner in lymphoid cancers. Nature. 2011;471(7338):377-381.

BLOOD, 17 OCTOBER 2013 x VOLUME 122, NUMBER 16 HELPER T CELLS IN HODGKIN LYMPHOMA 2863

For personal use only. by RAUL RIBEIRO on November 25, 2013. bloodjournal.hematologylibrary.orgFrom