Molecular pathology of Hodgkin’s lymphoma: prognostic implications

Hodgkin's Lymphoma

Molecular pathology of Hodgkin's lymphoma: prognostic implications

R.D. Gascoyne, D.W. Scott, C. Steidl

British Columbia Cancer Agency and the Centre for Lymphoid Cancer Vancouver, BC, Canada

Hematology Education: the education program for the annual congress of the European Hematology Association

2012;6:165-174

ABSTRACT

Classical Hodgkin Lymphoma (CHL) represents the most common subtype of malignant lymphoma in young people in the Western world. Most patients can be cured with modern treatment strategies, whereas about 10-15% will die following relapse or progressive disease. CHL is unique amongst all cancers because the malignant Hodgkin Reed?Sternberg (HRS) cells make up only about 1% of the tumor. The HRS cells appear to manipulate the microenvironment, permitting them to develop their malignant phenotype fully and to evade host immune attack. A number of phenotypic characteristics of the HRS cells are associated with prognosis, including expression of BCL2 protein and loss of HLAclass II expression. The non-neoplastic cells in the tumor microenvironment also contribute to prognosis, as gene expression signatures derived from non-neoplastic cells correlate well with response to initial and subsequent therapies, reflecting their functional relevance. Recent biomarker studies have added texture to clinical outcome predictors, and their incorporation in prognostic models may improve our ability to predict treatment failure. Macrophages have been repeatedly shown to be of prognostic relevance in CHL, and very recent low-density, gene expression outcome predictors are poised to penetrate routine clinical practice as a practical tool for up-front risk stratification.

Introduction

Hodgkin Lymphoma (HL) accounts for about 12% of all malignant lymphomas. The key morphologic features of this unique lymphoma were initially described over 100 years ago, characterized by the presence of the Hodgkin Reed-Sternberg (HRS) cells in Classical Hodgkin Lymphoma (CHL) and so called lymphocyte predominant (LP, previously called popcorn or L&H cells) cells in nodular lymphocyte-predominant HL (NLPHL).1 Typically, the malignant cells are greatly outnumbered by the reactive cells in a microenvironment that includes lymphocytes, macrophages, eosinophils, mast cells, plasma cells derived from hematopoietic elements, and stromal cells, including fibroblasts, endothelial cells, and fibroblastic reticular cells.2 Specifically in CHL, the frequencies of all these cellular components, including the HRS cells vary considerably between the CHL subtypes, lymphocyte-rich (LR), lymphocyte depleted (LD), mixed cellularity (MC), and nodular sclerosis (NS). NLPHL will not be further discussed in this review.

Although progress in understanding the fundamental biology of CHL was slow to emerge, in part because of the unique biology, major advances in the treatment of the disease with the introduction of polychemotherapy and radiation therapy have been achieved, such that the number of patients who succumb to CHL has been reduced by 60% since the early 1970s.3 Since its introduction in 1998, the International Prognostic Score (IPS)4 has become the gold

standard for risk assessment and stratification and is used to guide treatment decisions in advanced-stage CHL. Recent data suggest that the IPS may be losing some of its predictive power.5,6 This tool was developed using clinical cohorts, largely treated with ABVD and ABVD-like regimens, assembled during the late 1980s, optimized to predict freedom from disease progression as the endpoint. Today, with more of a focus on overall survival (OS) and the total treatment package, including treatment at relapse, biologically based outcome predictors are desperately needed in order to inform on up-front treatment decisions. Debate in the clinical literature continues between those who favor ABVD versus those who favor BEACOPP or escalated BEACOPP.7 Additionally, the precise role of radiotherapy in CHL remains ill defined. Although strong arguments can be made that BEACOPP and its variations likely produce more up-front cures in CHL compared with ABVD, the question remains at what cost? Both short-term toxicities and long-term sequelae that reduce OS and detract from quality of life are at the heart of the debate. Proponents of ABVD would argue that salvage therapy given to those patients whose first-line therapy fails can lead to cures in a substantial proportion of patients such that equivalent OS is achieved compared with BEACOPP.8 The introduction of novel agents, such as SGN-35, into the upfront setting will only further fuel the debate of optimal initial therapy for advanced-stage CHL. Thus, the field is ripe for new biological outcome predictors that could be used to

Hematology Education: the education programme for the annual congress of the European Hematology Association | 2012; 6(1) | 165 |

17th Congress of the European Hematology Association

risk-stratify patients prior to initial therapy, with the hope that such strategies might significantly reduce the 1015% of patients destined to die following relapse or progressive disease, while minimizing toxicity.9

This review focuses on the contribution of the HRS cells and the tumor microenvironment to prognosis in CHL. Serum-based biomarkers and implications for therapy will not be discussed. The intention is to provide an overview of current biomarkers in the field and lay the groundwork for future progress in this important area of clinical research.

The biology of Hodgkin Reed-Sternberg cells

Thomas Hodgkin described Hodgkin lymphoma (HL, originally called Hodgkin's disease) in 1832 in a small series of cases collected at Guy's Hospital in London. Sir Samuel Wilkes is credited with naming the disease after Thomas Hodgkin. Dorothy Reed and Carl Sternberg first described the malignant cells in CHL, the HRS cells and mononuclear variants. An elegant and comprehensive description of the biology of HL has recently been published.10 In the past 15 years, the detailed biological understanding of these enigmatic cells has been elucidated, in large part the result of two major advances: i) the development of technology that has allowed the study of these infrequent cells in clinical samples (laser-capture microdissection, LCM), and ii) the development of HL cell lines. LCM techniques emerged in the 1990s that facilitated isolation of sufficient numbers of HRS cells from clinical biopsy samples to allow their characterization at a genetic level. HL cell lines have recently been studied at basepair resolution and this analysis revealed a multitude of biological insights.11 Moreover, HL cell lines have allowed functional analyses to be performed.

HRS cells represent B cells that have undergone clonal rearrangement of their immunoglobulin heavy chain (IGH) genes, but fail to express surface IG and most of the usual genes associated with the B cell differentiation program.12 The molecular anatomy of HRS cells suggests that they arise in the germinal centre. Despite lacking the surface B cell receptor (IG) and harboring disadvantageous IGH mutations, they do not undergo apoptosis.13 The expression of a number of lineage-inappropriate genes has been characterized, and the mechanisms underlying this aberrant expression profile are beginning to emerge. HRS cells reveal perturbed signaling pathways, including constitutive NF-B signaling, JAK-STAT signaling, and abnormal NOTCH signaling among others.14 Recurrent mutations in key genes and abnormal copynumber changes help to explain some of the altered gene expression, as do genome-wide changes in methylation status15-18 HRS cells produce and release a vast array of cytokines and chemokines that cultivate their microenvironment.19 Together with characteristic receptor-ligand expression and recently described novel gene fusions, tissues involved by CHL represent a paradigm of cancer cell-induced immune privilege (see below). The EpsteinBarr virus (EBV) also plays a role in the pathogenesis of some cases of CHL, but is not described other than implications for prognosis (see below).

The biology of the microenvironment in CHL

CHL is unique amongst cancers because of the paucity of neoplastic HRS cells and the relative abundance of the inflammatory microenvironment. The most widely held theory proposes that genetic alterations harbored by the HRS cells are in large part responsible for the composition and function of the microenvironment, although the role of constitutional (host) genetics remains largely unexplored.20 Cytokines and chemokines are low molecular weight proteins with a wide variety of functions that work either in a paracrine manner to modulate the activity of surrounding cells or in an autocrine fashion to affect the cells that produce them directly. These molecules represent the primary language for the crosstalk that takes place between HRS cells and the wide array of immune and stromal cells in CHL tumors. Specifically, many of these molecules and their functional interaction networks underlie the specific cellular composition found in CHL and contribute to the proliferative advantage and anti-apoptotic phenotype of HRS cells (Figure 1). Many of the infiltrating immune cells also secrete both chemokines and cytokines that further amplify the inflammatory reaction.

HRS cells also foster immune privilege at sites of involvement and use a number of clever ways to do so. In aggregate, two major strategies are used by HRS cells: i) downregulation of key molecules that are required for immune cell recognition and, ii) they express on their surface and secrete factors that attract immune cells that foster a pro-tumoral immune microenvironment. Loss of both

Figure 1. The Hodgkin Reed-Sternberg cell and the tumor microenvironment. This schematic shows that major secreted and surface molecules of the HRS cells that attract and alter the function of cells in the tumor microenvironment. Arrows indicate interactions that result in increased cell number while lines ending in bars indicate interactions that reduce cell function. Arrows marked with a cross show a loss of normal stimulation.

| 166 | Hematology Education: the education programme for the annual congress of the European Hematology Association | 2012; 6(1)

Amsterdam, The Netherlands, June 14-17, 2012

HLA class I and II molecules is well described in CHL, although the mechanisms are poorly understood.21,22 These HLA antigens are required for recognition and activation of both cytotoxic T cells (CTLs, MHC class I) and helper CD4+ T cells (MHC class II), allowing escape from immunosurveillance when these antigens are not expressed. Additionally, expression of the non-classical MHC class I gene HLA-G might also foster immune privilege by allowing escape from NK cell and CTL killing.23

Most tumors in CHL reveal a Th2 environment with an abundance of regulatory T cells (Tregs) that are typically immunosuppressive.24 HRS cells secrete several chemokines (e.g., CCL5, CCL17/TARC, and CCL22/ MDC) that attract and retain Th2 and Treg cells.25-27 HRS cells also produce galactin-1 that promotes Treg infiltration and reduces CTL infiltration.28 HRS cells express PD-L1 (CD274) and to a lesser extent, PD-L2 (CD273), predominantly as a result of copy number gains,29 but these may also be overexpressed because of translocations.11 These ligands interact with the receptor molecule PD-1 on cytotoxic T cells, a member of the CD28 costimulatory receptor superfamily, resulting in T cell exhaustion and thus further promoting immune privilege.30 Prostaglandin E2 (PGE2) has recently been implicated in NSHL as another immunosuppressive molecule that decreases CD4+ T cell activation.31 Both the enriched Treg cells and the HRS also secrete IL10 and TGF, both cytokines that suppress CTL function.24,32,33 In aggregate, the HRS cells in CHL utilize a number of mechanisms to foster immune privilege that include downregulating surface receptors that prevent detection by immune cells and by cultivating a pro-tumoral microenvironment rich in immunosuppressive cells.

Prognostic biomarkers in classical Hodgkin Lymphoma

As the impact of the microenvironment on the specific pathobiology of HL becomes increasingly clear, more work is focused on correlation of biomarkers with treatment outcome. Although treatment regimens have been remarkably stable in the last two decades, specific aspects of patient selection, varying histologies and treatments might influence cross-comparability and in part explain differing conclusions from these studies. In general, although a plethora of biomarkers associated with clinical outcome have been described, to date none of these factors has penetrated clinical practice. The main reason for this lack of clinical translation lies with the lack of reproducibility between independent patient cohorts and prognostic implications that are not of significant magnitude to justify a change in clinical management. For individual biomarkers, reproducibility of immunohistochemistry (IHC) scoring has been suggested as a reason for inconclusive results and thus more robust multigene predictors based on expression profiling have been reported.34-36 Another barrier to widespread clinical implementation is that there has been no demonstration that any biomarker is predictive, identifying risk that can be overcome by altering the management. Table 1 summarizes some of the biomarkers for which outcome correlations have been described, separated into studies focused on the HRS cells versus those involving the microenvironment and includes IHC and gene expression profiling.

The HRS cells Phenotypic features of the HRS cells

A large number of studies have examined the expression of immunophenotypic markers by the HRS cells in CHL that have been linked to outcome. Perhaps the most convincing of these is the expression of BCL2 protein, where several different investigators have shown that cases with HRS cells that express BCL2 have inferior outcomes37,38 (Figure 2). This finding appears independent of the clinical IPS factors, but may not be applicable to pediatric cases of CHL. Downregulation of both class I and class II major histocompatibility antigens (HLA-class I and HLA-DR) have been described in CHL as a mechanism for HRS cells to escape immunosurveillance.39 More specifically, loss of HLA-DR expression in CHL has been shown to occur in approximately 30% of cases and is associated with inferior survival.22 The expression of CD20 by HRS cells occurs in roughly 20-40% of CHL cases and characteristically shows a variable expression pattern with some cells positive and others entirely negative in the same biopsy. It has been described to be both favorable and adverse, making interpretation of its role in CHL uncertain.40,41 The role of CD20-positive small B cell infiltrates in the microenvironment of CHL is completely unrelated to CD20 expression by HRS cells and is discussed below.

Similarly, MAL expression has been described to be associated with inferior survival, but these findings have not been validated.42 The expression of the germinal centre-associated protein HGAL has been shown to correlate with improved survival, and although it was not significant in multivariate analysis in the initial study, was shown to independently correlate with improved FFS in a subsequent study.43,44 Moreover, expression of topoisomerase II alpha (TOP2A) in HRS cells was repeatedly found to be an independent predictor of FFS.36,45 In 2010, Steidl and colleagues described a study using high-resolution array comparative genomic hybridization of microdissected HRS cells in a series of CHL cases

Figure 2. Biomarkers associated with adverse outcome measured in diagnostic formalin-fixed paraffin-embedded biopsies. IHC - immunohistochemistry; FISH - fluorescent in situ hybridization; RNA ISH - RNA in situ hybridization; qRTPCR - quantitative reverse-transcriptase polymerase chain reaction.

Hematology Education: the education programme for the annual congress of the European Hematology Association | 2012; 6(1) | 167 |

17th Congress of the European Hematology Association

Table 1. Biomarkers with outcome correlations described in the literature.

Marker

HRS cells Characteristic/Mechanism

EBV

EBER positivity in patients 50 years old

BCL2

Expression by HRS cells confers apoptosis resistance

TOPO IIa

Expression by HRS cells

HGAL

Expression by HRS cells

HLA class I and HLA-DR

Loss of MHC class I & II expression allows immune escape

c-MET

c-MET expression by HRS cells

ABCC1 gene

Copy number gains of 16p may predict multidrug resistance

CSF1R

Aberrant expression by ISH in HRS cells is additive with CD68 expression by macrophages

Marker

Microenvironment Characteristic/Mechanism

GrB/TIA-1

Decreased CTLs based on IHC

FOXP3

Decreased Treg cells based on IHC

FOXP3/TIA-1 ratio

Decreased ratio of Treg cells to CTLs using IHC

CD20/BCL11A

Decreased benign B cells based on GE and IHC

CD21

Disruption of FDC meshworks

Galectin-1

Increased expression of galectin-1 by macrophages and endothelial cells

CD68, CD163, LYZ, STAT1

Increased macrophages +/- myeloid suppressor cells based on IHC

Macrophage signature ALDH1A1, LYZ, STAT1, ITGA4, CCL13, MS4A4A, CCL23, VCAN, HSP90AB3P, HSP90AB1, CTSB, CFL1, JMJD6, MAPK7, IKBKG, RAB7A, RXRA, MAPK13

Angiogenic signature

ADH1B, CD93, SRPX, PLA2G2A, GPR126

Adipocyte signature

GLUL, MGST1, COL1A2, FABP4

Fibroblast function/ Extracellular matrix remodeling

MMP2, MMP3, TIMP1, COL1A1, COL4A1, COL4A2, COL5A1, COL18A1, COL16A1, MFAP2, THBS1/2, FN1, EDNRA, ITGB5, LAMA4 (adverse); TIMP4, SPON1, LAMB1, TACR1, CCL26 (favorable)

B cell signature

BCL11A, BANK1, STAP1, BLNK, FCER2, CD24, CCL21

Cytotoxic T cell signature

CD3D, CD8B1, CTSL, CD26,SH2D1A, IFI16, RGS13, CR2, ELL3, CCDC23, PPM1L, TRA@, PIK3CA

Plasmacytoid dendritic cells

ITM2A, SRPX, CTSB, APP

Outcome correlation

Ref.

Adverse (FFS )

70

Adverse (FFS)

37,38

Adverse (FFS, PFS, OS)

36,45

Favorable (EFS, OS)

43,44

Adverse ( FFS, RS )

22

Favorable (FFP)

49

Adverse (DSS)

46

Adverse (PFS, OS)

48

Outcome correlation

Ref.

Adverse (PFS)

86

Adverse (EFS, DFS)

88

Adverse (FFS)

89

Adverse (EFS, OS)

36

Adverse (OS)

94

Adverse (EFS, OS)

93

Adverse (DSS, OS, FFS)

35,78,80

Adverse (OS, PFS, DSS)

34,35,95

Adverse (primary treatment failure) 35

Adverse (primary treatment failure) 35

Adverse/Favorable

95,96

(primary treatment outcome)

Favorable (primary treatment outcome) 35,36

Adverse (primary treatment outcome ) 34,35,95

Adverse (primary treatment outcome) 35

enriched for primary treatment failure.46 They were able to show that cases with gains of chromosome 16p involving the locus for the ABCC1 gene were associated with inferior survival. Importantly, cases with primary progressive disease harbored the highest frequency of gains, implicating a prominent role for multidrug resistance underlying treatment failure in CHL. Two recent discoveries have also led to the identification of possible prognostic biomarkers expressed by HRS cells. HRS cells characteristically reveal lineage-inappropriate gene expression, including, for example CD15, a granulocyte

marker and CSF1R, a marker of macrophages.47 Steidl et al. recently described a gene expression study of microdissected HRS cells in which they showed that CSF1R expression, detected by RNA in situ hybridization, was associated with treatment failure (described further below).48 Visser and colleagues studied c-MET expression in CHL and showed that tumor cells were positive in 52% of cases. They could show that expression of c-MET correlated with improved survival and was an independent prognostic variable for freedom from progression.49 Lastly, EBV infection has been associated with

| 168 | Hematology Education: the education programme for the annual congress of the European Hematology Association | 2012; 6(1)

Amsterdam, The Netherlands, June 14-17, 2012

prognosis in CHL and will be described separately below.

EBV-association EBV infection of HRS cells occurs in up to 60% of

patients in some studies, but varies with geographical location, age, gender, clinical stage, and histology type.50 The impact of EBV infection on outcome is controversial, with some studies showing no impact,51-55 favorable,56-64 or unfavorable63,65-70 outcomes. These discrepancies might inpart be due to patient selection and confounding differences in the demographic factors, as there is an increasing body of work supporting an association between EBV positivity and adverse outcome in older adult patients (age 45 years)63,68,70 and a favorable effect in younger and specifically pediatric patients.59,63,64 The impact of age has hampered the easy translation of EBV status as a useful prognostic biomarker in CHL.

The microenvironment

Histopathology Routine histopathological studies represent the initial

attempts at correlative science in CHL. The separation of CHL into histologic subtypes was thought to impart survival information, particularly for lymphocyte-depleted and grade II NSHL. In the current era of therapy, these distinctions do not identify prognostic groups and thus the separation of CHL from nodular lymphocyte predominant HL is all that is deemed clinically relevant. Tissue eosinophilia and mast cell numbers were previously shown to have prognostic impact in CHL. Increased eosinophils had been linked to improved freedom from treatment failure and improved OS,71 but other studies have not confirmed this finding,72,73 In contrast, increased mast cell infiltration was associated with poor prognosis, but these data have not been validated.74 Importantly, issues related to establishing reproducible thresholds for these variables would likely preclude their use in routine practice.

Macrophages In 1985, Ree and colleagues showed that increased

numbers of tissue macrophages, as assessed by measuring peanut-agglutinin (PNA)-binding cells, was associated with more frequent B symptoms and relapse.75 This association is now well established, with 15 of 17 reported studies showing that increased macrophages in the diagnostic biopsies of patients with CHL is associated with inferior survival (Table 2). One recent study of 265 patients could not establish a prognostic role,76 while a study using gene expression found two macrophage genes, LYZ and STAT1, to be associated with favorable outcome.77 This same group had published twice previously that these same two genes were associated with adverse outcomes, a finding that remains unexplained. Moreover, a very recent paper from these authors describes inferior DSS associated with increased CD68+ cells (Table 2).78

Steidl and colleagues used Affymetrix gene expression profiling of 130 cases of CHL, including 92 patients with sustained remissions following primary therapy and 38 patients in which primary treatment failed.35 Several gene signatures related to monocytes/macrophages were associated with treatment failure. These authors then used immunohistochemistry to study an independent valida-

tion cohort of 166 patients using a tissue microarray with core sizes adequate for assessing the tumor microenvironment. They found that increased CD68+ cells (likely including tissue macrophages and possibly myeloidderived suppressor cells) was an independent predictor of inferior disease-specific survival. Subsequently, a large number of other groups have used similar IHC strategies for CD68 and/or CD163 and found similar results.79-83 These data in aggregate firmly establish that tumor-associated macrophages in CHL are associated with inferior outcome in the era of ABVD or ABVD-like chemotherapy regimens. Recently, Tan et al. used image analysis (Aperio) to measure macrophage content in diagnostic biopsies of CHL from a multicentre randomized phase III clinical trial comparing ABVD versus Stanford V (E2496 Intergoup trial).84 They were able to successfully interpret stains for CD68 and CD163 in 290 patients on this trial. By separating the cases into equal cohorts for training and validation, they developed a threshold in the training set that was carried forward into the validation set, and they could show that both markers were predictive of both FFS and OS, independent of the IPS. Kamper et al. demonstrated that latent EBV infection of the HRS cells was correlated with an increased percentage of CD68+ cells.80 The geographical differences in prevalence of EBV positive HRS cells along with heterogeneity of methods by which CD68+ cells were quantified may explain why there has been no consensus reached about the threshold at which the percentage of CD68+ cells impart poor prognosis. Lastly, Steidl et al. studied lineage-inappropriate gene expression that is characteristic of HRS cells in CHL.48 They found that microdissected HRS cells express both CSF1 and CSF1R, two related receptor-ligand macrophage genes, and that the expression of CSF1R when combined with CD68, was a powerful predictor of inferior survival in CHL. These findings further support the role of macrophages in CHL biology and suggest that both autocrine and paracrine mechanisms are utilized by HRS cells to sculpt their microenvironment and foster immune privilege.

In the 2010 report by Steidl and colleagues, an important finding related to salvage therapy was reported. They were able to show that the macrophage content in the diagnostic biopsy was also predictive of the results of salvage therapies, more specifically autologous bone marrow transplant.35 These data need to be confirmed by others and require validation using relapse biopsy material, but do suggest that important biology present at the time of diagnosis can predict the response to secondary therapies.

Macrophage polarization in the mouse is well characterized and includes so-called classically activated (M1) and alternatively activated (M2) subtypes. This dichotomy is based on functional and gene expression data generated in vitro using macrophages generated following cytokine stimulation of monocytes. Elegant work by Stables and colleagues has recently shown that in situ macrophage polarity is not so well-defined, with resolution-phase macrophages showing a hybrid phenotype with gene expression intermediate between M1 and M2 types.85 These new data provide some insight into macrophage biology and may in part explain some findings in CHL. Several recent IHC studies have used CD163 as a surrogate M2 marker, suggesting that an

Hematology Education: the education programme for the annual congress of the European Hematology Association | 2012; 6(1) | 169 |

17th Congress of the European Hematology Association

Table 2. Studies on the prognostic value of tumor-associated macrophages in CHL.

Markers used

Method

#

Outcome correlation

Reference

PNA

Histochemistry

43

Adverse (refractory disease, early relapse)

75

STAT1, ALDH1A1

GE, IHC

235

Adverse (DSS)

95

LYZ, STAT1, ALDH1A1

GE, IHC

194

Adverse (refractory disease, early relapse)

34

CD68

IHC

166

Adverse (PFS, DSS)

35

LYZ, STAT1

GE

262

Favorable (FFS)

77

CD68, CD163

IHC

288

Adverse (EFS, OS)

80

CD68

IHC

59

Adverse (refractory disease)

97

CD68 (also in combination with FOXP3)

IHC

122

Adverse (FFTF, OS)

98

CD68

IHC

105

Adverse (OS)

79

CD68

IHC

45

Adverse (PFS)

83

CD68

IHC

153

Adverse (OS, PFS)

99

CD68, CD163

IHC (double staining)

82

Adverse (OS)

100

CD68

IHC

52

Adverse (OS)

82

CD68, CD163

IHC

265

No survival impact

76

CD68, CD163

IHC

144

Adverse (OS)

81

CD68, CD163

IHC/Image analysis

287

Adverse (OS, FFS)

84

CD68, LYZ

IHC/Image analysis

266

Adverse (DSS)

78

PNA=peanut agglutinin, GE=gene expression (mRNA), IHC=immunohistochemistry, DSS=disease-specific survival, OS=overall survival, PFS=progression-free survival, FFS=failure-free survival, EFS=event-free survival, FFTF=freedom from treatment failure (for definition, see primary reference).

immunosuppressive role for macrophages in CHL must be mirrored by infiltration of involved tissues by M2 macrophages known to induce a pro-tumoral immunity. The questionable reliability of CD163 as a pure M2 marker and the recent data from Stables et al. suggest that macrophage polarity and correlates to the biology may require functional data using macrophages isolated from fresh CHL biopsies. A detailed list of the published literature is provided in Table 2.

T cell subsets Given that CTLs and NK cells represent a mechanism

of elimination of neoplastic cells, the finding by Oudejans and coworkers that an increased number of activated CTLs were associated with poor outcomes was surprising. Using IHC for granzyme B (GrB), they showed that patients, where the biopsy had more than 15% GrB+ cells, had significantly reduced PFS and OS.86 This was validated in a subsequent study by the same authors, where GrB was found to correlate with OS, independent of both age and stage.87 Regarding CTLs, several independent groups have confirmed these findings, although in some cases, TIA-1, another cytotoxic granule marker, has been shown to be prognostic but not GrB.88

The number of FOXP3 + Treg cells also correlates with

outcome, with increased numbers associated with prolonged event- and disease-free survival.88,89 Although this is contrary to findings in various solid tumors, this positive association with outcomes has been subsequently confirmed. The combination of markers as ratios, FOXP3+/TIA-1+ T88 cells, and FOXP3+/GrB+ cells,89 has been shown to be associated with improved event-free, disease-free and OS, and failure-free and OS, respectively. The most informative T cell marker combination for IHC is still uncertain.

The reasons for the unexpected link between reduced CTLs, increased FOXP3+ Treg cells, and outcome remain unclear. Specifically, the lack of HLA-class II cell-surface expression on HRS cells was frequently observed and is associated with adverse outcome.22 However, loss of HLA class II likely leads to a reduced cytotoxic antitumour response due to diminished activation of Th1 cells. It has been suggested that the prognostically favorable low number of CTLs was linked to overexpression of antiapoptotic proteins in HRS cells.90

Background B cells

Recently, two studies, two studies using gene expression profiling have identified that the number of benign B cells are significantly associated with outcome.35,36 They

| 170 | Hematology Education: the education programme for the annual congress of the European Hematology Association | 2012; 6(1)

Amsterdam, The Netherlands, June 14-17, 2012

showed that CD20 IHC significant correlations with OS and disease-specific survival, respectively, with one study showing that the impact on OS was independent of clinical parameters. Another B cell marker, BCL11A, which detects plasmacytoid dendritic cells, was an even better predictor for OS.36

Follicular dendritic cells

Finally, the number of follicular dendritic cells (FDCs) and the destruction of the normal nodal architecture have been associated with treatment outcomes in CHL.91 The absence of FDCs has been associated with unfavorable outcome and in a subsequent study, the significance of patterns of CD21+ FDC staining was examined. This study confirmed the poor prognosis of the absence of FDCs, with immediate prognosis with extensively disrupted FDC meshworks and best outcomes with welldefined follicle-like structures.

The whole tissue biopsy

Microarray based "genome-wide" expression profiling studies have been performed on whole tissue biopsies providing gene signatures that have been reported to correlate with prognosis. In the whole tissue studies, the signatures largely reflect the composition of the microenvironment related to the small representation by malignant cells. The requirement for fresh frozen material for these technologies limits their ability to penetrate clinical practice. However, many of the signatures have been validated by IHC, namely increased macrophages and CTLs and decreased Treg and B cells, as described above (Table 1).

Two low-density gene expression studies using formalin-fixed paraffin-embedded tissue (FFPET) biopsies have been reported in CHL. A TaqMan strategy based on 52 samples established a short list of 30 genes predictive of outcome in CHL.34 Using a model consisting of a pared-down list of 11 genes, a follow-up study of 262 patients, treated with ABVD or ABVD-like regimens, showed significant differences in FFS, but the predictor failed to validate for OS.77 More recently, a predictor model for OS has been described that uses NanoString technology, which allowed the digital quantitation of the 259 mRNA species of interest in the same assay. The 23 gene model, trained on 290 samples from the phase III randomized clinical trial (Intergroup trial E2496) led by ECOG, was validated in an independent cohort of 78 patients uniformed treated with ABVD.92

Finally, Kamper et al. have recently used proteomic analysis to correlate protein expression with outcome. They demonstrated that the expression of Galectin-1 in the microenvironment was associated with poor outcome, while expression in the HRS cells themselves was not.93

Conclusions

In summary, a number of biomarkers representing cellular components of the microenvironment in CHL have been shown to have a role in predicting prognosis. These include macrophages, T cells, and reactive B cells to name but a few. Although an assessment of CD68 has been well validated in a number of studies, it remains questionable whether reliable thresholds and reproducible IHC scoring will allow this approach to translate into the

clinic. The latest data emerging from robust low-density

gene expression platforms, such as NanoString, hold

promise to penetrate clinical practice and might, at last,

represent the implementation of the long-await biological

outcome predictor needed to allow personalized treatment

decisions to be made for all patients with advanced-stage

CHL.

References

1. Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H, et al., eds. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France: IARC 2008.

2. Pileri SA, Ascani S, Leoncini L, Sabattini E, Zinzani PL, Piccaluga PP, et al. Hodgkin's lymphoma: the pathologist's viewpoint. J Clin Pathol 2002;55(3):162-76.

3. Diehl V, Stein H, Hummel M, Zollinger R, Connors JM. Hodgkin's lymphoma: biology and treatment strategies for primary, refractory, and relapsed disease. Hematology Am Soc Hematol Educ Program 2003:225-47.

4. Hasenclever D, Diehl V. A prognostic score for advanced Hodgkin's disease. International Prognostic Factors Project on Advanced Hodgkin's Disease. N Engl J Med 1998;339(21): 1506-14.

5. Moccia MA, Donaldson J, Chhanabhai M, Hoskins P, Klasa R, Savage KJ, et al. The International Prognostic Factor Project Score (IPS) in Advanced Stage Hodgkin Lymphoma Has Limited Utility in Patients Treated in the Modern Era. ASH Annual Meeting Abstracts 2009;114(22):1554.

6. Guisado-Vasco P, Arranz-Saez R, Canales M, Canovas A, Garcia-Larana J, Garcia-Sanz R, et al. Stage IV and age over 45 years are the only prognostic factors of the International Prognostic Score for the outcome of advanced Hodgkin lymphoma in the Spanish Hodgkin Lymphoma Study Group series. Leuk Lymphoma 2012.

7. Connors JM. More is not necessarily better when treating Hodgkin's lymphoma. J Clin Oncol 2011;29(32):4215-6.

8. Lim SH, Johnson PW. Chemotherapy: Advanced Hodgkin lymphoma--balancing toxicity and cure. Nat Rev Clin Oncol 2011;8(11):634-6.

9. DeVita VT, Jr., Costa J. Toward a personalized treatment of Hodgkin's disease. N Engl J Med 2010;362(10):942-3.

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

11. Steidl C, Shah SP, Woolcock BW, Rui L, Kawahara M, Farinha P, et al. MHC class II transactivator CIITA is a recurrent gene fusion partner in lymphoid cancers. Nature 2011;471(7338):377-81.

12. Kuppers R, Rajewsky K, Zhao M, Simons G, Laumann R, Fischer R, et al. Hodgkin disease: Hodgkin and ReedSternberg cells picked from histological sections show clonal immunoglobulin gene rearrangements and appear to be derived from B cells at various stages of development. Proc Natl Acad Sci U S A 1994;91(23):10962-6.

13. Kanzler H, Kuppers R, Hansmann ML, Rajewsky K. Hodgkin and Reed-Sternberg cells in Hodgkin's disease represent the outgrowth of a dominant tumor clone derived from (crippled) germinal center B cells. J Exp Med 1996;184(4):1495-505.

14. Kuppers R, Klein U, Schwering I, Distler V, Brauninger A, Cattoretti G, et al. Identification of Hodgkin and ReedSternberg cell-specific genes by gene expression profiling. J Clin Invest 2003;111(4):529-37.

15. Ushmorov A, Leithauser F, Sakk O, Weinhausel A, Popov SW, Moller P, et al. Epigenetic processes play a major role in Bcell-specific gene silencing in classical Hodgkin lymphoma. Blood 2006;107(6):2493-500.

16. Joos S, Menz CK, Wrobel G, Siebert R, Gesk S, Ohl S, et al. Classical Hodgkin lymphoma is characterized by recurrent copy number gains of the short arm of chromosome 2. Blood 2002;99(4):1381-7.

17. Schmitz R, Hansmann ML, Bohle V, Martin-Subero JI, Hartmann S, Mechtersheimer G, et al. TNFAIP3 (A20) is a tumor suppressor gene in Hodgkin lymphoma and primary mediastinal B cell lymphoma. J Exp Med 2009;206(5):981-9.

18. Cabannes E, Khan G, Aillet F, Jarrett RF, Hay RT. Mutations in the IkBa gene in Hodgkin's disease suggest a tumour suppressor role for IkappaBalpha. Oncogene 1999;18(20):306370.

19. Maggio EM, Van Den Berg A, Visser L, Diepstra A, Kluiver J,

Hematology Education: the education programme for the annual congress of the European Hematology Association | 2012; 6(1) | 171 |

17th Congress of the European Hematology Association

Emmens R, et al. Common and differential chemokine expression patterns in rs cells of NLP, EBV positive and negative classical Hodgkin lymphomas. Int J Cancer 2002;99(5):66572. 20. Steidl C, Connors JM, Gascoyne RD. Molecular pathogenesis of Hodgkin's lymphoma: increasing evidence of the importance of the microenvironment. J Clin Oncol 2011;29(14):1812-26. 21. Oudejans JJ, Jiwa NM, Kummer JA, Horstman A, Vos W, Baak JP, et al. Analysis of major histocompatibility complex class I expression on Reed-Sternberg cells in relation to the cytotoxic T-cell response in Epstein-Barr virus-positive and negative Hodgkin's disease. Blood 1996;87(9):3844-51. 22. Diepstra A, van Imhoff GW, Karim-Kos HE, van den Berg A, te Meerman GJ, Niens M, et al. HLA class II expression by Hodgkin Reed-Sternberg cells is an independent prognostic factor in classical Hodgkin's lymphoma. J Clin Oncol 2007;25(21):3101-8. 23. Diepstra A, Poppema S, Boot M, Visser L, Nolte IM, Niens M, et al. HLA-G protein expression as a potential immune escape mechanism in classical Hodgkin's lymphoma. Tissue Antigens 2008;71(3):219-26. 24. Marshall NA, Christie LE, Munro LR, Culligan DJ, Johnston PW, Barker RN, et al. Immunosuppressive regulatory T cells are abundant in the reactive lymphocytes of Hodgkin lymphoma. Blood 2004;103(5):1755-62. 25. van den Berg A, Visser L, Poppema S. High expression of the CC chemokine TARC in Reed-Sternberg cells. A possible explanation for the characteristic T-cell infiltratein Hodgkin's lymphoma. Am J Pathol 1999;154(6):1685-91. 26. Hedvat CV, Jaffe ES, Qin J, Filippa DA, Cordon-Cardo C, Tosato G, et al. Macrophage-derived chemokine expression in classical Hodgkin's lymphoma: application of tissue microarrays. Mod Pathol 2001;14(12):1270-6. 27. Teruya-Feldstein J, Jaffe ES, Burd PR, Kingma DW, Setsuda JE, Tosato G. Differential chemokine expression in tissues involved by Hodgkin's disease: direct correlation of eotaxin expression and tissue eosinophilia. Blood 1999;93(8):246370. 28. Juszczynski P, Ouyang J, Monti S, Rodig SJ, Takeyama K, Abramson J, et al. The AP1-dependent secretion of galectin-1 by Reed Sternberg cells fosters immune privilege in classical Hodgkin lymphoma. Proc Natl Acad Sci U S A 2007;104(32):13134-9. 29. Green MR, Monti S, Rodig SJ, Juszczynski P, Currie T, O'Donnell E, et al. Integrative analysis reveals selective 9p24.1 amplification, increased PD-1 ligand expression, and further induction via JAK2 in nodular sclerosing Hodgkin lymphoma and primary mediastinal large B-cell lymphoma. Blood 2010;116(17):3268-77. 30. Yamamoto R, Nishikori M, Kitawaki T, Sakai T, Hishizawa M, Tashima M, et al. PD-1-PD-1 ligand interaction contributes to immunosuppressive microenvironment of Hodgkin lymphoma. Blood 2008;111(6):3220-4. 31. Chemnitz JM, Driesen J, Classen S, Riley JL, Debey S, Beyer M, et al. Prostaglandin E2 impairs CD4+ T cell activation by inhibition of lck: implications in Hodgkin's lymphoma. Cancer Res 2006;66(2):1114-22. 32. Herbst H, Foss HD, Samol J, Araujo I, Klotzbach H, Krause H, et al. Frequent expression of interleukin-10 by Epstein-Barr virus-harboring tumor cells of Hodgkin's disease. Blood 1996;87(7):2918-29. 33. Kadin ME, Agnarsson BA, Ellingsworth LR, Newcom SR. Immunohistochemical evidence of a role for transforming growth factor in the pathogenesis of nodular sclerosing Hodgkin's disease. Am J Pathol 1990;136(6):1209-14. 34. Sanchez-Espiridion B, Sanchez-Aguilera A, Montalban C, Martin C, Martinez R, Gonzalez-Carrero J, et al. A TaqMan low-density array to predict outcome in advanced Hodgkin's lymphoma using paraffin-embedded samples. Clin Cancer Res 2009;15(4):1367-75. 35. Steidl C, Lee T, Shah SP, Farinha P, Han G, Nayar T, et al. Tumor-associated macrophages and survival in classic Hodgkin's lymphoma. N Engl J Med 2010;362(10):875-85. 36. Chetaille B, Bertucci F, Finetti P, Esterni B, Stamatoullas A, Picquenot JM, et al. Molecular profiling of classical Hodgkin lymphoma tissues uncovers variations in the tumor microenvironment and correlations with EBV infection and outcome. Blood 2009;113(12):2765-3775. 37. Sup SJ, Alemany CA, Pohlman B, Elson P, Malhi S, Thakkar S, et al. Expression of bcl-2 in classical Hodgkin's lymphoma: an independent predictor of poor outcome. J Clin Oncol 2005;23(16):3773-9. 38. Canioni D, Deau-Fischer B, Taupin P, Ribrag V, Delarue R,

Bosq J, et al. Prognostic significance of new immunohistochemical markers in refractory classical Hodgkin lymphoma: a study of 59 cases. PLoS One 2009;4(7):e6341. 39. Diepstra A, Niens M, te Meerman GJ, Poppema S, van den Berg A. Genetic susceptibility to Hodgkin's lymphoma associated with the human leukocyte antigen region. Eur J Haematol Suppl 2005;(66):34-41. 40. Tzankov A, Krugmann J, Fend F, Fischhofer M, Greil R, Dirnhofer S. Prognostic significance of CD20 expression in classical Hodgkin lymphoma: a clinicopathological study of 119 cases. Clin Cancer Res 2003;9(4):1381-6. 41. Portlock CS, Donnelly GB, Qin J, Straus D, Yahalom J, Zelenetz A, et al. Adverse prognostic significance of CD20 positive Reed-Sternberg cells in classical Hodgkin's disease. Br J Haematol 2004;125(6):701-8. 42. Hsi ED, Sup SJ, Alemany C, Tso E, Skacel M, Elson P, et al. MAL is expressed in a subset of Hodgkin lymphoma and identifies a population of patients with poor prognosis. Am J Clin Pathol 2006;125(5):776-82. 43. Natkunam Y, Lossos IS, Taidi B, Zhao S, Lu X, Ding F, et al. Expression of the human germinal center-associated lymphoma (HGAL) protein, a new marker of germinal center Bcell derivation. Blood 2005;105(10):3979-86. 44. Azambuja D, Lossos IS, Biasoli I, Morais JC, Britto L, Scheliga A, et al. Human germinal center-associated lymphoma protein expression is associated with improved failurefree survival in Brazilian patients with classical Hodgkin lymphoma. Leuk Lymphoma 2009;50(11):1830-6. 45. Doussis-Anagnostopoulou IA, Vassilakopoulos TP, Thymara I, Korkolopoulou P, Angelopoulou MK, Siakantaris MP, et al. Topoisomerase IIalpha expression as an independent prognostic factor in Hodgkin's lymphoma. Clin Cancer Res 2008;14(6):1759-66. 46. Steidl C, Telenius A, Shah SP, Farinha P, Barclay L, Boyle M, et al. Genome-wide copy number analysis of Hodgkin ReedSternberg cells identifies recurrent imbalances with correlations to treatment outcome. Blood 2010;116(3):418-27. 47. Lamprecht B, Walter K, Kreher S, Kumar R, Hummel M, Lenze D, et al. Derepression of an endogenous long terminal repeat activates the CSF1R proto-oncogene in human lymphoma. Nat Med 2010;16(5):571-9. 48. Steidl C, Diepstra A, Lee T, Farinha P, Telenius A, Boyle M, et al. CSF1R Expression of Hodgkin Reed Sternberg Cells Is Associated with the Number of Macrophages in the Tumor Microenvironment and Is Correlated with Treatment Outcome. ASH Annual Meeting Abstracts 2011;118(21):427. 49. Xu C, Plattel W, van den Berg A, Ruther N, Huang X, Wang M, et al. Expression of the c-Met oncogene by tumor cells predicts favorable outcome in classical Hodgkin lymphoma. Haematologica 2011. 50. Glaser SL, Lin RJ, Stewart SL, Ambinder RF, Jarrett RF, Brousset P, et al. Epstein-Barr virus-associated Hodgkin's disease: epidemiologic characteristics in international data. Int J Cancer 1997;70(4):375-82. 51. Asano N, Oshiro A, Matsuo K, Kagami Y, Ishida F, Suzuki R, et al. Prognostic significance of T-cell or cytotoxic molecules phenotype in classical Hodgkin's lymphoma: a clinicopathologic study. J Clin Oncol 2006;24(28):4626-33. 52. Enblad G, Sandvej K, Lennette E, Sundstrom C, Klein G, Glimelius B, et al. Lack of correlation between EBV serology and presence of EBV in the Hodgkin and Reed-Sternberg cells of patients with Hodgkin's disease. Int J Cancer 1997;72(3):394-7. 53. Axdorph U, Porwit-MacDonald A, Sjoberg J, Grimfors G, Ekman M, Wang W, et al. Epstein-Barr virus expression in Hodgkin's disease in relation to patient characteristics, serum factors and blood lymphocyte function. Br J Cancer 1999;81(7):1182-7. 54. Flavell KJ, Billingham LJ, Biddulph JP, Gray L, Flavell JR, Constandinou CM, et al. The effect of Epstein-Barr virus status on outcome in age- and sex-defined subgroups of patients with advanced Hodgkin's disease. Ann Oncol 2003;14(2):28290. 55. Herling M, Rassidakis GZ, Medeiros LJ, Vassilakopoulos TP, Kliche KO, Nadali G, et al. Expression of Epstein-Barr virus latent membrane protein-1 in Hodgkin and Reed-Sternberg cells of classical Hodgkin's lymphoma: associations with presenting features, serum interleukin 10 levels, and clinical outcome. Clin Cancer Res 2003;9(6):2114-20. 56. Morente MM, Piris MA, Abraira V, Acevedo A, Aguilera B, Bellas C, et al. Adverse clinical outcome in Hodgkin's disease is associated with loss of retinoblastoma protein expression, high Ki67 proliferation index, and absence of Epstein-Barr virus-latent membrane protein 1 expression. Blood

| 172 | Hematology Education: the education programme for the annual congress of the European Hematology Association | 2012; 6(1)

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download