• diffuse large B-cell lymphoma;
  • immune-privileged site;
  • genomic aberration;
  • gene expression;
  • microarray;
  • tumour immunology;
  • apoptosis


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Supporting information
  9. References
  10. Supporting Information

Primary diffuse large B-cell lymphomas of different immune-privileged sites (IP-DLBCLs) share many clinical and biological features, such as a relatively poor prognosis, preferential dissemination to other immune-privileged sites, and deletion of the HLA region, which suggests that IP-DLBCL represents a separate entity. To further investigate the nature of IP-DLBCL, we investigated site-specific genomic aberrations in 16 testicular, nine central nervous system (CNS), and 15 nodal DLBCLs using array CGH. We also determined minimal common regions of gain and loss. Using robust algorithms including multiple testing procedures and the ACE-it script, which is specifically designed for this task, the array CGH data were combined with gene expression data to explore pathways deregulated by chromosomal aberrations. Loss of 6p21.32–p25.3, including the HLA genes, was associated with both types of IP-DLBCL, whereas gain of 2p16.1–p25.3 was associated with nodal DLBCL. Gain of 12q15–q21.1 and 12q24.32–q24.33 was associated with CNS DLBCL and gain of 19q13.12–q13.43 with testicular DLBCL. Analysis of candidate genes in site-specific regions and minimal common regions revealed two major groups of genes: one involved in the immune response, including regulation of HLA expression, and the other involved in apoptosis, including the p53 pathway. Many of these genes were also involved in homozygous deletions or high-level gains. The presence of both shared and site-specific aberrations in CNS and testicular DLBCLs underlines the concept of IP-DLBCL but also indicates that IP-DLBCLs of the CNS and testis do not form a single entity. The observed aberrations emphasize the importance of the deregulation of anti-tumour immune response and apoptosis pathways. Copyright © 2008 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Supporting information
  9. References
  10. Supporting Information

Primary diffuse large B-cell lymphoma (DLBCL) represents a heterogeneous group of lymphomas which is reflected by large variations in clinical features, (histo)pathology, genomic aberrations, and gene expression profiles (reviewed in ref 1). Two major subtypes of DLBCL are recognized, associated with differences in prognosis: the activated B-cell-like (ABC) and the germinal centre B-cell-like (GCB) subtypes 2. These subtypes can be distinguished by gene expression profile 3, 4 or immunohistochemistry 5, 6 and are associated with specific genomic aberrations 7–9.

Primary testicular DLBCL and primary DLBCL of the central nervous system (CNS) are part of the immune-privileged site-associated DLBCLs (IP-DLBCLs) and show very distinct and homogeneous characteristics that separate them from nodal (non-IP) DLBCLs. Besides having a different clinical behaviour 10, 11 and being predominantly of the ABC subtype 12–14, we demonstrated a prominent loss of HLA class I and II (gene) expression, often caused by small interstitial deletions of chromosome 6p21.3 15–17. This is associated with down-regulation of many immune-associated genes and diminished infiltration of T cells 17. Deletions at 6p21.3 account for loss of HLA expression in approximately 50% of IP-DLBCLs and in a much smaller number of non-IP DLBCL cases 18, indicating the presence of alternative mechanisms 17, 19.

The shared characteristics between testicular and CNS DLBCLs suggest that these IP-DLBCL subtypes belong to the same homogeneous entity, which is separate from nodal non-IP DLBCLs. So far, no studies have compared genomic aberrations and gene expression between these groups of DLBCLs. To that end, we explored genomic aberrations in testicular, CNS, and nodal DLBCLs. We combined these data with gene expression data obtained from the same cases using the ACE-it algorithm to identify deregulated candidate genes in the aberrant chromosomal regions.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Supporting information
  9. References
  10. Supporting Information

Tissue sample collection and selection

Forty-five frozen samples from 18 primary testicular, nine primary CNS, and 18 primary nodal DLBCLs were collected from tissue banks at the University Medical Center Groningen, Leiden University Medical Center, Josefine Nefkens Institute, Rotterdam; The Netherlands Cancer Institute, Amsterdam, The Netherlands; and the University of Würzburg, Germany. Only samples from patients with Ann Arbor stage I or II were selected. Using the linear predictor score (LPS) method 4 on Affymetrix gene expression data, 17 of 18 testicular, 7 of 9 CNS, and 9 of 18 nodal DLBCLs corresponded to the ABC subtype, while the remaining cases corresponded to the GCB subtype. Some of the cases (all the testicular and four of the nodal DLBCLs) had previously been characterized for ABC/GCB subtype on another microarray platform 12, yielding the same results. Two testicular and three nodal DLBCLs failed in the array CGH analysis, so further analyses were performed on 16 testicular, nine CNS, and 15 nodal DLBCLs. Gene expression array and array CGH data have been submitted to the GEO database under accession number GSE10524. This study was approved by the institutional board of the UMCG and was carried out in accordance with the modified Declaration of Helsinki.

Gene expression microarray hybridization

Total RNA was isolated using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. The samples were processed and hybridized on Affymetrix HG U133 plus 2.0 oligonucleotide arrays (Affymetrix, Santa Clara, CA, USA) following the protocol for eukaryotic samples in the Expression Analysis Technical Manual ( The average probe array signals were scaled to a target signal of 500 using the GeneChip operating software GCOS (Affymetrix). Probe sets with a signal below 50 were discarded, leaving a final data set of 29 381 probe sets.

Array CGH analysis

Genomic DNA was isolated using high salt after overnight SDS/proteinase K digestion and hybridized on an in-house printed CGH array containing ∼3700 large genomic insert clones (provided by Dr N Carter, Welcome Trust Sanger Institute), as previously described 20, 21. Log2 ratio values were analysed using R packages DNACopy 22 and aCGH/MergeLevels 23 to determine regions and levels of gain and loss. Hemi- and homozygous losses were defined as one and two levels lower than normal, respectively; gain was defined as one or two levels higher than normal; and high-level gain as three or more levels higher than normal 23. Chromosomes X and Y were excluded from further analysis, as were heterochromatic regions, centromeres, ribosomal gene clusters, and regions with known segmental duplications (grey areas in Figure 1). Minimal common regions (MCRs) were defined as the smallest regions of overlap that were smaller than 20 Mb and present in at least four cases, of which at least three showed a limited aberration smaller than a whole chromosome or chromosomal arm. Overrepresentation of aberrations in different localizations was assessed using the one-tailed Fisher exact test with a p value ≤0.05.

thumbnail image

Figure 1. Genomic aberrations found in CNS, testicular, and nodal DLBCLs using array CGH analysis. Losses (left) and gains (right) in testicular (orange), CNS (green), and nodal DLBCLs (blue). Homozygous deletions and high-level gains are indicated by thick coloured bars. MCRs are indicated by thick black bars on the left- and rightmost side for each chromosome. Greyed out areas were excluded

Download figure to PowerPoint

Combining expression and array CGH data

ACE-it version 2.1 24 was used to find genes for which expression levels correlate with genomic gain or loss. The process is explained in depth in the ACE-it manual 25. In short, for each lymphoma, a status of gain, normal or loss was assigned (without distinction between hemi- and homozygous loss, or between single-copy and high-level gain) to each base pair position based on the array CGH data. This status was then mapped back to the Affymetrix probe sets based on their chromosomal location. When all the probe sets in all lymphomas had been assigned a status of gain, loss or normal, the correlation of expression levels with a genomic aberration was analysed for each probe set showing an aberration in three or more cases. This was done by performing a non-parametric one-sided Wilcoxon test comparing expression levels of the probe set in samples with and without that aberration. When all the probe sets had been analysed, multiple testing correction was performed according to Benjamini and Hochberg 26. The p-value cut-off used was 0.05. Genes for which expression correlated with gain or loss were annotated through the Entrez ID using DAVID 27, focusing on candidate genes with a known function in oncogenesis, normal B cells or B-cell neoplasms for which loss or gain would be functionally relevant for the tumour cells.

Since ACE-it is a relatively new algorithm, we validated the results with the Bioconductor package multtest in R, using the two-sample unequal variance t-statistic, with FWER-controlling step-down minP multiple testing procedure. To estimate the test statistic null distribution, we used the bootstrap method. The p-value cut-off used was 0.05. The overlap between the significant genes determined with both methods was 94%. Within the chosen candidate genes with a known function in oncogenesis, normal B cells or B-cell neoplasms, the overlap was 100%.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Supporting information
  9. References
  10. Supporting Information

Differences in genomic aberrations between testicular, CNS, and nodal DLBCLs

To investigate specific genomic aberrations in testicular and CNS DLBCLs, we performed array CGH analysis (Figure 1). The average number of aberrations per case was 9.9 in testicular, 8.9 in CNS, and 9.4 in nodal DLBCLs. Homozygous losses were found in 16 regions and high-level gains in six regions. The most common aberrations, present in at least 30% of CNS, testicular or nodal DLBCL cases, are shown in Table 1. Table 2 summarizes the five regions showing significant overrepresentation in one or more localizations.

Table 1. The most common genomic aberrations in CNS, testicular, and nodal DLBCLs
 All cases (n = 40)CNS DLBCL (n = 9)Testicular DLBCL (n = 16)Nodal DLBCL (n = 15)
  1. Aberrations in 30% or more of CNS, testicular or nodal DLBCLs are listed. For each site, the number of cases (and the percentage) in which an aberration occurs is indicated.

 6p21.32–p25.218 (45%)5 (56%)11 (69%)2 (13%)
 6q20 (50%)5 (56%)10 (63%)5 (33%)
 15q12–q21.112 (30%)2 (22%)5 (31%)5 (33%)
 17p12–p13.313 (33%)5 (56%)4 (25%)4 (27%)
 1q21.3–q32.115 (38%)3 (33%)7 (44%)5 (33%)
 7/7q14 (35%)2 (22%)6 (38%)6 (40%)
 129 (23%)4 (44%)3 (19%)2 (13%)
 18p11 (28%)1 (11%)4 (25%)6 (40%)
 18q15 (38%)2 (22%)7 (44%)6 (40%)
 19q13.12–q13.4313 (33%)2 (22%)11 (69%)0 (0%)
Table 2. Site-specific aberrations and candidate genes
 Position (Mb)All cases (n = 40)CNS DLBCL (n = 9)Testicular DLBCL (n = 16)Nodal DLBCL (n = 15)Significantly higher frequency in*:Candidate genes
  • Aberrations that differ in frequency between CNS, testicular and nodal DLBCLs, and the candidate genes located therein. The number of cases (and percentage) in which an aberration occurs is indicated for each site. Mb = megabase.

  • *

    p≤0.05 using the one-tailed Fisher exact test.

  • Determined using ACE-it analysis; p≤0.05 after Benjamini and Hochberg correction. Only genes with a known function in oncogenesis, normal B cells or B-cell neoplasms for which loss or gain would be functionally relevant for the tumour cells are listed.

  • Gene expression in this region has been studied previously 17.

 6p21.32–p25.21–34.91918 (45%)5 (56%)11 (69%)2 (13%)IP-DLBCL
 2p16.1–p25.31–60.5604 (10%)0 (0%)0 (0%)4 (27%)Nodal DLBCLSUPT7L, BIRC6, BRE
 12q15–q21.166.251–71.74310 (25%)5 (56%)3 (19%)2 (13%)CNS DLBCLMDM2, YEATS4
 12q24.32–q24.33125.033–132.45010 (25%)5 (56%)3 (19%)2 (13%)CNS DLBCL
 19q13.12–q13.4337.665–63.81213 (33%)2 (22%)11 (69%)0 (0%)Testicular DLBCLLILRA3, SPIB, BCL2L12, PAK4, PPP5C

To identify candidate genes in these five localization-specific regions, we combined gene expression analysis with array CGH analysis using ACE-it. Significant genes whose expression levels correlated with genomic loss or gain were screened for a known function in oncogenesis, normal B cells or B-cell neoplasms, because deregulation of these genes could be functionally relevant for the tumour cells (Table 2; a full list of all significant genes is available in the Supporting information, Supplementary Table 1). Loss of 6p21.32–p25.2 was specifically observed in both DLBCLs of the testis and those of the CNS, and corroborates the previously described alterations of HLA and other genes in this region 17. Gain of 2p16.1–p25.3 was associated exclusively with nodal DLBCL. This region contained 32 significantly altered genes including candidates SUPT7L and anti-apoptotic BIRC6 and BRE. Gain of both 12q15–q21.1 and 12q24.23–q24.33 was associated with CNS DLBCL. The 12q15–q21.1 region contained eight significantly altered genes including candidates MDM2 and YEATS4, which are closely associated with p53. The 12q24.32–q24.33 region contained ten significantly altered genes but none with a relevant known function. Gain of 19q13.12–q13.43 was associated with testicular DLBCL. This region is very gene-dense and contained 83 significantly altered genes, including the five candidates LILRA3, SPIB, BCL2L12, PAK4, and PPP5C.

Analysis of gene expression in minimal common regions (MCRs)

We extended the combined array CGH/expression analysis to the MCRs, ie recurrent and distinct regions of overlap that were smaller than 20 Mb. In all 40 DLBCLs combined, 30 regions met our criteria for an MCR: 15 with gain and 15 with loss (black bars in Figure 1). Table 3 lists each region together with the number of genes significant in ACE-it analysis and the candidate genes of interest. (Full lists of significantly altered genes and site-specific frequency data are available in the Supporting information, Supplementary Tables 1 and 2, respectively.)

Table 3. MCR regions and candidate genes
CytobandPosition (Mb)Size (Mb)All cases (n = 40) (%)Annotated known genesCandidate genes*
Totalin expression data setcorrelating with gain/loss (%)
  • Regions containing homozygous losses or high-level gains are indicated in bold text. Mb = megabase.

  • *

    Determined using ACE-it analysis; p≤0.05 after Benjamini and Hochberg correction. Only genes with a known function in oncogenesis, normal B cells or B-cell neoplasms for which loss or gain would be functionally relevant for the tumour cells are listed.

  • Gene expression in this region has been studied previously 17.

 1p12–p13.2114.466–118.3013.847 (18)31174 (24) 
 1q32.2–q32.3203.602–208.0894.495 (13)28194 (21) 
 1q42.11–q42.13220.131–225.1955.065 (13)513911 (28)PARP1
 1q42.2–q43227.397–237.2179.825 (13)45339 (27)EGLN1
 2q14.3–q22.2127.254–143.61716.364 (10)633518 (51)ERCC3, MAP3K2
 5q13.1–q13.267.677–71.663.985 (13)30180 (0) 
 6p21.32–p22.129.496–33.4683.9715 (38)154111
 6q16.3–q21100.953–112.4311.4818 (45)523624 (67)REV3L, ATG5
 6q23.3–q24.2137.302–144.4327.1318 (45)261816 (89)HIVEP2, PERP, PLAGL1, TNFAIP3
 8q12.1–q12.258.379–61.8513.478 (20)965 (83) 
 8q13.1–q13.367.344–72.65.266 (15)23153 (20) 
 15q15.3–q21.141.852–45.1353.2812 (30)271910 (53)PDIA3
 17p13.23.399–6.0652.6712 (30)574024 (60)PFN1, ITGAE
 17p12–p13.17.438–12.925.488 (20)613813 (34)TP53, MAP2K4
 19p13.2–p13.35.893–12.0076.115 (13)15610432 (31)SMARCA4
 1q23.2–q23.3156.697–160.9104.2112 (30)695125 (49)CD48, FREB, PEA15
 1q31.1–q31.2185.925–198.88412.9612 (30)513415 (44) 
 1q32.1199.641–202.9733.3313 (33)50329 (28) 
 1q44241.691–245.5233.838 (20)54144 (29) 
 2p15–p16.160.754–62.6541.906 (15)12124 (33) 
 3q13.12–q13.13107.671–111.5763.908 (20)1473 (43)CD47
 3q13.13–q13.2110.272–114.7474.488 (20)27215 (24) 
 3q26.33–q28183.353–189.7326.389 (23)634219 (45)PARL
 9p24.15.121–8.3993.287 (18)20164 (25)PDCD1LG2
 9q34.2–q34.3133.750–138.4294.687 (18)86611 (2) 
 11q22.3–q23.1108.393–112.5714.186 (15)27204 (20) 
 18q11.1–q11.216.100–21.4315.3313 (33)22203 (15) 
 18q21.1–q21.3345.188–59.05913.8715 (38)503419 (56)POLI, BCL2, MALT1
 18q22.3–q2366.930–76.1179.1913 (33)252312 (52) 
 19q13.41–q13.4356.504–63.8127.3113 (33)21713642 (31)LILRA3


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Supporting information
  9. References
  10. Supporting Information

Primary IP-DLBCLs share many features including frequent dissemination to other immune-privileged sites 10, 28–32. This suggests that these lymphomas share important biological features such as similar expression of adhesion molecules, homing receptors or chemokines and their receptors 33–35. We have previously demonstrated that many IP-DLBCLs share prominent down-regulation of HLA class I and class II proteins, caused by small interstitial deletions at chromosome 6p21.3 in approximately 50% of cases 15–17, as a possible mechanism of immune escape. Other IP-DLBCLs may exhibit alternative mechanisms. On the basis of these and other observations, we hypothesized that these tumours readily arise in an immune sanctuary, but due to a shift in the micro-environment from immune-inhibitory towards-immune activating mechanisms, these tumours can only survive by additional immuno-editing such as down-regulation of HLA expression. These observations led us to investigate whether IP-DLBCLs share more functional and genomic features that separate them from nodal non-IP-DLBCLs.

Our approach focused on genomic gains and losses defined by array CGH and the concomitant alterations in gene expression. This approach enriches for gene expression alterations in the lymphoma cells themselves, as opposed to gene expression alterations related to the micro-environment. Five chromosomal regions showed significant differences in copy number between testicular, CNS, and nodal DLBCLs (Table 2). This included loss of 6p21.32–p25.2, which has been found previously in more than 50% of IP-DLBCLs 15–17 and in mediastinal large B-cell lymphomas (PMLBCLs) 36 and causes loss of HLA class I and II expression.

Aberrations that were more frequently found in CNS DLBCL than in other cases were gain of 12q15–q21.1 and 12q24.32–q24.33. We identified two candidate genes in the first region: YEATS4, a repressor of the p53 pathway 37, and MDM2, an inhibitor of p53 via ubiquitin-dependent proteasomal degradation 38, 39. High expression of MDM2 was recently reported to be associated with copy number gains of the MDM2 locus and poor prognosis in mantle cell lymphoma 40 and was also suggested to be a candidate in PMLBCL 36. Weber et al41 described a similar gain of 12q13–14 and 12q24 in CNS DLBCL but they did not study other DLBCLs, nor did they correlate their results with gene expression. They also identified a frequent high-level gain of 18q21; however, in our series, gain of this region, with BCL2 and MALT1 being candidate genes, was not specific for CNS DLBCL (Table 3 and Supporting information, Supplementary Table 2). According to the literature, gain of 18q21 is often associated with ABC-type DLBCL rather than with a specific site of DLBCL 8, 39, 42. High-level gain of 9p23–p24 has also been described in CNS DLBCL 41. We observed this gain in one CNS, four testicular, and two nodal DLBCLs (Supporting information, Supplementary Table 2), again indicating that many of the previously reported aberrations are not site-specific.

An aberration that was more frequently found in testicular DLBCL than in other cases was gain of 19q13.12–q13.43. One array CGH study 8 reported on this gain in approximately 50% of nodal ABC-type DLBCLs, which could not be confirmed in our nodal cases and was rarely identified in other series of B-cell lymphomas analysed by array CGH 36, 43. Eighty-three genes in this region showed higher expression in cases with gain, of which five were interesting candidate genes. LILRA3 is part of the leukocyte immunoglobulin-like receptor (LILR) gene family that contains both activating and inhibitory LILR genes (reviewed in ref 44). LILR3 exists in a soluble form 45, 46. High expression by the lymphoma cells might therefore act as an antagonist (by competing for ligand) for activating LILRs on tumour-infiltrating lymphocytes. Three other candidate genes in this region have a role in inhibition of apoptosis: BCL2L12, a recently discovered member of the BCL2 gene family 47; PAK4, which functions in TNFα-induced pro-survival pathways 48; and PPP5C, which inhibits p53-induced apoptosis as well as the MAP2K4/JNK pathway 49, 50. The fifth potential target gene is SPIB, which is deregulated by chromosomal translocation t(14;19) in the ABC-type DLBCL cell line OCI-Ly3 51 and is a direct target of the transcription factor BOB1 in normal B cells 52. Interestingly, in two of our testicular DLBCLs, SPIB is located at the transition point of normal gain to high-level gain, suggesting that deregulated expression is triggered by a rearrangement.

An aberration that was more frequently found in nodal DLBCL than in other cases was gain of 2p16.1–p25.3. The candidates BIRC6 and BRE are anti-apoptotic genes 53, 54 and SUPT7L (ARTC1) has been reported to be a ligand for anti-tumour regulatory T cells in melanoma 55, overexpression leading to suppression of the local anti-tumour immune response. Two well-known potential targets of 2p amplification in DLBCL, REL and BCL11A56, 57, were not marked as potential targets in our series since REL expression did not correlate with gain, and BCL11A was located outside both the nodal-specific region and the MCR of 2p15–p16.1. Apart from these site-specific chromosomal regions, we identified MCRs and the candidate genes therein. Combining both of these analyses, a large proportion of the candidate genes is involved in either apoptosis or regulation of the immune response/HLA (Table 4).

Table 4. Functionally interesting genes whose expression correlates with gain or loss
  1. Genes involved in homozygous loss or high-level gain are indicated in bold type. The genes are derived from Tables 2 and 3.

ApoptosisImmune responseProliferation
 ATG5and HLA regulationand migration
 MAP2K4 SMARCA4NFκB pathway
 PAK4B-cell signalling 
 PERPSomatic hypermutation 
 TP53 REV3L 

Regarding apoptosis, interesting candidates showing loss are PARP1 (at 1q42.12), which is fundamentally involved in DNA damage control and apoptosis in B-cell lymphomas 58; the pro-apoptotic tumour suppressor gene MAP2K459, 60; and TP53 (both located in the MCR at 17p12–p13.1). Loss of TP53 in DLBCL is associated with an aggressive clinical behaviour 61. Loss of 6q23.3–q24.2 (including homozygous deletion in two IP-DLBCLs) was associated with down-regulation of PERP, a direct effector of apoptosis downstream of TP53. Using SNP analysis, PERP was recently described as a candidate tumour suppressor gene for follicular lymphoma 62. Interestingly, in our series, 27 of 40 cases (8/9 CNS, 12/16 testicular, and 7/15 nodal) showed genomic aberrations affecting the p53 pathway (loss of TP53 or PERP or gain of MDM2), suggesting that this pathway is an important target for deregulation in DLBCL.

Deletion of 6p21.32 is the major mechanism by which HLA class II expression is lost in IP-DLBCL 15–17. However, low expression without deletion is also found in IP-DLBCL and nodal DLBCL 17, 18. Loss of SMARCA4 and disruption of the interferon-gamma pathway might be an alternative mechanism for loss of HLA expression. SMARCA4 is located in the MCR at 19p13.2 and is essential for the interferon-gamma-induced chromatin remodelling of class II transactivator (CIITA), which is the major regulator of HLA class II expression 63. Another factor in the interferon-gamma pathway, IFNGR1, is located in the MCR at 6q23.3–q24 64, but its expression did not correlate with genomic loss in our data. We also observed loss (including homozygous deletion in two cases) of PDIA3 at 15q15.3, which interacts with calnexin/calreticulin and tapasin in the MHC class I peptide loading complex 65, 66.

Apart from down-regulation of HLA, other mechanisms are deployed to escape the anti-tumour immune response. Gain (including high-level gain) of 9p24.1 was associated with higher expression of PDCD1LG2, which is a ligand of the ‘programmed cell death receptor’ on T cells; binding to this receptor inhibits T-cell activation 67. Consistently increased expression of this gene, often caused by gain of 9p24, has been described in PMLBCL 68. Gain of LILRA3 at 19q13, discussed earlier, also belongs in this category.

In conclusion, genomic differences between IP-DLBCL and nodal DLBCL exist, the most prominent being the hemi- and homozygous deletions of 6p21.32 in IP-DLBCL. The presence of different aberrations in CNS and testicular DLBCLs implies that these DLBCLs do not form one homogeneous entity. Using robust ACE-it and multiple testing algorithms that selected for alterations within tumour cells, we found a striking selection for genes involved in apoptosis, including the p53 pathway, as well as for genes that can modulate the anti-tumour immune response. Further studies will be necessary to prove a causal relationship between the genetic alterations, gene expression, and functional alterations of the tumour cells.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Supporting information
  9. References
  10. Supporting Information

We would like to thank Marja van der Burg for hybridizing the CGH arrays. AR and EH are supported by the Interdisciplinary Center for Clinical Research (IZKF).

Supporting information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Supporting information
  9. References
  10. Supporting Information

Supporting information may be found in the online version of this article.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Supporting information
  9. References
  10. Supporting Information
  • 1
    De Paepe P, De Wolf-Peeters C. Diffuse large B-cell lymphoma: a heterogeneous group of non-Hodgkin lymphomas comprising several distinct clinicopathological entities. Leukemia 2007; 21: 3743.
  • 2
    Alizadeh AA, Eisen MB, Davis RE, Ma C, Lossos IS, Rosenwald A, et al. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 2000; 403: 503511.
  • 3
    Rosenwald A, Wright G, Chan WC, Connors JM, Campo E, Fisher RI, et al. The use of molecular profiling to predict survival after chemotherapy for diffuse large-B-cell lymphoma. N Engl J Med 2002; 346: 19371947.
  • 4
    Wright G, Tan B, Rosenwald A, Hurt EH, Wiestner A, Staudt LM. A gene expression-based method to diagnose clinically distinct subgroups of diffuse large B cell lymphoma. Proc Natl Acad Sci U S A 2003; 100: 99919996.
  • 5
    Hans CP, Weisenburger DD, Greiner TC, Gascoyne RD, Delabie J, Ott G, et al. Confirmation of the molecular classification of diffuse large B-cell lymphoma by immunohistochemistry using a tissue microarray. Blood 2004; 103: 275282.
  • 6
    Chang CC, McClintock S, Cleveland RP, Trzpuc T, Vesole DH, Logan B, et al. Immunohistochemical expression patterns of germinal center and activation B-cell markers correlate with prognosis in diffuse large B-cell lymphoma. Am J Surg Pathol 2004; 28: 464470.
  • 7
    Bea S, Zettl A, Wright G, Salaverria I, Jehn P, Moreno V, et al. Diffuse large B-cell lymphoma subgroups have distinct genetic profiles that influence tumor biology and improve gene-expression-based survival prediction. Blood 2005; 106: 31833190.
  • 8
    Tagawa H, Suguro M, Tsuzuki S, Matsuo K, Karnan S, Ohshima K, et al. Comparison of genome profiles for identification of distinct subgroups of diffuse large B-cell lymphoma. Blood 2005; 106: 17701777.
  • 9
    Iqbal J, Sanger WG, Horsman DE, Rosenwald A, Pickering DL, Dave B, et al. BCL2 translocation defines a unique tumor subset within the germinal center B-cell-like diffuse large B-cell lymphoma. Am J Pathol 2004; 165: 159166.
  • 10
    Zucca E, Conconi A, Mughal TI, Sarris AH, Seymour JF, Vitolo U, et al. Patterns of outcome and prognostic factors in primary large-cell lymphoma of the testis in a survey by the International Extranodal Lymphoma Study Group. J Clin Oncol 2003; 21: 2027.
  • 11
    Schlegel U, Schmidt-Wolf IGH, Deckert M. Primary CNS lymphoma: clinical presentation, pathological classification, molecular pathogenesis and treatment. J Neurol Sci 2000; 181: 112.
  • 12
    Booman M, Douwes J, Glas AM, de Jong D, Schuuring E, Kluin PM. Primary testicular diffuse large B cell lymphomas have activated B cell-like subtype characteristics. J Pathol 2006; 210: 163171.
  • 13
    Camilleri-Broet S, Criniere E, Broet P, Delwail V, Mokhtari K, Moreau A, et al. A uniform activated B-cell-like immunophenotype might explain the poor prognosis of primary central nervous system lymphomas: analysis of 83 cases. Blood 2006; 107: 190196.
  • 14
    Lin CH, Kuo KT, Chuang SS, Kuo SH, Chang JH, Chang KC, et al. Comparison of the expression and prognostic significance of differentiation markers between diffuse large B-cell lymphoma of central nervous system origin and peripheral nodal origin. Clin Cancer Res 2006; 12: 11521156.
  • 15
    Jordanova ES, Riemersma SA, Philippo K, Giphart-Gassler M, Schuuring E, Kluin PM. Hemizygous deletions in the HLA region account for loss of heterozygosity in the majority of diffuse large B-cell lymphomas of the testis and the central nervous system. Genes Chromosomes Cancer 2002; 35: 3848.
  • 16
    Riemersma SA, Jordanova ES, Schop RF, Philippo K, Looijenga LH, Schuuring E, et al. Extensive genetic alterations of the HLA region, including homozygous deletions of HLA class II genes in B-cell lymphomas arising in immune-privileged sites. Blood 2000; 96: 35693577.
  • 17
    Booman M, Douwes J, Glas AM, Riemersma SA, Jordanova ES, Kok K, et al. Mechanisms and effects of loss of HLA class II expression in immune privileged site-associated B-cell lymphoma. Clin Cancer Res 2006; 12: 26982705.
  • 18
    Rimsza LM, Roberts RA, Campo E, Grogan TM, Bea S, Salaverria I, et al. Loss of major histocompatibility class II expression in non-immune privileged site diffuse large B cell lymphoma is highly coordinated and not due to chromosomal deletions. Blood 2006; 107: 11011107.
  • 19
    Jordanova ES, Riemersma SA, Philippo K, Schuuring E, Kluin PM. Beta2-microglobulin aberrations in diffuse large B-cell lymphoma of the testis and the central nervous system. Int J Cancer 2003; 103: 393398.
  • 20
    Knijnenburg J, Szuhai K, Giltay J, Molenaar L, Sloos W, Poot M, et al. Insights from genomic microarrays into structural chromosome rearrangements. Am J Med Genet A 2005; 132: 3640.
  • 21
    Fiegler H, Carr P, Douglas EJ, Burford DC, Hunt S, Scott CE, et al. DNA microarrays for comparative genomic hybridization based on DOP-PCR amplification of BAC and PAC clones. Genes Chromosomes Cancer 2003; 36: 361374.
  • 22
    Venkatraman ES, Olshen AB. A faster circular binary segmentation algorithm for the analysis of array CGH data. Bioinformatics 2007; 23: 657663.
  • 23
    Willenbrock H, Fridlyand J. A comparison study: applying segmentation to array CGH data for downstream analyses. Bioinformatics 2005; 21: 40844091.
  • 24
    van Wieringen WN, Belien JAM, Vosse SJ, Achame EM, Ylstra B. ACE-it: a tool for genome-wide integration of gene dosage and RNA expression data. Bioinformatics 2006; 22: 19191920.
  • 25
    Vosse SJ, van Wieringen WN. ACE-it: Array CGH expression integration tool. 2007.
  • 26
    Benjamini Y, Hochberg Y. Controlling the false discovery rate—a practical and powerful approach to multiple testing. J R Stat Soc Ser B—Method 1995; 57: 289300.
  • 27
    Dennis G Jr, Sherman BT, Hosack DA, Yang J, Gao W, Lane HC, et al. DAVID: database for annotation, visualization, and integrated discovery. Genome Biol 2003; 4: R60.
  • 28
    Booman M, Douwes J, Legdeur MC, van Baarlen J, Schuuring E, Kluin P. From brain to testis: immune escape and clonal selection in a B cell lymphoma with selective outgrowth in two immune sanctuaries. Haematologica 2007; 92: e69e71.
  • 29
    Fonseca R, Habermann TM, Colgan JP, O'Neill BP, White WL, Witzig TE, et al. Testicular lymphoma is associated with a high incidence of extranodal recurrence. Cancer 2000; 88: 154161.
  • 30
    Harney J, Pope A, Short SC. Primary central nervous system lymphoma with testicular relapse. Clin Oncol (R Coll Radiol) 2004; 16: 193195.
  • 31
    Rajappa SJ, Uppin SG, Digumarti R. Testicular relapse of primary central nervous system lymphoma. Leuk Lymphoma 2007; 48: 10231025.
  • 32
    Jahnke K, Thiel E, Martus P, Herrlinger U, Weller M, Fischer L, et al. Relapse of primary central nervous system lymphoma: clinical features, outcome and prognostic factors. J Neurooncol 2006; 80: 159165.
  • 33
    Horstmann WG, Timens W. Lack of adhesion molecules in testicular diffuse centroblastic and immunoblastic B cell lymphomas as a contributory factor in malignant behaviour. Virchows Arch 1996; 429: 8390.
  • 34
    Smith JR, Braziel RM, Paoletti S, Lipp M, Uguccioni M, Rosenbaum JT. Expression of B-cell-attracting chemokine 1 (CXCL13) by malignant lymphocytes and vascular endothelium in primary central nervous system lymphoma. Blood 2003; 101: 815821.
  • 35
    Rubenstein J, Fridlyand J, Shen A, Aldape K, Ginzinger D, Batchelor T, et al. Gene expression and angiotropism in primary CNS lymphoma. Blood 2006; 107: 37163723.
  • 36
    Wessendorf S, Barth TF, Viardot A, Mueller A, Kestler HA, Kohlhammer H, et al. Further delineation of chromosomal consensus regions in primary mediastinal B-cell lymphomas: an analysis of 37 tumor samples using high-resolution genomic profiling (array-CGH). Leukemia 2007; 21: 24632469.
  • 37
    Park JH, Roeder RG. GAS41 is required for repression of the p53 tumor suppressor pathway during normal cellular proliferation. Mol Cell Biol 2006; 26: 40064016.
  • 38
    Coutts AS, La Thangue NB. Mdm2 widens its repertoire. Cell Cycle 2007; 6: 827829.
  • 39
    Rao PH, Houldsworth J, Dyomina K, Parsa NZ, Cigudosa JC, Louie DC, et al. Chromosomal and gene amplification in diffuse large B-cell lymphoma. Blood 1998; 92: 234240.
  • 40
    Hartmann E, Fernandez V, Stoecklein H, Hernandez L, Campo E, Rosenwald A. Increased MDM2 expression is associated with inferior survival in mantle-cell lymphoma, but not related to the MDM2 SNP309. Haematologica 2007; 92: 574575.
  • 41
    Weber T, Weber RG, Kaulich K, Actor B, Meyer-Puttlitz B, Lampel S, et al. Characteristic chromosomal imbalances in primary central nervous system lymphomas of the diffuse large B-cell type. Brain Pathol 2000; 10: 7384.
  • 42
    Monni O, Joensuu H, Franssila K, Klefstrom J, Alitalo K, Knuutila S. BCL2 overexpression associated with chromosomal amplification in diffuse large B-cell lymphoma. Blood 1997; 90: 11681174.
  • 43
    Baudis M. Progenetix—database of cytogenetic abnormalities in cancer. [October 2007].
  • 44
    Brown D, Trowsdale J, Allen R. The LILR family: modulators of innate and adaptive immune pathways in health and disease. Tissue Antigens 2004; 64: 215225.
  • 45
    Borges L, Hsu ML, Fanger N, Kubin M, Cosman D. A family of human lymphoid and myeloid Ig-like receptors, some of which bind to MHC class I molecules. J Immunol 1997; 159: 51925196.
  • 46
    Colonna M, Navarro F, Bellon T, Llano M, Garcia P, Samaridis J, et al. A common inhibitory receptor for major histocompatibility complex class I molecules on human lymphoid and myelomonocytic cells. J Exp Med 1997; 186: 18091818.
  • 47
    Stegh AH, Kim H, Bachoo RM, Forloney KL, Zhang J, Schulze H, et al. Bcl2L12 inhibits post-mitochondrial apoptosis signaling in glioblastoma. Genes Dev 2007; 21: 98111.
  • 48
    Li X, Minden A. PAK4 functions in tumor necrosis factor (TNF) {alpha}-induced survival pathways by facilitating TRADD binding to the TNF receptor. J Biol Chem 2005; 280: 4119241200.
  • 49
    Urban G, Golden T, Aragon IV, Cowsert L, Cooper SR, Dean NM, et al. Identification of a functional link for the p53 tumor suppressor protein in dexamethasone-induced growth suppression. J Biol Chem 2003; 278: 97479753.
  • 50
    Zhou G, Golden T, Aragon IV, Honkanen RE. Ser/Thr protein phosphatase 5 inactivates hypoxia-induced activation of an apoptosis signal-regulating kinase 1/MKK-4/JNK signaling cascade. J Biol Chem 2004; 279: 4659546605.
  • 51
    Lenz G, Nagel I, Siebert R, Roschke AV, Sanger W, Wright GW, et al. Aberrant immunoglobulin class switch recombination and switch translocations in activated B cell-like diffuse large B cell lymphoma. J Exp Med 2007; 204: 633643.
  • 52
    Bartholdy B, Du Roure C, Bordon A, Emslie D, Corcoran LM, Matthias P. The Ets factor Spi-B is a direct critical target of the coactivator OBF-1. Proc Natl Acad Sci U S A 2006; 103: 1166511670.
  • 53
    Hao Y, Sekine K, Kawabata A, Nakamura H, Ishioka T, Ohata H, et al. Apollon ubiquitinates SMAC and caspase-9, and has an essential cytoprotection function. Nature Cell Biol 2004; 6: 849860.
  • 54
    Li Q, Ching AK-K, Chan BC-L, Chow SK-Y, Lim PL, Ho TC-Y, et al. A death receptor-associated anti-apoptotic protein, BRE, inhibits mitochondrial apoptotic pathway. J Biol Chem 2004; 279: 5210652116.
  • 55
    Wang HY, Peng G, Guo Z, Shevach EM, Wang RF. Recognition of a new ARTC1 peptide ligand uniquely expressed in tumor cells by antigen-specific CD4 + regulatory T cells. J Immunol 2005; 174: 26612670.
  • 56
    Fukuhara N, Tagawa H, Kameoka Y, Kasugai Y, Karnan S, Kameoka J, et al. Characterization of target genes at the 2p15–16 amplicon in diffuse large B-cell lymphoma. Cancer Sci 2006; 97: 499504.
  • 57
    Houldsworth J, Olshen AB, Cattoretti G, Donnelly GB, Teruya-Feldstein J, Qin J, et al. Relationship between REL amplification, REL function, and clinical and biologic features in diffuse large B-cell lymphomas. Blood 2004; 103: 18621868.
  • 58
    Dukers DF, Oudejans JJ, Vos W, ten Berge RL, Meijer CJ. Apoptosis in B-cell lymphomas and reactive lymphoid tissues always involves activation of caspase 3 as determined by a new in situ detection method. J Pathol 2002; 196: 307315.
  • 59
    Su GH, Hilgers W, Shekher MC, Tang DJ, Yeo CJ, Hruban RH, et al. Alterations in pancreatic, biliary, and breast carcinomas support MKK4 as a genetically targeted tumor suppressor gene. Cancer Res 1998; 58: 23392342.
  • 60
    Teng DHF, Perry WL III, Hogan JK, Baumgard M, Bell R, Berry S, et al. Human mitogen-activated protein kinase kinase 4 as a candidate tumor suppressor. Cancer Res 1997; 57: 41774182.
  • 61
    Bea S, Colomo L, Lopez-Guillermo A, Salaverria I, Puig X, Pinyol M, et al. Clinicopathologic significance and prognostic value of chromosomal imbalances in diffuse large B-cell lymphomas. J Clin Oncol 2004; 22: 34983506.
  • 62
    Ross CW, Ouillette PD, Saddler CM, Shedden KA, Malek SN. Comprehensive analysis of copy number and allele status identifies multiple chromosome defects underlying follicular lymphoma pathogenesis. Clin Cancer Res 2007; 13: 47774785.
  • 63
    Ni Z, Karaskov E, Yu T, Callaghan SM, Der S, Park DS, et al. Apical role for BRG1 in cytokine-induced promoter assembly. Proc Natl Acad Sci U S A 2005; 102: 1461114616.
  • 64
    Bosga-Bouwer AG, Kok K, Booman M, Boven L, van der Vlies P, van den Berg A, et al. Array comparative genomic hybridization reveals a very high frequency of deletions of the long arm of chromosome 6 in testicular lymphoma. Genes Chromosomes Cancer 2006; 45: 976981.
  • 65
    Santos SG, Campbell EC, Lynch S, Wong V, Antoniou AN, Powis SJ. Major histocompatibility complex class I–ERp57–tapasin interactions within the peptide-loading complex. J Biol Chem 2007; 282: 1758717593.
  • 66
    Zhang Y, Williams DB. Assembly of MHC class I molecules within the endoplasmic reticulum. Immunol Res 2006; 35: 151162.
  • 67
    Saunders PA, Hendrycks VR, Lidinsky WA, Woods ML. PD-L2 : PD-1 involvement in T cell proliferation, cytokine production, and integrin-mediated adhesion. Eur J Immunol 2005; 35: 35613569.
  • 68
    Rosenwald A, Wright G, Leroy K, Yu X, Gaulard P, Gascoyne RD, et al. Molecular diagnosis of primary mediastinal B cell lymphoma identifies a clinically favorable subgroup of diffuse large B cell lymphoma related to Hodgkin lymphoma. J Exp Med 2003; 198: 851862.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Supporting information
  9. References
  10. Supporting Information
path_2399_sm_descriptions.doc22KSupporting Information
path_2399_sm_tableS1.xls49KSupporting Information
path_2399_sm_tableS2.doc71KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.