Abstract Background Primary central nervous system lymphoma (PCNSL) represents a particular entity within non-Hodgkin lymphomas and is associated with poor outcome. The present study addresses the potential clinical relevance of chimeric transcripts in PCNSL discovered by using RNA sequencing (RNA-seq). Methods Seventy-two immunocompetent and newly diagnosed PCNSL cases were included in the present study. Among them, 6 were analyzed by RNA-seq to detect new potential fusion transcripts. We confirmed the results in the remaining 66 PCNSL. The gene fusion was validated by fluorescence in situ hybridization (FISH) using formalin-fixed paraffin-embedded (FFPE) samples. We assessed the biological and clinical impact of one new gene fusion. Results We identified a novel recurrent gene fusion, E26 transformation-specific translocation variant 6–immunoglobulin heavy chain (ETV6-IgH). Overall, ETV6-IgH was found in 13 out of 72 PCNSL (18%). No fusion conserved an intact functional domain of ETV6, and ETV6 was significantly underexpressed at gene level, suggesting an ETV6 haploinsufficiency mechanism. The presence of the gene fusion was also validated by FISH in FFPE samples. Finally, PCNSL samples harboring ETV6-IgH showed a better prognosis in multivariate analysis, P = 0.03, hazard ratio = 0.33, 95% CI = 0.12–0.88. The overall survival at 5 years was 69% for PCNSL harboring ETV6-IgH versus 29% for samples without this gene fusion. Conclusions ETV6-IgH is a new potential surrogate marker of PCNSL with favorable prognosis with ETV6 haploinsufficiency as a possible mechanism. The potential clinical impact of ETV6-IgH should be validated in larger prospective studies. ETV6-IgH, fusion gene, haploinsufficiency, primary CNS lymphoma, RNA sequencing Importance of the study Primary central nervous system lymphoma is a rare entity with heterogeneous clinical evolution. Chimeric genes are interesting molecular markers because they may allow detection of novel oncogenic pathways and could be used as biomarkers. We analyzed 6 fresh-frozen PCNSL by RNA-seq and detected a recurrent chimeric fusion involving ETV6-IgH. The prevalence of this gene fusion has been established using 66 fresh-frozen PCNSL samples by direct sequencing. We have analyzed the potential functional impact of this gene fusion by western blot of transfected COS-7 cells with ETV6-IgH gene fusion. Finally, we found that PCNSL harboring this chimeric gene are associated with a better prognosis in the multivariate analysis as well as low ETV6 expression, suggesting a haploinsufficiency mechanism. Primary central nervous system lymphoma (PCNSL) is an intriguing entity currently classified according to World Health Organization (WHO) criteria as a diffuse large B-cell lymphoma (DLBCL) restricted to the CNS.1 PCNSL are extranodal, malignant non-Hodgkin lymphomas that are confined to the brain, eyes, leptomeninges, or spinal cord, in the absence of systemic lymphoma.1 The particular tropism of PCNSL to the CNS as well as the reason this neoplasm exclusively manifests in the immunoprivileged brain in the absence of systemic spread is still unclear.2 Although PCNSL is associated with a dismal prognosis, the prognosis has been substantially improved by using high-dose methotrexate.3 However, treatment of this disease remains challenging because remissions are frequently short-lasting with substantial toxicity.4 The rarity of this disease and the small amount of tissue obtained in the vast majority of cases from stereotaxic biopsies have delayed understanding of the oncogenesis of PCNSL. The expression profiling of PCNSL with expression of B-cell lymphoma 6 (BCL6), interferon regulatory factor 4 (IRF4) together with an aberrant somatic hypermutation (aSHM) indicates that PCNSL cells belong to a late germinal center B cell.2,5 We and others have reported recurrent copy number aberrations using high-density comparative genomic hybridization or single nucleotide polymorphism arrays and described the mutational landscape of PCNSL using whole-exome sequencing.6–12 The most striking alterations reported to date are (i) frequent chromosomal deletions affecting HLA locus (6p21.32), 6q22 chromosome, and CDKN2A locus (9p21.3) and (ii) somatic mutations in genes involved in B-cell receptor/Toll-like receptor/nuclear factor-kappaB pathways, especially MYD88 and CD79B.6,13–15 The present study addresses the potential clinical relevance of chimeric transcripts in PCNSL discovered by using RNA-seq. We have identified several new fusion genes and have focused on the most frequent one involving E26 transformation-specific translocation variant 6 (ETV6) and immunoglobulin heavy chain (IgH) as a novel gene fusion that could be potentially used as a prognostic marker in PCNSL. Materials and Methods PCNSL Samples Seventy-two immunocompetent (HIV negative and no history of immunosuppressive drugs or organ transplantation) and newly diagnosed PCNSL cases homogeneously treated with high-dose methotrexate regimen (3.5 g/m2) were included in the present study. Tumors were selected on the basis of fresh frozen tissue availability. All tumors were PCNSL classified as CD20+ DLBCL according to the WHO criteria1 and demonstrated to contain at least 80% tumor cells. For all cases, systemic lymphoma was excluded by extensive investigation. This project was approved by the local ethics committee (CPPRB Pitié-Salpêtrière). Written consent for sample collection and genetic analysis was obtained from all the participants. Details about PCNSL cases investigated in the present study are provided in Supplementary Table S1. RNA Extraction and Quality Assessment Total RNA from cryopreserved samples was extracted using the iPrep Trizol Plus RNA kit (Life Technologies). Tumor lysis was first performed in Trizol (Invitrogen) lysis buffer using the FastPrep system (MP Biomedicals). After chloroform addition, total RNA was purified using the iPrep Trizol Plus RNA kit. RNA was quantified using a NanoDrop spectrophotometer, and the quality, depending on RNA Integrity Number, RNA concentration, and 28S:18S rRNA ratio, was assessed using an Agilent BioAnalyzer. RNA Sequencing RNA sequencing was performed for cases with a minimal amount of RNA of 1.5 µg and an RNA Integrity Number of at least 7. Library was prepared using the TruSeq Stranded mRNA kit protocol (Illumina Technology) with an input total RNA of 1 µg. Capture of polyadenylated RNA was realized using oligodeoxythymidine beads. Captured RNA was fragmented in approximately 400 bp. After DNA synthesis, Illumina adaptors ligation, and library amplification by PCR, 100 bp paired-end sequencing was performed on an Illumina HiSEQ 2000. Data Analysis and Detection of Putative Fusion Transcripts Data analysis was realized by GenoSplice technology (ICM). Data quality control was performed using FastQC v0.10.1 (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/). Fusion transcripts were detected using 3 different approaches: tophat-fusion, defuse, and EASANA-fusion (GenoSplice). We considered only chimeric transcripts that were commonly detected by at least 2–3 algorithms. Further details on the bioinformatics analysis are found in the Supplementary methods. ETV6 Expression in PCNSL and Transfected Cells ETV6 expression was assessed by quantitative PCR. Primer and probes were synthesized using Universal Probe Library (Roche) software (primers and probes are provided in Supplementary materials). Quantitative PCR was performed on LightCycler 480 (Roche) using the following conditions: 10 minutes at 95°C for 1 cycle, 10 seconds at 95°C, 30 seconds at 60°C and 1 second at 72°C for 45 cycles, and 30 seconds at 40°C. Expression levels were normalized to peptidylprolyl isomerase A and relative expression of ETV6 was calculated using the ΔΔCt method. Cell Culture The monkey kidney COS-7 cell line was obtained from the American Type Culture Collection and supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin (15140, ThermoFisher Scientific). The cells were cultured in a humidified incubator with 5% CO2 at 37°C. Plasmid Construction ETV6 wild-type (wt), ETV6-IgHG4, ETV6-delta (a truncated version of ETV6 without IgH), and green fluorescent protein control were cloned into lentiviral vector control using a CMV-3HA-pPGK-puromycin selection. COS-7 cells expressing ETV6 wt, ETV6-delta (ETV6 lacking the last 4 exons), ETV6-IgHG4, or green fluorescent protein control were generated by lentiviral transduction and subsequent puromycin selection. Western Blot Analysis Immunoblotting of COS-7 cells was performed with the following antibodies: anti-hemagglutinin (ab18181, Abcam, diluted 1:5000) and anti-ETV6 (ab151698, Abcam, diluted 1:5000) in at first, and then anti-cyclophylin B (PA1-027A, Pierce, diluted 1:2000). After overnight incubation at 4°C with primary antibodies IRDye 680RD goat anti-rabbit IgG (Li-Cor, diluted 1:5000), membranes were washed again and scanned on an Odyssey CLx Imaging System. Scan settings were high quality, 169 µm resolution, intensity 5 for both channels without focus offset. Further details are provided in the Supplementary methods. Interphase Fluorescence In Situ Hybridization on Formalin-Fixed Paraffin-Embedded Sections ETV6-IgH fusion was confirmed using 3-µm formalin-fixed paraffin-embedded (FFPE) tissue sections using ETV6-IgH positive PCNSL samples detected by RNA-seq or by Sanger sequencing that were deparaffinized with the histology FISH Accessory Kit (Dako). Slides were visualized using a fluorescence scanner (Pathscan, Excilone). Hybridizing signals in at least 100 non-overlapping nuclei were counted. The presence of the break-apart probe signal in greater than 15% of tumor cells was defined as positive for ETV6-IgH fusions. Direct Sequencing of MYD88 and CD79B Somatic Mutations The hotspot mutations of MYD88 (L265P) and CD79B (Y196) were investigated by Sanger as previously described.8 Shortly, the amplification conditions were 94°C for 3 min followed by 45 cycles of 94°C × 15 sec, 60°C × 45 sec, and 72°C × 1 min, with a final step at 72°C for 8 min. The somatic DNA was amplified using the following primers: for MYD88 L265P: TGTGTGAGTGAATGTGTGCC (forward) and GAGTCCAGAACCAAGATTTGGT (added); and for CD79B Y196: CACCCCTCTCCCTGGCCCTC (forward) and CGGGACCACACCCCAACCAC (reverse). Validation The validation of the putative fusion transcripts identified by RNA-seq was performed using reverse transcriptase (RT)-PCR. Five hundred nanograms of total RNA were retrotranscribed using the Maxima first strand cDNA synthesis kit (Thermo Scientific) following the manufacturer’s instructions. PCR was performed using primers designed according to predicted fusion transcript sequence with the forward primer located within the 5ʹ end of the ETV6 transcript and the reverse primer within the 3ʹ end of the IgH transcript. Primer sequences are listed in Supplementary Table S2. The amplification conditions were as already described.8 The purified sequences were addressed to GATC Biotech for conventional Sanger sequencing. All transcriptome sequencing data have been deposited at the Gene Expression Omnibus, which is hosted by the National Center for Biotechnology Information, under the accession code GSE81816. The investigation of additional cases with ETV6-IgH fusion gene was assessed using an optimized RT-PCR assay. Further details are provided in Supplementary Table S2. Statistical Analyses We applied unpaired Wilcoxon Mann–Whitney tests for comparing ETV6 expression levels obtained by quantitative (q)RT-PCR, age and Karnofsky perfomance status (KPS), both as continuous variables, in PCNSL samples according to ETV6-IgH status. Kaplan–Meier analysis and the log-rank test were used to explore differences between overall survival (OS) according to ETV6-IgH status, age (≥60 vs <60 y), and KPS (≥70% vs <70%). Cox proportional hazards regression models were used to obtain hazard ratios (HRs) with Wald 95% CIs for the relationship between OS and ETV6-IgH status, age, and KPS in the patient cohorts. We assessed the proportionality of the hazards for Cox regression with the Schoenfeld residuals. All P-values were 2-sided and P-values less than 0.05 were interpreted as statistically significant. Analyses were performed using R statistical software, version 3.3 (Free Software Foundation available at http://www.r-project.org). Results Gene Fusion Identification Using RNA-Seq We collected a cohort of 6 PCNSL samples on which we performed transcriptome sequencing with the aim of identifying new chimer alterations. We applied 3 different gene fusion algorithms and only those fusion genes detected by all of them were further considered. We identified a total of 1827 putative fusion transcripts in the 6 PCNSL samples (Supplementary Table S3). Thirty-two putative fusions involving 57 distinct genes were commonly detected by at least 2 of the 3 fusion detection algorithms (Fig. 1, Supplementary Figures S1 and S2, Supplementary Table S3), including 3 interchromosomal and 29 intrachromosomal fusions. Only 3 fusions were commonly detected by the 3 pipelines: SSR2-GON4L, ETV6-IgH, and WHSC2-LETM1. Among them, we selected the most frequent chimeric transcript, ETV6-IgH, detected in 2 of the 6 cases investigated by RNA-seq. This fusion raised our interest because ETV6 is frequently involved in different hematological diseases and has a prominent role in hematopoietic stem cell homeostasis.16,17 In addition, focal deletions of ETV6 locus and recurrent somatic mutations have been recently identified in 2 different PCNSL studies.18,19 In the same line, there are many studies suggesting that ETV6 could act in some setting as a tumor suppressor gene.20 Fig. 1 View largeDownload slide Overview of the 32 putative chimeric transcripts identified by at least 2 fusion detection algorithms. Inner arcs represent rearrangements from the 6 cases analyzed by RNA-seq. Interchromosomal fusions are shown in purple and intrachromosomal fusions are shown in red. Fig. 1 View largeDownload slide Overview of the 32 putative chimeric transcripts identified by at least 2 fusion detection algorithms. Inner arcs represent rearrangements from the 6 cases analyzed by RNA-seq. Interchromosomal fusions are shown in purple and intrachromosomal fusions are shown in red. Validation of the ETV6-IgH Gene Fusion by Sequencing in 66 PCNSL The fusion is a somatic genomic event as ETV6 break-apart fluorescence in situ hybridization (FISH) and FISH with custom ETV6 and IgH probes revealed rearrangements in the respective chromosomal regions in the tumor cells, but not in surrounding nontumoral cells (Fig. 2E). Fig. 2 View largeDownload slide ETV6-IgH fusion transcripts identified by RNA sequencing of PCNSL and FISH. (A) ETV6-IgH specific PCR from cDNA derived from the 6 PCNSL cases of the RNA-seq cohort showing 2 different ETV6 breakpoints (red arrows) detected in 2 patients. (B, C) Schematics of the 2 fusion transcripts identified in 2 cases using RNA-seq. Regions corresponding to ETV6 or IgH are shown in blue or purple, respectively. Vertical red lines show breakpoints and horizontal dotted lines indicate open reading frame for each fusion transcript. (D) Chromosomal rearrangements detected by FISH using custom ETV6 and IgH probes showing (white arrows). Fig. 2 View largeDownload slide ETV6-IgH fusion transcripts identified by RNA sequencing of PCNSL and FISH. (A) ETV6-IgH specific PCR from cDNA derived from the 6 PCNSL cases of the RNA-seq cohort showing 2 different ETV6 breakpoints (red arrows) detected in 2 patients. (B, C) Schematics of the 2 fusion transcripts identified in 2 cases using RNA-seq. Regions corresponding to ETV6 or IgH are shown in blue or purple, respectively. Vertical red lines show breakpoints and horizontal dotted lines indicate open reading frame for each fusion transcript. (D) Chromosomal rearrangements detected by FISH using custom ETV6 and IgH probes showing (white arrows). We next performed RT-PCR using primers specific for the chimeric transcript to identify additional tumors bearing the fusion in a set of 66 PCNSL, in addition to the 6 PCNSL tested by RNA-seq. We identified 11 additional tumors carrying the fusion (Supplementary Table S1). All breakpoints that we identified on ETV6 were located on the 5ʹ side of the transcript (ie, before the third exon), while IgH breakpoints were distributed all along the transcripts. Some preferential clusterings of breakpoints were identified at the ends of exons 1 and 2 for ETV6 and in the middle of exon 4 for IgHG4 (Fig. 2B–D and Supplementary Figure S3). The predicted fusion proteins indicated that none preserved an entire functional domain of ETV6 protein (Supplementary Figure S3). Four ETV6-IgH proteins were predicted to conserve a part of the PNT (or pointed) domain responsible for protein-protein interactions, including one conserving more than half of its domain. Clinical Impact of ETV6 Gene Fusion Age and sex were equally distributed in PCNSL carrying ETV6-IgH and in ETV6 wt PCNSL counterparts (Supplementary Table S4A). Interestingly, univariate survival analysis pinpointed that PCNSL harboring ETV6-IgH had a better prognosis than their ETV6 wt counterparts (P = 0.04; Fig. 3A). Moreover, multivariate analysis using the Cox proportional hazards model confirmed that ETV6-IgH was independently associated with favorable prognosis after adjusting for age and KPS (P = 0.03, HR = 0.33, 95% CI = 0.12–0.88) (Supplementary Table S4B) with an OS at 5 years of 69% for PCNSL harboring ETV6-IgH versus 29%. Fig. 3 View largeDownload slide (A) Kaplan–Meier plot showing OS according to ETV6-IgH status. (B) Western blot using COS-7 cell lines in the presence of an empty vector, no transfected cell line, transfection with ETV6, ETV6-IgHG4, and ETV6 truncated constructions using a lentivirus. (C) The boxes represent the median (black middle line) limited by the 25th (Q1) and 75th (Q3) percentiles of ETV6 expression according to ETV6-IgH fusion status in arbitrary units. Significance of the differences of ETV6 expression was determined using the Wilcoxon Mann–Whitney test. Fig. 3 View largeDownload slide (A) Kaplan–Meier plot showing OS according to ETV6-IgH status. (B) Western blot using COS-7 cell lines in the presence of an empty vector, no transfected cell line, transfection with ETV6, ETV6-IgHG4, and ETV6 truncated constructions using a lentivirus. (C) The boxes represent the median (black middle line) limited by the 25th (Q1) and 75th (Q3) percentiles of ETV6 expression according to ETV6-IgH fusion status in arbitrary units. Significance of the differences of ETV6 expression was determined using the Wilcoxon Mann–Whitney test. We also analyzed the prognostic impact of ETV6 expression. Patients with high ETV6 expression levels (according to the median) had lower KPS compared with the low ETV6 expression samples (P = 0.02), and age was equally distributed (Supplementary Table S4C). Low ETV6 expression in the overall cohort was associated with a better prognosis in univariate (P = 0.007; Supplementary Figure S4) and in multivariate analysis (P = 0.01, HR = 0.44, 95% CI = 0.24–0.83; Supplementary Table S4D) with an OS at 5 years of 55% for PCNSL with low ETV6 expression levels versus 20%. However, when only ETV6 wt samples (ie, without ETV6 fusion) were analyzed, we did not find any prognostic impact of ETV6 gene expression (P = 0.17; Supplementary Figure S5). Functional Impact of ETV6-IgH Fusion In most of the cases, ETV6 fusions involved the first 2 exons, potentially altering the expression of ETV6. To validate this prediction, we transduced COS-7 cells with ETV6-IgHG4 and ETV6-delta lentiviruses and performed western blot of COS-7 cells to determine the expression. We also transduced this cell line with either an empty vector or a virus containing normal ETV6 (ETV6 wt) (Fig. 3B). We did not find any difference in ETV6 protein expression compared with the different ETV6 constructions (Fig. 3B). We next analyzed the expression of ETV6 in COS-7 cells using qRT-PCR showing an underexpression of ETV6 3p compared with control constructions (P < 0.05, data not shown), arguing in favor of a potential haploinsufficiency of ETV6 expression. Likewise, qRT-PCR in PCNSL samples showed a significant ETV6 underexpression in ETV6-IgH positive samples compared with those with ETV6 wt (P < 0.05; Fig. 3C). Taken together these data suggest that ETV6-IgH leads to a single-allele loss of ETV6, reducing its gene expression (Fig. 3C) but without significantly modifying its protein expression (Fig. 3B). Therefore, haploinsufficiency may have a potential impact on the mechanism involved in this gene fusion. Correlation of ETV6 Fusion with Other Molecular Features We have also screened the most frequent hotspot mutations described in PCNSL: MYD88 L265P and CD79B Y196.18,19 Overall, 29/72 (40.3%) harbored MYD88 L265P, 19/72 (26.4%) CD79B Y196 mutation, and 15/72 (20.8%) both of them (Supplementary Table S1). In addition, the distribution of MYD88 L265P and CD79B Y196 mutations was similar according to ETV6-IgH gene fusion status. Indeed, according to ETV6-IgH gene fusion status, 5/13 (38.6%) harbored MYD88 L265P mutation versus 24/59 (40.7%), P = 1, Fisher’s exact test, CD79B Y196 in 2/13 (15.4%) versus 13/59 (22%), P = 0.7, as well as in 2/13 (15.4%) versus 13/59 (22%) both of them (Supplementary Table S1). Discussion We have characterized a small cohort of PCNSL by RNA-seq to discover new chimeric transcripts. We have identified several potential interesting gene fusions and have further estimated the frequency and the clinical impact in a larger series of PCNSL using fresh-frozen tissue. All ETV6-IgH gene fusions were validated by cDNA sequencing. Overall, we identified 13 cases with the ETV6-IgH fusion gene in our whole cohort of 72 PCNSL. We estimate the frequency of ETV6-IgH in PCNSL to be approximately 18%. Therefore, ETV6-IgH is the most frequently reported fusion gene in PCNSL. We provide evidence that ETV6-IgH leads to a decrease of expression (at mRNA), suggesting a potential role of haploinsufficiency of ETV6. In the same line, there are many studies suggesting that ETV6 could be a tumor suppressor gene also by a haploinsufficiency mechanism.21 Haploinsufficiency occurs when the amount of protein product created from the remaining wild-type allele is not sufficient for normal cellular function. Therefore, ETV6 could be considered as “haplo-insufficient” to indicate that one copy of the gene is insufficient for proper function.22 The ETV6 protein contains 2 major domains, the helix-loop-helix (HLH) domain, encoded by exons 3 and 4, and the E26 transformation-specific (ETS) domain, encoded by exons 6 through 8, with in-between the internal domain encoded by exon 5. ETV6 is a strong transcriptional repressor, acting through its HLH and internal domains.16 This transcription factor is frequently rearranged in childhood pre-B acute lymphoblastic leukemia and leukemia of myeloid or lymphoid origins.23,24 It is important to emphasize that ETV6 is known to be fused with a wide range of genes encoding receptor tyrosine kinase genes, transcription factors, homeobox genes, and many others.25 Interestingly, the mentioned fusions, as the one described in this study, do not include the full-length ETV6 protein. Remarkably, several gene fusions involving ETV6 have been associated with a haploinsufficiency mechanism.26,27 Furthermore, even fusions of ETV6 with the same target will not always have the same breakpoints in ETV6 protein.25 Interestingly, in a recent PCNSL study, ETV6 was found to be statistically significantly associated as a target of the aSHM phenotype in 22/41 cases (53.7%).19 The fusion partner of ETV6, IgH, is a frequently rearranged locus in DLBCL and PCNSL, and in both diseases these rearrangements could be associated with aSHM.28 IgH translocations have been found in 13% of PCNSL and are less frequent than in DLBCL (45%).14 In addition, the most common IgH translocation partner in PCNSL is BCL6 (80%), while in DLBCL it is more frequently linked to BCL2 (15%).14 Furthermore, ETV6-IgH samples harbored a favorable prognosis in multivariate analysis, with an OS at 5 years of 69% for PCNSL harboring ETV6-IgH versus 29% for samples without this gene fusion after adjusting for age and KPS (Supplementary Table S4). Different prognostic scores using clinical characteristics have been proposed but age and KPS seem to be the strongest independent predictors in PCNSL.29 However, it should be noted that further molecular alterations might impact the clinical evolution of PCNSL. Accordingly, another gene fusion involving ETV6, ETV6-RUNX1, is the most frequent genomic aberration found in pre-B acute lymphoblastic leukemia, occurring in approximately 25% of cases, and is associated with favorable prognosis.30 Different potential biomarkers of prognosis in PCNSL have been described during the last years. Overexpression of BCL6 was associated with improved survival compared with tumors that did not express BCL6.31 However, other studies did not corroborate these findings.32 More recently, recurrent somatic nonsynonymous mutations in MYD88 and CD79B genes were found in approximately two-thirds of PCNSL.9,18,19 Interestingly, the blockade of B-cell receptor signals with an inhibitor of Bruton tyrosine kinase (ibrutinib) has shown clinical efficacy against activated B-cell DLBCL, notably in DLBCL with double mutations (CD79B and MYD88), showing a potential prediction biomarker for a target therapy.33 In our study the double mutations of MYD88 L265P and CD79B Y196 were equally distributed according to ETV6-IgH gene fusion status (2/13 [15.4%] vs 13/59 [22%], P = 0.7, Fisher’s exact test). It is also important to highlight that all the patients included in this study were treated with a high-dose methotrexate regimen without any prior chemotherapy or radiotherapy.34 We have validated the presence of ETV6-IgH gene fusion by FISH in FFPE samples. This technique could be used to detect this chimeric transcript in the clinical setting and to be screened in PCNSL samples in order to validate this potential new biomarker. Recent studies have pinpointed recurrent chromosomal rearrangements in PCNSL with highly heterogeneous results.18,19 Among the recently described gene fusions one study found: BCL6-IgH (17%) and programmed cell death ligand (PD-L) foci (PD-L1 or PD-L2) translocations (6%).18 We found a common gene fusion with this study involving BCL6-IGL (Supplementary Table S3). Conversely, in another recent study, only one rare fusion gene was found in a series of 30 PCNSL.19 These divergent results could be explained in part due to different pipeline analysis, next-generation sequencing approaches, and different tissue samples (ie, fresh-frozen and FFPE). Interestingly, one of these studies using whole-exome and RNA-seq analysis of PCNSL had also identified inactivating alterations of ETV6 in 3 out of 24 cases (12.5%), with deletions of exon 2 or exons 2–5 that modified the reading frame.18 Therefore, it is tempting to speculate that these single-allele deletions of ETV6 may also be involved in loss-of-function of this gene, leading to a reduction of the amount of ETV6 within the cell, as we showed in ETV6-IgH chimeric transcript. Furthermore, the mutational landscape of DLBCL using whole-genome analysis has also highlighted the presence of a rare gene fusion involving ETV6 with an IgH in 1 out of 40 (2.5%) that was further validated by RNA-seq.35 Consequently, we can hypothesize that due to the higher frequency found in this study, this gene fusion could be more frequently found in PCNSL (13 out of 72, 18%, vs 1 out of 40, 2.5%, P-value = 0.017, Fisher’s exact test). It is worth mentioning that our study has some limitations. This is a small retrospective dataset and the potential clinical impact should be validated in larger prospective studies. The impact of intratumoral heterogeneity of ETV6-IgH has not been thoroughly assessed. Further studies analyzing larger cohorts of PCNSL using FISH are warranted to better characterize the potential impact of intratumoral heterogeneity in ETV6-IgH gene fusion. It should also be noted that other genetic alterations (ie, mutations and copy number alterations) of the ETV6 wt allele may modify the impact of this gene fusion. These alterations should be further evaluated in future studies. Finally, we cannot formally exclude a potential role of dominant-negative in ETV6-IgH. However, the loss of both oligomerization and DNA-binding domains in ETV6-IgH fusion makes it unlikely that this molecular mechanism has a major effect. To the best of our knowledge, this is the first study showing a novel fusion gene in PCNSL that could be used as a potential biomarker to detect a subset of PCNSL patients with less severe disease. Supplementary Material Supplementary material is available at Neuro-Oncology online. Funding This work is part of the national program Cartes d’Identité des Tumeurs (CIT) funded and developed by the Ligue nationale contre le cancer, the Institut National du Cancer, Association pour la recherche sur les tumeurs cérébrales (ARTC), Cancéropôle Île-de-France “Emergence 2015-1” (2015-1-EMERG-05-INSERM 6-1), Ligue nationale contre le cancer (Comité du Val d’Oise, R14044DD), Ligue Nationale contre le cancer “Recherche épidemiologique” (N° PRE2015.LNCC), Fondation pour la Recherche Médicale (FDT20140930968), and the program “Investissements d’avenir” ANR-10-IAIHU-06, and the Institut National du Cancer (INCa) (Réseau Expert National LOC, Lymphomes Oculo-Cérébraux). This study was supported by the Lymphomes Oculo-Cérébraux (LOC) study group network (Réseau national de centres experts des lymphomes primitifs du système nerveux central). Conflict of interest statement. The authors declare no conflicts of interest. Acknowledgments We would like to thank the French LOC Network investigators. Specimens from Marseille were retrieved from the AP-HM tumor bank AC 2013-1786. References 1. Campo E, Swerdlow SH, Harris NL, Pileri S, Stein H, Jaffe ES. The 2008 WHO classification of lymphoid neoplasms and beyond: evolving concepts and practical applications. Blood . 2011; 117( 19): 5019– 5032. Google Scholar CrossRef Search ADS PubMed 2. Deckert M, Montesinos-Rongen M, Brunn A, Siebert R. Systems biology of primary CNS lymphoma: from genetic aberrations to modeling in mice. Acta Neuropathol . 2014; 127( 2): 175– 188. Google Scholar CrossRef Search ADS PubMed 3. Hoang-Xuan K, Bessell E, Bromberg J et al. ; European Association for Neuro-Oncology Task Force on Primary CNS Lymphoma. Diagnosis and treatment of primary CNS lymphoma in immunocompetent patients: guidelines from the European Association for Neuro-Oncology. Lancet Oncol . 2015; 16( 7): e322– e332. Google Scholar CrossRef Search ADS PubMed 4. Ricard D, Idbaih A, Ducray F, Lahutte M, Hoang-Xuan K, Delattre JY. Primary brain tumours in adults. Lancet . 2012; 379( 9830): 1984– 1996. Google Scholar CrossRef Search ADS PubMed 5. Montesinos-Rongen M, Küppers R, Schlüter D et al. Primary central nervous system lymphomas are derived from germinal-center B cells and show a preferential usage of the V4-34 gene segment. Am J Pathol . 1999; 155( 6): 2077– 2086. Google Scholar CrossRef Search ADS PubMed 6. Braggio E, McPhail ER, Macon W et al. Primary central nervous system lymphomas: a validation study of array-based comparative genomic hybridization in formalin-fixed paraffin-embedded tumor specimens. Clin Cancer Res . 2011; 17( 13): 4245– 4253. Google Scholar CrossRef Search ADS PubMed 7. Braggio E, Van Wier S, Ojha J et al. Genome-wide analysis uncovers novel recurrent alterations in primary central nervous system lymphomas. Clin Cancer Res . 2015; 21( 17): 3986– 3994. Google Scholar CrossRef Search ADS PubMed 8. Bruno A, Boisselier B, Labreche K et al. Mutational analysis of primary central nervous system lymphoma. Oncotarget . 2014; 5( 13): 5065– 5075. Google Scholar CrossRef Search ADS PubMed 9. Gonzalez-Aguilar A, Idbaih A, Boisselier B et al. Recurrent mutations of MYD88 and TBL1XR1 in primary central nervous system lymphomas. Clin Cancer Res . 2012; 18( 19): 5203– 5211. Google Scholar CrossRef Search ADS PubMed 10. Schwindt H, Vater I, Kreuz M et al. Chromosomal imbalances and partial uniparental disomies in primary central nervous system lymphoma. Leukemia . 2009; 23( 10): 1875– 1884. Google Scholar CrossRef Search ADS PubMed 11. Sung CO, Kim SC, Karnan S et al. Genomic profiling combined with gene expression profiling in primary central nervous system lymphoma. Blood . 2011; 117( 4): 1291– 1300. Google Scholar CrossRef Search ADS PubMed 12. Vater I, Montesinos-Rongen M, Schlesner M et al. The mutational pattern of primary lymphoma of the central nervous system determined by whole-exome sequencing. Leukemia . 2015; 29( 3): 677– 685. Google Scholar CrossRef Search ADS PubMed 13. Booman M, Douwes J, Glas AM et al. Mechanisms and effects of loss of human leukocyte antigen class II expression in immune-privileged site-associated B-cell lymphoma. Clin Cancer Res . 2006; 12( 9): 2698– 2705. Google Scholar CrossRef Search ADS PubMed 14. Cady FM, O’Neill BP, Law ME et al. Del(6)(q22) and BCL6 rearrangements in primary CNS lymphoma are indicators of an aggressive clinical course. J Clin Oncol . 2008; 26( 29): 4814– 4819. Google Scholar CrossRef Search ADS PubMed 15. Montesinos-Rongen M, Godlewska E, Brunn A, Wiestler OD, Siebert R, Deckert M. Activating L265P mutations of the MYD88 gene are common in primary central nervous system lymphoma. Acta Neuropathol . 2011; 122( 6): 791– 792. Google Scholar CrossRef Search ADS PubMed 16. Bohlander SK. ETV6: a versatile player in leukemogenesis. Semin Cancer Biol . 2005; 15( 3): 162– 174. Google Scholar CrossRef Search ADS PubMed 17. Hock H, Meade E, Medeiros S et al. Tel/Etv6 is an essential and selective regulator of adult hematopoietic stem cell survival. Genes Dev . 2004; 18( 19): 2336– 2341. Google Scholar CrossRef Search ADS PubMed 18. Chapuy B, Roemer MG, Stewart C et al. Targetable genetic features of primary testicular and primary central nervous system lymphomas. Blood . 2016; 127( 7): 869– 881. Google Scholar CrossRef Search ADS PubMed 19. Fukumura K, Kawazu M, Kojima S et al. Genomic characterization of primary central nervous system lymphoma. Acta Neuropathol . 2016; 131( 6): 865– 875. Google Scholar CrossRef Search ADS PubMed 20. Van Vlierberghe P, Ambesi-Impiombato A, Perez-Garcia A et al. ETV6 mutations in early immature human T cell leukemias. J Exp Med . 2011; 208( 13): 2571– 2579. Google Scholar CrossRef Search ADS PubMed 21. Fenrick R, Wang L, Nip J et al. TEL, a putative tumor suppressor, modulates cell growth and cell morphology of ras-transformed cells while repressing the transcription of stromelysin-1. Mol Cell Biol . 2000; 20( 16): 5828– 5839. Google Scholar CrossRef Search ADS PubMed 22. Berger AH, Pandolfi PP. Haplo-insufficiency: a driving force in cancer. J Pathol . 2011; 223( 2): 137– 146. Google Scholar CrossRef Search ADS PubMed 23. Golub TR, Barker GF, Bohlander SK et al. Fusion of the TEL gene on 12p13 to the AML1 gene on 21q22 in acute lymphoblastic leukemia. Proc Natl Acad Sci U S A . 1995; 92( 11): 4917– 4921. Google Scholar CrossRef Search ADS PubMed 24. Golub TR, Barker GF, Lovett M, Gilliland DG. Fusion of PDGF receptor beta to a novel ets-like gene, tel, in chronic myelomonocytic leukemia with t(5;12) chromosomal translocation. Cell . 1994; 77( 2): 307– 316. Google Scholar CrossRef Search ADS PubMed 25. De Braekeleer E, Douet-Guilbert N, Morel F, Le Bris MJ, Basinko A, De Braekeleer M. ETV6 fusion genes in hematological malignancies: a review. Leuk Res . 2012; 36( 8): 945– 961. Google Scholar CrossRef Search ADS PubMed 26. Panagopoulos I, Strömbeck B, Isaksson M, Heldrup J, Olofsson T, Johansson B. Fusion of ETV6 with an intronic sequence of the BAZ2A gene in a paediatric pre-B acute lymphoblastic leukaemia with a cryptic chromosome 12 rearrangement. Br J Haematol . 2006; 133( 3): 270– 275. Google Scholar CrossRef Search ADS PubMed 27. Belloni E, Trubia M, Mancini M et al. A new complex rearrangement involving the ETV6, LOC115548, and MN1 genes in a case of acute myeloid leukemia. Genes Chromosomes Cancer . 2004; 41( 3): 272– 277. Google Scholar CrossRef Search ADS PubMed 28. Montesinos-Rongen M, Van Roost D, Schaller C, Wiestler OD, Deckert M. Primary diffuse large B-cell lymphomas of the central nervous system are targeted by aberrant somatic hypermutation. Blood . 2004; 103( 5): 1869– 1875. Google Scholar CrossRef Search ADS PubMed 29. Abrey LE, Ben-Porat L, Panageas KS et al. Primary central nervous system lymphoma: the Memorial Sloan-Kettering Cancer Center prognostic model. J Clin Oncol . 2006; 24( 36): 5711– 5715. Google Scholar CrossRef Search ADS PubMed 30. Rubnitz JE, Downing JR, Pui CH et al. TEL gene rearrangement in acute lymphoblastic leukemia: a new genetic marker with prognostic significance. J Clin Oncol . 1997; 15( 3): 1150– 1157. Google Scholar CrossRef Search ADS PubMed 31. Levy O, Deangelis LM, Filippa DA, Panageas KS, Abrey LE. Bcl-6 predicts improved prognosis in primary central nervous system lymphoma. Cancer . 2008; 112( 1): 151– 156. Google Scholar CrossRef Search ADS PubMed 32. Camilleri-Broët S, Crinière E, Broët P 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( 1): 190– 196. Google Scholar CrossRef Search ADS PubMed 33. Wilson WH, Young RM, Schmitz R et al. Targeting B cell receptor signaling with ibrutinib in diffuse large B cell lymphoma. Nat Med . 2015; 21( 8): 922– 926. Google Scholar CrossRef Search ADS PubMed 34. Hoang-Xuan K, Taillandier L, Chinot O et al. ; European Organization for Research and Treatment of Cancer Brain Tumor Group. Chemotherapy alone as initial treatment for primary CNS lymphoma in patients older than 60 years: a multicenter phase II study (26952) of the European Organization for Research and Treatment of Cancer Brain Tumor Group. J Clin Oncol . 2003; 21( 14): 2726– 2731. Google Scholar CrossRef Search ADS PubMed 35. Morin RD, Mungall K, Pleasance E et al. Mutational and structural analysis of diffuse large B-cell lymphoma using whole-genome sequencing. Blood . 2013; 122( 7): 1256– 1265. Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Neuro-Oncology. All rights reserved. For permissions, please e-mail: firstname.lastname@example.org This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
Neuro-Oncology – Oxford University Press
Published: Feb 8, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera