In-Depth Genetic Analysis of Sclerosing Epithelioid

Published OnlineFirst September 22, 2017; DOI: 10.1158/1078-0432.CCR-17-1856

Biology of Human Tumors Clinical Cancer Research In-depth Genetic Analysis of Sclerosing Epithelioid Fibrosarcoma Reveals Recurrent Genomic Alterations and Potential Treatment Targets Elsa Arbajian1, Florian Puls2, Cristina R. Antonescu3, Fernanda Amary4, Raf Sciot5, Maria Debiec-Rychter6, Vaiyapuri P. Sumathi7, Marcus Jara€ s1, Linda Magnusson1, Jenny Nilsson1, Jakob Hofvander1, and Fredrik Mertens1,8

Abstract

Purpose: Sclerosing epithelioid fibrosarcoma (SEF) is a highly line; these cells were subsequently analyzed by RNA-seq, and aggressive soft tissue sarcoma closely related to low-grade fibro- expression of the CD24 protein was assessed by FACS analysis. myxoid sarcoma (LGFMS). Some tumors display morphologic Results: The SNP array analysis detected a large number of characteristics of both SEF and LGFMS, hence they are known as structural aberrations in SEF and SEF/LGFMS, many of which hybrid SEF/LGFMS. Despite the overlap of gene fusion variants were recurrent, notably DMD microdeletions. RNA-seq identified between these two tumor types, SEF is much more aggressive. The FUS-CREM and PAX5-CREB3L1 as alternative fusion genes in one current study aimed to further characterize SEF and hybrid SEF/ SEF each. CD24 was strongly upregulated, presumably a direct LGFMS genetically to better understand the role of the character- target of the fusion proteins. This was further confirmed by the istic fusion genes and possible additional genetic alterations in gene expression analysis and FACS analysis on Tet-On 3G cells tumorigenesis. expressing EWSR1-CREB3L1. Experimental Design: We performed whole-exome sequenc- Conclusions: Although gene fusions are the primary tumori- ing, SNP array analysis, RNA sequencing (RNA-seq), global gene genic events in both SEF and LGFMS, additional genomic changes expression analyses, and/or IHC on a series of 13 SEFs and 6 explain the differences in aggressiveness and clinical outcome hybrid SEF/LGFMS. We also expressed the FUS-CREB3L2 and between the two types. CD24 and DMD constitute potential EWSR1-CREB3L1 fusion genes conditionally in a fibroblast cell therapeutic targets. Clin Cancer Res; 23(23); 7426–34. 2017 AACR.

Introduction location, no strong predictive markers of metastasis have been identified (1–5). MUC4 is a useful IHC marker for diagnosis (4). Sclerosing epithelioid fibrosarcoma (SEF) is a malignant soft In some cases, more or less extensive areas reminiscent of low- tissue tumor that preferentially affects middle-aged and elderly grade fibromyxoid sarcoma (LGFMS) can be seen, either in individuals. It is highly aggressive, with high rates (>50%) of local synchronous or metachronous combination with SEF areas; such recurrence and distant spreading; apart from tumor size and tumors are known as hybrid SEF/LGFMS (1, 5). LGFMS is a morphologically low-grade fibroblastic tumor that typically affects younger adults and children. Also, LGFMS cells are positive for MUC4, in accordance with the fact that it was the most strongly 1Division of Clinical Genetics, Department of Laboratory Medicine, Lund Uni- differentially expressed gene compared with various histological- 2 versity, Lund, Sweden. Department of Pathology and Clinical Genetics, Sahl- ly similar tumors (6, 7). Although it has a much lower potential 3 grenska University Hospital, Gothenburg, Sweden. Department of Pathology, for recurrence (10%) and metastasis (5%) than SEF within the first Memorial Sloan Kettering Cancer Center, New York, New York. 4Department of Histopathology, Royal National Orthopaedic Hospital, Stanmore, United King- 5 years, it is notorious for late occurring metastases (8, 9). dom. 5Department of Pathology, KU Leuven and University Hospitals, Leuven, The intriguing morphologic and clinical overlaps and differ- Belgium. 6Center for Human Genetics, KU Leuven and University Hospitals, ences between SEF and LGFMS could perhaps be explained by Leuven, Belgium. 7Department of Musculoskeletal Pathology, Royal Orthopae- shared and distinct genetic features, respectively. LGFMS is by far 8 dic Hospital NHS Foundation Trust, Birmingham, United Kingdom. Department the more extensively analyzed subtype of the two. Around 95% of fi of Clinical Genetics, Of ce for Medical Services, Division of Laboratory Medicine, the cases show a FUS-CREB3L2 fusion gene, typically on the basis Lund, Sweden. of a balanced translocation t(7;16)(q33;p11) (10, 11). The Note: Supplementary data for this article are available at Clinical Cancer remaining cases display either a t(11;16)(p11;p11), resulting in Research Online (http://clincancerres.aacrjournals.org/). a FUS-CREB3L1 fusion (12) or, in rare cases, an EWSR1-CREB3L1 Corresponding Author: Elsa Arbajian, Lund University Hospital, Klinikgatan 28 fusion (13, 14). Cytogenetic and genomic array data have iden- BMC C13, Lund 22184, Sweden. Phone: 467-0025-3200; E-mail: tified few additional changes, suggesting that the fusion gene is [email protected] both necessary and sufficient for tumorigenesis, and global gene doi: 10.1158/1078-0432.CCR-17-1856 expression analysis has identified a distinct profile (7, 10). The 2017 American Association for Cancer Research. genetic information on SEF is limited to data on fusion gene

7426 Clin Cancer Res; 23(23) December 1, 2017

Genetic Characterization of SEF

addition, tumors from 18 patients with LGFMS were included for Translational Relevance comparison of gene fusion status, GGE profiles, genomic imbal- Sclerosing epithelioid fibrosarcoma (SEF) is a highly malig- ances, and/or whole-exome mutation profiles. Finally, 40 other nant soft tissue sarcoma; although patients with localized soft tissue tumors, 11 myxofibrosarcomas (MFS), six desmoid tumors may be cured by surgery, the prognosis for patients tumors (DFM), 13 solitary fibrous tumors (SFT), and six extra- with disseminated disease is dismal. Previous studies have skeletal myxoid chondrosarcomas (EMC), were used for compar- shown that most SEFs display gene fusions that, to some isons of GGE profiles. For filtering of mutations at WES, mutation extent, overlap with those found in low-grade fibromyxoid data on five simultaneously analyzed cases of ossifying fibromyx- sarcoma (LGFMS), a tumor type associated with much better oid tumor were used. All tumors were diagnosed according to outcome than SEF. Here we show that the clinical differences established criteria (19). Clinical data and information on the between SEF and LGFMS can largely be explained by complex analyses performed in each case are summarized in Supplemen- genomic rearrangements, including recurrent intragenic dele- tary Table S1. tions of the DMD gene encoding dystrophin. Furthermore, The studies were conducted in accordance with the Declaration both in vivo and in vitro, the EWSR1-CREB3L1 fusion, which of Helsinki. All studies were performed after approval by the predominates in SEF, leads to higher expression of the CD24 Institutional Review Board and after obtaining written consent gene than does the FUS-CREB3L2 fusion, which predominates from the subjects. in LGFMS. Both DMD and CD24 constitute promising treatment targets. IHC CD24 IHC on formalin-fixed paraffin-embedded (FFPE) tissue was performed following pressure cooker epitope retrieval (Novo- castra Epitope Retrieval Solution pH 6), using a rabbit polyclonal status, revealing a predominance (80%–90% of the cases) of antibody raised against KLH-conjugated synthetic peptide derived EWSR1-CREB3L1 (15–17), but occasionally showing recurrent from human CD24 (dilution 1:100, overnight incubation at 4 C, alternative EWSR1-CREB3L2 and FUS-CREB3L2 fusions (17, 18). Biorbyt Ltd.). The EnVision Dual Link system (Dako) was used for Furthermore, a similar FUS-CREB3L2 fusion as seen in LGFMS has visualization. been also found in most hybrid SEF/LGFMS cases (4, 15, 17), but some cases may display rearrangements of EWSR1 and/or RNA-seq CREB3L1 (4). RNA was extracted from 12 fresh-frozen tumor samples and To understand better why SEF and LGFMS show different from 25 cell lines. mRNA libraries were prepared for sequencing clinical phenotypes and why some LGFMS recur as SEF, we using the TruSeq RNA Sample Preparation Kit v2 (Illumina), as analyzed pure SEFs and hybrid SEF/LGFMS, using high-resolution described previously (20). Paired-end 151 bp reads were gener- SNP arrays, global gene expression (GGE) profiling, transcrip- ated from the mRNA libraries on a NextSeq 500 (Illumina). From fi tome sequencing [RNA sequencing (RNA-seq)], and whole- another ve tumors, mRNA with DV200 values 40 could be exome sequencing (WES). In addition, the impact of the FUS- extracted from FFPE blocks using Qiagen's RNeasy FFPE Kit CREB3L2 and EWSR1-CREB3L1 chimeras was further investigated (Qiagen), as described previously (21). mRNA libraries were in vitro using the Tet-On 3G inducible gene expression system. prepared from 20 to 50 ng of RNA, depending on the DV200 value, using the capturing chemistry of the TruSeq RNA Access Library Prep Kit (Illumina). Paired-end 85 nt reads were generated Materials and Methods from the mRNA libraries on a NextSeq 500 (Illumina). Sectioning, Patients and tumors RNA extraction, library preparation, and sequencing were per- The study included 15 tumors from 13 patients classified as formed as described previously (21). pure SEF and six tumors from 4 patients with hybrid SEF/LGFMS; ChimeraScan and FusionCatcher, using default settings, were of the latter six tumors, five showed SEF morphology and one used to identify candidate fusion transcripts from the sequence LGFMS morphology. The histologic appearance of case 17, with data (22, 23). The GRCh37/hg19 build was used as the human synchronous SEF and LGFMS areas, can be seen in Fig. 1. In reference genome.

Figure 1. Histologic appearances of hybrid SEF/ LGFMS with FUS-CREB3L2 fusion (case 17). A, Low magniﬁcation with abrupt transition from LGFMS (left) to SEF (right). B, High magniﬁcation appearances of increasing cellularity, atypia, and decreasing extracellular matrix during progression of LGFMS to SEF (left to right) within the same lesion.

www.aacrjournals.org Clin Cancer Res; 23(23) December 1, 2017 7427

Arbajian et al.

For gene expression studies, data were normalized using Tet-On 3G setup Cufflinks with default settings (24). Correlation-based principal An hTert immortalized Bj5ta human fibroblast cell line component analysis (PCA) and hierarchical clustering analysis (cultured in 4:1 DMEM: 199 medium with 10% Tet-approved were performed using the Qlucore Omics Explorer version 3.2 FBS þ 0.01 mg/mL hygromycin B) was used throughout the (Qlucore AB). Differences between cell clones in log2-trans- experiments. Bj5ta was sequentially transduced with the pLVX- formed expression data were calculated using a t test, and correc- Tet3G vector encoding the regulator Tet-On 3G protein and tions for multiple testing were based on the Benjamini–Hochberg with one of the response plasmids pLVX-TRE3G-FUS-CREB3L2, method (Qlucore AB). Genes with P < 0.01 and an FDR < 0.07 pLVX-TRE3G-EWSR1-CREB3L1, pLVX-TRE3G-EWSR1-NR4A3, were considered significantly altered. or pLVX-TRE3G-EMPTY, encoding the different fusion genes of interest. Expression of the inserted gene fusion was turned on RT-PCR analyses to confirm gene fusions by addition of doxycycline to the culture medium. The EWSR1- The same RNA that had been used for RNA-seq was reverse NR4A3 expressing cell line was used for comparison in subse- transcribed and PCR amplified as described previously (7, 25). quent expression studies (see Supplementary Methods for Primers specific for FUS, EWSR1, PAX5, CREB3L1, CREB3L2, and further details). CREM were designed to detect fusion transcripts (Supplementary Table S2). Transcripts were amplified using an initial denaturation Flow cytometry for 5 minutes at 94C, followed by 30 cycles of 1 minute at 94C, 1 To evaluate the expression of the CD24 protein in the fusion minute at 58C, and 2 minutes at 72C, and a final extension for 5 gene-expressing Tet-On 3G transduced cells cultured with or minutes at 72C. Amplified fragments were purified from agarose without doxycycline, flow cytometry was performed (see Supple- gels and directly sequenced using the Big Dye v1.1 Cycle Sequenc- mentary Methods for further details). ing Kit (Applied Biosystems) on an ABI-3130 genetic analyzer (Applied Biosystems). The BLASTN software (http://www.ncbi. Results nlm.nih.gov/blast) was used for the analysis of sequence data. Identification of SEF-specific expression patterns At unsupervised PCA, the data were variance filtered until the SNP array analysis and WES 63 samples formed clusters that corresponded to the different DNA was extracted from eight fresh-frozen tumor samples (six tumor types. In this setting (variance ratio, F ¼ 0.4, 882 genes), the pure SEF, two hybrid SEF/LGFMS) and prepared for SNP array LGFMS and SEF groups were clearly different from the other analysis using the Affymetrix Cytoscan HD array (Affymetrix), as groups and appeared most similar to each other and then to the described previously (26). From another four tumor samples (one MFS group. The filtered dataset was then subjected to ANOVA pure SEF, three hybrid SEF/LGFMS), DNA was extracted from tests, generating 424 genes with a significant (P < 0.01) differential FFPE blocks using QIAamp DNA FFPE Tissue Kit and prepared for expression pattern between SEF and the control tumors; a second SNP array analysis using the Oncoscan CNV array (Affymetrix), ANOVA test generated 375 genes with a significant (P < 0.01) according to the manufacturer's instructions. The position of the differential expression pattern between SEF and LGFMS. In the SNPs was based on the GRCh37/hg19 sequence assembly. SNP PCA, the data were then subjected to one-sided t tests to extract the array analysis was done using the Affymetrix Chromosome Anal- most up- and downregulated genes. The cluster of the 100 most ysis Suite (ChAS) software. The SNP array data on SEF and hybrid upregulated transcripts in SEF versus controls and SEF versus SEF/LGFMS cases were compared to previously published, and LGFMS are listed in Supplementary Table S3. here reevaluated, SNP array data on 10 cases of LGFMS (7) that CD24 was the most upregulated gene (80) in SEF (Fig. 2A). had been analyzed using the Illumina HumanOmni-Quad ver- Consistent with this finding, IHC for CD24 on SEF cases showed sion 1.0 array (Illumina). The LGFMS data were converted to the cytoplasmic and membranous staining (Fig. 2C). CD24 was also hg19 genome assembly using UCSC LiftOver. upregulated (37) in LGFMS, in accordance with the findings by See Supplementary Methods for details regarding WES. Moller€ and colleagues (7). Also, MUC4 was upregulated in both LGFMS and SEF (Supplementary Fig. S1). Global gene expression microarray analysis The two hybrid SEF/LGFMS cases appeared to be most closely RNA from 10 tumors (eight pure SEF, two hybrid SEF/LGFMS) related to each other and then to pure SEF cases. Among pure SEF, fi was of suf cient quality for GGE analysis using the Human EWSR1-CREB3L1- and EWSR1-CREB3L2–positive tumors did not GeneChip Gene 1.0 ST Array (Affymetrix). Extraction of total cluster separately, whereas the single case with a FUS-CREM RNA from frozen tumor biopsies, RNA concentration and quality fusion clustered further apart from the other SEF cases. measurements, and hybridization were carried out as described previously (27). Gene expression data were normalized, back- Fusion gene status ground-corrected, and summarized by using the Robust Multi- The gene fusion status of the SEF and hybrid SEF/LGFMS cases chip Analysis algorithm implemented in the Expression Console was assessed through RNA-seq, SNP array, RT-PCR, and/or FISH. version 1.4 software (Affymetrix). These data were compared with Among the 13 cases of pure SEF, eight were positive for EWSR1- previously published (7) gene expression data from 17 LGFMS, 11 CREB3L1, two for EWSR1-CREB3L2, and one each for FUS-CREM, MFS, six DFM, 13 SFT, and 6 EMC. Correlation-based PCA and FUS-CREB3L2, and PAX5-CREB3L1. All six samples, including five hierarchical clustering analysis were performed using the Qlucore with SEF morphology and one with LGFMS morphology, from Omics Explorer version 3.2 (Qlucore AB). Differences between four cases of hybrid SEF-LGFMS displayed a FUS-CREB3L2 fusion tumor groups in log2-transformed expression data were calculated (Supplementary Table S1). The previously undescribed FUS- using a t test, and corrections for multiple testing were based on CREM and PAX5-CREB3L1 transcripts were confirmed by RT-PCR, the Benjamini–Hochberg method (Qlucore AB). Genes with P < confirming the presence of an inframe fusion of FUS ex 11 with 0.01 and FDR < 0.07 were considered significantly altered. CREM ex 7 and PAX5 ex 2 with CREB3L1 ex 6, respectively.

7428 Clin Cancer Res; 23(23) December 1, 2017 Clinical Cancer Research

Genetic Characterization of SEF

Figure 2. CD24 expression. A, By global gene expression, CD24 is found to be highly upregulated in SEF (80)andin LGFMS (37) as compared with a set of control tumors (left box plot). In line with that, Tet-On 3G cells expressing the EWSR1-CREB3L1 fusion (E-C Dox cells) and Tet-On 3G cells expressing the FUS-CREB3L2 fusion (F-C Dox cells) also show upregulation of CD24 (125 and 28, respectively) as compared with the No Dox counterpart and the controls (right box plot). B, CD24 FACS on Tet-On 3G cells expressing EWSR1-CREB3L1 (E-C Dox) and FUS-CREB3L2 (F-C Dox) was well in line with the RNA-seq results visualized in the box plots above. Cell surface CD24 expression was 42 higher, and total CD24 expression was 50 higher in E-C Dox (red line) than in E-C No Dox (blue line) in the cell clone illustrated. C, IHC showed cytoplasmic and membranous staining of CD24 in cases 5 (left) and 6A (right).

SNP array rearrangements; only 11 of 197 aberrations affected whole chro- Detailed information on all imbalances exceeding 100 kb, as mosomes. Several imbalances were recurrent: loss of several well as smaller (10 kb) homozygous deletions and imbalances regions in 22q12-qter (minimal overlapping regions 38.12– affecting genes involved in fusions, is given in Supplementary 40.93 Mb, 41.78–42.56 Mb, 44.97–49.46 Mb, 49.92–50.26 Mb, Table S4 and visualized in Fig. 3. In all cases, the proﬁles were and 50.85–51.15 Mb) in six cases, loss of 11p (1–45.96 Mb) and compatible with a near-diploid chromosome count. proximal 22q (16.89–23.05 Mb) in ﬁve cases each, and gain of Using the 100 kb cutoff, imbalances were found in 7 of 10 1q31-qter (188.32–248.84 Mb), loss of 10q21 (56.18–56.65 LGFMS cases (mean, 1.4; range, 0–3). The only recurrent (2–3 Mb), 13q21 (71.31–72.01 Mb), 22q12 (29.29–29.68 Mb), and cases) aberrations were gain of material from chromosomes/arms Xp21 (32.48–32.70 Mb) in three cases each. High-level (5 7q, 12, and 15q. Three and four cases, respectively, displayed copies) gain was rare, involving three different regions in three microdeletions involving the FUS and CREB3L2 loci. cases. Complete loss (0 alleles) was seen in four cases, affecting the All seven samples from pure SEF showed multiple imbalances DMD gene in Xp21 in two of them. One or more copy number (mean, 28; range, 6–78). Most imbalances were due to structural transitions in the EWSR1, CREB3L1, and CREB3L2 loci, typically

www.aacrjournals.org Clin Cancer Res; 23(23) December 1, 2017 7429

Arbajian et al.

Figure 3. SNP array results. Circos plot showing SNP array results in LGFMS (inner 10 circles, light gray), hybrid SEF/LGFMS (middle ﬁve circles, gray), and pure SEF (outer seven circles, dark gray). Gains, losses, and copy-neutral loss of heterozygosity exceeding 100 kb, as well as smaller homozygous deletions or deletions affecting genes involved in fusions, are indicated.

deleting the parts not included in the fusion genes, were seen in Two imbalances were present in all 3 patients with hybrid five, four, and one case, respectively. The single case with a SEF/LGFMS: loss and/or loss of heterozygosity (LOH) for the FUS-CREM fusion showed copy number transitions in both loci. entire 11p, and loss of one copy of chromosome 22. In addition, All five samples from 3 patients with hybrid SEF/LGFMS both samples from both cases 15 and 17 showed homozygous showed multiple imbalances (mean, 20; range, 17–22). The two deletions of the DMD gene (Fig. 4). All three cases showed copy samples from case 15 (a local recurrence and a metastasis) had less number shifts in the FUS and CREB3L2 loci. than 15% of the imbalances in common, whereas the two samples In summary, the SNP array profiles of SEF and hybrid SEF/ from the primary tumor of case 17 (one showing SEF morphology LGFMS were very similar to each other and much more complex and one LGFMS morphology) shared >80% of the aberrations. than those in LGFMS.

7430 Clin Cancer Res; 23(23) December 1, 2017 Clinical Cancer Research

Genetic Characterization of SEF

Figure 4. DMD deletions in case 15. Homozygous DMD deletions (i.e., 0 copies), detected by SNP array analysis in cases 15A and B (top and bottom, respectively), illustrate convergent clonal evolution; two separate events target the same gene. Case 15A harbors a 130-kb deletion; case 15B harbors two deletions of 172 and 257 kb each.

Mutation pattern in tumors CREB3L1 and FUS-CREB3L2, respectively, these chimeras are so WES on five SEF and two hybrid SEF/LGFMS achieved a mean closely related that they could be predicted to have similar coverage >100, with 99% of the targeted bases covered to a downstream effects. Indeed, the comprehensive genetic analy- depth of >20. Filtering of the data resulted in 260 potential ses presented here strongly suggest that the poor outcome for mutations (Supplementary Table S5). Targeted DNA sequencing SEF patients could largely be accounted for by extensive sec- of the tumor samples, with a median depth of 204 reads per ondary genomic alterations. Although LGFMS has few aberra- amplicon, confirmed 72 of 125 tested variants, of which 50 were tions at the genome or nucleotide level in addition to the unique mutations (Supplementary Table S5). No gene was con- characteristic FUS-CREB3L2 chimera, SEF consistently displays sistently involved. complex genomic alterations. Strong support for the relative WES was also performed on five cases of LGFMS with paired importance for outcome of genomic imbalances compared blood samples, achieving a mean coverage 85. One to 12 single- with gene fusion status was derived from the hybrid SEF/ nucleotide variants (SNV) and 0 to 1 insertion/deletion were LGFMS cases. These tumors have the same FUS-CREB3L2 fusion found per case (Supplementary Table S5). None of the mutations as LGFMS, but display the same types of genomic alterations was recurrent. Two samples from a hybrid SEF/LGFMS case were and show the same poor outcome as pure SEF. analyzed: 17A showing LGFMS morphology and 17B showing Point mutations seem to have little impact on the clinical SEF morphology. This resulted in 7 SNVs each (Supplementary outcome, as no consistent mutations were identified, in either Table S6). SEF, LGFMS, or hybrid SEF/LGFMS. WES on LGFMS, performed with paired blood samples, showed no recurrent mutation and no Cells with stable expression of FUS-CREB3L2, recurrently mutated gene. The results of WES on SEF and SEF/ EWSR1-CREB3L1, or EWSR1-NR4A3 LGFMS, which resulted in 260 mutations, were harder to interpret Of the 100 most highly expressed genes in SEF tumors com- because no paired blood sample was available, and we can fi pared with controls, 38 were also signi cantly upregulated (fold therefore not rule out that some of the detected mutations were > P < change 2, 0.05) in Tet-On 3G EWSR1-CREB3L1 cells cultured constitutional. CD24 with doxycycline (Supplementary Table S6). Among these, Some of the recurrent genomic imbalances in SEF were was upregulated 135 (Fig. 2A). CD24 was also upregulated related to the underlying gene fusion, notably deletions of (28 ) in TetOn 3G FUS-CREB3L2 cells cultured with doxycycline. 11p (where CREB3L1 is located) and 22q (where EWSR1 is CD24 protein expression in the transduced cells cultured with located). However, at least the 11p deletions could not simply fl or without doxycycline and in Bj5ta cells was assessed by ow be considered side effects of translocations giving rise to the cytometry. TetOn 3G EWSR1-CREB3L1 cells cultured with doxy- EWSR1-CREB3L1 chimera; copy-neutral LOH or deletions of fi cycline showed signi cant upregulation of cell surface CD24 (9 ) 11p were found also in SEFs without CREB3L1 fusions. Loss/ and of total CD24 (50 ), as compared with the no doxycycline LOH of 11p is frequently seen in many other sarcomas, notably counterparts (Fig. 2B). TetOn 3G FUS-CREB3L2 cells cultured embryonal rhabdomyosarcomas (26). The expression pattern with doxycycline showed a 7 upregulation of cell surface CD24 of imprinted genes in 11p15 (Supplementary Fig. S1B) indi- and a 3 upregulation of total CD24 compared with the no cates that the maternal copy of 11p is lost in SEF. However, the doxycycline counterparts (Supplementary Fig. S2). correlation between 11p deletion and gene expression is unclear; cases without 11p deletion displayed the same pattern Discussion of expression and only two genes in 11p, PAMR1 and ADM, On the basis of the limited genetic information previously were among the 100 most downregulated genes in SEF versus available on SEF and LGFMS, the prognostic difference between controls (Supplementary Table S3C). the two tumor types was difficult to explain; although two Another link with myogenic sarcomas was the frequent finding different gene fusions predominate in these tumors, EWSR1- of intragenic deletions of the DMD gene, encoding the dystrophin

www.aacrjournals.org Clin Cancer Res; 23(23) December 1, 2017 7431

Arbajian et al.

protein that is causally involved in inherited muscular dystro- (30, 31). Indeed, according to numerous studies, several tumor phies. Although no such aberrations were seen in LGFMS, 2 of 7 types display increased expression of CD24, typically at the pure SEF and 2 of 3 hybrid SEF/LGFMS with SNP array data protein level, and increased expression is often associated with showed DMD deletions of variable size (Supplementary Table more aggressive phenotype, increased risk of metastasis develop- S4). The deletions always started after exon 1 and extended no ment, and worse clinical outcome (31, 32). longer than intron 74. In all cases, the deletion on the X-chro- It was recently shown that CD24 is most abundant intracellu- mosome was quite distinct, targeting only the DMD locus, imply- larly, where it can be found both in the cytoplasm and in the ing that these deletions are not by chance, but selected for during nucleus (33). This is in agreement with our IHC results on SEF tumor growth. From 2 of the patients with hybrid SEF/LGFMS, we showing both cytoplasmic and membranous staining (Fig. 2C), had two samples each with DMD deletions. In case 15, the and with FACS data on EWSR1-CREB3L1–expressing cells (Fig. deletions were different in the local recurrence (15A) and the 2B). When introducing ectopic expression of CD24 in an osteo- metastasis (15B), strongly arguing for late and independent sarcoma cell line, it was found that CD24 inhibits the binding of mutations in the two samples, while identical deletions were ARF to nucleophosmin (NPM1), a key regulator of mitosis. In seen in the two portions from a primary hybrid SEF/LGFMS, one turn, this increased the levels of MDM2 and decreased the levels of showing LGFMS morphology (case 17A) and one SEF morphol- TP53 and its target CDKN1A (33). Here, we could show that CD24 ogy (17B). In case 17, the deletion must therefore have occurred is transcriptionally activated also in the fibroblast cell lines relatively early and cannot explain the morphologic transition expressing EWSR1-CREB3L1 and FUS-CREB3L2. Interestingly, the from LGFMS to SEF. No effect on the transcription levels of the CD24 expression was higher in EWSR1-CREB3L1 expressing cells, DMD gene was seen at global gene expression profiling, which in line with the higher expression levels in SEF samples compared strongly suggests that a truncated mRNA is expressed, in keeping with LGFMS (Fig. 2A). The high expression of CD24 in both with the fact that none of the deletions involved the entire coding LGFMS and SEF has promising therapeutic implications. SEF is a sequence. highly malignant tumor, as exemplified by the currrent study in Similar intragenic DMD deletions were recently described in which 14 of 17 patients with SEF or hybrid SEF/LGFMS developed advanced sarcomas showing myogenic differentiation (28). On metastases and/or died of disease. CD24 has been suggested as a the basis of various types of supportive data, the authors con- potential treatment target in several other tumor types. For cluded that intragenic DMD deletions, abrogating expression of instance, in both in vitro and mouse models of colorectal, pan- full-length dystrophin, are an important mechanism behind creatic, and bladder cancer, silencing of CD24 through siRNA or tumor progression in myogenic sarcomas (28). As far as we are an mAB resulted in reduced tumor growth and/or reduced risk for aware, SEF does not show any myogenic differentiation, which lung metastases (34, 35). combined with the occasional finding of DMD deletions also in In addition to CD24 and MUC4, 36 more genes among the 100 other malignancies, such as malignant melanomas, suggests that a top upregulated genes in SEF tumors were also significantly tumor suppressor function of DMD is not limited to myogenic upregulated in Tet-On 3G cells expressing EWSR1-CREB3L1 cells. The DMD locus encodes a variety of protein isoforms, some (Supplementary Table S6). Thus, in spite of the extensive genomic of which are expressed by nonmuscle cells. Notably, the Dp71 alterations in SEF, a significant part of the characteristic gene protein, encoded by exons in the 30-end of DMD, is involved in expression profile can be recapitulated simply by expressing the many cellular processes, including cell adhesion, mitosis, and underlying gene fusion, providing strong evidence for its impor- nuclear architecture (29). This notwithstanding, the finding of tance in SEF tumorigenesis. DMD deletions in yet another highly malignant type of sarcoma Detection of fusion genes is important for clinical, diagnostic provides further incentive for evaluating means to restore dystro- purposes (36). It is hence of interest that we here detected some phin expression as a treatment option. Another possible scenario novel fusions in SEF. One MUC4-positive SEF (case 11) showed is that DMD does not act as a tumor suppressor in nonmyogenic an inframe fusion of exon 11 of FUS with exon 7 of CREM at RNA- cells and that instead, the remaining translated part of DMD Seq; the fusion was verified by RT-PCR, and SNP array analysis should be the focus in these cases. revealed copy number transitions in both loci (Supplementary Also, GGE identified a potential therapeutic target: CD24. Tables S1 and S4). Interestingly, this case had the lowest mRNA Indeed, CD24 was the top upregulated gene in SEF compared expression of MUC4 of all SEFs, in line with a scattered staining of with the control tumors, showing on average 80 higher expres- tumor cells (Supplementary Figure S1A). The breakpoint in FUS sion (Fig. 2A). We previously showed that CD24 was among the was unusually distal, but breakpoints as distal as exon 14 have most strongly differentially expressed genes also in LGFMS and been described in other FUS fusions (37). CREM (cAMP-respon- that the high transcriptional activity of CD24 could be directly sive element modulator) belongs to a large family of basic leucine linked to the presence of chimeric CREB3L transcription factors zipper (bZIP)-containing proteins, including also CREB3L1 and (7). The CD24 locus harbors several CRE half-sites (TGACG), and CREB3L2. The b-ZIP domain of CREM, which is encoded by exons in vitro studies showed that truncated FUS/CREB3L2 and 7–8 and thus retained in the fusion chimera, preferentially binds CREB3L2 constructs in HEK 293 cells activated transcription of to palindromic CRE (50-TGACGTCA-30) or half-CRE (50-TGACG- CD24 (7). 30 or 50-CGTCA-30) sites in promoters of target genes. However, CD24 encodes a mucin-like protein that is heavily glycosylated studies using chromatin immunoprecipitation followed by and may be found both in the cell membrane and in the cyto- sequencing have revealed that CREM occupies promoters of plasm. When CD24 is expressed at the cell surface, it is known that thousands of genes, that more than 15% of the binding sites do P-selectin can act as a ligand. This binding has been shown to not contain CRE or half-CRE sites, and that binding sites vary in a affect the adherence of CD24-expressing lymphocytes to activated cell-specific way (38). thrombocytes and endothelial cells. Possibly, this mechanism is Another MUC4-positive pure SEF had a PAX5-CREB3L1 fusion; also of importance for lung colonization of solid tumor cells the breakpoint in CREB3L1 (exon 6) was similar to that in the

7432 Clin Cancer Res; 23(23) December 1, 2017 Clinical Cancer Research

Genetic Characterization of SEF

more common EWSR1-CREB3L1 fusions. PAX5, which has not Analysis and interpretation of data (e.g., statistical analysis, biostatistics, before been reported in sarcoma-associated gene fusions, is a computational analysis): E. Arbajian, C.R. Antonescu, M. Debiec-Rychter, € recurrent 50-partner to many different 30-genes in acute lympho- M. Jaras, J. Hofvander, F. Mertens PAX5 Writing, review, and/or revision of the manuscript: E. Arbajian, F. Puls, blastic leukemia, where the breakpoints in range from exon C.R. Antonescu, R. Sciot, M. Debiec-Rychter, V.P. Sumathi, J. Hofvander, 4 to 9; the cellular consequences of these translocations are not F. Mertens fully understood (39). Administrative, technical, or material support (i.e., reporting or organizing In conclusion, we provide evidence for the respective fusion data, constructing databases): E. Arbajian, L. Magnusson, J. Nilsson, F. Mertens genes in SEF and LGFMS being the main tumorigenic drivers. We Study supervision: E. Arbajian, F. Mertens also show that additional genomic imbalances found in SEF and SEF/LGFMS can account for the significant difference in clinical Acknowledgments outcome between the highly aggressive SEF and SEF/LGFMS and The authors thank Dr. Philippe Taniere, University Hospitals Birmingham, LGFMS, morphologically low grade but notorious for late metas- for providing tumor material. tases. Also, CD24 and DMD were identified as potential therapeutic targets for SEF. Grant Support This work was supported by the Swedish Cancer Society, the Swedish fl Disclosure of Potential Con icts of Interest Childhood Cancer Foundation, and Region Skane (to F. Mertens); P50 No potential conflicts of interest were disclosed. CA140146-01 and P30-CA008748 (to C.R. Antonescu). The costs of publication of this article were defrayed in part by the Authors' Contributions payment of page charges. This article must therefore be hereby marked advertisement Conception and design: E. Arbajian, F. Mertens in accordance with 18 U.S.C. Section 1734 solely to indicate Development of methodology: E. Arbajian, F. Puls, J. Hofvander this fact. Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): F. Puls, C.R. Antonescu, F. Amary, R. Sciot, Received June 29, 2017; revised August 1, 2017; accepted September 15, 2017; M. Debiec-Rychter published OnlineFirst September 22, 2017.

References 1. Meis-Kindblom JM, Kindblom LG, Enzinger FM. Sclerosing epithelioid ization of low-grade fibromyxoid sarcoma, and cloning of a novel FUS/ fibrosarcoma. A variant of fibrosarcoma simulating carcinoma. Am J Surg CREB3L1 fusion gene. Lab Invest 2005;85:408–15. Pathol 1995;19:979–93. 13. Lau PP, Lui PC, Lau GT, Yau DT, Cheung ET, Chan JK. EWSR1-CREB3L1 2. Antonescu CR, Rosenblum MK, Pereira P, Nascimento AG, Woodruff gene fusion: a novel alternative molecular aberration of low-grade fibro- JM. Sclerosing epithelioid fibrosarcoma: a study of 16 cases and con- myxoid sarcoma. Am J Surg Pathol 2013;37:734–8. firmation of a clinicopathologically distinct tumor. Am J Surg Pathol 14. Rubinstein JC, Visa A, Zhang L, Antonescu CR, Christison-Lagay ER, 2001;25:699–709. Morotti R. Primary low-grade fibromyxoid sarcoma of the kidney in a 3. Weiss SW, Goldblum J. Soft tissue tumors. St Louis, MO: Mosby; 2001. child with the alternative EWSR1-CREB3L1 gene fusion. Pediatr Dev Pathol 4. Doyle LA, Wang WL, Dal Cin P, Lopez-Terrada D, Mertens F, Lazar AJ, et al. 2014;17:321–6. MUC4 is a sensitive and extremely useful marker for sclerosing epithelioid 15. Arbajian E, Puls F, Magnusson L, Thway K, Fisher C, Sumathi VP, et al. fibrosarcoma: association with FUS gene rearrangement. Am J Surg Pathol Recurrent EWSR1-CREB3L1 gene fusions in sclerosing epithelioid fibro- 2012;36:1444–51. sarcoma. Am J Surg Pathol 2014;38:801–8. 5. Kindblom L-G, Mertens F, Coindre J-M, Hornick JL, Meis JM. Sclerosing 16. Stockman DL, Ali SM, He J, Ross JS, Meis JM. Sclerosing epithelioid epithelioid fibrosarcoma. In: Fletcher CDM, Bridge JA, Hogendoorn PCW, fibrosarcoma presenting as intraabdominal sarcomatosis with a novel Mertens F, editors. WHO Classification of Tumours of Soft Tissue and Bone. EWSR1-CREB3L1 gene fusion. Hum Pathol 2014;45:2173–8. Lyon, France: IARC Press; 2013. p. 97–8. 17. Prieto-Granada C, Zhang L, Chen HW, Sung YS, Agaram NP, Jungbluth AA, 6. Doyle LA, Moller€ E, Dal Cin P, Fletcher CDM, Mertens F, Hornick JL. MUC4 et al. A genetic dichotomy between pure sclerosing epithelioid fibrosar- is a highly sensitive and specific marker for low-grade fibromyxoid sarco- coma (SEF) and hybrid SEF/low-grade fibromyxoid sarcoma: a pathologic ma. Am J Surg Pathol 2011;35:733–41. and molecular study of 18 cases. Genes Chromosomes Cancer 2015;54: 7. Moller€ E, Hornick JL, Magnusson L, Veerla S, Domanski HA, Mertens F. 28–38. FUS-CREB3L2/L1-positive sarcomas show a specific gene expression pro- 18. Guillou L, Benhattar J, Gengler C, Gallagher G, Ranchere-Vince D, Collin F, file with upregulation of CD24 and FOXL1. Clin Cancer Res 2011; et al. Translocation-positive low-grade fibromyxoid sarcoma: clinicopath- 17:2646–56. ologic and molecular analysis of a series expanding the morphologic 8. Evans HL. Low-grade fibromyxoid sarcoma: a clinicopathologic study spectrum and suggesting potential relationship to sclerosing epithelioid of 33 cases with long-term follow-up. Am J Surg Pathol 2011;35: fibrosarcoma: a study from the French Sarcoma Group. Am J Surg Pathol 1450–62. 2007;31:1387–402. 9. Folpe AL, Hornick JL, Mertens F. Low-grade fibromyxoid sarcoma. In: 19. Fletcher CDM, Bridge JA, Hogendoorn P, Mertens F. WHO Classifi- Fletcher CDM, Bridge JA, Hogendoorn PCW, Mertens F, editors. WHO cation of Tumours of Soft Tissue and Bone. Lyon, France: IARC Press; Classification of Tumours of Soft Tissue and Bone. Lyon, France: IARC 2013. Press; 2013. p. 95–6. 20. Walther C, Tayebwa J, Lilljebjorn H, Magnusson L, Nilsson J, von Steyern 10. Panagopoulos I, Storlazzi CT, Fletcher CDM, Fletcher JA, Nascimento A, FV, et al. A novel SERPINE1-FOSB fusion gene results in transcriptional up- Domanski HA, et al. The chimeric FUS/CREB3L2 gene is specificforlow- regulation of FOSB in pseudomyogenic haemangioendothelioma. J Pathol grade fibromyxoid sarcoma. Genes Chromosomes Cancer 2004;40: 2014;232:534–40. 218–28. 21. Walther C, Hofvander J, Nilsson J, Magnusson L, Domanski HA, Gisselsson 11. Matsuyama A, Hisaoka M, Shimajiri S, Hayashi T, Imamura T, Ishida T, D, et al. Gene fusion detection in formalin-fixed paraffin-embedded benign et al. Molecular detection of FUS-CREB3L2 fusion transcripts in low-grade fibrous histiocytomas using fluorescence in situ hybridization and RNA fibromyxoid sarcoma using formalin-fixed, paraffin-embedded tissue spe- sequencing. Lab Invest 2015;95:1071–6. cimens. Am J Surg Pathol 2006;30:1077–84. 22. Iyer MK, Chinnaiyan AM, Maher CA. ChimeraScan: a tool for identifying 12. Mertens F, Fletcher CDM, Antonescu CR, Coindre J-M, Colecchia M, chimeric transcription in sequencing data. BMC Bioinformatics 2011;27: Domanski HA, et al. Clinicopathologic and molecular genetic character- 2903–4.

www.aacrjournals.org Clin Cancer Res; 23(23) December 1, 2017 7433

Arbajian et al.

23. Nicorici D, Satalan M, Edgren H, Kangaspeska S, Murumagi A, Kallioniemi 31. Kristiansen G, Sammar M, Altevogt P. Tumour biological aspects of CD24, O, et al. FusionCatcher - a tool for finding somatic fusion genes in paired- a mucin-like adhesion molecule. J Mol Histol 2004;35:255–62. end RNA-sequencing data. bioRxiv 2014. Available from: https://www. 32. Lee J, Kim S, Lee E, Kim Y. CD24 overexpression in cancer development and biorxiv.org/content/early/2014/11/19/011650. progression: a meta-analysis. Oncol Rep 2009;22:1149–56. 24. Trapnell C, Williams BA, Pertea G, Mortazavi A, Kwan G, van Baren MJ, 33.WangL,LiuR,YeP,WongC,ChenG-Y,ZhouP,etal.Intracellular et al. Transcript assembly and quantification by RNA-Seq reveals unan- CD24 disrupts the ARF-NPM interaction and enables mutational and notated transcripts and isoform switching during cell differentiation. Nat viral oncogene-mediated p53 inactivation. Nat Commun 2015;6: Biotechnol 2010;28:511–5. 5909. 25. Panagopoulos I, Mertens F, Domanski HA, Isaksson M, Brosjo€ O, Gus- 34. Sagiv E, Starr A, Rozovski U, Khosravi R, Altevogt P, Wang T, et al. Targeting tafson P, et al. No EWS/FLI1 fusion transcripts in giant-cell tumors of bone. CD24 for treatment of colorectal and pancreatic cancer by monoclonal Int J Cancer 2001;93:769–72. antibodies or small interfering RNA. Cancer Res 2008;68:2803–12. 26. Walther C, Mayrhofer M, Nilsson J, Hofvander J, Jonson T, Mandahl N, 35. Overdevest JB, Thomas S, Kristiansen G, Hansel DE, Smith SC, Theodor- et al. Genetic heterogeneity in rhabdomyosarcoma revealed by SNP array escu D. CD24 offers a therapeutic target for control of bladder cancer analysis. Genes Chromosomes Cancer 2016;55:3–15. metastasis based on a requirement for lung colonization. Cancer Res 27. Hallor KH, Sciot R, Staaf J, Heidenblad M, Rydholm A, Bauer HC, et al. Two 2011;71:3802–11. genetic pathways, t(1;10) and amplification of 3p11–12, in myxoinflam- 36. Mertens F, Antonescu CR, Mitelman F. Gene fusions in soft tissue tumors: matory fibroblastic sarcoma, haemosiderotic fibrolipomatous tumour, Recurrent and overlapping pathogenetic themes. Genes Chromosomes and morphologically similar lesions. J Pathol 2009;217:716–27. Cancer 2016;55:291–310. 28. Wang Y, Marino-Enriquez A, Bennett RR, Zhu M, Shen Y, Eilers G, et al. 37. Panagopoulos I, Mertens F, Isaksson M, Mandahl N. A novel FUS/CHOP Dystrophin is a tumor suppressor in human cancers with myogenic chimera in myxoid liposarcoma. Biochem Biophys Res Commun 2000; programs. Nat Genet 2014;46:601–6. 279:838–45. 29. Tadayoni R, Rendon A, Soria-Jasso LE, Cisneros B. Dystrophin Dp71: the 38. Martianov I, Choukrallah MA, Krebs A, Ye T, Legras S, Rijkers E, et al. Cell- smallest but multifunctional product of the Duchenne muscular dystrophy specific occupancy of an extended repertoire of CREM and CREB binding gene. Mol Neurobiol 2012;45:43–60. loci in male germ cells. BMC Genomics 2010;11:530. 30. Friederichs J, Zeller Y, Hafezi-Moghadam A, Grone€ H-J, Ley K, Altevogt P. 39. Coyaud E, Struski S, Prade N, Familiades J, Eichner R, Quelen C, et al. Wide The CD24/P-selectin binding pathway initiates lung arrest of human A125 diversity of PAX5 alterations in B-ALL: a Groupe Francophone de Cyto- adenocarcinoma cells. Cancer Res 2000;60:6714–22. genetique Hematologique study. Blood 2010;115:3089–97

7434 Clin Cancer Res; 23(23) December 1, 2017 Clinical Cancer Research

In-depth Genetic Analysis of Sclerosing Epithelioid Fibrosarcoma Reveals Recurrent Genomic Alterations and Potential Treatment Targets

Elsa Arbajian, Florian Puls, Cristina R. Antonescu, et al.

Clin Cancer Res 2017;23:7426-7434. Published OnlineFirst September 22, 2017.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-17-1856

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2017/09/22/1078-0432.CCR-17-1856.DC1

Cited articles This article cites 34 articles, 5 of which you can access for free at: http://clincancerres.aacrjournals.org/content/23/23/7426.full#ref-list-1

Citing articles This article has been cited by 4 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/23/23/7426.full#related-urls

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/23/23/7426. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.