Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Genomic EWS-FLI1 Fusion Sequences in Ewing Sarcoma Resemble Breakpoint Characteristics of Immature Lymphoid Malignancies

  • Manfred Berger,

    Affiliation University Hospital Erlangen, Department of Pediatrics, Erlangen, Germany

  • Uta Dirksen,

    Affiliation University Hospital Muenster, Department of Pediatric Hematology and Oncology, Muenster, Germany

  • Andreas Braeuninger,

    Affiliation University Hospital Muenster, Department of Pathology, Muenster, Germany

  • Gabriele Koehler,

    Affiliation University Hospital Muenster, Department of Pathology, Muenster, Germany

  • Heribert Juergens,

    Affiliation University Hospital Muenster, Department of Pediatric Hematology and Oncology, Muenster, Germany

  • Manuela Krumbholz ,

    Contributed equally to this work with: Manuela Krumbholz, Markus Metzler

    Affiliation University Hospital Erlangen, Department of Pediatrics, Erlangen, Germany

  • Markus Metzler

    Contributed equally to this work with: Manuela Krumbholz, Markus Metzler

    Affiliation University Hospital Erlangen, Department of Pediatrics, Erlangen, Germany

Genomic EWS-FLI1 Fusion Sequences in Ewing Sarcoma Resemble Breakpoint Characteristics of Immature Lymphoid Malignancies

  • Manfred Berger, 
  • Uta Dirksen, 
  • Andreas Braeuninger, 
  • Gabriele Koehler, 
  • Heribert Juergens, 
  • Manuela Krumbholz, 
  • Markus Metzler


Chromosomal translocations between the EWS gene and members of the ETS gene family are characteristic molecular features of the Ewing sarcoma. The most common translocation t(11;22)(q24;q12) fuses the EWS gene to FLI1, and is present in 85–90% of Ewing sarcomas. In the present study, a specifically designed multiplex long-range PCR assay was applied to amplify genomic EWS-FLI1 fusion sites from as little as 100 ng template DNA. Characterization of the EWS-FLI1 fusion sites of 42 pediatric and young adult Ewing sarcoma patients and seven cell lines revealed a clustering in the 5′ region of the EWS-breakpoint cluster region (BCR), in contrast to random distribution of breakpoints in the FLI1-BCR. No association of breakpoints with various recombination-inducing sequence motifs was identified. The occurrence of small deletions and duplications at the genomic junction is characteristic of involvement of the non-homologous end-joining (NHEJ) repair system, similar to findings at chromosomal breakpoints in pediatric leukemia and lymphoma.


Reciprocal balanced chromosomal translocations are recurrent and specific somatic aberrations in a wide variety of tumors, and are particularly associated with hematological malignancies [1]. Sequence analysis of individual DNA breakpoints reveals fingerprints of different DNA recombination and repair mechanisms involved in chromosomal translocation, depending on the differentiation stage of the target cell at initiation. For example, chromosomal fusion sites of lymphoid malignancies are often linked to aberrant recombination activating gene (RAG)-mediated V(D)J recombination [2]. In multiple myeloma, incorrect class switch recombination by activation-induced deaminase (AID) during secondary antibody diversification induces translocations within the immunoglobulin switch regions [3]. Chromosomal breakpoints in therapy-related acute lymphoblastic leukemia are clustered in proximity to topoisomerase II binding sites of the MLL gene [4], [5]. By contrast, more immature leukemia cells show predominantly sequence-independent breakages [6].

Far fewer breakpoint DNA sequences have been deciphered from solid tumors to allow comparative analyses, although there is increasing evidence of recurrent chromosomal translocations in non-hematological tumors of both mesenchymal and epithelial origin [7][10]. One of the first solid tumors found to carry characteristic reciprocal chromosomal translocations was Ewing sarcoma [11]. Ewing sarcoma family (ES) are the second most common solid bone and soft tissue malignancies in children, adolescents and young adults. ES are associated with chromosomal rearrangements that result in the fusion of the EWS gene with one gene of the ETS family of transcription factors. The most common translocation found in ES is t(11;22)(q24;q12), which fuses the EWS gene with the FLI1 gene. More than 85% of Ewing sarcoma patients carry a EWS-FLI1 fusion sequence. In 10% of cases, the EWS gene is fused to the ERG gene resulting in the t(21;22)(q22;q12) translocation [12]. Fusion transcript type does not appear to imply a prognostic impact with current treatment regimens [13]. The frequency at which characteristic fusion genes were being identified led to the incorporation of molecular analysis, i.e., fluorescence in situ hybridization (FISH) or reverse transcription-PCR (RT-PCR), in the routine diagnostic workup for ES [14], [15]. The large introns within the genomic breakpoint cluster regions (BCRs), however, mean that identification of genomic breakpoints is limited by the amplification range of conventional PCR, and detection of genomic fusion sites is not yet a routine diagnostic test. The BCR of the EWS gene spans a 5.7 kb region between exon 7 and exon 11, and the BCR of the FLI1 gene extends from exon 4 to exon 9, encompassing 38.2 kb.

Using primers covering both of the BCRs a nested multiplex long-range PCR (MLR-PCR) assay was established for reliable identification of EWS-FLI1 fusion sites. Unlike leukemia, where diagnostic material is easily accessible from blood or bone marrow, diagnostic samples of ES- are usually obtained by tumor biopsy and, therefore, the amount of diagnostic DNA for genomic studies is often very limited. Using the MLR-PCR assay described herein, it was possible to identify genomic EWS-FLI1 fusion sequences from as little as 100 ng DNA from the tumors of all the pediatric and young adult Ewing-sarcoma patients investigated.

In addition to diagnostic aspects, sequencing of genomic fusion sites is a prerequisite for detailed breakpoint characterization to identify breakpoint initiation mechanisms, and could also provide complementary information about the cellular origin of ES-.

Materials and Methods

Patients and Cell Lines

Genomic DNA was analyzed from t(11;22)(q24;q12) translocation-positive Ewing- sarcoma cell lines (n = 7), and 42 individuals (median age 14 years) whose cryopreserved tumor biopsies were sent to the Euro E:W:I:N:G:- 99 trial reference laboratory for pathological review and molecular diagnostics. Clinical and molecular parameters are summarized in Table 1. Informed consent was obtained from all patients or their legal guardians, in accordance with the declaration of Helsinki.

Table 1. Patientś characteristics and breakpoint coordinates of tumor samples and cell lines.

Cell lines were cultured in RPMI medium supplemented with 10% fetal bovine serum, L-glutamine, and antibiotics at 37°C in 5% CO2 on collagen-coated flasks. DNA from cell lines and tumor tissues was isolated using a QIAamp® DNA Blood and Tissue Mini Kit (Qiagen GmbH, Hilden, Germany).

Detection and Analysis of Genomic EWS-FLI1 Fusion Sites

Genomic EWS-FLI1 fusion sequences (derivative (der)22) were amplified by nested MLR-PCRs using 11 nested primer pairs covering both of the EWS gene and FLI1 gene BCRs (Figure 1A). Due to the different BCR length of the EWS and FLI1 genes, the average distance between primers was 500 bp and 3000 bp, respectively. In addition, an analogous primer set in the opposite direction was designed for the amplification of FLI1-EWS fusion sites (der11) (Figure S1). Primer sequences and positions are shown in Table S1.

Figure 1. Genomic fusion site sequencing.

(A) Genomic organization of the EWS and FLI1 genes and corresponding breakpoint cluster regions (BCR). Nested primer sets for der22 are shown as double headed arrows. (B) Representative breakpoint sequencing workflow. Left: Gel electrophoresis of MLR-PCR products from two tumor samples in lane 1 and 2 (lane 3 negative control DNA; lane 4 ddH2O; lane 5 positive control DNA; M = DNA ladder). Center: Gel electrophoresis of single long-range PCR products from 1st round MLR-PCR product of sample 1 (lane 1–11; lane 12 positive control) to identify FLI1 and EWS primers next to the fusion sites and to reduce amplification product size for direct sequencing. Right: Sequencing of the shortest amplification product and alignment to EWS and FLI1 reference sequences.

All PCRs were performed using the AccuPrime™ Taq DNA Polymerase System (Invitrogen, Karlsruhe, Germany) according to the manufacturer’s instructions. For the first round, MLR-PCR 100 ng template DNA was combined with the most 5′ EWS sense primer and 11 antisense FLI1 primers. If no specific amplification product was visible by gel electrophoresis, a second round MLR-PCR was carried out with corresponding internal primers and 1 µl of first round MLR-PCR product as template DNA (Figure 1B left side). To identify the FLI1 primer positioned next to the fusion site and, therefore, responsible for product generation, a series of single LR-PCRs was set up with the most 5′ internal EWS sense primer, one of each internal antisense FLI1 primers, and again, with first round MLR-PCR product as template DNA. Subsequently, the appropriate FLI1 primer was used in a series of single LR-PCRs in combination with additional internal sense EWS primers to further reduce the size of the specific amplification product for direct sequencing on a Beckman Coulter CEQ 8800 Genetic Analysis System (Figure 1B middle, right side). Patient-specific breakpoints were confirmed by an independent PCR using specific primer sets next to the patient’s fusion site and 10 ng original tumor DNA.

Statistical Breakpoint Analysis

Patient-specific fusion sequences were aligned to the NCBI reference sequences of EWS (NC_000022.10) and FLI1 (NC_000011.9) using VectorNTI® software for sequence editing and analysis. Breakpoint and primer positions are numbered according to their positions in the reference sequences, starting with number one for the first nucleotide.

Repeat elements in the EWS- and FLI1-BCR were identified with the RepeatMasker tool ( A Chi-square test was used to test for significant colocalization of patient-specific fusion sites and repeat elements or recombination-related DNA sequence motifs.

Components of the free software environment R ( were applied to Kernel density analysis. Bandwidth selection was performed according to Sheather and Jones [16]. Clusters were defined as regions in which the lower limit of the 95% confidence band, determined by a bootstrapping procedure, was higher than a density function resulting from simulations at randomly distributed pseudo-breakpoints [17]. Both bootstrapping and simulations used 1000 permutations.


Detection and Sequencing of Genomic EWS-FLI1 Fusion Sites

A nested MLR-PCR assay was developed for detection of genomic EWS-FLI1 and FLI1-EWS fusion sites from diagnostic tumor biopsy specimens. PCR conditions were optimized using DNA from seven EWS-FLI1-positive Ewing- sarcoma cell lines (Table 1). Sensitivity of the MLR-PCR was determined by a series of 2-fold dilutions of EWS-FLI1-positive A673 Ewing- sarcoma cells in EWS-FLI1-negative HL-60 cells. With the initial multiplex PCR, the EWS-FLI1 fusion gene was detectable up to a 1∶64 dilution, corresponding to a tumor cell proportion of 1.6%. A subsequent quantitative real-time PCR, applying individually designed fusion site-specific primer and probe sets, allowed breakpoint detection with a sensitivity of 10−4–10−5 (data not shown).

A major advantage of the assay is the requirement for a minimal amount of DNA (100 ng) as both of the EWS and FLI1 BCRs are covered by one single initial multiplex PCR. The genomic EWS-FLI1 fusion sites (der22) were successfully amplified from all seven cell lines and 42 tumor samples (Table 1). The reciprocal genomic FLI1-EWS fusion sites (der11) were detected in 27/42 Ewing sarcoma patients and five out of seven cell lines (overall 65%). The lower detection rate of the reciprocal FLI1-EWS fusion site is in line with results from other recurrent chromosomal translocations and is attributable to large deletions or complex rearrangements on der11 [18][20].

Alignments of individual fusion sites to the reference sequences are shown in Figure S2. The breakpoint sequences were deposited in the NCBI GenBank (accession numbers JX266448 to JX266528).

Distribution of Genomic Breakpoints

Breakpoint distribution analysis of all 49 EWS-FLI1 fusion sites revealed two cluster regions within the EWS-BCR. Forty-three of 49 breaks (88%) were located in a cluster region spanning from intron 7 to the 5′-region of intron 8; the remaining six breaks (12%) were positioned in a region spanning from intron 9 to exon 11, constituting a second cluster region (Figure 2A). Kernel density analysis confirmed this observation and showed a statistically significant overrepresentation of breakpoints within the cluster region at the 5′-region of the ESW-BCR (Figure 2B). Due to the low number of breakpoints in the second cluster at the 3′-region of the EWS-BCR no statistical significance was obtained by Kernel density analysis. By contrast, breakpoints in the FLI1-BCR were randomly distributed (Figures 2A and 2B). Localization of EWS breakpoints showed no significant correlation with the corresponding FLI1 breakpoints (Figure 2C).

Figure 2. Breakpoint distribution in the BCR of EWS and FLl1.

(A) Vertical bars above or below the breakpoint regions indicate individual breakpoint positions of Ewing sarcoma patients. Black boxes represent exons, and gray boxes correspond to repeat elements. (B) Results of Kernel density analysis (dashed line = breakpoint density; gray line = lower limit of 95% confidence band determined by bootstrapping procedure; black line = 95% confidence interval of a density function resulting from simulations at randomly distributed pseudo-breakpoints). X-axes indicate the BCR nucleotide positions within the respective reference gene. (C) Scatterblot of gender-specific EWS-FLI1 breakpoints. Circles represent female, and squares represent male subjects. (D) Number of microhomologies and filler nucleotides at EWS-FLI1 (der22; black bar) and FLI1-EWS (der11; gray bar) fusion sites. Each bar on the x-axis represents one individual.

Patients with breakpoints in the smaller EWS cluster region revealed no differences with respect to age, gender (Figure S3A) and tumor localization. Similar results were obtained for the FLI1-BCR; no correlation between breakpoint localization and gender or age at time of diagnosis was observed (Figure S3B).

In 34/49 (69%) samples, genomic breakpoints within the FLI1 gene resulted in an EWS-FLI1 fusion transcript of EWS exons 1–7 fused to FLI1 exons 6–9, representing Ewing sarcoma type 1. In seven out of 49 (14%), there was an EWS-FLI1 fusion transcript of EWS exons 1–7 and FLI1 exons 5–9, corresponding to Ewing sarcoma type 2. In 15 patients and four cell lines with genomic breakpoints located in EWS exon 8 or intron 8, exon 8 was spliced out at the transcriptional level (Table 1). The splicing out of exon 8 is essential to form a functionally active EWS-FLI1 fusion transcript with an intact reading frame [21]. The frequencies of the different chimeric EWS-FLI1 fusion transcript types in this study are consistent with published data [22].

Sequence Analysis of EWS-FLI1 and FLI1-EWS Fusion Sites

Genomic fusion sites of der22 and der11 were analyzed for colocalization with the following repeat elements and recombination-related DNA sequence motifs: Alu repeats, topoisomerase II binding sites, translin binding sites, chi-like sequences, heptamer/nonamer recombination signals, palindromic sequences, hypervariable minisatellite core sequence, hypervariable minisatellite recombination sequence, DNA polymerase frameshift hotspots, CpG islands, human replication origin consensus sequence, and repeat sequences such as low complexity, SINE/MIR, DNA/hAT-Charlie, and LINE/L2 (Table S2). No deviation between the expected and observed breakpoint within the DNA sequence motifs and repeat elements of EWS- and FLI1-BCR was observed (Table S2).

Detailed sequence analysis of the 49 identified EWS-FLI1 and 32 FLI1-EWS genomic fusion sites revealed clean transitions between the two contributing genes in 30 cases (37%), 37 fusion sites (46%) featured small microhomologies (<17 bp), and another 14 breakpoints (17%) presented with filler nucleotides at the fusion sites (<16 bp) (Figure 2D). Patient UPN 5 showed an extended inverted insertion of a FLI1 gene segment between the 5′ to 3′ EWS and FLI1 portion at the genomic breakpoint.

These findings are comparable to breakpoint structures observed in hematological malignancies. Immature leukemia cells in particular show predominantly sequence-independent breakage with small microhomologies or deletions at the fusion sites [4], [6].


Chromosomal fusion of the EWS gene to one member of the ETS family of transcription factors is the genetic hallmark of ES-. Detection of the chromosomal translocation by FISH, or amplification of the resulting fusion transcripts by RT-PCR, has become a well-established diagnostic component for molecular confirmation of histopathological tumor classification. Chromosomal breakage and re-fusion, however, occur in large intronic parts of the respective genes. Although the large size of the BCRs complicates the amplification of the breakpoint-spanning site, identification of the individual intronic DNA fusion sequence may unravel additional information associated with mechanisms involved in the rearrangement formation.

DNA-based tumor diagnostics has general advantages and disadvantages. The chemical stability of DNA facilitates storage and transport of patient material, and in contrast to RT-PCR methods, enables the detection of fusion genes independent of their gene expression. As a complementary diagnostic tool, DNA-based minimal residual disease detection could improve the quantification of resting residual tumor cells from different specimens including bone marrow, peripheral blood stem cell collection products, and paraffin-embedded, fixed-tissue sections. Fusion of EWS to one of its partner genes is an early event in tumorigenesis and an essential oncogenic factor for maintaining the malignant transformation of ES- [23], [24]. In contrast to secondary genetic aberrations occurring during clonal evolution and contributing to the genetic heterogeneity within an individual ES- [25], the genomic fusion site remains a common and consistent molecular marker of tumor cells, unaffected by clonal selection from therapeutic intervention, and is currently evaluated as an additional tool for tumor cell quantification during treatment.

A major obstacle in quantifying DNA breakpoints in ES is the identification of the genomic EWS-FLI1 fusion sequence, because of large intronic regions within the BCRs. A classical approach, using multiple single PCRs to cover the complete BCRs, requires large amounts of patient material. This aspect and its consequent limit on the availability of adequate material for additional molecular diagnostics are probably the main reasons why only very few studies on genomic EWS-FLI1 fusion sequences are published, in contrast to numerous studies on EWS-FLI1 transcripts and genomic breakpoints in fusion genes in diseases with more freely available material, e.g., acute leukemia [13], [26]. Two studies have identified the genomic fusion sequence in a total of eight individual cases of ES on the basis of transcript amplification, using primers specifically designed for the region of interest [27], [28]. A larger cohort of 77 Ewing- sarcoma patients was analyzed by Zucman-Rossi et al. using 15 single PCRs, but no patient characteristics are available and primer sequences have not been made permanently available [18].

In the present study, to overcome the problem of high DNA consumption, a nested MLR-PCR assay was established for reliable amplification of genomic EWS-FLI1 fusion sequences. Highly stringent primer selection and the use of advanced polymerases facilitated the development of a detection assay based on a single initial MLR-PCR. Thus, only minimal patient material is required to extract template DNA (100 ng), enabling the detection of tumor genomic fusion sequences from small tumor samples, e.g., fine needle biopsies. Using this assay EWS-FLI1 fusion sequences were identified from all 42 pediatric and young adult ES patients investigated. In addition, an analogous primer set was designed for detection of the second most common fusion gene in ES, the EWS-ERG gene. The assay can be readily adapted to include rare fusion partner genes occurring in the remaining 2–5% of cases [12].

The patient-specific breakpoints are distributed in two subclusters within the EWS-BCR, and this is in line with results from the cohort studied by Zucman-Rossi et al. [18]. Despite searching for an extensive spectrum of sequence motifs and repeat elements, no DNA motif associated with either of the breakpoint clusters or with a significant subgroup of Ewing sarcoma was identified. However, small microhomologies and filler nucleotides at the fusion sites, as well as deletions or insertion of several nucleotides in the corresponding chromosomal derivatives (der22 and der11), resemble the characteristics of non-homologous end-joining (NHEJ) repair, and suggest that NHEJ repair is involved in translocation formation in ES [29].

A comparison of genomic breakpoint features of ES- with the far more extensively characterized breakpoints in leukemia and lymphoma reveals a number of similarities. In cells from hematological malignancy, different DNA double-strand break and repair mechanisms are associated with the lineage and differentiation stage of the cell population. Sequence-independent breakages as observed in ES- are also characteristic for immature lymphoid malignant cells [6]. This subtype of leukemia cells shows small microhomologies or non-template insertions, as well as small deletions and duplications at the chromosomal breakpoint regions, indicating that NHEJ repair is involved in breakpoint initiation [4], [30][32].

This observation of reciprocal balanced translocations with an NHEJ repair signature indicates molecular genetic similarities to mesenchymal tumors. It further supports results from functional and molecular genetic studies characterizing ES- by global gene expression profiling, which propose that ES- has its cellular origin in early mesenchymal progenitors [33], [34].

In summary, in this study, genomic EWS-FLI1 fusion sequences were identified in a cohort of pediatric and young adult Ewing sarcoma patients with a specifically designed MLR-PCR assay requiring only minimal patient DNA. Detailed characterization of the genomic fusion sites revealed similarities to fusion site characteristics identified in immature lymphoid leukemia cells, and suggests that NHEJ repair mechanisms are involved in breakpoint initiation.

Supporting Information

Figure S1.

Genomic organization of the FLI1 and EWS genes and corresponding breakpoint cluster regions (BCR). Nested primer sets for der11 are shown as double headed arrows.


Figure S2.

Alignments of the patients’ specific fusion sequences to EWS and FLI1 reference sequences.


Figure S3.

Scatterblots of the distribution of (A) EWS breakpoints and (B) FLI1 breakpoints in reference to the age at time of diagnosis. Circles represent female, squares represent male subjects. Y-axes indicate the BCR nucleotide positions within the respective reference gene.


Table S1.

Sequences of primers used in nested MLR-PCR assay.


Table S2.

Summary of sequence motifs and repeat elements tested for association with breakpoint localization.



We thank Ursula Jacobs and Sabine Semper for excellent technical assistance, and Dr. Thomas Keller (ACOMED Statistik, Leipzig) for statistical analysis.

Author Contributions

Conceived and designed the experiments: MK MM. Performed the experiments: MB MK AB. Analyzed the data: MB MM MK UD. Contributed reagents/materials/analysis tools: UD AB GK HJ. Wrote the paper: MB MK MM UD AB GK HJ.


  1. 1. Mitelman F, Johansson B, Mertens F (2007) The impact of translocations and gene fusions on cancer causation. Nat Rev Cancer 7: 233–245.
  2. 2. Raghavan SC, Kirsch IR, Lieber MR (2001) Analysis of the V(D)J recombination efficiency at lymphoid chromosomal translocation breakpoints. J Biol Chem 276: 29126–29133.
  3. 3. Aplan PD (2006) Causes of oncogenic chromosomal translocation. Trends Genet 22: 46–55.
  4. 4. Greaves MF, Wiemels J (2003) Origins of chromosome translocations in childhood leukaemia. Nat Rev Cancer 3: 639–649.
  5. 5. Jung R, Jacobs U, Krumbholz M, Langer T, Keller T, et al. (2010) Bimodal distribution of genomic MLL breakpoints in infant acute lymphoblastic leukemia treatment. Leukemia 24: 903–907.
  6. 6. Tsai AG, Lu H, Raghavan SC, Muschen M, Hsieh CL, et al. (2008) Human chromosomal translocations at CpG sites and a theoretical basis for their lineage and stage specificity. Cell 135: 1130–1142.
  7. 7. Xia SJ, Barr FG (2005) Chromosome translocations in sarcomas and the emergence of oncogenic transcription factors. Eur J Cancer 41: 2513–2527.
  8. 8. Kumar-Sinha C, Tomlins SA, Chinnaiyan AM (2008) Recurrent gene fusions in prostate cancer. Nat Rev Cancer 8: 497–511.
  9. 9. Tognon C, Knezevich SR, Huntsman D, Roskelley CD, Melnyk N, et al. (2002) Expression of the ETV6-NTRK3 gene fusion as a primary event in human secretory breast carcinoma. Cancer Cell 2: 367–376.
  10. 10. Pierotti MA (2001) Chromosomal rearrangements in thyroid carcinomas: a recombination or death dilemma. Cancer Lett 166: 1–7.
  11. 11. Sandberg AA, Turc-Carel C, Gemmill RM (1988) Chromosomes in solid tumors and beyond. Cancer Res 48: 1049–1059.
  12. 12. Riggi N, Stamenkovic I (2007) The Biology of Ewing sarcoma. Cancer Lett 254: 1–10.
  13. 13. Le Deley MC, Delattre O, Schaefer KL, Burchill SA, Koehler G, et al. (2010) Impact of EWS-ETS fusion type on disease progression in Ewing’s sarcoma/peripheral primitive neuroectodermal tumor: prospective results from the cooperative Euro-E.W.I.N.G. 99 trial. J Clin Oncol 28: 1982–1988.
  14. 14. Antonescu CR (2006) The role of genetic testing in soft tissue sarcoma. Histopathology 48: 13–21.
  15. 15. Lazar A, Abruzzo LV, Pollock RE, Lee S, Czerniak B (2006) Molecular diagnosis of sarcomas: chromosomal translocations in sarcomas. Arch Pathol Lab Med 130: 1199–1207.
  16. 16. Sheather SJ, Jones MC (1991) A reliable data-based bandwidth selection method for kernel density estimation. Journal of the Royal Statistical Society Series B 53: 683–690.
  17. 17. Segal MR, Wiemels JL (2002) Clustering of translocation breakpoints. J Am Stat Assoc 97: 66–76.
  18. 18. Zucman-Rossi J, Legoix P, Victor JM, Lopez B, Thomas G (1998) Chromosome translocation based on illegitimate recombination in human tumors. Proc Natl Acad Sci U S A 95: 11786–11791.
  19. 19. Delattre O, Zucman J, Plougastel B, Desmaze C, Melot T, et al. (1992) Gene fusion with an ETS DNA-binding domain caused by chromosome translocation in human tumours. Nature 359: 162–165.
  20. 20. Turc-Carel C, Aurias A, Mugneret F, Lizard S, Sidaner I, et al. (1988) Chromosomes in Ewing’s sarcoma. I. An evaluation of 85 cases of remarkable consistency of t(11;22)(q24;q12). Cancer Genet Cytogenet 32: 229–238.
  21. 21. Zucman J, Melot T, Desmaze C, Ghysdael J, Plougastel B, et al. (1993) Combinatorial generation of variable fusion proteins in the Ewing family of tumours. EMBO J 12: 4481–4487.
  22. 22. Lin PP, Brody RI, Hamelin AC, Bradner JE, Healey JH, et al. (1999) Differential transactivation by alternative EWS-FLI1 fusion proteins correlates with clinical heterogeneity in Ewing’s sarcoma. Cancer Res 59: 1428–1432.
  23. 23. Prieur A, Tirode F, Cohen P, Delattre O (2004) EWS/FLI-1 silencing and gene profiling of Ewing cells reveal downstream oncogenic pathways and a crucial role for repression of insulin-like growth factor binding protein 3. Mol Cell Biol 24: 7275–7283.
  24. 24. Takigami I, Ohno T, Kitade Y, Hara A, Nagano A, et al. (2011) Synthetic siRNA targeting the breakpoint of EWS/Fli-1 inhibits growth of Ewing sarcoma xenografts in a mouse model. Int J Cancer 128: 216–226.
  25. 25. Mackintosh C, Ordonez JL, Garcia-Dominguez DJ, Sevillano V, Llombart-Bosch A, et al. (2012) 1q gain and CDT2 overexpression underlie an aggressive and highly proliferative form of Ewing sarcoma. Oncogene 31: 1287–1298.
  26. 26. van Doorninck JA, Ji L, Schaub B, Shimada H, Wing MR, et al. (2010) Current treatment protocols have eliminated the prognostic advantage of type 1 fusions in Ewing sarcoma: a report from the Children’s Oncology Group. J Clin Oncol 28: 1989–1994.
  27. 27. Bhagirath T, Abe S, Nojima T, Yoshida MC (1995) Molecular analysis of a t(11;22) translocation junction in a case of Ewing’s sarcoma. Genes Chromosomes Cancer 13: 126–132.
  28. 28. Obata K, Hiraga H, Nojima T, Yoshida MC, Abe S (1999) Molecular characterization of the genomic breakpoint junction in a t(11;22) translocation in Ewing sarcoma. Genes Chromosomes Cancer 25: 6–15.
  29. 29. Rassool FV (2003) DNA double strand breaks (DSB) and non-homologous end joining (NHEJ) pathways in human leukemia. Cancer Lett 193: 1–9.
  30. 30. Langer T, Metzler M, Reinhardt D, Viehmann S, Borkhardt A, et al. (2003) Analysis of t(9;11) chromosomal breakpoint sequences in childhood acute leukemia: almost identical MLL breakpoints in therapy-related AML after treatment without etoposides. Genes Chromosomes Cancer 36: 393–401.
  31. 31. Metzler M, Brehm U, Langer T, Viehmann S, Borkhardt A, et al. (2004) Asymmetric multiplex-polymerase chain reaction - a high throughput method for detection and sequencing genomic fusion sites in t(4;11). Br J Haematol 124: 47–54.
  32. 32. von Goessel H, Jacobs U, Semper S, Krumbholz M, Langer T, et al. (2009) Cluster analysis of genomic ETV6-RUNX1 (TEL-AML1) fusion sites in childhood acute lymphoblastic leukemia. Leuk Res 33: 1082–1088.
  33. 33. Richter GH, Plehm S, Fasan A, Rossler S, Unland R, et al. (2009) EZH2 is a mediator of EWS/FLI1 driven tumor growth and metastasis blocking endothelial and neuro-ectodermal differentiation. Proc Natl Acad Sci U S A 106: 5324–5329.
  34. 34. Lin PP, Wang Y, Lozano G (2011) Mesenchymal Stem Cells and the Origin of Ewing’s Sarcoma. Sarcoma 2011.