A genomic case study of desmoplastic small round cell tumor: comprehensive analysis reveals insights into potential therapeutic targets and development of a monitoring tool for a rare and aggressive disease
Ferreira et al. Human Genomics
A genomic case study of desmoplastic small round cell tumor: comprehensive analysis reveals insights into potential therapeutic targets and development of a monitoring tool for a rare and aggressive disease
Elisa Napolitano Ferreira 0 3
Bruna Durães Figueiredo Barros 0 3
Jorge Estefano de Souza 2
Renan Valieris Almeida 0 3
Giovana Tardin Torrezan 0 3
Sheila Garcia 0 3
Ana Cristina Victorino Krepischi 1
Celso Abdon Lopes de Mello 6
Isabela Werneck da Cunha 5
Clóvis Antonio Lopes Pinto 5
Fernando Augusto Soares 5
Emmanuel Dias-Neto 0 3
Ademar Lopes 6
Sandro José de Souza 4
Dirce Maria Carraro 0 3
0 International Research Center/CIPE, A.C. Camargo Cancer Center , São Paulo, SP , Brazil
1 Institute of Biosciences, University of São Paulo , São Paulo, SP , Brazil
2 Instituto Metrópole Digital, Federal University of Rio Grande do Norte , Natal, RN , Brazil
3 International Research Center/CIPE, A.C. Camargo Cancer Center , São Paulo, SP , Brazil
4 Federal University of Rio Grande do Norte , Natal, RN , Brazil
5 Department of Anatomic Pathology, A.C. Camargo Cancer Center , São Paulo, SP , Brazil
6 Departament of Abdominal Surgery, A.C. Camargo Cancer Center , São Paulo, SP , Brazil
Background: Genome-wide profiling of rare tumors is crucial for improvement of diagnosis, treatment, and, consequently, achieving better outcomes. Desmoplastic small round cell tumor (DSRCT) is a rare type of sarcoma arising from mesenchymal cells of abdominal peritoneum that usually develops in male adolescents and young adults. A specific translocation, t(11;22)(p13;q12), resulting in EWS and WT1 gene fusion is the only recurrent molecular hallmark and no other genetic factor has been associated to this aggressive tumor. Here, we present a comprehensive genomic profiling of one DSRCT affecting a 26-year-old male, who achieved an excellent outcome. Methods: We investigated somatic and germline variants through whole-exome sequencing using a family based approach and, by array CGH, we explored the occurrence of genomic imbalances. Additionally, we performed mate-paired whole-genome sequencing for defining the specific breakpoint of the EWS-WT1 translocation, allowing us to develop a personalized tumor marker for monitoring the patient by liquid biopsy. Results: We identified genetic variants leading to protein alterations including 12 somatic and 14 germline events (11 germline compound heterozygous mutations and 3 rare homozygous polymorphisms) affecting genes predominantly involved in mesenchymal cell differentiation pathways. Regarding copy number alterations (CNA) few events were detected, mainly restricted to gains in chromosomes 5 and 18 and losses at 11p, 13q, and 22q. The deletions at 11p and 22q indicated the presence of the classic translocation, t(11;22)(p13;q12). In addition, the mapping of the specific genomic breakpoint of the EWS-WT1 gene fusion allowed the design of a personalized biomarker for assessing circulating tumor DNA (ctDNA) in plasma during patient follow-up. This biomarker has been used in four post-treatment blood samples, 3 years after surgery, and no trace of EWS-WT1 gene fusion was detected, in accordance with imaging tests showing no evidence of disease and with the good general health status of the patient. (Continued on next page) © The Author(s). 2016 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
(Continued from previous page)
Conclusions: Overall, our findings revealed genes with potential to be associated with risk assessment and tumorigenesis
of this rare type of sarcoma. Additionally, we established a liquid biopsy approach for monitoring patient follow-up based
on genomic information that can be similarly adopted for patients diagnosed with a rare tumor.
Comprehensive molecular profiling is an especially
important tool to gain insights on the biological pathways
involved in tumor onset and to improve the
management and treatment of rare tumors. Desmoplastic small
round cell tumor (DSRCT) is a very rare type of sarcoma,
with an age-adjusted incidence rate of 0.3 cases/million
, which typically arises from the abdominal or pelvic
peritoneum and occurs mainly in male adolescents and
young adults (peak incidence at 20–24 years of age) .
Current therapeutic approaches involve the use of
multimodal therapeutic regimen, including aggressive
polychemotherapy, debulking surgery, and whole abdominal
A specific translocation, t(11;22)(p13;q12), is detected
in DSRCT cases, juxtaposing the Ewing’s sarcoma gene
(EWSR1) to the Wilm’s tumor gene (WT1). The chimeric
transcript containing the 5′ region of the EWSR1, which
includes the N-terminal transactivation domain of EWS,
and the 3′ sequence of WT1 containing 2–4 zinc finger
domains have been shown to upregulate EGR-1  and
induce the expression of PDGFA  and IGF1R .
Apart from this translocation, no other recurrent
genomic alteration has been reported in DSRCT cases. Silva et
al.  detected a somatic amplification involving AURKB
and MCL1 genes in one patient, and La Starza et al. 
found specific genomic imbalances, including gain at
chromosome 3 reported in two cases and chromosome
5 polysomy in one case. In terms of point mutations,
the data is even scarcer. Variants of unknown clinical
significance were reported in ARID1A and RUNX1
genes in one patient , whereas in another study, no
mutations were detected in a panel of 29 genes evaluated
in a cohort of 24 DSRCT cases . This limited genomic
information about DSRCT impairs new and more efficient
therapeutic opportunities for the young patients affected
with this rare tumor.
Here, aiming to contribute with the knowledge of the
genomic abnormalities that underlies DSRCT, we
performed a comprehensive genomic profiling using a
family based approach in one case of DSRCT diagnosed
in a young male patient with pelvic tumor, who
presented excellent outcome sustained for above 3 years.
We identified somatic mutations in a genomic
background of rare germline variants either homozygous or
as compound heterozygous inheritance, which can
improve the understanding of the genetic basis of this rare
tumor. We have also generated the profile of genomic
imbalances, which was confirmed by whole-exome
sequencing. In addition, as we were able to define the
precise genomic breakpoints of the EWS-WT1
translocation by whole-genome and Sanger sequencing, we
managed to establish a personalized strategy for tracking
DNA tumor traces in plasma, allowing an accurate
monitoring of tumor recurrence.
Overall, our analysis revealed potential genes and
pathways associated with this rare sarcoma and demonstrated
the feasibility of using genomic profiling for the benefit of
patients affected by rare tumors by developing a
personalized monitoring strategy.
Tumor tissues and blood samples were collected following
the technical and ethical procedures of A.C. Camargo
Tumor Bank, registered at National Council for Ethics in
Research by the number B001 . Genomic DNA and
plasma DNA were extracted in DNA and RNA Bank 
using QIASymphony DNA Mini kit (QIAGEN, Hilden,
Germany) for tumor and leukocyte DNA and QIAamp
DNA blood Midi kit (QIAGEN, Hilden, Germany) for
plasma DNA, following standard procedures.
Target sequencing was performed using the Ion
AmpliSeq™ Comprehensive Cancer Panel, which comprises all
exons from 409 genes associated with different types of
tumors (AmpliSeq, Ion Torrent™). This panel, based on
multiplex PCR, was performed with as little as 40 ng of
DNA from the tumor sample. Library was prepared based
on Ion AmpliSeq™ Library Preparation protocol and
sequenced at Ion Proton™ platform (Ion Torrent™),
according to the manufacturer’s instructions.
Whole-exome sequencing of the tumor and leukocyte
DNA samples from the patient and his mother were
performed using the TargetSeq™ Exome Enrichment Kit (Life
Technologies), followed by paired-end sequencing (75 × 50)
in SOLiD 5500xl System (Life Technologies). Leukocyte
DNA sample from his father was submitted for
wholeexome sequencing using Ion Xpress™ Plus Fragment
Library kit and Ion TargetSeq™ Exome Enrichment Kit (Life
Technologies), followed by sequencing at Ion Proton™
platform (Ion Torrent™), according to the manufacturer’s
instructions. Single-end sequencing was performed on
an Ion PI™ Chip v2 with 200 pb sequencing kit (Ion
Whole-genome sequencing was performed by mate-paired
DNA libraries prepared using the 5500 SOLiD™
MatePaired Library Construction Kit (Life Technologies),
following the manufacturer’s instructions. Briefly, genomic
DNA from tumor and leukocyte of the patient was
sheared with a Covaris sonicator into approximately
2 Kb fragments, circularized with mate-paired adaptors,
nick-translated and digested, incorporated with sequencing
adaptors and individual barcodes (distinct barcodes were
used for the tumor and leukocyte DNA), and submitted to
emulsion PCR. Mate-paired sequencing (60 × 60) was
performed in a SOLiD 5500xl System (Life Technologies).
For validation, primers were designed flanking the
variants, in order to generate fragments of nearly 400 bp. PCR
reactions were performed using GoTaq® Green Master
Mix (Promega, Madison, WI, USA), using 15 ng of DNA,
with 300 nM of each primer, for a final reaction volume
of 20 μL. Approximately 200 ng of PCR-amplified
fragments were purified with ExoSAP-IT (USB Corporation,
Cleveland, OH, USA) and sequenced in both directions.
All alterations were evaluated in all four samples (mother,
father, and patient’s leukocyte and tumor). Products were
analyzed using an ABI 3130xl DNA sequencer (Applied
Biosystems, Foster City, CA, USA), and sequences were
aligned with the respective gene reference sequence using
CLC Genomics Workbench Software (QIAGEN, Hilden,
Comparative genome hybridization based on microarrays
Comparative genomic hybridization based on
microarrays was performed in a commercial whole-genome 180 K
platform containing 180,000 oligonucleotide probes
(Agilent Technologies; design 22060), using DNA from
the tumor sample. Reference DNA was a commercially
available human pool of samples from multiple anonymous
healthy donors (Promega Corporation). Technical
procedures are described in Torrezan et al. . Hybridization
and washing were performed as recommended by the
manufacturer. Scanned images were processed using
Feature Extraction 10.7.3.1 software (Agilent Technologies),
and array CGH analysis was conducted with Nexus Copy
Number software 7.0 (Biodiscovery). We used the FASST2
segmentation algorithm, according to the following settings:
minimum of five consecutive probes (effective resolution of
~70 Kb for CNA calling), significance threshold set at 10−8,
and threshold log2 Cy3/Cy5 of 0.33 and −0.3 for gains
for loss, respectively, and 1.2 and −1.1 for high copy
number gains and homozygous losses, respectively. All
copy number alterations are reported in the Database
of Genomic Variants .
Comprehensive cancer panel
Sequencing reads from Ion Proton™ were mapped to the
reference genome (GRCh37/hg19) with TMAP (torrent
mapper 4.2.18). Sequence variants (SNVs and indels) were
identified with Torrent Variant Caller 4.0-5, followed by
confirmation by GATK protocol vs3.2-2-gec30cee .
Variants were annotated using SnpEff version 3.5d (build
Sequencing reads from SOLiD 5500xl System were mapped
to the reference genome (GRCh37/hg19) with Lifescope
(LifeScope™ Genomic Analysis Software v2.5.1). Sequencing
reads from Ion Proton™ were mapped to the reference
genome (GRCh37/hg19) with TMAP (torrent mapper
4.2.18). Sequence variants (SNVs and indels) were
identified following the GATK protocol vs3.2-2-gec30cee
 for SOLiD 5500XL System and with Torrent
Variant Caller 4.0-5, followed by GATK protocol
vs3.2-2gec30cee for Ion Proton reads. Variants were annotated
using SnpEff version 3.5d (build 2014-03-05)  and an
in-house developed script. Identified variants were
compared to dbNSFP version 2.4 [15, 16]; COSMIC v69 ;
1000 Genomes ; NHLBI GO Exome Sequencing
Project version ESP6500SI-V2 ; HapMap ; and dbSNP
version 138 [21, 22] for further annotation. Somatic
variants were defined for regions with a minimum coverage of
10× for both tumor and leukocyte sample from the patient
and minimum variant frequency of 20% in the tumor only.
De novo variants were defined with a minimum coverage
of 10× for leukocyte samples of the patient and his parents
and with a minimum variant frequency of 30%. To identify
rare polymorphisms inherited in homozygosity, we
selected variants with a minimum coverage of 10× for
leukocyte samples of the patient and his parents, detected
in heterozygosis in both parents and homozygosis in the
patient presenting a minor allele frequency ≤10% in the
public databases (1000 Genome , NHLBI GO Exome
Sequencing Project , and HapMap ). To identify
germline compound heterozygosis cases, we identified
genes with two distinct heterozygous mutations in the
patient, where each variant was exclusively present in one of
his parents in heterozygosity and detected in regions with
a minimum coverage of 10× for all leukocyte samples. We
discarded variants that are detected with a minor allele
frequency above 10% in the 1000 Genome Project .
Copy number alterations were detected using the
bioinformatics packages Excavator (version 2.2)  and cn.mops
(version 1.8.9)  by comparing exome data from the
tumor to leukocyte from the patient. For visualization,
we used circos 0.67-7 package .
For detecting structural variations, mate-pair reads
obtained by SOLiD 5500xl System were analyzed by svdetect
(version r0.8b) .
Ingenuity Pathway Analysis
We applied the core analysis of Ingenuity Pathway Analysis
(IPA) system (QIAGEN, Germantown, MD, USA) to
identify gene interaction networks.
Digital droplet PCR
Digital droplet PCR assays were carried out using the
QX200™ Droplet Digital™ (ddPCR™) System (Bio-Rad). A
primer-probe assay labeled with FAM was designed for
the amplification of wild-type WT1 gene and a
primerprobe assay labeled with HEX was designed for the
amplification of the gene fusion event (EWS-WT1). For
the amplification reaction, we used 1× ddPCR
Supermix, 1× primer-probe assay (FAM), 1× primer-probe
assay (HEX), and 4 μl of DNA. Droplet generation, PCR
amplification, and droplet counting were performed
following the manufacturer’s recommendations.
DNA samples from tumor tissue (60 ng–6 pg) and from
leukocytes (6 ng) were used as positive and negative
controls, respectively. We loaded 4 μl of 30 μl (1/8) of cfDNA
samples from the patient. We also used DNA and cell-free
DNA extracted from leukocytes and plasma, respectively,
from healthy donors as negative controls and performed
non-template control. All reactions were performed with
at least two replicates.
In this study, we performed a comprehensive genomic
profiling of one case of desmoplastic small round cell
tumor (DSRCT) by a combination of targeted sequencing,
array CGH, whole genome, and whole exome applied in a
family based format. The patient studied here is a
26-yearold male, who presented at A.C. Camargo Cancer Center
in November, 2011, with a large abdominal mass and a
small nodule on the pelvic region. Staging images showed
that the disease was limited to the abdominal cavity. He
underwent a CT-guided biopsy that revealed a
desmoplastic small round cell tumor (DSRCT), showing
positivity for EMA, desmin, and nuclear staining for WT1
(carboxy-terminus antibody). FISH analysis was positive
for EWS translocation.
The patient started systemic treatment with 4 cycles of
vincristine, cyclophosphamide, and doxorubicin (VAC)
and alternated with ifosfamide, carboplatin, and
etoposide (ICE). After 4 cycles, the best response was stable
disease, with minor reduction in the tumor dimensions.
He underwent complete surgical cytoreduction, with
resection of the large mass and resection of peritoneal
implant on the pelvic region, and hyperthermic
intraperitoneal chemotherapy (HIPEC), with cisplatin and
doxorubicin. The patient presented a complete recovery
from these procedures. After that, the patient received
four more cycles of chemotherapy and total abdominal
irradiation (total of 30 Gy). After a follow-up of 48 months
since surgery, the patient is asymptomatic, with no signs
of disease (Additional file 1: Figure S1).
The comprehensive genomic profiling was carried out
in multiple fronts. To identify actionable mutations, we
carried out targeted sequencing of the most important
actionable genes in the tumor sample. Further, to
identify genes and mutations possibly involved with tumor
onset and predisposition, we performed whole exome
sequencing, using the DSRCT tumor sample and blood
samples from the patient and his both parents (Additional
file 2: Table S1). Array CGH was performed in tumor
sample to identify structural rearrangements and copy
number imbalances. Whole-genome sequencing defined
one tumor marker used to precisely monitor patient after
Producing a portrait of somatically acquired variants
To identify actionable mutations or pathways related to
DSRCT in this patient, we initially performed targeted
sequencing using a cancer-oriented gene panel composed of
409 genes in the tumor sample (Comprehensive Cancer
Panel – Thermo Scientific). Since no actionable mutation
for targeted therapy was detected in the tumor, we
investigated the complete landscape of somatic mutations
by whole-exome sequencing (WES) the tumor and the
patient’s leukocyte. The analysis of the tumor revealed
15 somatic acquired mutations, 12 of which were
proteinaffecting variants (validated by capillary sequencing)
including one non-sense mutation in the ZNF808 gene, one
nucleotide change at the 3’ splice site of RIMS4, and 10
missense mutations considered disease associated by at
least one pathogenicity prediction program (Table 1).
Gene ontology enrichment analysis of the 12 genes
harboring protein-affecting somatic mutations revealed
several biological processes such as muscle tissue/organ
development (ZFPM2 and MEGF10), which is related
to the mesothelium origin of this tumor, cell adhesion
(DPP4, CDH9, CNTNAP4, MEGF10), response to
mechanic stimulus (MEIS2, TRPA4), and response to abiotic
chr16:76495948 CNTNAP4 c.1210G>T, p.A404S
c.1288G>T, p.E430Ter Nonsense
28% (29×) 37×
20% (15×) 18×
37% (41×) 46×
22% (82×) 39×
25% (71×) 75×
40% (25×) 20×
28% (36×) 44×
36% (45×) 35×
26% (31×) 23×
33% (42×) 43×
26% (38×) 33×
43% (30×) 27×
Table 1 Description of somatically acquired point mutations detected in the DSRCT by whole-exome sequencing
Variant type Frequency Coverage
(tumor of leukocyte
Synonymous 35% (55×) 53×
Synonymous 23% (31×) 28×
Synonymous 43% (82×) 66×
stimulus (DPP4, MEIS2, TRPA1) (Additional file 3:
Table S2). Intriguingly, biological network analysis
obtained using Ingenuity Pathway Analysis (IPA)
interconnected the 15 genes harboring somatic mutations in
a single network associated with cell death and survival,
cell damage or degeneration, and nervous system
development and function (Fig. 1a).
Searching for germline variants associated with DSRCT
In an attempt to identify de novo germline mutations and
inherited variants potentially associated with DSRCT, we
performed whole-exome sequencing of the leukocyte DNA
from the patient’s parents (Additional file 2: Table S1). For
selecting genetic variants, a minimum coverage of 10× in
all samples with at least 30% frequency of the variant allele
was considered. Based on these criteria, no de novo
variants were found.
Next, we explored the possibility of finding genetic
variants associated with DSRCT in an autosomal
recessive model of inheritance. First, we searched for
polymorphisms (MAF ≤10%) occurring in homozygosity in
the patient that were inherited from both heterozygous
parents. Four polymorphisms inherited in homozygosis
were found in ADAMTS12, RASSF1, VEZT, and ISX genes
(population frequency ranging from 2.0 to 7.8%). All
polymorphisms lead to missense alterations, two of them
reported as possibly disease associated by at least one
pathogenicity prediction program (Table 2). Interestingly,
biological network analysis showed an interconnection
between the four genes in a single network from IPA,
showing association with cell cycle and digestive system
development and function (Fig. 1b).
Next, to identify candidate genes affected by compound
heterozygosity, we looked for genes containing two
distinct variants inherited independently from each parent.
We could confirm 11 genes affected by germline
compound heterozygous mutations, in which one variant allele
was inherited from the mother and the other variant allele
was inherited from the father (MAF ≤10%) (Table 3). We
speculate that these genes could be associated with risk
to DSRCT development in an autosomal recessive
model of inheritance.
Gene ontology analysis of these 11 genes showed
enrichment of biological processes related to muscle tissue
development (LAMB2, SYNE1, and TTN),
morphogenesis (C2CD3 and TTN), and cell cycle (RSPH1, TTN, and
SPICE1) (Additional file 4: Table S4). Functional analysis
of IPA revealed that 9 of the 11 genes are interconnected
in a single network related to cancer, organismal injury
and abnormalities, and gastrointestinal disease.
Copy number alterations
Genomic copy number alterations (CNA) were
investigated in the DSRCT tumor sample using array CGH in
a 180-K platform. Few copy number alterations were
detected (Fig. 2a), including aneuploidies such as gain of
chromosomes 5 and 18, and 11p, 13q, and 22q deletions. Only
one small focal homozygous deletion was identified in a
segment of ~1.3 Mb at 9p22.2 (chr9:17,106,384-18,449,088;
hg19), encompassing the CNTLN and SH3GL2 genes.
Deleterious Damaging Disease causing
Deleterious Damaging Disease causing
Deleterious Damaging Disease causing
Damaging Disease causing
Fig. 1 Network analysis by IPA. a Interaction network of genes harboring protein-affecting somatic mutations (network score = 45) is
associated with the top disease and functions: cell death and survival, nervous system development and function, cellular compromise. b
Interaction network of genes harboring rare polymorphisms detected in homozygosis in the patient (network score = 12) is associated with
the top disease and functions: cell cycle, digestive system development and function, hair and skin development and function. c Interaction
network of genes affected by compound heterozygous variants (network score = 25) is associated with the top disease and functions: cancer,
organismal injury, abnormalities, and gastrointestinal disease. Continuous and dashed lines indicate direct and indirect interactions between
molecules, respectively. Blue molecules represent the genes encountered in our analysis and blank molecules represent other genes
automatically included by IPA. Molecules are displayed by various shapes depending on the functional class of the gene product, according
to IPA Path designer shapes (Additional file 7)
The detection of copy number losses affecting
terminal segments of 11p and 22q suggested the presence
of the chromosomal translocation t(11;22)(p13;q12)
involving EWSR1 and WT1 genes.
Additionally, we used NGS data from the WES of the
tumor and patient’s leukocyte to search for CNAs, and
98.4% of the events detected by array CGH were validated,
demonstrating the efficacy of WES for identification of a
Table 2 Description of rare polymorphisms detected in homozygosity in the DSRCT patient. All variants were validated by sanger
c.1486G > A p.V496I
Missense rs10507051 (0.0302)
Missense rs8140287 (0.0308)
Missense rs2073498 (0.0711)
ADAMTS12 c.3529 T > C p.W1177R Missense rs3813474 (0.0513)
DC P P
MAF Minor allele frequency, PrD Probably damaging, B Benign, PD Possibly damaging, D Damaging, T Tolerated, DC Disease causing, P Polymorphism. (*) low
wide range of somatic events, besides point mutations and
indels (Fig. 2b; Additional file 5: Table S3).
Establishment of a personalized monitoring strategy
based on detection of circulating tumor DNA (ctDNA)
The genomic translocation t(11;22)(p13;q12), which is
considered the molecular hallmark of this tumor type, was
initially detected by FISH analysis for diagnostic purposes.
Furthermore, the same translocation was also detected in
the array CGH and confirmed by whole-genome
sequencing. The mate-pair approach used for whole-genome
sequencing followed by validation by PCR and Sanger
sequencing allowed the precise delimitation of the
chromosomal breakpoints (Fig. 3). This somatic fusion
event was then used as a tumor biomarker for
monitoring the patient along the follow-up period using a liquid
biopsy-based approach. We searched for traces of
circulating tumor DNA in plasma samples collected 17,
36, 42, and 47 months after surgery. No signs of the
translocation in any of the post-treatment plasma
samples were detected (Fig. 3) using digital droplet PCR
(ddPCR). The absence of the biomarker in any of the
post-treatment plasma samples is in agreement with
favorable clinical response of the patient, showing
longterm disease-free survival and no sign of disease
recurrence. As a positive control, we confirmed the presence
of cell-free circulating DNA by detecting non-rearranged
WT1 gene in all plasma samples.
Despite the unavailability of plasma samples at the
moment of diagnosis and before treatment (neoadjuvant/
adjuvant chemotherapy and surgery) to be used as a
baseline positive control, we carried out stringent control
assays. We tested the robustness of the approach by
evaluating six different amounts of tumor DNA, ranging from
60 ng to 6 pg, and searched for both the tumor-specific
fusion event and for the non-rearranged WT1 gene. We
achieved a linear quantification of tumor marker and
wild-type DNA, and we were able to discriminate the
presence of the fusion event even in extremely low
quantity of DNA (6 pg) (Fig. 3d). To test the specificity
of the ddPCR assay, we used leukocyte sample of the
patient and also DNA from plasma and leukocyte
samples of healthy donors, and the absence of detection of
tumor-specific fusion confirmed the assay specificity
for tumor DNA, without false positive signals (Additional
file 6: Figure S2). Further, we mimicked a situation of
circulating tumor DNA in body fluid by mixing 1 part of
tumor DNA to 100 and 1000 parts of leukocyte DNA,
then screened for the fusion event using 1 ng of this DNA
mix. Typically, the detection of circulating tumor DNA
has been reported as a fraction between 0.1 and 90% of
plasma DNA for cancer patients, depending on tumor
type and patient characteristics . Here, we were able to
detect the fusion event in the 1.0 and 0.1% fractions,
supporting the reliability of our established approach
for patient surveillance based on liquid-biopsy screening.
DSRCT is an aggressive tumor not yet broadly
investigated with the powerful genomic tools available. Mainly
due to the rarity of this disease, only a few studies
investigated molecular alterations in this tumor type. Shukla
et al.  screened the occurrence of 275 COSMIC
mutations in 29 oncogenes and found no alterations in any of
the 24 DSRCT samples investigated. More recently, a study
based on targeted exome sequencing of six adult patients
with pediatric-type malignancies found AURKB and MCL1
amplifications and variants of unknown clinical significance
in ARID1A and RUNX1 genes in one DSRCT .
Here, by performing a comprehensive screening of the
genomic alterations in a family based approach, we
identified somatic and germline variants possibly associated
with DSRCT. In our study, the use of a commercially
available gene panel did not show to be an adequate
strategy. Based on this finding, one can argue that screening
by commercially available gene panels is not an effective
approach for most cases of rare tumors, since the targeted
genes represented in these panels are usually those well
characterized in common solid tumors. On the other
hand, the use of WES not only revealed mutated genes
but also showed robustness for detecting DNA copy
number alterations. Concordance rates between WES and
GOLGA3 c.209G > A
MTMR6 c.685C > G
SLC9A9 c.1765A > G
c.65147C > T p.S21716L rs13021201
MAF Minor allele frequency, PrD Probably damaging, B Benign, PD Possibly damaging, D Damaging, T Tolerated, DC Disease causing, P Polymorphism. (*) low
array CGH (the gold standard for CNA screening) were
above 98% (Fig. 2). Among the CNAs not detected in the
WES analysis, two of them were mapped to regions not
covered by the library probes and five were low-level
mosaic alterations (Additional file 3: Table S2). Thus, if we
consider CNAs mapping to WES target regions and
nonmosaic CNAs, concordance rates were above 99.8%.
In total, 38 somatically acquired alterations, including
point mutations (15) and CNAs (23), were detected. The
small number of somatic alterations identified here is in
agreement with what is expected for pediatric tumors
. Moreover, given the occurrence of the driver
EWSWT1 fusion protein, additional oncogenic mutations for
tumor onset is probably less necessary.
Fig. 2 (See legend on next page.)
(See figure on previous page.)
Fig. 2 Array CGH profile showing the pattern of somatic copy number alterations detected in the DSRCT genome. a Copy number alterations detected
by array CGH analysis using a 180-K platform with an effective resolution of ~70 Kb: aneuploidy of chromosomes 5 and 18 (gains, in blue), and partial
losses of chromosome 13q, 11p, and 22q (in red). The green circle indicates the focal deletion of a segment of 1.3 Mb at 9p24.1. Arrows indicate
chromosome 11 and chromosome 22 breakpoints, 11p13 and 22q12.2, respectively. b Circus plot shows the copy number alterations detected by array
CGH and WES. Only the genomic regions affected by CNA events are represented. The numbers on each chromosome region are described in
megabases. In blue, data from array CGH and in green data from WES. A great overlap of CNA detection can be observed using both approaches
Fig. 3 Use of the chromosomal translocation t(11;22)(p13;q12) as a personalized tool for patient monitoring along follow-up. a FISH analysis shows
break apart probes for WT1 gene, indicating the occurrence of the fusion. b Mate-pair whole-genome sequencing detected paired reads mapping to
EWS and WT1 genes. c PCR amplification followed by Sanger sequencing confirmed the breakpoint region involving intronic regions of EWS and WT1
genes. d Digital droplet PCR assays for detection of the somatic rearrangement EWS-WT1. Left panel—Screening of ctDNA from plasma samples
collected serially along patient follow-up by ddPCR. No gene fusion was detected in ctDNA from the patient collected in four different time
points after surgery, suggesting no relapse, recurrence, or progression of the disease. Presence of cell-free DNA is shown by detection of
non-rearranged WT1 probes in the plasma samples from the patient and from control plasma sample. Middle panel—Serial dilutions of tumor
DNA to check the sensibility of the approach in detecting the fusion event, starting from 60 ng of input following five dilution series of
tenfold as indicated. Right panel—Detection of somatic rearrangement in different tumor DNA fractions, 1.0 and 0.1%
We identified 12 genes affected by somatic mutations
possibly involved with the disease. These genes are
involved with cellular development and morphology that are
pathways in which WT1 gene plays an important role.
Similarly, the genes affected by compound heterozygous
mutations showed enrichment in biological processes of
muscle tissue development and morphogenesis. Altogether,
these data suggests that disruption of the embryonic
cellular development process is involved with DSRCT onset,
which is commonly seen in pediatric tumors.
Another interesting finding is that among the set of
somatically mutated genes, 8 showed to be mutated in
desmoplastic melanoma samples from TCGA project
[29, 30], in frequencies ranging from 5 to 25% of the 20
samples interrogated. This data suggests that mutation
in these genes might be involved with the desmoplastic
phenotype, seen in both tumor types.
Additionally, 7 out of 15 genes harboring somatic
mutations (CHL1, MEGF10, MEIS2, MYH8, RIMS4, TBPL1,
and ZFPM2) are regulated by the same transcription
factor, LEF1 (p < 0.001 by enrichment analysis), which, in
turn, is regulated by WT1 . We therefore postulate
that DSRCT tumors presenting increased activity of WT1
due to EWS-WT1 fusion might upregulate the expression
of several genes mediated by LEF1 transcription factor.
However, the accumulation of mutations in this set of genes
regulated by LEF1 activation and its interrelation with the
EWS-WT1 fusion protein remains to be addressed.
Finally, the definition of the precise genomic breakpoint
of the t(11;22)(p13;q12) translocation by whole-genome
and Sanger sequencing enabled the development of a
personalized tool to precisely monitor the presence of ctDNA
in plasma samples during the patient’s follow-up.
Detection of tumor-specific genomic rearrangements has been
shown as a sensitive and specific method for monitoring
of disease status of cancer patients [32–35] and has clear
advantages over point mutations concerning the
specificity of detection . Here, we applied ctDNA screening
in plasma samples collected in three clinical appointments
during 3 years after surgery and, up to now, we did not
detect the presence of this tumor marker. These results
are in agreement with imaging exams (CT scan and
PET scan) showing no signs of disease and also with
good overall clinical condition of the patient and
highlight the applicability of using genomic rearrangement
for building personalized tool for patient surveillance.
After defining the genomic breakpoint of EWS-WT1
fusion, DSRCT patients can benefit from a highly
specific test that has the advantage of a rapid turnaround
time and potentially higher sensitivity in detecting
disease progression earlier than imaging exams or other
cancer antigens measurements, as reported for other
tumor types [36, 37]. Monitoring through liquid biopsy
is particularly attractive for solid tumors, which cannot
be repeatedly sampled without more invasive
procedures. Considering the rarity of this subtype of sarcoma
and the lack of effective treatment, the detection of
specific tumor marker and the monitoring of its persistence
can improve the identification of patients with worse
prognosis to tailor the treatment more properly. Thus,
the perspective is to employ the approach used here for
new patients and improve the outcome for those with
To our knowledge, this is the first comprehensive
genomic characterization of one DSRCT case. Continuous
efforts to establish the genomic landscape of rare
diseases, frequently neglected in large sequencing
consortiums, are highly significant to improve the knowledge
of defective pathways involved with tumor onset in
general, in addition to the strong potential of revealing
druggable targets for clinical use.
Additional file 1: Figure S1. Patient medical history and sample collection
time points. Chemotherapy treatment marked in green consisted of 4 cycles
of vincristine, cyclophosphamide, and doxorubicin (VAC) alternated with
ifosfamide, carboplatin, and etoposide (ICE). Chemotherapy treatment marked
in red consisted of hyperthermic intraperitoneal chemotherapy (HIPEC) with
cisplatin and doxorubicin. Abdominal irradiation total of 30 Gy. (JPG 96 kb)
Additional file 2: Table S1. Statistics of sequencing results. Sequence
coverage by Comprehensive Cancer Panel (Thermo Scientific), Whole
Exome Sequencing (Thermo Scientific) and Whole Genome Sequencing
(Mate-Paired approach) (Thermo Scientific). (DOC 38 kb)
Additional file 3: Table S2. Gene Ontology enriched categories of
genes affected by somatic mutations. Biological processes with a p-value
<0,001 was considered based on Webgestalt annotation tool [38, 39].
(DOC 31 kb)
Additional file 4: Table S4. Gene Ontology-enriched categories of
genes affected by compound heterozygous mutations. Biological
processes with a p value <0.001 was considered based on WebGestalt
annotation tool [38, 39]. (DOC 32 kb)
Additional file 5: Figure S3. Copy Number Alterations detected by
array CGH and confirmed by WES. (DOC 57 kb)
Additional file 6: Figure S2. Screening of ctDNA in leukocyte
samples. Pre-surgery sample collected at day of surgery. Post-surgery
sample collected at 22 months after diagnosis (17 months after
surgery). (JPG 75 kb)
Additional file 7: Description of the functional class of each molecule
shape used by IPA. (PDF 99 mb)
CGH: Comparative genomic hybridization; CNA: Copy number alterations;
COSMIC: Catalogue of Somatic Mutations in Cancer; CT: Computed
tomography; ctDNA: Circulating tumor DNA; ddPCR: Digital droplet PCR;
DSRCT: Desmoplastic small round cell tumor; GATK: Genome Analysis
Toolkit; HIPEC: Hyperthermic intraperitoneal chemotherapy; ICE: Ifosfamide,
carboplatin, and etoposide; Indel: Small insertions and deletions; IPA: Ingenuity
pathway analysis; MAF: Minor allele frequency; PET: Positron emission
tomography; SNP: Simple nucleotide polymorphism; TCGA: The cancer
genome atlas; VAC: Vincristine, cyclophosphamide, and doxorubicin;
WES: Whole-exome sequencing
The authors thank the patient and his parents for their collaboration, the A.C.
Camargo Cancer Center Biobank for providing the tumor sample, and Louise D.
C. Mota for the technical support for plasma DNA collection and extraction.
This work was supported by São Paulo Research Foundation—FAPESP
(2013/23277-8)—and by the National Council of Technological and Scientific
Availability of data and materials
Data generated during the current study are available in the SRA repository,
under the provisional project accession number PRJNA340004.
ENF, BDFB, and GTT carried out the sequencing experiments, and BDFB
and SG performed the validation experiments. ENF performed and
analyzed the ddPCR assays. ACVK performed and analyzed the CGH array
experiments and revised the manuscript. JES and RVA conducted the
bioinformatics analyses, and SJS supervised the bioinformatics analysis.
CCF and AL were responsible for patient treatment and care. IWC, CP,
and FAS supervised immunohistochemical reactions and performed the
histopathological analyses. SJS and EDN contributed to the analysis of the
results and revised the manuscript. DMC and ENF designed the study,
analyzed the results, and wrote the initial manuscript. DMC conceived and
supervised the study. All authors read and approved the final manuscript.
The authors declare that they have no competing interests.
Ethics approval and consent to participate
Both the patient and his parents signed an informed consent form. This
study was performed in compliance with the Helsinki Declaration and was
approved by the ethics committee of the A. C. Camargo Cancer Center
under number 1819/13.
1. Lettieri CK , Garcia-Filion P , Hingorani P. Incidence and outcomes of desmoplastic small round cell tumor: results from the surveillance, epidemiology, and end results database . J Cancer Epidemiol . 2014 ; 2014 : 680126 .
2. Zhang S , Zhang Y , Yu YH , Li J. Results of multimodal treatment for desmoplastic small round cell tumor of the abdomen and pelvis . Int J Clin Exp Med . 2015 ; 8 : 9658 - 66 .
3. Liu J , Nau MM , Yeh JC , Allegra CJ , Chu E , Wright JJ. Molecular heterogeneity and function of EWS-WT1 fusion transcripts in desmoplastic small round cell tumors . Clin Cancer Res . 2000 ; 6 : 3522 - 9 .
4. Lee SB , Kolquist KA , Nichols K , Englert C , Maheswaran S , Ladanyi M , Gerald WL , Haber DA . The EWS-WT1 translocation product induces PDGFA in desmoplastic small round-cell tumour . Nat Genet . 1997 ; 17 : 309 - 13 .
5. Karnieli E , Werner H , Rauscher 3rd FJ , et al. The IGF-I receptor gene promoter is a molecular target for the Ewing's sarcoma-Wilms' tumor 1 fusion protein . J Biol Chem . 1996 ; 271 : 19304 - 9 .
6. Silva JG , Corrales-Medina FF , Maher OM , Tannir N , Huh WW , Rytting ME , Subbiah V. Clinical next generation sequencing of pediatric-type malignancies in adult patients identifies novel somatic aberrations . Oncoscience . 2015 ; 2 : 187 - 92 .
7. La Starza R , Barba G , Nofrini V , Pierini T , Pierini V , Marcomigni L , Perruccio K , Matteucci C , Storlazzi CT , Daniele G , Crescenzi B , Giansanti M , Giovenali P , Dal Cin P , Mecucci C. Multiple EWSR1-WT1 and WT1- EWSR1 copies in two cases of desmoplastic round cell tumor . Cancer Genet . 2013 ; 206 : 387 - 92 .
8. Shukla N , Ameur N , Yilmaz I , Nafa K , Lau CY , Marchetti A , Borsu L , Barr FG , Ladanyi M. Oncogene mutation profiling of pediatric solid tumors reveals significant subsets of embryonal rhabdomyosarcoma and neuroblastoma with mutated genes in growth signaling pathways . Clin Cancer Res . 2012 ; 18 : 748 - 57 .
9. Campos AH , Silva AA , Mota LD , Olivieri ER , Prescinoti VC , Patrão D , Camargo LP , Brentani H , Carraro DM , Brentani RR , Soares FA . The value of a tumor bank in the development of cancer research in Brazil: 13 years of experience at the A C Camargo hospital . Biopreserv Biobank . 2012 ; 10 : 168 - 73 .
10. Olivieri EH , Franco Lde A , Pereira RG , Mota LD , Campos AH , Carraro DM . Biobanking practice: RNA storage at low concentration affects integrity . Biopreserv Biobank . 2014 ; 12 : 46 - 52 .
11. Torrezan GT , da Silva FC , Santos EM , Krepischi AC , Achatz MI , Aguiar Jr S , Rossi BM , Carraro DM . Mutational spectrum of the APC and MUTYH genes and genotype-phenotype correlations in Brazilian FAP, AFAP, and MAP patients . Orphanet J Rare Dis . 2013 ; 8 : 54 .
12. Database of genomic variants . http://projects.tcag.ca/variation/.
13. McKenna A , et al. The genome analysis toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data . Genome Res . 2012 ; 20 : 1297 - 303 .
14. Cingolani P , Platts A , le Wang L , Coon M , Nguyen T , Wang L , Land SJ , Lu X , Ruden DM . A program for annotating and predicting the effects of single nucleotide polymorphisms, SnpEff: SNPs in the genome of Drosophila melanogaster strain w1118; iso-2; iso-3 . Fly (Austin). 2012 ; 6 : 80 - 92 .
15. Liu X , Jian X , Boerwinkle E. dbNSFP: a lightweight database of human non-synonymous SNPs and their functional predictions . Hum Mutat . 2011 ; 32 : 894 - 9 .
16. Liu X , Jian X , Boerwinkle E. dbNSFP v2.0: a database of human nonsynonymous SNVs and their functional predictions and annotations . Hum Mutat . 2013 ; 34 : E2393 - 402 .
17. Forbes SA , Bindal N , Bamford S , Cole C , Kok CY , Beare D , Jia M , Shepherd R , Leung K , Menzies A , Teague JW , Campbell PJ , Stratton MR , Futreal PA. COSMIC: mining complete cancer genomes in the catalogue of somatic mutations in cancer . Nucleic Acids Res . 2011 ; 39 : D945 - 50 .
18. 1000 Genomes Project Consortium, Auton A , Brooks LD , Durbin RM , Garrison EP , Kang HM , Korbel JO , Marchini JL , McCarthy S , McVean GA , Abecasis GR . A global reference for human genetic variation . Nature . 2015 ; 526 : 68 - 74 .
19. Exome Variant Server , NHLBI GO Exome Sequencing Project (ESP) , Seattle. http://evs.gs.washington.edu/EVS/. Version: ESP6500SI - V2 .
20. International HapMap Consortium. The International HapMap Project . Nature . 2003 ; 426 : 789 - 96 .
21. Sherry ST , Ward MH , Kholodov M , Baker J , Phan L , Smigielski EM , Sirotkin K. dbSNP: the NCBI database of genetic variation . Nucleic Acids Res . 2001 ; 29 : 308 - 11 .
22. Database of Single Nucleotide Polymorphisms (dbSNP). Bethesda: National Center for Biotechnology Information, National Library of Medicine. (dbSNP Build ID : 138 ). Available from: http://www.ncbi. nlm.nih.gov/SNP/.
23. Magi A , Tattini L , Cifola I , et al. EXCAVATOR: detecting copy number variants from whole-exome sequencing data . Genome Biol . 2013 ; 14 :R120.
24. Klambauer G , Schwarzbauer K , Mayr A , Clevert DA , Mitterecker A , Bodenhofer U , Hochreiter S. cn. MOPS: mixture of Poissons for discovering copy number variations in next-generation sequencing data with a low false discovery rate . Nucleic Acids Res . 2012 ; 40 :e69.
25. Krzywinski M , Schein J , Birol I , Connors J , Gascoyne R , Horsman D , Jones SJ , Marra MA . Circos: an information aesthetic for comparative genomics . Genome Res . 2009 ; 19 : 1639 - 45 .
26. Zeitouni B , Boeva V , Janoueix-Lerosey I , Loeillet S , Legoix-ne P , Nicolas A , Delattre O , Barillot E. SVDetect: a tool to identify genomic structural variations from paired-end and mate-pair sequencing data . Bioinformatics . 2010 ; 26 : 1895 - 6 .
27. Taniguchi K , Uchida J , Nishino K , Kumagai T , Okuyama T , Okami J , Higashiyama M , Kodama K , Imamura F , Kato K. Quantitative detection of EGFR mutations in circulating tumor DNA derived from lung adenocarcinomas . Clin Cancer Res . 2011 ; 17 : 7808 - 15 .
28. Vogelstein B , Papadopoulos N , Velculescu VE , Zhou S , Diaz Jr LA , Kinzler KW . Cancer genome landscapes . Science . 2013 ; 339 : 1546 - 58 .
29. Gao J , Aksoy BA , Dogrusoz U , Dresdner G , Gross B , Sumer SO , Sun Y , Jacobsen A , Sinha R , Larsson E , Cerami E , Sander C , Schultz N. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal . Sci Signal . 2013 ; 6 : l1 .
30. Cerami E , Gao J , Dogrusoz U , Gross BE , Sumer SO , Aksoy BA , Jacobsen A , Byrne CJ , Heuer ML , Larsson E , Antipin Y , Reva B , Goldberg AP , Sander C , Schultz N. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data . Cancer Discov . 2012 ; 2 : 401 - 4 .
31. Gao Y , Toska E , Denmon D , Roberts SG , Medler KF. WT1 regulates the development of the posterior taste field . Development . 2014 ; 141 : 2271 - 8 .
32. McBride DJ , Orpana AK , Sotiriou C , Joensuu H , Stephens PJ , Mudie LJ , Hämäläinen E , Stebbings LA , Andersson LC , Flanagan AM , Durbecq V , Ignatiadis M , Kallioniemi O , Heckman CA , Alitalo K , Edgren H , Futreal PA , Stratton MR , Campbell PJ . Use of cancer-specific genomic rearrangements to quantify disease burden in plasma from patients with solid tumors . Genes Chromosomes Cancer . 2010 ; 49 : 1062 - 9 .
33. Leary RJ , Sausen M , Kinde I , Papadopoulos N , Carpten JD , Craig D , O'Shaughnessy J , Kinzler KW , Parmigiani G , Vogelstein B , Diaz Jr LA , Velculescu VE . Detection of chromosomal alterations in the circulation of cancer patients with whole-genome sequencing . Sci Transl Med . 2012 ; 4 : 162ra154 .
34. Donnard ER , Carpinetti PA , Navarro FC , Perez RO , Habr-Gama A , Parmigiani RB , Camargo AA , Galante PA. ICRmax: an optimized approach to detect tumorspecific interchromosomal rearrangements for clinical application . Genomics . 2015 ; 105 : 265 - 72 .
35. Olsson E , Winter C , George A , Chen Y , Howlin J , Tang MH , Dahlgren M , Schulz R , Grabau D , van Westen D , Fernö M , Ingvar C , Rose C , Bendahl PO , Rydén L , Borg Å , Gruvberger-Saal SK , Jernström H , Saal LH. Serial monitoring of circulating tumor DNA in patients with primary breast cancer for detection of occult metastatic disease . EMBO Mol Med . 2015 ; 7 : 1034 - 47 .
36. Dawson SJ , Tsui DW , Murtaza M , Biggs H , Rueda OM , Chin SF , Dunning MJ , Gale D , Forshew T , Mahler-Araujo B , Rajan S , Humphray S , Becq J , Halsall D , Wallis M , Bentley D , Caldas C , Rosenfeld N. Analysis of circulating tumor DNA to monitor metastatic breast cancer . N Engl J Med . 2013 ; 368 : 1199 - 209 .
37. Pereira E , Camacho-Vanegas O , Anand S , Sebra R , Catalina Camacho S , Garnar-Wortzel L , Nair N , Moshier E , Wooten M , Uzilov A , Chen R , PrasadHayes M , Zakashansky K , Beddoe AM , Schadt E , Dottino P , Martignetti JA . Personalized circulating tumor DNA biomarkers dynamically predict treatment response and survival in gynecologic cancers . PLoS One . 2015 ; 10 :e0145754.
38. Zhang B , Kirov SA , Snoddy JR . WebGestalt: an integrated system for exploring gene sets in various biological contexts . Nucleic Acids Res . 2005 ; 33 : W741 - 8 .
39. Wang J , Duncan D , Shi Z , Zhang B. WEB-based GEne SeT AnaLysis Toolkit (WebGestalt): update 2013 . Nucleic Acids Res . 2013 ; 41 : W77 - 83 .