Review Article
Full text online at http://www.jiaps.com
Pediatric rhabdomyosarcomas and nonrhabdomyosarcoma soft tissue
S. Agarwala
Department of Pediatric Surgery, All India Institute of Medical Sciences, New Delhi, India
Correspondence: Sandeep Agarwala, Department of Pediatric Surgery, All India Institute of Medical Sciences, New Delhi - 110 029,
India. E-mail: [email protected]
Tumors arising from the soft tissues are uncommon in children, accounting for about 6% of all childhood malignancies. More than half
(53%) of these originate from the striated muscles and are called rhabdomyosarcomas (RMS) the remaining are nonrhabdomyosarcoma
soft tissue sarcomas (NRSTS). Almost two-thirds of RMS cases are diagnosed in children <6 years of age. They can arise at varied
locations like the head and neck region, genitourinary tract, extremities, trunk and retroperitoneum. Pathologically RMS is now classified
as superior, intermediate and poor outcome histologies. For stratification of treatment and also comparison of results the RMS are now
staged both by the clinical grouping and the TNM staging systems. The ultimate outcome depends on the site, extent of disease and
histology. Currently, approximately 70% of the patients survive for 5 years or more and are probably cured. This is credited to the use of
multi-modal, risk-adapted therapy, refinements in tumor grouping and better supportive care which has emerged out of cooperative
studies like Intergroup Rhabdomyosarcoma Study (IRS) and the International Society of Pediatric Oncology studies (SIOP). The treatment
involves chemotherapy, radiotherapy and organ/function preserving surgery. The gold standard chemotherapy is still vincristine, actinomycin
D and cyclophosphamide (VAC) regime with high doses of intensity bone marrow rescue with colony stimulating factors. The NRSTS
are rare and of heterogenous histologies and so it has been difficult to arrive at a treatment strategy for these. What is definitely
understood is that these are usually immature and poorly differentiated tumors that respond poorly to chemotherapy and so surgical
resection forms the mainstay of treatment with adjuvant radiotherapy and chemotherapy to prevent local recurrences. In all likelihood,
the molecular analysis of RMS will further refine current classification schemes and knowledge of genetic features of the tumors will
significantly improve the ability of investigators to identify patients at lower or higher risk of treatment failures, thus paving the way for
advances in risk-based therapy.
KEY WORDS: Children, nonrhabdomyosarcoma soft tissue sarcomas, rhabdomyosarcomas.
The soft tissues refer to a wide range of different cell types
and include connective tissues, lymphatics, vessels,
smooth and striated muscles, fat, facsia, synovium,
endothelium and reticuloendothelium. Tumors arising
from these soft tissues are uncommon in children,
accounting for about 6% of all childhood malignancies.
More than half (53%) of these soft tissue sarcomas (STS)
originate from the striated muscles and are called
rhabdomyosarcomas (RMS).[1,2] The remaining group
(47%) consists of a heterogenous collection of subtypes
referred to as nonrhabdomyosarcoma soft tissue sarcomas
(NRSTS). Pediatric STS shows a striking difference in
the incidence as compared to their adult counterparts.
RMS, by far the commonest STS in children, is rare in
adults. In children RMS is commonly of the embryonal
histology as compared to pleomorphic variety in adults.
Similarly among the NRSTS, malignant fibrous
histiocytoma (MFH) comprises the most common
histology in adults, but is exceedingly rare in children. Of
the MFH also, only the angiomatoid variety, a low grade
lesion of borderline behavior, occurs in children.
Rhabdomyosarcoma is thought to arise from immature
mesenchymal cells that are committed to skeletal muscle
lineage, but these tumors are also known to arise in tissues
in which striated muscle is not normally found, such as
urinary bladder. Among the extracranial solid tumors of
childhood, RMS is the third most common neoplasm after
neuroblastoma and Wilms’ tumor, comprising 15% of all
solid tumors. Almost two-thirds of cases of RMS are
diagnosed in children <6 years of age although there is
another mid-adolescence peak. It is slightly more common
in males than in females (1.3-1.4: 1). It is ubiquitous
occurring almost everywhere but most commonly in the
J Indian Assoc Pediatr Surg / Jan-Mar 2006 / Vol 11 / Issue 1
Agarwala S: Pediatric soft tissue sarcomas
head and neck and the genitourinary (GU) areas. There
are certain distinctive clusters of features regarding age
at diagnosis, site of primary and histology. The head and
neck tumors are most common in children younger than
8 years of age and if arising in the orbit, are almost always
of embryonal histology. On the other hand, the extremity
tumors are more commonly seen in adolescents and are
more frequently of alveolar histology.
Currently local control in achieved in 80-90% of patients[3]
and approximately 70% of the patients survive for 5 years
or more and are probably cured.[4,5] Recent reports from
IRS-IV showed a 3-year failure free survival (FFS) of 83%
for nonmetastatic embryonal RMS.[6] This is credited to
the use of multi-modal, risk-adapted therapy, refinements
in tumor grouping and better supportive care which has
emerged out of cooperative studies like Intergroup
rhabdomyosarcoma study (IRS). In all likelihood, the
molecular analysis of RMS will further refine current
classification schemes and knowledge of genetic features
of the tumors will significantly improve the ability of
investigators to identify patients at lower or higher risk
of treatment failures, thus paving the way for advances
in risk-based therapy.
Rhabdomyosarcomas are grossly firm, nodular and of
variable size and consistency. They are well circumscribed
but not encapsulated and often tend to infilterate
extensively into adjacent tissues. Sarcoma botryoides
subtype has characteristic grape-like appearance with its
grape like clusters of tumors arising from a mucosa lined
area. Histologically RMS falls into the broad category of
small blue round cell tumor. The standard classification
is still the one proposed by Horn and Enterline[7] in 1958
which divided the tumor into four subgroups: embryonal,
alveolar, botryoid and pleomorphic and noted that
botryoid was actually a subtype of embryonal. Since there
was no overall agreement among the pathologists using
the conventional classification therefore an international
classification system for childhood RMS was proposed.[8]
This system is being used in all new IRS studies beginning
IRS IV [Table 1]. The histologic distribution of the tumor
in IRS III[3] is shown in [Table 2]. Light microscopy
diagnosis of RMS is based on the identification of cross­
striations, characteristic of skeletal muscle, or
characteristic rhabdomyoblasts. Cross-striations are seen
in 50-60% of the cases. Histologically embryonal
rhabdomyosarcoma (ERMS) is composed of
rhabdomyoblasts and small round cells. Rhabdomyoblast,
the more mature of the embryonal component, is
characterized by bright eosinophilic cytoplasm. Sarcoma
botryoides and spindle cell variant are two subtypes of
J Indian Assoc Pediatr Surg / Jan-Mar 2006 / Vol 11 / Issue 1
Table 1: International classification system for childhood
Superior prognosis
a) Botryoid
b) Spindle cell
II Intermediate prognosis
a) Embryonal
III Poor prognosis
a) Alveolar
b) Undifferentiated sarcomas
Table 2: Histologic distribution of childhood
Histologic type
RMS, type undetermined
Extraosseous Ewing’s sarcoma*
RMS, rhabdomyosarcoma, *No longer defined as rhabdomyosarcoma.
embryonal RMS. Alveolar RMS (ARMS) consists of
rhabdomyoblasts mixed with a larger round cells with
prominent eosinophilic cytoplasm. The tumor grows in
cords and produces cleft like spaces, namely alveoli. The
pleomorphic RMS, which is extremely rare in children
show anaplastic cells present in large aggregates or as
diffuse sheets. It occurs in the extremities and the trunk.
Electron microscopy and immunohistochemical analysis
of tumors are now useful tools for demonstrating
characteristics of RMS, especially when light microscopy
is inconclusive. The diagnostic EM features of RMS are
visible z-bands. Skeletal muscle or muscle-specific
proteins, like antidesmin, muscle-specific actin and Myo
D can be identified by immunohistochemical staining.
Monoclonal antibodies, like those to desmin, muscle­
specific actin, sarcomeric actin and myoglogin have also
been used to confirm the myogenic lineage with very good
specificity and sensitivity. [10] Monoclonal antibodies
against Myo D can be used in frozen section analysis also.
The ERMS and ARMS have been associated with distinct
clinical characteristics and genetic alterations. And ARMS
is associated with 2;13 or 1;13 chromosomal
translocations, which generate PAX3-FKHR and PAX7FKHR fusion proteins respectively. In contrast, ERMS
have allelic loss at chromosome 11p15.5. Therefore
despite similarities, the striking cytogenetic and molecular
differences between ARMS and ERMS indicate distinct
molecular etiologies in these tumor subtypes). [11]
Clinical staging
It is critical to assess the extent of tumor in every patient
as the therapy and prognosis depends on the degree to
which the mass has spread beyond the primary site.
Several surgico-pathologic staging systems have been used
Agarwala S: Pediatric soft tissue sarcomas
historically, but the clinical group staging system,
developed by IRS in 1972 has been most widely used.
IRS committee has now adopted a modification of the
so-called TNM system[4] [Table 3].
Patterns of spread
The tumor spreads locally to invade adjacent structures
and may also spread distantly via lymphatics and
hematogenous routes. Approximately 15% children with
RMS present with metastatic disease and their prognosis
has not improved over the last 15 years.[3,4,12,13] The most
Table 3a: Clinical grouping system used by the IRS I
through III[4]
Group I:
Localized disease, completely resected
A Confined to the organ or muscle of origin.
B Infiltration outside organ or muscle of origin;
regional nodes not involved.
Group II: Total gross resection with evidence of regional spread
A Grossly resected tumors with “microscopic” residual
B Regional disease completely resected with regional
nodes involved, tumor extension into adjacent organs
or both.
Group III: Incomplete resection or biopsy with gross residual
disease remaining.
A Localized or locally extensive tumor, gross residual
disease after biopsy only.
B Localized or locally extensive tumor, gross residual
disease after “major” resection (>50% debulking).
Group IV: Any size primary tumor, with or without regional
lymphnode involvement, with distant metastases,
irrespective of surgical approach to the primary tumor.
frequent sites of distant metastases are regional lymph
nodes, lungs, bone marrow, bones, central nervous system,
heart, liver and the breast. [14] The lung is the most
frequent site of metastases (40-50%) and majority (74%)
of these have bilateral metastases.[15,16] Less common sites,
either isolated or in conjunction with multi-metastatic
disease, are bone marrow (20-30%), bone (10%) and
depending on the site of primary tumor, lymph node (up
to 20%). Though visceral metastases is rare, but in cases
of treatment failures, predominantly, visceral metastasis
(e.g. brain or liver) may be seen. IRS-IV has reported that
among children with metastases, 46% had lung as the
only site of metastases while 50% had other sites.
Interestingly, 29% of the patients with metastases had
two sites of metastases and 16% had three sites
Prognostic factors
The most important prognostic variables identified appear
to be the extent of disease, i.e. patients with CG-IV or
TNM stage 4 fare worse than others. [17,18] Amongst
localized RMS, those tumors which are completely excised
surgically (CG-I) have better survival rate than those with
microscopic residue or those which are locally extensive
lesions (CG-II). Those with gross residual tumors (CGIII) fare less well but are much better of than CG-IV
patients. Histology is also an important prognostic factor.
Pooled data from IRS I, II and III show that 5-year survival
is related to the histology with 95% for sarcoma botryoides,
Table 3b: TNM pretreatment staging classification for RMS [4]
Staging prior to treatment requires thorough clinical examination and laboratory and imaging examinations. Biopsy is required to
establish the histologic diagnosis. Pretreatment size is determined by external measurement or MRI or CT, depending upon the
anatomic location. For less accessible primary sites, CT is employed as a means of lymphnode assessment as well. Metastatic
sites require some form of imaging (but not histologic confirmation, except for bone marrow examination) confirmation.
T1 or T2
T1 or T2
T1 or T2
a or b
a or b
N0 or N1 or Nx
N0 or Nx
N0 or N1 or Nx
N0 or N1
T1 or T2
Favorable sites are orbit, head and neck (excluding parameningeal), or genitourinary (excluding bladder/prostate)
Unfavorable sites are bladder/prostate, parameningeal, extremities, trunk and all others
T1 = Tumor confined to anatomic site of origin
a) <5 cm in diameter
b) >5 cm in diameter
T2 = Extension and/or fixation to surrounding tissues
a) <5 cm in diameter
b) >5 cm in diameter
Regional nodes:
N0, Regional nodes not clinically involved
N1, Regional nodes clinically involved by tumor
Nx, Clinical status of regional nodes unknown (specially sites which preclude lymph node evaluation)
M0, No distant metastases
M1, Metastases present
J Indian Assoc Pediatr Surg / Jan-Mar 2006 / Vol 11 / Issue 1
Agarwala S: Pediatric soft tissue sarcomas
75% for pleomorphic sarcoma, 66% for embryonal, 54%
for alveolar and 40% for undifferentiated RMS. [4,7,12,19]
Other unfavorable prognostic variables are older age at
diagnosis, [20] presence of regional LN metastases for
extremity and paratesticular tumors,[21,22] presence of
extensive bony erosions in cranial parameningeal
tumors,[23] DNA proliferative activity (S-phase fraction
>15%),[24,25] and diploid embryonal tumors. The most
meaningful prognostic variable is the response to
treatment, because those who never achieve complete
obliteration of the tumor do not survive.
Metastatic disease is the single most important predictor
of clinical outcome in patients with RMS.[16] Children
with metastatic RMS have a poorer prognosis with a 3­
year FFS of only 25%.[3,12] IRS-IV studies have shown that
children with lung-only metastases have a slightly better
prognosis as compared with metastases at other sites.[16]
Principles of treatment of RMS
There are three modalities of treating children with RMS.
These are surgery, radiation therapy for control of residual
bulk or microscopic tumor and systemic combination
chemotherapy for primary cytoreduction and eradication
of gross and micrometastases.
Principles of surgical management
Surgery is the most rapid way to ablate the disease and
should always be done if subsequent function or cosmesis
will not be greatly impaired. Surgery includes complete
resection of the primary tumor with surrounding margins
of uninvolved tissue during the initial surgery and any
subsequent operation. If microscopic residual disease is
found after initial resection, re-excision of the area is
indicated before any other nonsurgical management.
Debulking procedures have no value as initial biopsy and
neo-adjuvant therapy results in shrinkage of the tumor
allowing complete resection at second look operation.
This is better than partial or marginal resections. Overall
surgical resections for childhood RMS today are less
exenterative or mutilating than those employed three to
four decades earlier.[26]
In sites such as vagina, bladder and most of head and
neck an incisional biopsy may be the only feasible surgical
procedure. Open biopsy is the preferred technique of
obtaining a biopsy specimen and needle biopsy of fine
needle aspiration cytology is to be used only if open biopsy
is surgically hazardous. Clinically involved LN should
always be sampled, while the histologic evaluation of the
clinically uninvolved nodes is site specific as detailed
J Indian Assoc Pediatr Surg / Jan-Mar 2006 / Vol 11 / Issue 1
Second-look procedures are required in two clinical
situations: (1) to pathologically verify the completeness
of an apparently complete clinical (radiologic) remission
and (2) to resect any residual viable tumor cells that have
survived induction chemotherapy and local RT.
Principles of radiation therapy
Radiation therapy (RT) is a major tool in the treatment
of RMS. It can eradicate residual tumor cells, especially
in the head and neck region and the pelvis. According to
IRSG protocols all RMS should get RT to achieve long­
term local control of tumors. The dose delivered may
differ depending on many factors. The International
Society of Pediatric Oncology (SIOP) differs in this as
shown clearly by their study MMT-84. The major
difference in therapeutic approach between MMT-84 and
IRSG studies was the omission of radiotherapy (and also
of second look surgery) in patients with nonmetastatic
RMS who achieved complete remission with
chemotherapy alone. However, RT was given to patients
<5 years of age and having parameningeal tumors and
to patients more than 12 years of age having tumors at
any site. In this study with a median follow-up of 8 years,
the 5-year event-free survival (EFS) was 53% and the
overall survival was 68%. The EFS was inferior to IRSG
reports because of a higher rate of local recurrences. There
is also controversy regarding the role of pulmonary
radiotherapy for those with pulmonary metastases. IRSIV report[16] seems to suggest that pulmonary radiotherapy
does provide some benefit as evidenced by decreased
incidence of pulmonary recurrences and improved FFS
and overall survival.
Principles of chemotherapy
All patients with RMS must receive combination
chemotherapy as there is ample evidence that the
adjuvant or neoadjuvant therapy significantly improves
survival. The chemotherapy should begin as soon as
possible after diagnostic studies are completed or primary
excision done, as the major role of chemotherapy is the
eradication of microscopic foci of disease (local and
distant), thereby improving both local control and
survival. Preoperative chemotherapy may also be used in
unresectable tumors to reduce them to resectable size.
The current gold standard frontline chemotherapy
consists of vincristine, actinomycin D and
cyclophosphamide (VAC) was pioneered by Wilbur et al.
Two conventional regimes, which are most commonly
used are: intensive vincristine/actinomycin D (intensive
VA) and pulse vincristine/actinomycin D/
cyclophosphamide (pulse VAC).
Phase II trials from Europe and USA have shown that
Agarwala S: Pediatric soft tissue sarcomas
ifosfamide, as a single agent, is an active drug against
RMS. In combination with other drugs the response rate
is even better.[27] Irnotecan, a topoisomerase I inhibitor
also appears to have promising activity against RMS, with
minimal hematopoietic toxicity.[28]
Clinical presentation/management and outlook for
specific sites
The clinically evident signs and symptoms of RMS are in
two main ways: The appearance of a mass lesion and
disturbance of a normal body function by an unsuspected,
critically located enlarging mass. The common sites where
RMS is reported in children as per the IRS group III report
is shown in [Table 4]. Thirty-five percent of the cases
involve the head and neck region while the second most
common site is the GU tract (26%).[3]
Head and neck
The head and neck tumors are further divided into those
that arise in the parameningeal region (50%), orbit (25%)
and head and neck superficial, i.e. nonparameningeal
(25%) [Table 4].[29]
Orbital RMS produce proptosis, chemosis, eyelid or
conjunctival mass, opthalmoplegia and rarely blindness.
These are usually diagnosed early before distant
dissemination has occurred. Parameningeal tumors
usually cause nasal, aural or sinus obstruction. These are
often associated with cranial bone erosions that can
manifest as cranial nerve palsies. Erosion of contiguous
bone at the cranial base and intracranial extension may
lead to headache, vomiting and systemic hypertension.
Nasophranygeal tumors can cause voice changes, airway
obstruction, dysphagia and epistaxis while sinus tumor
can be painful in addition to be having persistent nasal
discharge and occasion epistaxis. Tumors of the middle
ear or mastoid can present as a polypoidal growth from
the ear, otitis media, or facial palsy. Laryngeal tumors can
present with hoarseness. Regional lymph node metastases
to cervical lymph nodes may be present in upto 20% cases
depending on the site. Distant metastasis is primarily to
the lungs or the bones. While the orbital tumors have
Table 4: The site of primary tumor as reported in IRS-III[21]
All Head/neck
GU (not BP)
BP, bladder/prostate, Figures in parenthesis are in percentage
375 (35)
109 (10)
111 (10)
155 (15)
167 (16)
110 (10)
202 (19)
208 (20)
very good prognosis, the parameningeal tumors have the
poorest prognosis.
For orbital tumors nonexcisional therapy is standard.
Initial biopsy followed by chemotherapy and radiation
leads to survival rates of >90%. Routine lymph node
sampling is not indicated as incidence of nodal spread is
only 3%. Orbital exenteration is now recommended only
for recurrent disease. Nonorbital, nonparameningeal
tumors and head and neck tumors (superficial) are mostly
unresectable. These patients are best managed by an
incisional biopsy followed by chemotherapy appropriate
for their group[3] Cervical lymph node dissection is not
warranted, however, the clinically suspicious nodes must
be biopsied and if histologically positive they must be
included in the RT portal. The 5-year survival rates are
approximately 80%. These patients are treated with
systemic chemotherapy and RT according to the group.
RT is also necessary for the spinal cord if CSF is +ve for
tumor cells. The 5-year survival rate is almost 70% for
nonmetastatic disease, but only 43% for metastatic
Genitourinary tract
These tumors arise in the bladder, prostate, vagina, uterus,
vulva, paratesticular regions and rarely the kidneys and
ureter and constitute nearly 26% of all RMS cases[3]
[Table 4]. The embryonal histology is the commonest in
this region and the most frequent of the genitourinary
rhabdomyosarcomas (GU RMS) are those of the bladder/
prostate (BP) [Table 4]. Within this category of GU RMS
are tumors with good prognosis, namely vulva, vagina and
paratesticular and those with poorer prognosis, namely
bladder and prostate (GU-BP). The bladder tumors
usually grow intra-luminally, in the region of the trigone
and have a polypoidal appearance on gross or endoscopic
examination. Tumors arising from the dome of the
bladder are uncommon as compared to those from the
trigone, but the dome tumors have a better outcome.
Children with bladder RMS are usually under 4 years of
age and may present with hematuria, urinary obstruction
and rarely extrusion of tumor tissue. Prostatic tumors can
occur in relatively older children and usually present as
large pelvic masses with or without urethral strangury and/
or constipation. Initial diagnostic workup includes an
ultrasound of the pelvis, micturating cystourethrogram
and a CECT or MRI of the abdomen. Workup for
metastatic disease should include an X-ray chest, bone
marrow aspiration and biopsy from both iliac crests and a
bone scan besides evaluation for retroperitoneal nodes
and liver secondaries in the CECT or MRI. IRS-IV
recommends initial endoscopic, perineal, or suprapubic
diagnostic biopsy followed by intensive chemotherapy and
J Indian Assoc Pediatr Surg / Jan-Mar 2006 / Vol 11 / Issue 1
Agarwala S: Pediatric soft tissue sarcomas
early RT. Initial complete resection is done only for those
patients who have a tumor of the dome of the bladder
and in whom the preservation of bladder and urethral
function can be assured. Anterior pelvic exenteration and
total cystectomy is reserved for patients who do not
achieve local control after chemotherapy and RT.
Currently, 60% of the patients retain a functional bladder
and the overall survival rate exceeds 85%. [30,31]
Paratesticular RMS arises in the distal area of the
spermatic cord and may invade the testis and the
surrounding tissues. These usually present as a
unilateral painless scrotal swelling or a mass above
the testis in pre- or postpubertal boys. Almost 30%
of the paratesticular tumors are of the spindle cell
variety which has an excellent prognosis. [32] Initial
inguinal orchiectomy with the removal of the entire
spermatic cord should be done. Scrotal violation or
trans-scrotal biopsy should be avoided to prevent
scrotal contamination. In case there is scrotal
violation then hemiscrotectomy should be done to
prevent metastases to the inguinal nodes (which are
considered distant metastases and not local). Now,
because of the availability of sensitive imaging
techniques like spiral CT and MRI, IRS-IV does not
recommend routine RPLN dissection for patients
with completely resected localized tumors and
negative imaging studies. The survival rates are now
around 90% with adjuvant chemotherapy and RT
directed to known nodal or residual disease. [22,33]
Vaginal tumors present at younger age than those of the
uterine (mean age 2 years vs 14 years). Vaginal RMS is
usually of the botryoid variety and present as
mucosanguinous discharge, bleeding, or a prolapsing
polypoidal mass. Cervical and uterine sarcomas are
diagnosed in older children who present with a mass and
history of vaginal discharge. Vulval RMS is usually seen
in older children as a mass lesion extending from the
vulval and can often be mistaken for an abscess. At present
the IRS approach for vaginal RMS is biopsy, to confirm
diagnosis, cystoscopy and CT scan of the pelvis to rule
out local spread. Chemotherapy is followed by repeat
vaginal examinations and biopsy without resectional
surgery. Persistent disease is managed by local limited
resection or partial vaginectomy. Uterine tumors are also
initially treated with chemotherapy and second look
surgery and radical resection (hysterectomy + proximal
vaginectomy) is required only for gross residual disease
who have failed to achieve a complete radiographic
response within 6 months of induction chemotherapy
and RT, or those who have early progression.[34]
J Indian Assoc Pediatr Surg / Jan-Mar 2006 / Vol 11 / Issue 1
RMS involving the extremities comprise 19% of all RMS
[Table 4] and are characterized by a swelling in the
affected body part. They involve the lower extremity more
than the upper and distal limb involvement is more
common than proximal. Pain, tenderness and redness may
occur and almost 45% of these are of alveolar histology.[35]
Limb sparing, wide resection of the tumor is
recommended whenever feasible and without loss of
function since excision results in improved results. [36]
Amputation should be avoided. Since more than one­
fourth of these patients have metastases to the regional
lymph nodes therefore IRS-IV recommends routine LN
sampling, even if clinically negative. Positive lymph nodes
warrant their inclusion in the RT portal. If surgical margins
are microscopically positive the re-excision should be done
prior to chemotherapy and RT. [37] Postoperatively all
patients get chemotherapy and radiotherapy. The survival
rate was 80% when lymph nodes were not involved as
compared to 46% when involved. [36] There is also an
increased rate of distant metastases when regional LN
was involved.[38]
Retroperitoneal (RP) tumors, excluding the GU tract,
account for 11% of cases and can be either embryonal or
alveolar. Complete resection of these tumors is often
impossible for technical reasons and regional LN are often
involved and distant metastases present at the time of
diagnosis. Treatment includes chemotherapy and RT
according to the clinical grouping. Patients with RP-RMS
have the worst prognosis with a 5-year survival rate, of
nonmetastatic tumor, being only 50%.
These sites include tumors of the chest wall, paraspinal
region and the abdominal wall in decreasing order of
frequency and constitute 10% of all cases of RMS. Most
of these are of alveolar histology and <30% are amenable
to complete resection at diagnosis.[32] Whenever excised
they have a tendency to recur locally. Treatment includes
chemotherapy and RT according to the clinical grouping.
The rarity and histologic heterogenecity [Table 5] of
NRSTS in children preclude careful study of their natural
history and response to therapy. NRSTS like RMS can
arise in any part of the body, but the most common sites
are extremities, trunk, abdomen and pelvis. [39] The most
frequent histologic types are synovial sarcoma,
Agarwala S: Pediatric soft tissue sarcomas
Table 5: Some common and uncommon NRSTS seen in
Synovial sarcoma
Peripheral neuroectodermal tumor
Malignant fibrous histiocytoma
Neurogenic tumors
- Malignant schwanoma
- Neurofibrosarcoma
- Malignant peripheral
nerve sheath tumor
Clear cell sarcoma
Epitheloid cell sarcoma
Intra-abdominal desmoplastic
Malignant mesenchymoma
Extra skeletal chondrosarcoma
neurofibrosarcoma and fibrosarcoma. In the extremities
the tumor occurs mostly in the lower limbs. While most
of the extremity NRSTS in children are synovial sarcomas,
tumors of the trunk are predominantly malignant fibrous
histiocytomas (MFH), or neurogenic in origin.
Histiology and grading of NRSTS
Most of the NRSTS in children except for the MFH and
fibrosarcomas are immature and poorly differentiated with
approximately half of the tumors having a histiologic
grade G-3.[40] Because of the inconsistencies in predicted
behavior, a grading scheme for pediatric NRSTS is used,
which takes into account the cytohistiologic features that
are used for adult sarcomas, but with caveats of the
childhood lesions [Table 6]. This system is not used for
RMS or for primitive neuroectodermal lesions, which are
always considered high grade tumors.
Prognostic factors for NRSTS
The prognostic factors for children with NRSTS include
the presence of or absence of metastatic disease, surgical
respectability of the lesion, tumor histiologic grade, tumor
invasiveness and size of the lesion.[39] In a review of 154
children with NRSTS treated at a single institution,[39]
Table 6: POG schema for grading of NRSTS in children
Grade 1
- Myxoid well-differentiated liposarcoma
- Deep seated dermatofibrosarcoma protuberans
- Well-differentiated or infantile (age <5 years) fibrosarcoma.
- Well-differentiated or infantile (age <5 years)
- Well-differentiated malignant peripheral nerve sheath tumor.
- Angiomatoid malignant fibrous histiocytoma
Grade 2
Sarcomas not specifically included in grade 1 or 3; <15% of
tumor showing geographic necrosis or the mitotic index is <5
per 10 high power fields.
Grade 3
- Pleomorphic or round cell liposarcoma
- Mesenchymal chondrosarcoma
- Extra-skeletal osteosarcoma
- Malignant triton tumor
- Alveolar soft part sarcoma
31% of those with grade 1 or 2 lesions had treatment
failures, while 73% of the children with grade 3 disease
developed recurrent disease.
Clinical evaluation and staging of NRSTS
In all children suspected to have NRSTS the clinical
evaluation should include routine hemograms, renal and
liver function tests, bone scans and bone marrow
examination. MRI scan is considered the imaging
modality of choice for the evaluation of local and regional
disease, particularly in the extremities, the pelvis and head
and neck regions. The staging system currently used for
NRSTS is the same as the modified TNM staging system
used for RMS. The optimal method for obtaining tissue
for diagnosis in patients with NRSTS is again datable.
Although fine needle aspiration cytology (FNAC) is a
useful diagnostic tool in the initial evaluation of NRSTS
or a possible metastatic lesion, needle core biopsy (NCB)
is better for providing enough tissue to permit accurate
histologic subtyping of a sarcoma. Excisional biopsy rarely
should be used in the initial evaluation of these tumors.
Simple excision of the tumor violates the tissue planes
and results in dissemination of the tumor cells throughout
the operative field. Subsequent surgery in the region is
thereby compromised. An excisional biopsy is undertaken
only in those instances when the tumor is small
(<2.5 cm) or situated so that an eventual wide local
resection can be done without risk or functional deformity.
When an incisional biopsy is done it should be properly
planned. For an extremity lesion, the incision should be
planned longitudinally or parallel to the neurovascular
Treatment and outcome of NRSTS
During the past few years the surgical management of
these tumors has undergone a considerable evolution with
the realization that multimodal therapy provides the best
chance for survival. Unlike that for RMS, which is a highly
chemosensitive tumor, the mainstay of treatment of
NRSTS is complete surgical resection with or without
adjuvant radiotherapy to prevent local recurrence. Several
prospective adult trials have failed to document survival
benefit of adjuvant chemotherapy.[9] The only prospective
pediatric trial addressing the value of adjuvant
chemotherapy in patients with NRSTS was conducted
by the Pediatric Oncology Group (POG). In this trial, 75
children with completely resected NRSTS lesions were
assigned to receive observation vs adjuvant chemotherapy
with VAC and doxorubicin. The 3-year disease-free
survival rate for the two groups did not differ (74% vs
76%). Sub-group analysis disclosed that patients with
grade 3 lesions fared significantly worse than those with
grades 1 and 2 lesions (3-year EFS, 75% vs 91%;
J Indian Assoc Pediatr Surg / Jan-Mar 2006 / Vol 11 / Issue 1
Agarwala S: Pediatric soft tissue sarcomas
P = 0.018).[9] Distant relapses accounted for more than
80% of the failures in the high grade group. The outcome
for children with metastatic NRSTS continues to be poor;
fewer than 20% of the patients are disease free at 3 years.[9]
The most active drugs against NRSTS include ifosfamide
and doxorubicin. Currently POG is investigating the
clinical activity of some combination chemotherapies for
unresectable or metastatic NRSTS. If these prospective
trials identify some beneficial outcomes then these active
agents may be also tried as adjuvant chemotherapy in
children with completely resected high grade NRSTS.
Wide local excision or en bloc resection should be the
primary form of treatment in children with NRSTS. All
attempts should be made to obtain negative margins.
What consists of an adequate margin of tissue is still
debated. In some areas such as head and neck,
mediastinum and retroperitoneum, wide local excision
with clear margins may be impossible to achieve [41]
without mutilating resections. The finding of microscopic
involvement of surgical margins is highly predictive for
local disease recurrence, distant disease recurrence and
diminished overall survival.[42,43] This is why primary re­
excision should have priority over any adjuvant therapy.[44]
Adjuvant RT is recommended in all cases of NRSTS in
adults as these tumors respond to RT. This is not the case
in childhood NRSTS. RT has been used sparingly in
children because of its long-term effects. [45] Current
recommendations for children are to avoid RT for grades
1 and 2 completely resected tumors, however,
incompletely resected tumors require additional therapy
for local control.
Miller RW, Young JL, Jr, Novakovic B. Childhood cancer. Cancer
Miser JS, Triche TJ, Kinsella TJ, Pritchard DJ. Other soft tissue
sarcomas of childhood. In: Pizzo PA, Poplack DG (editors) Principales
and practice of pediatric oncology. Lippincott-Raven: Philadelphia;
1997. p. 865-88.
Crist W, Gehan EA, Ragab AH, Dickman PS, Donaldson SS, Fryer
C, et al. The third Intergroup Rhabdomyosarcoma Study (IRS-III).
J Clin Oncol 1995;13:610-30
Maurer HM, Beltangady M, Gehan EA, Crist W, Hammond D, Hays
D, et al. The Intergroup Rhabdomyosarcoma Study I: A final report.
Cancer 1988;61:209-20.
Pinkel D, Pickren J. Rhabdomyosarcoma in children. JAMA
Ruymann FB. The development of VAC chemotherapy in
rhabdomyosarcoma: what does one do for an encore? Curr Oncol
Rep 2003;5:505-9
Horn RC, Enterline HT. Rhabdomyosarcoma: a clinicopathologic
study and classification of 39 cases. Cancer 1958;11:181-99
Asmar L, Gehan EA, Newton WA, Webber BL, Marsden HB, van
Unnik AJ, et al. Agreement among and within groups of pathologists
in the classification of rhabdomyosarcoma and related childhood
sarcomas: a report of an international study of four pathology
J Indian Assoc Pediatr Surg / Jan-Mar 2006 / Vol 11 / Issue 1
classifications. Cancer 1994;74:2579-88.
Pappo AS, Shapiro DN, Cris WM. Rhabdomyosarcoma> Biology
and treatment [Review]. Pediatr Clin North Am 1997;44:953-72.
Tsokos M. The diagnosis and classification of childhood
rhabdomyosarcoma. Semin Diagn Pathol 1994;11:26-38.
Xia SJ, Pressey JG, Barr FG. Molecular pathogenesis of
rhabdomyosarcoma. Cancer Biol Ther 2002;1:97-104.
Maurer HM, Gehan EA, Beltangady M, Crist W, Dickman PS,
Donaldson S, et al. The Intergroup Rhabdomyosarcoma Study II.
Cancer 1993;71:1904-22.
Pappo A, Lyden E, Breneman J, Wiener E, Teot L, Meza J, et al.
Upfront window trial of topotecan in previously untreated children
and adolescents with metastatic rhabdomyosarcoma: an intergroup
rhabdomyosarcoma study. J Clin Oncol 2001;19:213-9.
Ruymann FB, Newton WA, Ragab A, Donaldson MH, Foulkes M.
Bone marrow metastasis at diagnosis in children and adolescents
with rhabdomyosarcoma: a report from the Intergroup
Rhabdomyosarcoma Study. Cancer 1984;53:368-73.
Breneman J, Lyden E, Pappo A. Prognostic factors and clinical
outcome in children and adolescents with metastatic
rhabdomyosarcoma - a report from IRS-IV. J Clin Oncol 2003;2178­
Rodeberg D, Arndt C, Breneman J, Lyden E, Donaldson S, Paidas
C, et al. Characteristics and outcome of rhabdomyosarcoma patients
with isolated lung metastasis from intergroup rhabdomyosarcoma
study-IV. J Pediatr Surg 2005;40:256-62.
Gehan EA, Glover FN, Maurer HM, Sutow WW, Hays DM, Lawrence
W Jr, et al. Prognostic factors in children with rhabdomyosarcoma.
Monogr Natl Cancer Inst 1981;56:83-90.
Okamura J, Sutow WW, Moon TE. Prognosis in children with
metastatic sarcoma. Med Pediatr Oncol 1977;3:243.
Newton WA, Soule EH, Hamoudi AB, Reiman HM, Shimada H,
Beltangady M, et al. Histopathology of childhood sarcomas,
Intergroup Rhabdomyosarcoma Studies I and II: clinicopathologic
correlation. J Clin Oncol 1988;6:67.
La Quaglia MP, Heller G, Ghavimi F, Casper ES, Vlamis V, Hajdu S,
et al. The effect of age at diagnosis on outcome in
rhabdomyosarcoma. Cancer 1994;73:109-17.
Rodary C, Gehan EA, Flamant F, Treuner J, Carli M, Auquier A, et
al. Prognostic factors in 951 non-metastatic rhabdomyosarcoma in
children: a report of the Intergroup Rhabdomyosarcoma Workshop.
Med Pediatr Oncol 1991;19:89-95.
Wiener ES, Lawrence W, Hays D, Lobe TE, Andrassy R, Donaldson
S, et al. Retroperitoneal node biopsy in paratesticular
rhabdomyosarcoma. J Pediatr Surg 1994;29:171-7, Discussion 178.
Mandell LR, Massey V, Ghavani F. The influence of extensive bone
erosion on local control in nonorbital rhabdomyosarcoma of the head
and neck. Int J Ratiat Oncol Biol Phys 1989;17:649-53.
Niggli FK, Powell JE, Parkes SE, Ward K, Raafat F, Mann JR, et al.
DNA ploidy and proliferative activity (S-phase) in childhood soft­
tissue sarcomas: their value as prognostic indicators. Br J Cancer
Wijnaendts LC, van der Linden JC, van Diest P, van Unnik AJ,
Delemarre JF, Voute PA, et al. Prognostic importance of DNA flow
cytometric variables in rhabdomyosarcoma. J Clin Pathol
Schalow EL, Broecker BH. Role of surgery in children with
rhabdomyosarcoma. Med Pediatr Oncol. 2003;41:1-6.
Carli M, Passone E, Perilongo G, Bisogno G. Ifosfamide in pediatric
solid tumors. Oncology. 2003;65:99-104.
Cosetti M, Wexler LH, Calleja E, Trippett T, LaQuaglia M, Huvos
AG, et al. Irinotecan for pediatric solid tumors: the Memorial SloanKettering experience. J Pediatr Hematol Oncol 2002;24:101-5.
Wiener ES. Rhabdomypsarcoma. In: O’Neill JA, Rowe MI, Grosfeld
JL, Fonkalsrud EW, Coran AG (eds) Pediatric Surgery 5th edn, Mosby:
St Louis; 1998: p. 431-45.
Hays DM. Bladder/prostate rhabdomyosarcoma: results of multi­
institutional trials of the Intergroup Rhabdomyosarcoma Study.
Semin Surg Oncol 1993;9:520-3.
Agarwala S: Pediatric soft tissue sarcomas
31. Fryer CJ. Pelvic rhabdomyosarcoma: paying the price of bladder
preservation. (editorial). Lancet 1995;345:141-2.
32. Maurer HM, Donaldson SS, Wiener ES. Rhabdomyosarcoma in
children. In: Holland JF, Bast RC (Jr), Morton DL et al (editors)
Cancer Medicine 4th edn, Wilams and Wilkins: Baltimore; 1997: p.
33. Raney RB Jr, Tefft M, Lawrence W Jr, Ragab AH, Soule EH,
Beltangady M, et al. Paratesticular sarcomas in childhood and
adolescence: a report from intregroup rhabdomyosarcoma studies I
and II, 1973-1983. Cancer 1987;60:2337-43.
34. Wexler LH, Helman LJ. Rhabdomyosarcoma and the
undifferentiated sarcomas. In: Pizzo PA, Poplack DG (editors)
Principles and practice of pediatric oncology, 3rd edition. LippincottRaven Publishers: Philadelphia; 1997: p. 799-829.
35. Hays DM, Soule EH, Lawrence W Jr, Gehan EA, Maurer HM,
Donaldson M, et al. Extremity lesions in the Intergroup
Rhabdomyosarcoma Study (IRS-I): a preliminary report. Cancer
36. Lawrence W, Hays DM, Heyn R, Beltangady M, Maurer HM. Surgical
lesions from the intergroup rhabdomyosarcoma study pertaining to
extremity tumors. World J Surg 1988;12:676-84.
37. Hays DM, Lawrence W Jr, Wharam M, Newton W Jr, Ruymann FB,
Beltangady M, et al. Primary re-excision for patients with
“microscopic residual” following initial excision of sarcomas of the
trunk and extremity sites. J Pediatr Surg 1989;24:5-10.
38. Mandell L, Ghavimi H, LaQuagila M, Exelby P. Prognostic
significance of regional lymph node involvement in childhood
extermity rhabdomyosarcoma. Med Pediatr Oncol 1990;18:466-71.
39. Rao BN. Nonrhabdomyosarcoma in children: prognostic factors
influencing survival. Semin Surg Oncol 1993;9:524-31.
40. Dillon PW, Maurer H, Jenkins J. A prospective study of
nonrhabdomyosarcoms soft tissue sarcomas in pediatric age group.
J Pediatr Surg 1992;27:241-5.
41. Pappo AS, Shapiro DN, Cris WM. Rhabdomyosarcoma> Biology
and treatment [Review]. Pediatr Clin North Am 1997;44:953-72.
42. Andrassy RJ, Corpron CA, Hay D, Raney RB, Wiener ES, Lawrence
W Jr, et al. Extremity sarcomas. An analysis of prognostic factors
from the intergroup rhabdomyosarcoma study III. J Pediatr Surg
43. Lawrence W Jr anderson JR, Gehan EA, Maurer H. Pretreatment
TNM staging of childhood rhabdomyosarcoma: a report of Intergroup
Rhabdomyosarcoma Study Group. Cancer 1997;80:1165-70.
44. Crist WM, Garnsey L, Beltangady MS, Gehan E, Ruymann F,
Webber B, et al. Prognosis in children with rhabdomyosarcoma: a
report of Intergroup Rhabdomyosarcoma Studies I and II. J Clin
Oncol 19920;8:443-52.
45. Pinkel D, Piekren J. Rhabdomyosarcoma in children. JAMA
J Indian Assoc Pediatr Surg / Jan-Mar 2006 / Vol 11 / Issue 1