Osteosarcoma

“Osteosarcoma and Undifferentiated Pleomorphic Sarcoma of Bone Treatment (PDQ®)–Health

Professional Version was originally published by the National Cancer Institute.” (Reviewed: 8 th Oct
2021)

General Information About Osteosarcoma

Disease Overview

Osteosarcoma occurs predominantly in adolescents and young adults. Review of data from the
National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) Program resulted in
an estimated osteosarcoma incidence rate of 4.4 cases per 1 million each year in people aged 0 to 24
years.[1] The U.S. Census Bureau estimated that there were 110 million people in this age range in
the year 2010, resulting in an incidence of roughly 450 cases per year in children and young adults
younger than 25 years.

Osteosarcoma accounts for approximately 5% of childhood tumors. In children and adolescents,

more than 50% of these tumors arise from the long bones around the knee. Osteosarcoma is rarely
observed in soft tissue or visceral organs. There appears to be no difference in presenting symptoms,
tumor location, and outcome for younger patients (<12 years) compared with adolescents.[2,3]

Two trials conducted in the 1980s were designed to determine whether chemotherapy altered the
natural history of osteosarcoma after surgical removal of the primary tumor. The outcome of
patients in these trials who were treated with surgical removal of the primary tumor recapitulated
the historical experience before 1970; more than one-half of these patients developed metastases
within 6 months of diagnosis, and overall, approximately 90% developed recurrent disease within 2
years of diagnosis.[4] Overall survival (OS) for patients treated with surgery alone was statistically
inferior.[5] The natural history of osteosarcoma has not changed over time, and fewer than 20% of
patients with localized resectable primary tumors treated with surgery alone can be expected to
survive free of relapse.[4,6]; [7][Level of evidence: 1iiA]

In 2013, the World Health Organization (WHO) published an update to the Classification of Tumors
of Soft Tissue and Bone.[8] They removed the term malignant fibrous histiocytoma (MFH) and
replaced it with undifferentiated pleomorphic sarcoma (UPS). This type of sarcoma is much more
common in soft tissues; however, it does arise in bone. In bone, it has features that are histologically
similar to osteosarcoma but it does not produce osteoid. Most of the literature describing the
clinical behavior and response to therapy for this histology in bone was published before the 2013
WHO update, and a search for UPS of bone will not retrieve these articles. The citations in this
summary appear with their titles as published; therefore, many references will describe MFH of
bone, a condition now called UPS of bone.

Diagnostic Evaluation

Osteosarcoma can be diagnosed by core needle biopsy or open surgical biopsy. It is preferable that
the biopsy be performed by a surgeon skilled in the techniques of limb sparing (removal of the
malignant bone tumor without amputation and replacement of bones or joints with allografts or
prosthetic devices). In these cases, the original biopsy incision placement is crucial. Inappropriate
alignment of the biopsy or inadvertent contamination of soft tissues can render subsequent limb-
preserving reconstructive surgery impossible.

Prognostic Factors

Dramatic improvements in survival have been achieved for children and adolescents with cancer.
Between 1975 and 2010, childhood cancer mortality decreased by more than 50%.[9] For
osteosarcoma, the 5-year survival rate increased over the same time from 40% to 76% in children
younger than 15 years and from 56% to approximately 66% in adolescents aged 15 to 19 years, but
has seen no substantial improvement since the 1980s.[10]
In general, prognostic factors for osteosarcoma have not been helpful in identifying patients who
might benefit from treatment intensification or who might require less therapy while maintaining an
excellent outcome.

Factors that may influence outcome include the following:[11]

Primary tumor site.

Size of the primary tumor.
Presence of clinically detectable metastatic disease.
Surgical resectability of primary tumor.
Degree of tumor necrosis after administration of neoadjuvant chemotherapy.
Age and sex.
Other possible prognostic factors.

1. Primary tumor site

The site of the primary tumor is a significant prognostic factor for patients with localized disease.
Among extremity tumors, distal sites have a more favorable prognosis than do proximal sites. Axial
skeleton primary tumors are associated with the greatest risk of progression and death, primarily
related to the inability to achieve a complete surgical resection.

Prognostic considerations for the axial skeleton and extraskeletal sites are as follows:

Pelvis: Pelvic osteosarcomas make up 7% to 9% of all osteosarcomas; survival rates for patients with
pelvic primary tumors are 20% to 47%.[12-14] Complete surgical resection is associated with a
positive outcome for osteosarcoma of the pelvis in some cohorts of patients.[12,15]

Craniofacial/head and neck: In patients with craniofacial osteosarcoma, those with primary sites in
the mandible and maxilla have a better prognosis than do patients with other primary sites in the
head and neck.[16-18] For patients with osteosarcoma of craniofacial bones, complete resection of
the primary tumor with negative margins is essential for cure.[19-21] When treated with surgery
alone, patients who have osteosarcoma of the head and neck have a better prognosis than do those
who have appendicular lesions.

Despite a relatively high rate of inferior necrosis after neoadjuvant chemotherapy, fewer patients
with craniofacial primaries develop systemic metastases than do patients with osteosarcoma
originating in the extremities.[22-24]

A meta-analysis concluded that systemic adjuvant chemotherapy improves the prognosis for
patients with osteosarcoma of the head and neck, while small series have not shown a benefit for
using adjuvant chemotherapy in these patients.[22-24] Another large meta-analysis detected no
benefit of chemotherapy for patients with osteosarcoma of the head and neck, but suggested that
incorporating chemotherapy into the treatment plan for patients with high-grade tumors may
improve survival.[21] A retrospective analysis identified a trend toward better survival in patients
with high-grade osteosarcoma of the mandible and maxilla who received adjuvant
chemotherapy.[21,25]

Radiation therapy was found to improve local control, disease-specific survival, and OS in a
retrospective study of patients with osteosarcoma of the craniofacial bones who had positive or
uncertain margins after surgical resection.[26][Level of evidence: 3iiA] Radiation-associated
craniofacial osteosarcomas are generally high-grade lesions, usually fibroblastic, and tend to recur
locally with a high rate of metastasis.[27]

Extraskeletal: Osteosarcoma in extraskeletal sites is rare in children and young adults. With current
combined-modality therapy, the outcome of patients with extraskeletal osteosarcoma appears to be
similar to that of patients with primary tumors of bone.[28]

2. Size of the primary tumor

In some series, patients with larger tumors appeared to have a worse prognosis than did patients
with smaller tumors.[11,29,30] Tumor size has been assessed by longest single dimension, cross-
sectional area, or estimate of tumor volume; all assessments have correlated with outcome.

Serum lactate dehydrogenase (LDH), which also correlates with outcome, is a likely surrogate for
tumor volume.[13]

3. Presence of clinically detectable metastatic disease

Patients with localized disease have a much better prognosis than do patients with overt metastatic
disease. As many as 20% of patients will have radiographically detectable metastases at diagnosis,
with the lung being the most common site.[31] The prognosis for patients with metastatic disease
appears to be determined largely by site(s) of metastases, number of metastases, and surgical
resectability of the metastatic disease.[32,33]

Site of metastases: Prognosis appears more favorable for patients with fewer pulmonary nodules
and for those with unilateral rather than bilateral pulmonary metastases;[32] not all patients with
suspected pulmonary metastases at diagnosis have osteosarcoma confirmed at the time of lung
resection. In one large series, approximately 25% of patients had exclusively benign lesions removed
at the time of surgery.[33]

Number of metastases: Patients with skip metastases (at least two discontinuous lesions in the
same bone) have been reported to have inferior prognoses.[34] However, an analysis of the German
Cooperative Osteosarcoma Study experience suggests that skip lesions in the same bone do not
confer an inferior prognosis if they are included in planned surgical resection. Skip metastasis in a
bone other than the primary bone should be considered systemic metastasis.[35]

Historically, metastasis across a joint was referred to as a skip lesion. Skip lesions across a joint might
be considered hematogenous spread and have a worse prognosis.[35]

Patients with multifocal osteosarcoma (defined as multiple bone lesions without a clear primary
tumor) have an extremely poor prognosis.[36,37]
Surgical resectability of metastases: Patients who have complete surgical ablation of the primary
and metastatic tumor (when confined to the lung) after chemotherapy may attain long-term
survival, although overall event-free survival (EFS) rates remain about 20% to 30% for patients with
metastatic disease at diagnosis.[32,33,38,39]

4. Surgical resectability of primary tumor

Resectability of the tumor is a critical prognostic feature because osteosarcoma is relatively resistant
to radiation therapy. Complete resection of the primary tumor and any skip lesions with adequate
margins is generally considered essential for cure. A retrospective review of patients with
craniofacial osteosarcoma performed by the cooperative German-Austrian-Swiss osteosarcoma
study group reported that incomplete surgical resection was associated with inferior survival
probability.[16][Level of evidence: 3iiB] In a European cooperative study, the size of the margin was
not significant. However, prognosis was better when both the biopsy and resection were performed
at a center with orthopedic oncology experience.[13]

For patients with axial skeletal primaries who either do not undergo surgery for their primary tumor
or who undergo surgery that results in positive margins, radiation therapy may improve
survival.[15,40]

5. Degree of tumor necrosis after administration of neoadjuvant chemotherapy

Most treatment protocols for osteosarcoma use an initial period of systemic chemotherapy before
definitive resection of the primary tumor (or resection of sites of metastases). The pathologist
assesses necrosis in the resected tumor. Patients with at least 90% necrosis in the primary tumor
after induction chemotherapy have a better prognosis than do patients with less necrosis.[29]
Patients with less necrosis (<90%) in the primary tumor after initial chemotherapy have a higher rate
of recurrence within the first 2 years than do patients with a more favorable amount of necrosis
(≥90%).[41]

Less necrosis should not be interpreted to mean that chemotherapy has been ineffective; cure rates
for patients with little or no necrosis after induction chemotherapy are much higher than cure rates
for patients who receive no chemotherapy. The EFS rate for patients who receive no adjuvant
chemotherapy is approximately 11%.[42] Many large published series of patients treated with
chemotherapy have reported EFS rates of 40% to 50% for patients with little or no necrosis in the
primary tumor after initial systemic chemotherapy.[43-45] A review of two consecutive prospective
trials performed by the Children’s Oncology Group showed that histologic necrosis in the primary
tumor after initial chemotherapy was affected by the duration and intensity of the initial period of
chemotherapy. More necrosis was associated with better outcome in both trials, but the magnitude
of the difference between patients with more and less necrosis was diminished with a longer and
more intensive period of initial chemotherapy.[46][Level of evidence: 1iiD]
6. Age and sex

Patients in the older adolescent and young adult age group, typically defined as age 18 to 40 years,
tend to have a worse prognosis. In addition, male sex has been associated with a worse
prognosis.[30,47,48] Compared with the other prognostic factors listed, both age and sex have a
relatively minor impact on outcome.

7. Other possible prognostic factors

Other factors that may be prognostic but with either limited or conflicting data include the following:

Subsequent neoplasms. Patients with osteosarcoma as a subsequent neoplasm, including tumors

arising in a radiation field, share the same prognosis as patients with de novo osteosarcoma if they
are treated aggressively with complete surgical resection and multiagent chemotherapy.[49-52]

In a German series, approximately 25% of patients with craniofacial osteosarcoma had

osteosarcoma as a second tumor, and in 8 of these 13 patients, osteosarcoma arose after treatment
for retinoblastoma. In this series, there was no difference in outcome for primary or secondary
craniofacial osteosarcoma.[16]

Laboratory abnormalities. Possible prognostic factors identified for patients with conventional
localized high-grade osteosarcoma include LDH level, alkaline phosphatase level, and histologic
subtype.[29,44,47,53-56]

Body mass index. Higher body mass index at initial presentation is associated with worse OS.[57]

Pathologic fracture. Some studies have suggested that a pathologic fracture at diagnosis or during
preoperative chemotherapy does not have adverse prognostic significance.[6]; [58,59][Level of
evidence: 3iiiA]; [60][Level of evidence: 3iiD]

However, a systematic review of nine cohort studies examined the impact of pathologic fractures on
outcome in patients with osteosarcoma. The review included 2,187 patients, and 311 of these
patients had a pathologic fracture. The presence of a pathologic fracture correlated with decreased
EFS and OS.[61] In two additional series, a pathologic fracture at diagnosis was associated with a
worse overall outcome.[62]; [63][Level of evidence: 3iiA] A retrospective analysis of 2,847 patients
with osteosarcoma from the German cooperative group identified 321 patients (11.3%) with a
pathological fracture before the initiation of systemic therapy.[64][Level of evidence: 3iA] In
pediatric patients, OS and EFS did not differ significantly between patients with and without a
pathologic fracture. In adults, the 5-year OS rate in patients with a pathologic fracture was 46%
versus 69% for patients without a pathologic fracture (P < .001). The 5-year EFS rate in adults was
36% for patients with a pathologic fracture versus 56% for patients without a pathologic fracture (P <
.001). In a multivariable analysis, the presence of a pathologic fracture was not a statistically
significant factor for OS or EFS in the total cohort or in pediatric patients. In adult patients, presence
of a pathologic fracture remained an independent prognostic factor for OS (hazard ratio, 1.893; P =
.013).

Time to definitive surgery. In a large series, a delay of 21 days or longer from the time of definitive
surgery to the resumption of chemotherapy was an adverse prognostic factor.[65]

Genetic factors.

ERBB2 expression. There are conflicting data concerning the prognostic significance of this human
epidermal growth factor.[66-68]

Tumor cell ploidy.[69]

Specific chromosomal gains or losses.[70]

Loss of heterozygosity of the RB1 gene.[71,72]

Loss of heterozygosity of the p53 locus.[73]

Increased expression of p-glycoprotein.[74,75] A prospective analysis of p-glycoprotein expression

determined by immunohistochemistry failed to identify prognostic significance for patients with
newly diagnosed osteosarcoma, although earlier studies suggested that overexpression of p-
glycoprotein predicted poor outcome.[76]

Genomics of Osteosarcoma
Molecular Features of Osteosarcoma
Genetic predisposition to osteosarcoma
Molecular Features of Osteosarcoma

The genomic landscape of osteosarcoma is distinctive from that of other childhood cancers. It is
characterized by an exceptionally high number of structural variants with relatively small numbers of
single nucleotide variants compared with many adult cancers.[1,2]

Key observations regarding the genomic landscape of osteosarcoma are summarized below:

The number of structural variants observed for osteosarcoma is very high, at more than 200
structural variants per genome;[1,2] thus, osteosarcoma has the most chaotic genome among
childhood cancers. The Circos plots shown in Figure 1 illustrate the exceptionally high numbers of
intra- and inter-chromosomal translocations that typify osteosarcoma genomes.

The number of mutations per osteosarcoma genome that affect protein sequence (approximately 25
per genome) is higher than that of some other childhood cancers (e.g., Ewing sarcoma and rhabdoid
tumors) but is far below that for adult cancers such as melanoma and non-small cell lung
cancer.[1,2]
Genomic alterations in TP53 are present in most osteosarcoma cases. A distinctive form of TP53
inactivation occurs through structural variations in the first intron of TP53 that lead to disruption of
the TP53 gene.[1] Other mechanisms of TP53 inactivation are also observed, including missense and
nonsense mutations and deletions of the TP53 gene.[1,2] The combination of these various
mechanisms for loss of TP53 function leads to biallelic inactivation in most cases of osteosarcoma.

MDM2 amplification is observed in a minority of osteosarcoma cases (approximately 5%) and

provides another mechanism for loss of TP53 function.[1,2]

RB1 is commonly inactivated in osteosarcoma, sometimes by mutation but more commonly by

deletion.[1,2]

Other genes with recurrent alterations in osteosarcoma include ATRX and DLG2.[1] Additionally,
pathway analysis showed that the PI3K/mammalian target of rapamycin (mTOR) pathway was
altered by mutation/loss/amplification in approximately one-fourth of patients, with PTEN
mutation/loss being the most common alteration.[2]

A retrospective review of 71 osteosarcoma tumors from 66 patients identified theoretically

targetable alterations in 14 patients (21%), including amplification of CDK4 (n = 9) and/or MDM2 (n =
9) and somatic truncating mutations/deletions in BRCA2 (n = 3) and PTCH1 (n = 1). The authors
reported mutually exclusive patterns of alterations, suggesting distinct biological subsets defined by
gains at 4q12 and 6p12-21. Specifically, potentially targetable gene amplifications at 4q12 involving
KIT, KDR, and PDGFRA were identified in 13 of 66 patients (20%), which showed strong PDGFRA
expression by immunohistochemistry. In another largely nonoverlapping subset of 14 patients (24%)
with gains at 6p12-21, VEGFRA amplification was identified.[3]

The range of mutations reported for osteosarcoma tumors at diagnosis do not provide obvious
therapeutic targets, as they primarily reflect loss of tumor suppressor genes (e.g., TP53, RB1, PTEN)
rather than activation of targetable oncogenes.

Genetic predisposition to osteosarcoma

Several germline mutations are associated with susceptibility to osteosarcoma; Table 1 summarizes
the syndromes and associated genes for these conditions. The frequency of pathogenic or likely
pathogenic cancer-susceptibility gene variants in individuals with osteosarcoma is 28% greater than
that in control groups.[4]

TP53 mutations

Mutations in TP53 are the most common germline alterations associated with osteosarcoma.
Mutations in this gene are found in approximately 70% of patients with Li-Fraumeni syndrome (LFS),
which is associated with increased risk of osteosarcoma, breast cancer, various brain cancers, soft
tissue sarcomas, and other cancers. While rhabdomyosarcoma is the most common sarcoma arising
in patients aged 5 years and younger with TP53-associated LFS, osteosarcoma is the most common
sarcoma in children and adolescents aged 6 to 19 years.[5] One study observed a high frequency of
young osteosarcoma cases (age <30 years) carrying a known LFS-associated or likely LFS-associated
TP53 mutation (3.8%) or rare exonic TP53 variant (5.7%), with an overall TP53 mutation frequency of
9.5%.[6] Another study observed germline TP53 mutations in 7 of 59 osteosarcoma cases (12%)
subjected to whole-exome sequencing.[2] Other groups have reported lower rates (3%–7%) of TP53
germline mutations in patients with osteosarcoma.[7,8]

RECQL4 mutations

Investigators analyzed whole-exome sequencing from the germline of 4,435 pediatric cancer
patients at the St. Jude Children’s Research Hospital and 1,127 patients from the National Cancer
Institute's Therapeutically Applicable Research to Generate Effective Treatment (TARGET) database.
They identified 24 patients (0.43%) who harbored loss-of-function RECQL4 variants, including 5 of
249 patients (2.0%) with osteosarcoma.[9] These RECQL4 variants were significantly
overrepresented in children with osteosarcoma, the cancer most frequently observed in patients
with Rothmund-Thomson syndrome, as compared with 134,187 noncancer controls in the Genome
Aggregation Database (gnomAD v2.1; P = .00087; odds ratio, 7.1; 95% confidence interval, 2.9–17).
Nine of the 24 individuals (38%) possessed the same c.1573delT (p.Cys525Alafs) variant located in
the highly conserved DNA helicase domain, suggesting that disruption of this domain is central to
oncogenesis.

Table 1. Genetic Diseases That Predispose to Osteosarcomaa

Syndrome Description Location Gene Function

AML = acute myeloid leukemia; IL-1 = interleukin-1; MDS = myelodysplastic

syndrome; RANKL = receptor activator of nuclear factor kappa beta ligand; TNF =
tumor necrosis factor.

aAdapted from Kansara et al.[10]

Bloom Rare inherited 15q26.1 BLM DNA

syndrome [11] disorder helicase
characterized by
short stature and
sun-sensitive
skin changes.
Often presents
with a long,
narrow face,
small lower jaw,
large nose, and
prominent ears.

Diamond- Inherited pure Ribosomal Ribosome

Blackfan anemia red cell aplasia. proteins production
[12] Patients at risk [12,13]
for MDS and
 Syndrome Description Location Gene Function

AML. Associated
with skeletal
abnormalities
such as abnormal
facial features
(flat nasal bridge,
widely spaced
eyes).

Li-Fraumeni Inherited 17p13.1 P53 DNA

syndrome [14] mutation damage
in TP53 gene. response
Affected family
members at
increased risk of
bone tumors,
breast cancer,
leukemia, brain
tumors, and
sarcomas.

Paget disease Excessive 18q21- LOH18CR1 IL-1/TNF

[15] breakdown of qa22 signaling;
bone with RANKL
abnormal bone signaling
formation and 5q31 pathway
remodeling,
resulting in pain
from weak,
malformed bone. 5q35-qter

Retinoblastoma Malignant tumor 13q14.2 RB1 Cell-cycle

[16] of the retina. checkpoint
Approximately
66% of patients
are diagnosed by
age 2 years and
95% of patients
by age 3 years.
Patients with
heritable germ
cell mutations at
greater risk of
subsequent
neoplasms.
 Syndrome Description Location Gene Function

Rothmund- Autosomal 8q24.3 RECQL4 DNA

Thomson recessive helicase
syndrome (also condition.
called Associated with
poikiloderma skin findings
congenitale) (atrophy,
[17,18] telangiectasias,
pigmentation),
sparse hair,
cataracts, small
stature, and
skeletal
abnormalities.
Increased
incidence of
osteosarcoma at
a younger age.

Werner Patients often 8p12-

syndrome [19] have short p11.
stature and in
their early
twenties, develop
signs of aging,
including graying
of hair and
hardening of skin.
Other aging
problems such as
cataracts, skin
ulcers, and
atherosclerosis
develop later.
Cellular Classification of Osteosarcoma
Central (Medullary) Tumors
Surface (Peripheral) Tumors
Extraosseous Osteosarcoma
Undifferentiated Pleomorphic Sarcoma (UPS) of Bone

Osteosarcoma is a malignant tumor that is characterized by the direct formation of bone or osteoid
tissue by the tumor cells. The World Health Organization’s histologic classification [1] of bone
tumors separates the osteosarcomas into central (medullary) and surface (peripheral) tumors [2,3]
and recognizes a number of subtypes within each group.

Central (Medullary) Tumors

Conventional central osteosarcomas. The most common pathologic subtype is conventional central
osteosarcoma, which is characterized by areas of necrosis, atypical mitoses, and malignant osteoid
tissue and/or cartilage. The other subtypes are much less common, each occurring at a frequency of
less than 5%.

Telangiectatic osteosarcomas.[4,5] Telangiectatic osteosarcoma may be confused radiographically

with an aneurysmal bone cyst or giant cell tumor. This variant should be managed the same as a
conventional osteosarcoma.[4,5]

Intraosseous well-differentiated (low-grade) osteosarcomas.

Small-cell osteosarcomas.

Surface (Peripheral) Tumors

The terms parosteal and periosteal osteosarcoma are embedded in the literature and widely used.
They are confusing to patients and practitioners. It would be more helpful to divide osteosarcoma by
location and histological grade. High-grade osteosarcoma, sometimes referred to as conventional
osteosarcoma, typically arises centrally and grows outward, destroying surrounding cortex and soft
tissues, but there are unequivocal cases of high-grade osteosarcoma in surface locations.[6]
Similarly, there are reports of low-grade osteosarcoma arising in the medullary cavity.

Parosteal (juxtacortical) well-differentiated (low-grade) osteosarcomas.[7,8] Parosteal osteosarcoma

is defined as a lesion arising from the surface of the bone with a well-differentiated appearance on
imaging and low-grade histological features.[9] The most common site for parosteal osteosarcoma is
the posterior distal femur. Parosteal osteosarcoma occurs more often in older patients than does
conventional high-grade osteosarcoma and is most common in patients aged 20 to 30 years.
Parosteal osteosarcoma can be treated successfully with wide excision of the primary tumor
alone.[7,10]

Periosteal osteosarcomas (low-grade to intermediate-grade osteosarcomas).[11-13] Periosteal

osteosarcoma typically appears as a broad-based soft tissue mass with extrinsic erosion of the
underlying bony cortex.[12] Pathology shows an intermediate grade of differentiation. Wide
resection is essential.
A single-institution retrospective review identified 29 patients with periosteal osteosarcoma.[11]
The 5-year disease-free survival rate was 83%. The authors could not make a definitive statement
regarding the benefits of adjuvant chemotherapy.

Another single-institution retrospective review identified 33 patients with periosteal

osteosarcoma.[13] The 10-year overall survival (OS) rate was 84%. The 10-year OS rate was 83% for
patients who were treated with surgery alone and 86% for patients who were treated with surgery
and chemotherapy.

The European Musculoskeletal Oncology Society retrospectively analyzed 119 patients with
periosteal osteosarcoma; 17 patients had metastasis.[12] The OS rate was 89% at 5 years and 83% at
10 years. Eighty-one patients received chemotherapy; 50 of those patients received chemotherapy
before definitive surgical resection. There was no difference in outcome between the patients who
received chemotherapy and the patients who did not receive chemotherapy.

High-grade surface osteosarcomas.[3,6,14]

Extraosseous Osteosarcoma

Extraosseous osteosarcoma is a malignant mesenchymal neoplasm without direct attachment to the

skeletal system. Previously, treatment for extraosseous osteosarcoma followed soft tissue sarcoma
guidelines,[15] although a retrospective analysis of the cooperative German-Austrian-Swiss
osteosarcoma study group identified a favorable outcome for extraosseous osteosarcoma treated
with surgery and conventional osteosarcoma therapy.[16]

Undifferentiated Pleomorphic Sarcoma (UPS) of Bone

UPS of bone should be distinguished from angiomatoid fibrous histiocytoma, a low-grade tumor that
is usually noninvasive, small, and associated with an excellent outcome using surgery alone.[17] One
study suggests similar event-free survival rates for UPS and osteosarcoma.[18]

Staging and Site Information for Osteosarcoma

Localized Osteosarcoma
Metastatic Osteosarcoma
Staging Evaluation

Historically, the Enneking staging system for skeletal malignancies was used.[1] This system inferred
the aggressiveness of the primary tumor by the descriptors intracompartmental or
extracompartmental. The American Joint Committee on Cancer's tumor-node-metastasis (TNM)
staging system for malignant bone tumors is not widely used for pediatric osteosarcoma, and
patients are not stratified on the basis of prognostic stage groups.
For the purposes of treatment, osteosarcoma is described as one of the following:

Localized. Patients without clinically detectable metastatic disease are considered to have localized
osteosarcoma.

Metastatic. Patients with any site of metastasis at the time of initial presentation detected by
routine clinical studies are considered to have metastatic osteosarcoma.

Localized Osteosarcoma

Localized tumors are limited to the bone of origin. Patients with skip lesions confined to the bone
that includes the primary tumor are considered to have localized disease if the skip lesions can be
included in the planned surgical resection.[2] Approximately one-half of the tumors arise in the
femur; of these, 80% are in the distal femur. Other primary sites, in descending order of frequency,
are the proximal tibia, proximal humerus, pelvis, jaw, fibula, and ribs.[3] Osteosarcoma of the head
and neck is more likely to be low grade [4] and to arise in older patients than is osteosarcoma of the
appendicular skeleton.

Metastatic Osteosarcoma

Radiologic evidence of metastatic tumor deposits is found in approximately 20% of patients at

diagnosis, with 85% to 90% of metastatic disease presenting in the lungs. The second most common
site of metastasis is another bone, which may be solitary or multiple.[5]

The syndrome of multifocal osteosarcoma refers to a presentation with multiple foci of

osteosarcoma without a clear primary tumor, often with symmetrical metaphyseal involvement.[3]

Staging Evaluation

For patients with confirmed osteosarcoma, in addition to plain radiographs of the primary site that
include a single-plane view of the entire affected bone to assess for skip metastasis, pretreatment
staging studies should include the following:[6]

Magnetic resonance imaging (MRI) of the primary site to include the entire bone.
Computed tomography (CT) scan, if MRI is not available.
Bone scan.
Posteroanterior and lateral chest radiograph.
CT scan of the chest.
Positron emission tomography (PET) using fluorine F 18-fludeoxyglucose is an optional staging
modality.

A retrospective review of 206 patients with osteosarcoma compared bone scan, PET scan, and PET-
CT scan for the detection of bone metastases.[7] PET-CT was more sensitive and accurate than bone
scan, and the combined use of both imaging studies achieved the highest sensitivity for diagnosing
bone metastases in osteosarcoma.
Treatment Option Overview for Osteosarcoma
It is imperative that patients with proven or suspected osteosarcoma have an initial evaluation by an
orthopedic oncologist familiar with the surgical management of this disease. This evaluation, which
includes imaging studies, should be done before the initial biopsy, because an inappropriately
performed biopsy may jeopardize a limb-sparing procedure. Additionally, protective weight bearing
is recommended for patients with tumors of weight-bearing bones to prevent pathological fractures
that could preclude limb-preserving surgery.

Successful treatment generally requires the combination of effective systemic chemotherapy and
complete resection of all clinically detectable disease.

Randomized clinical trials have established that both neoadjuvant and adjuvant chemotherapy are
effective in preventing relapse in patients with clinically nonmetastatic tumors.[1]; [2][Level of
evidence: 1iiA] The Pediatric Oncology Group (POG) conducted a study in which patients were
randomly assigned to either immediate amputation or amputation after neoadjuvant therapy. A
large percentage of patients declined to be assigned randomly, and the study was terminated
without approaching the stated accrual goals. In the small number of patients treated, there was no
difference in outcome between those who received preoperative chemotherapy and those who
received postoperative chemotherapy.[3]

The treatment of osteosarcoma also depends on the histologic grade, as follows:

Low-grade osteosarcoma. Patients with low-grade osteosarcoma can be treated successfully by

wide surgical resection alone, regardless of site of origin.

Intermediate-grade osteosarcoma. Pathologists sometimes characterize tumors as intermediate-

grade osteosarcoma. It is difficult to make treatment decisions for patients with intermediate-grade
tumors. When a tumor biopsy suggests an intermediate-grade osteosarcoma, an option is to
proceed with wide resection. The availability of the entire tumor allows the pathologist to examine
more tissue and evaluate soft tissue and lymphovascular invasion, which can often clarify the nature
of the lesion.

If the lesion proves to have high-grade elements, systemic chemotherapy is indicated, just as it
would be for any high-grade osteosarcoma. The POG performed a study in which patients with high-
grade osteosarcoma were randomly assigned to either immediate definitive surgery followed by
adjuvant chemotherapy or to an initial period of chemotherapy followed by definitive surgery.[3]
The outcome was the same for both groups. Although the strategy of initial chemotherapy followed
by definitive surgery has become an almost universally applied approach for osteosarcoma, this
study suggests that there is no increased risk of treatment failure if definitive surgery is done before
chemotherapy begins; this can help to clarify equivocal diagnoses of intermediate-grade
osteosarcoma.

High-grade osteosarcoma. Patients with high-grade osteosarcoma require surgery and systemic
chemotherapy whether the tumor arises in the conventional central location or on a bone surface.

Recognition of intraosseous well-differentiated osteosarcoma and parosteal osteosarcoma is

important because patients with these tumor types have the most favorable prognosis and can be
treated successfully with wide excision of the primary tumor alone.[4,5] Patients with periosteal
osteosarcoma have a generally good prognosis [6] and treatment is guided by histologic grade.[5,7]
Patients with undifferentiated pleomorphic sarcoma (UPS) of bone are treated according to
osteosarcoma treatment protocols.[8]

Imaging modalities such as dynamic magnetic resonance imaging or positron emission tomography
scanning are under investigation as noninvasive methods to assess response.[9-17]

Table 2. Treatment Options for Osteosarcoma and Undifferentiated Pleomorphic Sarcoma

(UPS) of Bone

Treatment Group Treatment Options

Localized osteosarcoma and Surgical removal of primary tumor.

UPS of bone

Chemotherapy.

Radiation therapy, if surgery is not feasible or surgical

margins are inadequate.

Osteosarcoma and UPS of bone Chemotherapy.

with metastatic disease at
diagnosis:

Lung-only metastases Preoperative chemotherapy followed by surgery to

remove the tumor followed by postoperative
combination chemotherapy.

Bone metastases with Preoperative chemotherapy followed by surgery to

or without lung remove the primary tumor and all metastatic disease
metastases followed by postoperative combination chemotherapy.

Surgery to remove the primary tumor followed by

chemotherapy and then surgical resection of
metastatic disease followed by postoperative
combination chemotherapy.

Resection of metastatic bone lesions if possible.

 Treatment Group Treatment Options

Radiation therapy to the extremities (may offer some

local control).

Recurrent osteosarcoma and Surgery to remove all sites of metastatic disease.

UPS of bone:

Chemotherapy and targeted therapy.

Radiopharmaceuticals and radiation therapy.

Local recurrence Surgery to remove the tumor.

Lung-only recurrence Surgery to remove the tumor.

Recurrence with bone- Surgery to remove the tumor.

only metastases

153Sm-EDTMP with or without stem cell support.

Second recurrence of Surgery to remove the tumor and/or chemotherapy.

osteosarcoma

153Sm-EDTMP = samarium Sm 153-ethylenediamine tetramethylene phosphonic acid.

Special Considerations for the Treatment of Children With Cancer
Cancer in children and adolescents is rare, although the overall incidence has been slowly increasing
since 1975.[1] Children and adolescents with cancer should be referred to medical centers that have
a multidisciplinary team of cancer specialists with experience treating the cancers that occur during
childhood and adolescence. This multidisciplinary team approach incorporates the skills of the
following health care professionals and others to ensure that children receive treatment, supportive
care, and rehabilitation that will achieve optimal survival and quality of life:

Primary care physicians.

Orthopedic surgeon experienced in bone tumors.
Pathologists.
Radiation oncologists.
Pediatric oncologists.
Rehabilitation specialists.
Pediatric nurse specialists.
Social workers.
Child-life professionals.
Psychologists.

The American Academy of Pediatrics has outlined guidelines for pediatric cancer centers and their
role in the treatment of pediatric patients with cancer.[2] At these pediatric cancer centers, clinical
trials are available for most types of cancer that occur in children and adolescents, and the
opportunity to participate is offered to most patients and their families. Clinical trials for children
and adolescents with cancer are generally designed to compare potentially better therapy with
current standard therapy. Most of the progress made in identifying curative therapies for childhood
cancers has been achieved through clinical trials. Information about ongoing clinical trials is available
from the National Cancer Institute's website.

Childhood and adolescent cancer survivors require close monitoring because side effects of cancer
therapy may persist or develop months or years after treatment. (Refer to the PDQ summary on Late
Effects of Treatment for Childhood Cancer for specific information about the incidence, type, and
monitoring of late effects in childhood and adolescent cancer survivors.)
Treatment of Localized Osteosarcoma
Surgical removal of primary tumor
Chemotherapy
Radiation therapy
Osteosarcoma of the Head and Neck

Patients with localized osteosarcoma who undergo surgery and chemotherapy have a 5-year overall
survival (OS) rate of 62% to 65%.[1] Complete surgical resection is crucial for patients with localized
osteosarcoma; however, at least 80% of patients treated with surgery alone will develop metastatic
disease.[2] Randomized clinical trials have established that adjuvant chemotherapy is effective in
preventing relapse or recurrence in patients with localized resectable primary tumors.[2]; [3][Level
of evidence: 1iiA]

Undifferentiated pleomorphic sarcoma (UPS) of bone is seen more commonly in older adults.
Patients with UPS of bone are treated according to osteosarcoma treatment protocols, and the
outcome for patients with resectable UPS is similar to the outcome for patients with
osteosarcoma.[4] As with osteosarcoma, patients with favorable necrosis (≥90% necrosis) have a
longer survival than do those with an inferior necrosis (<90% necrosis).[5] Many patients with UPS
will need preoperative chemotherapy to achieve a wide local excision.[6]

Treatment options for patients with localized osteosarcoma include the following:

Surgical removal of primary tumor.

Chemotherapy (may start before or after definitive surgical resection of the primary tumor).
Radiation therapy, if surgery is not feasible or surgical margins are inadequate.

1. Surgical removal of primary tumor

Surgical resection of the primary tumor with adequate margins is an essential component of the
curative strategy for patients with localized osteosarcoma. The type of surgery required for
complete ablation of the primary tumor depends on a number of factors that must be evaluated on
a case-by-case basis.[7]

Limb-sparing procedures

In general, more than 80% of patients with extremity osteosarcoma can be treated using a limb-
sparing procedure and do not require amputation.[8] Limb-sparing procedures are planned only
when the preoperative staging indicates that it would be possible to achieve wide surgical margins.
In one study, patients who underwent limb-salvage procedures who had poor histologic response
and close surgical margins had a high rate of local recurrence.[9]

Reconstruction after limb-sparing surgery can be accomplished with many options, including metallic
endoprosthesis, allograft, vascularized autologous bone graft, and rotationplasty. An additional
option, osteogenesis distraction bone transport, is available for patients whose tumors do not
involve the epiphysis of long bones.[10] This procedure results in a stable reconstruction that
functionally restores the normal limb.
The choice of optimal surgical reconstruction involves many factors, including the
following:[11][Level of evidence: 1iiA]

Site and size of the primary tumor.

Ability to preserve the neurovascular supply of the distal extremity.
Age of the patient and potential for additional growth.
Needs and desires of the patient and family for specific functions such as sports participation.

If a complicated reconstruction delays or prohibits the resumption of systemic chemotherapy, limb

preservation may endanger the chance for cure. In retrospective analyses of 703 patients with
localized osteosarcoma, the resumption of chemotherapy 21 days or more after definitive surgery
was associated with an increased risk of death and adverse events (hazard ratio [HR], 1.57; 1.04–
2.36).[11] Delays in total time to completion of chemotherapy have also been associated with
inferior OS and event-free survival (EFS). In a retrospective multivariate analysis of 113 patients with
localized osteosarcoma, a delay of time to completion of chemotherapy greater than 4 weeks was
associated with an OS HR of 2.70 (1.11–6.76, P = .003) and an EFS HR of 1.13 (1.00–1.26, P =
.016).[12]

Amputation

For some patients, amputation remains the optimal choice for management of the primary tumor. A
pathologic fracture noted at diagnosis or during preoperative chemotherapy does not preclude limb-
salvage surgery if wide surgical margins can be achieved.[13] If the pathologic examination of the
surgical specimen shows inadequate margins, an immediate amputation should be considered,
especially if the histologic necrosis after preoperative chemotherapy was poor.[14]

Factors associated with an increased risk of local recurrence

Patients who undergo amputation have lower local recurrence rates than do patients who undergo
limb-salvage procedures.[15] There is no difference in OS between patients initially treated with
amputation and those treated with a limb-sparing procedure. Patients with tumors of the femur
have a higher local recurrence rate than do patients with primary tumors of the tibia or fibula.
Rotationplasty and other limb-salvage procedures have been evaluated for both their functional
outcome and their effect on survival. While limb-sparing resection is the current practice for local
control at most pediatric institutions, there are few data to indicate that salvage of the lower limb is
substantially superior to amputation with regard to patient quality of life.[16]

The German Cooperative Osteosarcoma Study Group performed a retrospective analysis of 1,802
patients with localized and metastatic osteosarcoma who underwent surgical resection of all
clinically detectable disease.[17][Level of evidence: 3iiA] Local recurrence (n = 76) was associated
with a high risk of death from osteosarcoma. Factors associated with an increased risk of local
recurrence included nonparticipation in a clinical trial, pelvic primary site, limb-preserving surgery,
soft tissue infiltration beyond the periosteum, poor pathologic response to initial chemotherapy,
failure to complete planned chemotherapy, and performing the biopsy at an institution different
from where the definitive surgery is being performed.
2. Chemotherapy

Preoperative chemotherapy

Almost all patients receive intravenous preoperative chemotherapy as initial treatment. However, a
standard chemotherapy regimen has not been determined. Current chemotherapy protocols include
combinations of the following agents: high-dose methotrexate, doxorubicin, cyclophosphamide,
cisplatin, ifosfamide, etoposide, and carboplatin.[18-26]

Evidence (preoperative chemotherapy):

A meta-analysis of protocols for the treatment of osteosarcoma concluded the following:[27]

Regimens containing three active chemotherapy agents were superior to regimens containing two
active agents.

Regimens with four active agents were not superior to regimens with three active agents.

Three-drug regimens that did not include high-dose methotrexate were inferior to three-drug
regimens that did include high-dose methotrexate.

An Italian study that used regimens containing fewer courses of high-dose methotrexate observed
the following results:[28][Level of evidence: 2A]

There was a lower probability for EFS than found in earlier studies that used regimens containing
more courses of high-dose methotrexate.

The Children's Oncology Group (COG) performed a prospective randomized trial in newly diagnosed
children and young adults with localized osteosarcoma. All patients received cisplatin, doxorubicin,
and high-dose methotrexate. One-half of the patients were randomly assigned to receive ifosfamide.
In a second randomization, one-half of the patients were assigned to receive the biological
compound muramyl tripeptide-phosphatidyl ethanolamine encapsulated in liposomes (L-MTP-PE)
beginning after definitive surgical resection. Results showed that:[29][Level of evidence: 1iiA]

The addition of ifosfamide did not improve outcome.

The addition of L-MTP-PE produced improvement in the EFS rate, which did not meet the
conventional test for statistical significance (P = .08), and a significant improvement in the OS rate
(78% vs. 70%; P = .03).

There has been speculation regarding the potential contribution of postrelapse treatment, although
there were no differences in the postrelapse surgical approaches in the relapsed patients. The
appropriate role of L-MTP-PE in the treatment of osteosarcoma remains under discussion.[30]

The COG performed a series of pilot studies in patients with newly diagnosed localized
osteosarcoma.[31][Level of evidence: 2A]

In pilot study 1, patients with lower degrees of necrosis after three-drug initial therapy received
subsequent therapy with a higher cumulative dose of doxorubicin of 600 mg/m2.

In pilot study 2, all patients received four-drug initial chemotherapy with cisplatin, doxorubicin, high-
dose methotrexate, and ifosfamide. Patients with lower degrees of necrosis received subsequent
chemotherapy with a higher cumulative dose of doxorubicin of 600 mg/m2.
In pilot study 3, all patients received the same four-drug initial chemotherapy as pilot study 2.
Patients with lower degrees of necrosis received higher doses of ifosfamide with the addition of
etoposide in subsequent therapy.

The results of these pilot studies were as follows:

Outcomes for all three pilot studies were similar to each other and to historical controls.

All patients received dexrazoxane before each dose of doxorubicin. The addition of dexrazoxane did
not appear to decrease the rate of good necrosis after initial therapy or EFS.

Left ventricular fractional shortening, as measured by echocardiography, was minimally affected at

78 weeks from study entry.

There was no evidence for an increased risk of secondary leukemia.

The international European and American Osteosarcoma Study (EURAMOS) Group consortium was
formed to conduct a large, prospective, randomized trial to help determine whether modifying the
chemotherapy regimen on the basis of the degree of necrosis would improve EFS. All patients
received initial therapy with cisplatin, doxorubicin, and high-dose methotrexate (MAP). Patients with
more than 90% necrosis were randomly assigned to continue the same chemotherapy after surgery
or to receive the same chemotherapy with the addition of interferon alpha-2B. Patients with less
than 90% necrosis were randomly assigned to continue the same chemotherapy or to receive the
same chemotherapy with the addition of high-dose ifosfamide and etoposide (MAPIE).[32][Level of
evidence: 1iiA]

At a median follow-up of 54 months for all registered patients (N = 2,260), the 3-year EFS rate was
59% (95% confidence interval [CI], 57%–61%), and the 5-year EFS rate was 54% (95% CI, 52%–56%).

The 3-year OS rate was 79% (95% CI, 77%–81%), and the 5-year OS rate was 71% (95% CI, 68%–73%).

Patients with localized disease at diagnosis (M0) who underwent complete surgical resection (n =
1,549) had 3-year EFS and OS rates from surgery of 70% and 88%, respectively; the 5-year EFS and
OS rates from surgery were 64% and 79%, respectively.

Forty percent of patients/families who participated in the study declined randomization after
definitive surgical resection, making interpretation of the outcome of the randomized study
questions more difficult and challenging the generalizability of the results.

Of patients with more than 90% necrosis, 716 were randomly assigned to continue the same
chemotherapy after surgery with or without the addition of interferon alpha-2B. The 3-year EFS
rates were 74% (95% CI, 69%–79%) for patients who received MAP and 77% (95% CI, 72%–81%) for
patients who received MAP plus interferon alpha-2B. The HR was 0.83 (95% CI, 0.61–1.12; P =
.2).[33]

Of patients with less than 90% necrosis, 618 were randomly assigned to continue MAP
chemotherapy or to receive MAPIE chemotherapy. The 3-year EFS estimates for patients with
localized disease were 60% (95% CI, 54%–66%) for the MAP group and 57% (95% CI, 51%–63%) for
the MAPIE group. The HR was 1.03 (0.81–1.33, P = .80).[34]
Postoperative chemotherapy

Historically, the extent of tumor necrosis was used in some clinical trials to determine what type of
postoperative chemotherapy would be given. In general, if tumor necrosis exceeded 90%, the
preoperative chemotherapy regimen was continued. If tumor necrosis was less than 90%, some
groups incorporated drugs not previously utilized in the preoperative therapy.

Patients with less necrosis after initial chemotherapy have a prognosis that is inferior to the
prognosis for patients with more necrosis. The prognosis is still substantially better than the
prognosis for patients treated with surgery alone and no adjuvant chemotherapy.

Based on the following evidence, it is inappropriate to conclude that patients with less necrosis have
not responded to chemotherapy and that adjuvant chemotherapy should be withheld for these
patients. Chemotherapy after definitive surgery should include the agents used in the initial phase of
treatment unless there is clear and unequivocal progressive disease during the initial phase of
therapy.

Evidence (using the same agents for postoperative chemotherapy):

Early reports from the Memorial Sloan Kettering Cancer Center (MSKCC) that suggested that adding
cisplatin to postoperative chemotherapy improved the outcome for patients with less than 90%
tumor necrosis.[35]

With longer follow-up, the outcome for patients with less than 90% tumor necrosis treated at
MSKCC was the same whether they did or did not receive cisplatin in the postoperative phase of
treatment.[36]

In an early experience, the German cooperative osteosarcoma group performed a trial in which the
chemotherapy regimen for patients with poor necrosis was changed after initial treatment.[37] The
agents used before surgery were discontinued and other agents were substituted.

The results were substantially poorer for these patients than for patients who continued to receive
the same agents.

Subsequent trials performed by other groups failed to demonstrate improved EFS when drugs not
included in the preoperative regimen (cisplatin) were added to postoperative therapy.[19,38]

A limited-institution pilot trial tested the strategy of discontinuing the agents used in the initial
phase of therapy for patients with poorer necrosis; postoperative therapy consisted of melphalan
with autologous stem cell reconstitution.[39]

The 5-year EFS rate for this group was 28%, which was lower than the EFS rates observed in many
large series in which agents were continued despite a lesser degree of necrosis.

The international European and American Osteosarcoma Study (EURAMOS) group consortium was
formed to conduct a large, prospective, randomized trial to help determine whether modifying the
chemotherapy regimen on the basis of the degree of necrosis would improve EFS. All patients
received initial therapy with cisplatin, doxorubicin, and high-dose methotrexate (MAP). Patients with
more than 90% necrosis were randomly assigned to continue the same chemotherapy after surgery
or to receive the same chemotherapy with the addition of interferon alpha-2b.[33][Level of
evidence: 1iiDi] In the same EURAMOS trial, patients with less than 90% necrosis were randomly
assigned to continue the same chemotherapy or to receive the same chemotherapy with the
addition of high-dose ifosfamide and etoposide (MAPIE).[34][Level of evidence: 1iiDi]
The addition of interferon alpha-2b did not improve the probability of EFS.

With a median follow-up of over 61 months, the EFS did not differ between the two groups.

The intensification of treatment in the MAPIE group resulted in greater toxicity than did the
treatment in the standard MAP arm.

Other chemotherapy approaches not considered effective

The Italian Sarcoma Group and the Scandinavian Sarcoma Group performed a clinical trial in patients
with osteosarcoma who presented with clinically detectable metastatic disease.[40] Consolidation
with high-dose etoposide and carboplatin followed by autologous stem cell reconstitution did not
appear to improve outcome and the investigators did not recommend this strategy for the
treatment of osteosarcoma.

Laboratory studies using cell lines and xenografts suggested that bisphosphonates had activity
against osteosarcoma.[41] A single-institution clinical trial demonstrated that pamidronate could
safely be administered along with multiagent chemotherapy to patients with newly diagnosed
osteosarcoma.[41] The French pediatric and adult sarcoma cooperative groups performed a
prospective trial for the treatment of osteosarcoma.[42] All patients received multiagent
chemotherapy, and patients were randomly assigned to receive or not to receive zoledronate. The
addition of zoledronate did not improve EFS.

3. Radiation therapy

If complete surgical resection is not feasible or if surgical margins are inadequate, radiation therapy
may improve the local control rate.[43,44]; [45][Level of evidence: 3iiA] Radiation therapy should be
considered in patients with osteosarcoma of the head and neck who have positive or uncertain
resection margins.[46][Level of evidence: 3iiA]

Evidence (radiation therapy for local control):

While it is accepted that the standard approach is primary surgical resection, a retrospective analysis
of a small group of highly selective patients reported long-term EFS with external-beam radiation
therapy for local control in some patients.[47][Level of evidence: 3iiiA]

Investigators from a single institution reported on 28 children and young adults with osteosarcoma
who were treated with radiation therapy for local control. Sixteen patients received radiation
therapy during the primary treatment course, and 12 patients received radiation therapy as part of
therapy after recurrence.[48]

For patients who received radiation therapy during primary treatment, the cumulative incidence of
local failure at 5 years was 25%.

For patients with recurrent disease, the cumulative incidence of local failure at 5 years was 44%.

Local tumor progression was observed in 3 of 13 patients (23%) who were treated with adjuvant
radiation therapy after resection, while three of six patients (50%) who received definitive radiation
therapy as a sole modality of local control experienced local progression.
4. Osteosarcoma of the Head and Neck

Osteosarcoma of the head and neck occurs in an older population than does osteosarcoma of the
extremities.[46,49-52] In the pediatric age group, osteosarcomas of the head and neck are more
likely to be low-grade or intermediate-grade tumors than are osteosarcomas of the
extremities.[53,54] All reported series emphasize the need for complete surgical resection.[46,49-
54][Level of evidence: 3iiiA] The probability for cure with surgery alone is higher for osteosarcoma of
the head and neck than it is for extremity osteosarcoma. When surgical margins are positive, there is
a trend for improved survival with adjuvant radiation therapy.[46,51][Level of evidence: 3iiiA]

There are no randomized trials to assess the efficacy of chemotherapy in patients with osteosarcoma
of the head and neck, but several series suggest a benefit.[49,55] Chemotherapy should be
considered for younger patients with high-grade osteosarcoma of the head and neck.[53,56]

Patients with osteosarcoma of the head and neck have a higher risk of developing a local recurrence
and a lower risk of having distant metastasis than do patients with osteosarcoma of the
extremities.[49,51,52,57]

Treatment of Osteosarcoma With Metastatic Disease at Diagnosis

Treatment Options for Osteosarcoma With Metastatic Disease at Diagnosis
Treatment Options for Lung-Only Metastases at Diagnosis
Treatment Options for Bone Metastases With or Without Lung Metastases

Approximately 20% to 25% of patients with osteosarcoma present with clinically detectable
metastatic disease. For patients with metastatic disease at initial presentation, roughly 20% will
remain continuously free of disease, and roughly 30% will survive 5 years from diagnosis.[1]

The lung is the most common site of initial metastatic disease.[2] Patients with metastases limited to
the lungs have a better outcome than do patients with metastases to other sites or to the lungs
combined with other sites.[1,3]

Treatment Options for Osteosarcoma With Metastatic Disease at Diagnosis

Treatment options for patients with osteosarcoma with metastatic disease at diagnosis include the
following:

Chemotherapy.

The chemotherapeutic agents used include high-dose methotrexate, doxorubicin, cisplatin, high-
dose ifosfamide, etoposide, and, in some reports, carboplatin or cyclophosphamide.
Evidence (chemotherapy):

In a trial that investigated high-dose ifosfamide (17.5 g per course) in combination with etoposide
for patients with newly diagnosed metastatic osteosarcoma, the following was observed:[4]

The combination produced a complete response in 10% of the patients and a partial response in 49%
of the patients.

However, similar to localized disease, there is no evidence that the addition of ifosfamide or
etoposide contributes to improved event-free survival (EFS) or overall survival (OS) in patients with
metastatic disease.

A study using a factorial design in patients with metastatic osteosarcoma (n = 91) evaluated the
addition of either muramyl tripeptide or ifosfamide to a standard chemotherapy regimen that
included cisplatin, high-dose methotrexate, and doxorubicin.[5]

There was a nominal advantage for the addition of muramyl tripeptide (but not for ifosfamide) in
terms of EFS and OS, but criteria for statistical significance were not met.

In the international European and American Osteosarcoma Study (EURAMOS) group consortium, 362
of 2,186 patients (17%) presented with metastasis at diagnosis. Patients were randomly assigned to
receive either treatment with cisplatin, doxorubicin, and high-dose methotrexate or cisplatin,
doxorubicin, high-dose methotrexate, and ifosfamide.[6][Level of evidence: 1iiA]

At a median follow-up of 47 months, the 3-year EFS rate was 32% (95% confidence interval [CI],
27%–37%), and the 5-year EFS rate was 28% (95% CI, 23%–33%).

The 3-year OS rate was 56% (95% CI, 50%–61%), and the 5-year OS rate was 45% (95% CI, 39%–50%).

The treatment options for UPS of bone with metastasis at initial presentation are the same as the
treatment for osteosarcoma with metastasis. Patients with unresectable or metastatic UPS have a
very poor outcome.[7]

Treatment Options for Lung-Only Metastases at Diagnosis

Treatment options for patients with metastatic lung lesions at diagnosis include the following:

Preoperative chemotherapy followed by surgery to remove the tumor followed by postoperative

combination chemotherapy.

Patients with metastatic lung lesions as the sole site of metastatic disease should have the lung
lesions resected if possible. Generally, this is performed after the administration of preoperative
chemotherapy. After definitive surgical resection of the primary tumor, most clinicians resume
systemic chemotherapy before initiating lung surgery to avoid longer delays in the resumption of
chemotherapy. In approximately 10% of patients, all lung lesions disappear after preoperative
chemotherapy.[3] Complete resection of pulmonary metastatic disease can be achieved in a high
percentage of patients with residual lung nodules after preoperative chemotherapy. The long-term
survival is poor for patients who do not undergo complete surgical resection of pulmonary
metastatic disease.[8,9][Level of evidence: 2A]
For patients who present with primary osteosarcoma and metastases limited to the lungs and who
achieve complete surgical remission, the 5-year EFS rate is approximately 20% to 25%. Multiple
metastatic nodules confer a worse prognosis than do one or two nodules, and bilateral lung
involvement is worse than unilateral.[1] Patients with peripheral lung lesions may have a better
prognosis than patients with central lesions.[10] Patients with fewer than three nodules confined to
one lung may achieve a 5-year EFS rate of approximately 40% to 50%.[1]

Treatment Options for Bone Metastases With or Without Lung Metastases

The second most common site of metastasis is another bone that is distant from the primary tumor.
Patients with metastasis to other bones distant from the primary tumor experience Event Free
Survival and Overall Survival rates of approximately 10%.[1] In a study of patients who presented
with primary extremity tumors and synchronous metastasis to other bones, only 3 of 46 patients
remained continuously disease-free 5 years later.[11] Patients with transarticular skip lesions have a
poor prognosis.[12]

Multifocal osteosarcoma is different from osteosarcoma that presents with a clearly delineated
primary lesion and limited bone metastasis. Multifocal osteosarcoma classically presents with
symmetrical, metaphyseal lesions, and it may be difficult to determine the primary lesion. Patients
with multifocal bone disease at presentation have an extremely poor prognosis, but treatment with
systemic chemotherapy and aggressive surgical resection may significantly prolong life.[13,14]

Treatment options for patients with bone metastases with or without lung metastases include the
following:

Preoperative chemotherapy followed by surgery to remove the primary tumor. After definitive
surgical resection of the primary tumor, most clinicians resume systemic chemotherapy. The timing
of surgery to remove metastatic tumors is not well defined. It is usually not attempted at the time of
primary surgery because delays of more than 21 days until resumption of chemotherapy can
increase the risk of adverse events and death.[15]

Surgery to remove the primary tumor followed by chemotherapy and then surgical resection of
metastatic disease followed by postoperative combination chemotherapy.

When the usual treatment course of preoperative chemotherapy followed by surgical ablation of the
primary tumor and resection of all overt metastatic disease followed by postoperative combination
chemotherapy cannot be used, an alternative treatment approach may be used. This alternative
treatment approach begins with surgery for the primary tumor, followed by chemotherapy, and
then surgical resection of metastatic disease. This approach may be appropriate in patients with
intractable pain, pathologic fracture, or uncontrolled infection of the tumor when initiation of
chemotherapy could create risk of sepsis.[16]

Resection of metastatic bone lesions if possible.

Radiation therapy to the extremities. There is some evidence that radiation therapy to the
extremities may offer some local control.[17]
Treatment of Recurrent Osteosarcoma
Prognostic Factors for Recurrence
Treatment Options for Recurrent Osteosarcoma and UPS of Bone
Surgery
Chemotherapy and targeted therapy
Radiopharmaceuticals and radiation therapy
Treatment Options for Local Recurrence
Treatment Options for Lung-Only Recurrence
Treatment Options for Recurrence With Bone-Only Metastases
Treatment Options for Second Recurrence of Osteosarcoma
Treatment Options Under Clinical Evaluation

Approximately 50% of relapses in patients with recurrent osteosarcoma occur within 18 months of
therapy termination, and only 5% of recurrences develop beyond 5 years.[1-4]

Prognostic Factors for Recurrence

Prognostic factors for recurrent osteosarcoma or undifferentiated pleomorphic sarcoma (UPS) of

bone include the following:

Time from diagnosis. In 564 patients with a recurrence, patients whose disease recurred within 2
years of diagnosis had a worse prognosis than did patients whose disease recurred after 2 years.[1]
In another series of 431 patients, recurrences occurring less than 2 years from diagnosis were also
associated with worse outcomes.[5]

Age at initial diagnosis. Older age at initial study enrollment was associated with a worse
prognosis.[5]

Presence of metastatic disease at diagnosis. The presence of metastatic disease at initial

presentation was associated with a worse prognosis.[5]

Tumor response to preoperative chemotherapy. Patients with a good histologic response to initial
preoperative chemotherapy had a better overall survival (OS) after recurrence than did poor initial
responders.[1]

Site of metastases. The probability of developing lung metastases at 5 years is 28% in patients
presenting with localized disease.[6] In two large series, the incidence of recurrence by site was as
follows: lung only (65%–80%), bone only (8%–10%), local recurrence only (4%–7%), and combined
relapse (10%–15%).[4,7] Abdominal metastases are rare but may occur as late as 4 years after
diagnosis.[8] The Children's Oncology Group (COG) reported the outcomes of 431 young patients
with recurrent osteosarcoma.[5][Level of evidence: 3iiDiv] Patients with recurrences in both lung
and bone had worse outcomes than did patients with recurrences in lung only (P = .005).

Surgical resectability. Patients with recurrent osteosarcoma should be assessed for surgical
resectability, because they may sometimes be cured with aggressive surgical resection with or
without chemotherapy.[9,7,10-13]

Control of osteosarcoma after recurrence depends on complete surgical resection of all sites of
clinically detectable metastatic disease. If surgical resection is not attempted or cannot be
performed, progression and death are certain. The ability to achieve a complete resection of
recurrent disease is the most important prognostic factor at first relapse, with a 5-year survival rate
of 20% to 45% after complete resection of metastatic pulmonary tumors and a 20% survival rate
after complete resection of metastases at other sites.[4,7,13,14]

Treatment Options for Recurrent Osteosarcoma

Treatment options for patients with recurrent osteosarcoma or UPS of bone include the following:

Surgery to remove all sites of metastatic disease.

Chemotherapy and targeted therapy.
Radiopharmaceuticals and radiation therapy.

Surgery
Control of osteosarcoma after recurrence depends on complete surgical resection of all sites of
clinically detectable metastatic disease.

Chemotherapy and targeted therapy

The role of systemic chemotherapy for the treatment of patients with recurrent osteosarcoma is not
well defined. The selection of further systemic treatment depends on many factors, including the
site of recurrence, the patient’s previous primary treatment, and individual patient considerations.

The COG reported the outcomes of patients with recurrent osteosarcoma from seven phase II trials,
all of which were assessed to have shown no treatment benefit.[15]

The event-free survival (EFS) rate for 96 patients with osteosarcoma and measurable disease was
12% at 4 months (95% confidence interval [CI], 6%–19%).

There was no significant difference in EFS between the trials according to the number of previous
treatment regimens or patient age, sex, and ethnicity.

One additional phase II trial with a different study design was reported. In this trial, patients with
osteosarcoma and metastases to the lung underwent surgical resection of all lung nodules and then
were treated with adjuvant inhaled granulocyte-macrophage colony-stimulating factor (GM-
CSF).[15]

The 12-month EFS rate for the 42 evaluable patients enrolled in this study was 20% (95% CI, 10%–
34%).

The following chemotherapy and targeted therapy agents have been studied to treat recurrent
osteosarcoma and UPS of bone:

Ifosfamide alone with mesna uroprotection, or in combination with etoposide. Ifosfamide alone with
mesna uroprotection, or in combination with etoposide, has been active in as many as one-third of
patients with recurrent osteosarcoma who have not previously received this drug.[16-19]

Gemcitabine and docetaxel. A nonrandomized comparison of two doses of gemcitabine, both given
with docetaxel, suggested that a higher dose of gemcitabine (900 mg/m2) was associated with a
better response rate and longer survival than was a lower dose of gemcitabine (675 mg/m2) for
recurrent or refractory osteosarcoma.[20][Level of evidence: 3iiA] The combination of gemcitabine
(at a dose of 900 mg/m2) and docetaxel has also been reported to have activity in some studies that
included patients with unresectable disease.[21-23]; [24][Level of evidence: 3iiDiv]

Cyclophosphamide and etoposide. Cyclophosphamide and etoposide have been shown to have
activity in recurrent osteosarcoma.[21]

Sorafenib. The Italian Sarcoma Group reported rare objective responses and disease stabilization
with sorafenib in patients with recurrent osteosarcoma.[25]

Sorafenib and everolimus. The Italian Sarcoma Group also reported the outcome of patients with
metastatic recurrent osteosarcoma treated with the combination of sorafenib and everolimus. They
observed two partial responses and two minor responses in 38 patients; 17 of 38 patients were
progression free at 6 months from study entry but toxicity was greater than with sorafenib
monotherapy.[26][Level of evidence: 2Diii]

Regorafenib. Two prospective, randomized, double-blind trials have evaluated the role of
regorafenib in the treatment of metastatic recurrent osteosarcoma. Both studies used the approved
treatment regimen of 160 mg by mouth daily for 21 days followed by 7 days without treatment. The
French trial randomly assigned patients 2:1 between regorafenib and placebo and allowed crossover
for patients assigned to placebo.[27] Seventeen of 26 patients (65%; one-sided 95% CI, 47%) in the
regorafenib group did not have disease progression at 8 weeks, compared with 0 of 12 patients in
the placebo group. The Sarcoma Alliance for Research Collaboration (SARC) group randomly
assigned adult patients 1:1 between regorafenib and placebo.[28] Median progression-free survival
was significantly improved with regorafenib versus placebo: 3.6 months (95% CI, 2.0–7.6 months)
versus 1.7 months (95% CI, 1.2–1.8 months), respectively (hazard ratio, 0.42; 95% CI, 0.21–0.85; P =
.017).

Radiopharmaceuticals and radiation therapy

High-dose samarium Sm 153-ethylenediamine tetramethylene phosphonic acid (153Sm-EDTMP)

coupled with peripheral blood stem cell support may provide significant pain palliation in patients
with bone metastases.[29-32] Toxicity of 153Sm-EDTMP is primarily hematologic.[33][Level of
evidence: 3iiDiii]

A single-institution retrospective review reported that high-dose fraction radiation therapy (2

Gy/fraction) was a useful form of palliation for patients with recurrent osteosarcoma.[34][Level of
evidence: 3iiiDiv] Thirty-two courses of palliative radiation therapy were given to 20 patients with
symptomatic metastatic and/or locally recurrent primary disease. Twenty-four of the 32 courses
(75%) were associated with symptom improvement. Higher doses of radiation therapy correlated
with longer durations of symptom response.

Treatment Options for Local Recurrence

Treatment options for patients with osteosarcoma or UPS of bone that has recurred locally include
the following: Surgery to remove the tumor.

The postrelapse outcome of patients who have a local recurrence is quite poor,[35-37] and survival
of patients with local recurrences and either previous or concurrent systemic metastases is poor.[38]
Two retrospective, single-institution series reported a survival rate of 10% to 40% after local
recurrence without associated systemic metastasis.[38-41]

A retrospective review from the Italian Sarcoma Group identified 62 patients (median age, 21 years)
with local recurrences.[42] With a median follow-up of 43 months (range, 5–235 months), the 5-year
post–local relapse survival rate was 37%, significantly better for patients with a longer local
recurrence–free interval (≤24 months, 31% vs. >24 months, 61.5%; P = .03), absence of distant
metastases (no distant metastases, 56% vs. distant metastases, 11.5%; P = .0001), and achievement
of second complete remission (CR2) by surgical resection (no CR2, 0% vs. CR2, 58.5%; P = .0001). No
difference in post–local relapse survival was found according to age, and there was no benefit from
chemotherapy administration.

The incidence of local relapse was higher in patients who had a poor pathologic response to
chemotherapy in the primary tumor and in patients with inadequate surgical margins.[35,40]

Treatment Options for Lung-Only Recurrence

Treatment options for patients with osteosarcoma and UPS of bone that has recurred in the lung
only include the following:

Surgery to remove the tumor.

Repeated resections of pulmonary recurrences can lead to extended disease control and possibly
cure for some patients.[14,43] Survival for patients with unresectable metastatic disease is less than
5%.[7,44] The 5-year EFS rate for patients who have complete surgical resection of all pulmonary
metastases ranges from 20% to 45%.[4,13,14]; [45][Level of evidence: 3iiiA]

Factors associated with a better outcome include the following:[4,6,7,46,47]

Fewer pulmonary nodules.

Unilateral pulmonary metastases.

Longer intervals between primary tumor resection and metastases.

Tumor location in the periphery of the lung.

Approximately 50% of patients with one isolated pulmonary lesion more than 1 year after diagnosis
were long-term survivors after metastasectomy. Chemotherapy did not appear to offer an
advantage.[48][Level of evidence: 3iiiA]

Control of osteosarcoma requires surgical resection of all macroscopic tumors; however,

recommendations are conflicting regarding the surgical approach to the treatment of pulmonary
metastases in osteosarcoma. Several options are available to resect pulmonary nodules in a patient
with osteosarcoma, including thoracoscopy and thoracotomy with palpation of the collapsed lung.
When patients have nodules identified only in one lung, some surgeons advocate thoracoscopy;
some advocate unilateral thoracotomy; and some advocate bilateral thoracotomy. Bilateral
thoracotomy can be performed as a single surgical procedure with a median sternotomy or a
clamshell approach, or by staged bilateral thoracotomies.
Evidence (surgical approach for lung-only recurrence of osteosarcoma):

The St. Jude Children’s Research Hospital reported on 81 patients who had pulmonary nodules
identified at initial presentation in only one lung by computed tomography (CT) scan.[49] They
performed unilateral thoracotomy and did not explore the contralateral hemithorax. At the time of
thoracotomy, 44 of 81 patients had a solitary nodule identified; 15 of 81 patients had two nodules
identified; 16 of 81 patients had three to five nodules identified; and 6 of 81 patients had more than
five nodules identified. Additional patients who were considered unresectable were not included in
the analysis.

Thirty-nine of 81 patients had subsequent pulmonary recurrence; for most patients, recurrence
occurred within 6 months.

Within the first 6 months, 9 of 81 patients had ipsilateral recurrence, and 10 of 81 patients had a
contralateral recurrence. By 2 years after initial thoracotomy, 13 of 81 patients had ipsilateral
recurrence; 19 of 81 patients had contralateral recurrence; and 2 of 81 patients had bilateral
recurrence.

OS was similar for patients with ipsilateral and bilateral recurrence.

The Memorial Sloan Kettering Cancer Center reported on pulmonary metastatic disease recurrence
after initial therapy for osteosarcoma. Fourteen patients had pulmonary nodules identified in only
one lung by CT scan; nine patients were identified less than 2 years from initial diagnosis (early
metastases), and five patients were identified more than 2 years from initial diagnosis (late
metastases).[46] Seven of nine patients with early metastases had staged contralateral
thoracotomies, and six of seven had nodules removed from the contralateral lung, despite negative
CT scan findings.

The lack of a comparison group precludes evaluation of the impact of the contralateral thoracotomy
on subsequent EFS or OS.

The same group expanded their analysis to include a retrospective review of 161 thoracotomies
performed in 88 patients with osteosarcoma metastatic to the lung.[50] In this expanded series, CT
failed to identify one-third of pulmonary metastases confirmed by pathologic examination.

Treatment Options for Recurrence With Bone-Only Metastases

Treatment options for patients with osteosarcoma that has recurred in the bone only include the
following:

Surgery to remove the tumor.

153Sm-EDTMP with or without stem cell support.

Patients with osteosarcoma who develop bone metastases have a poor prognosis. In one large
series, the 5-year EFS rate was 11%.[51] Patients with late solitary bone relapse have a 5-year EFS
rate of approximately 30%.[51-54]

For patients with multiple unresectable bone lesions, 153Sm-EDTMP with or without stem cell
support may produce stable disease and/or provide pain relief.[33]
Treatment Options for Second Recurrence of Osteosarcoma

Treatment options for patients with osteosarcoma or UPS of bone that has recurred twice include
the following:

Surgery to remove the tumor and/or chemotherapy.

Evidence (surgery and/or chemotherapy):

The cooperative German-Austrian-Swiss osteosarcoma study group reported on 249 patients who
had a second recurrence of osteosarcoma. The main feature of therapy was repeated surgical
resection of recurrent disease.[55]

Of these patients, 197 died and 37 were alive in CR (24 patients after a third complete response and
13 patients after a fourth or subsequent complete response).

Fifteen patients who did not achieve surgical remission remained alive, but follow-up for these
patients was extremely short.

The Spanish Group for Research on Sarcoma reported the results of a phase II trial of patients with
relapsed or refractory osteosarcoma who were treated with gemcitabine and sirolimus.[56][Level of
evidence: 3iiDiv]

The progression-free survival (PFS) rate at 4 months was 44%.

After central radiologic review of 33 assessable patients, 2 partial responses and 14 disease
stabilizations (48.5%) were reported.

The COG reported the outcomes of patients with recurrent osteosarcoma from seven phase II trials,
all of which were assessed to have shown no treatment benefit.[15]

The EFS rate for 96 patients with osteosarcoma and measurable disease was 12% at 4 months (95%
CI, 6%–19%).

There was no significant difference in EFS between the trials according to the number of previous
treatment regimens or patient age, sex, and ethnicity.

One additional phase II trial with a different study design was reported. In this trial, patients with
osteosarcoma and metastases to the lung underwent surgical resection of all lung nodules and then
were treated with adjuvant inhaled GM-CSF.[15]

The 12-month EFS rate for the 42 evaluable patients enrolled in this study was 20% (95% CI, 10%–
34%).