Bone and Soft Tissue Tumors: A Multidisciplinary Review with Case Presentations Therese J. Bocklage, Robert H. Quinn, Berndt P. Schmit, Claire F. Verschraegen
Chapter Notes

Save Clear

Pathology overview of musculoskeletal tumors and principles of diagnosisChapter 1

Individuals learn the practice of medicine not only through lectures and laboratory sessions but also through daily experiential learning guided by senior practitioners. This form of apprenticeship learning defines this text, which combines succinct overviews of the tumors followed by actual patient cases.
Problem-based learning
The current term for such case-based learning is problem-based learning (PBL). This book parallels the PBL curriculum popular in many medical schools.
Problem-based learning not only recognizes but also capitalizes on the concept that human beings are storytellers and story listeners by nature. Consequently, students naturally absorb key medical elements of a personal story and may find learning passive sets of data both more difficult and less engaging. One complements the other, and this book presents both. The text is arranged in the flow in which patients are treated. The histories are trimmed to key facets.
Evidence-based data
Where feasible, the information provided in this text is evidence based. However, extensively verified facts are few in rare entities, and the realm of bone and soft tissue tumors is replete with such rarities. Thus, the approach to treating a patient with a rare tumor is one that is comparative, sometimes requiring a trial and error approach. However, with careful documentation of these patients’ care along with data aggregation, even rare tumors become better characterized, advancing the field of musculoskeletal oncology.
Team approach
Close collaboration that bridges medical disciplines yields optimal care of a patient with a mesenchymal tumor. This is facilitated by regular review of patient cases in the setting of a working tumor board, including a surgical oncologist, medical oncologist, radiation oncologist, radiologist and pathologist. Students should always be welcome.
The treatment of a complex patient also relies on contributions from nurses, social workers, nutritionists, counselors, psychiatrists, laboratorians, pathologists’ assistants, physical and occupational therapists and others, and this book acknowledges the dedicated attention of these medical professionals.
The value of exploring a disease course in a single patient
In many books on bone and soft tissue pathology, tumors often appear clear and easily identifiable as the images shown are the result of the cumulative experience of the author, who has distilled his or her knowledge to present ‘perfect’ examples of the tumor(s) in question.
The reality is that tumors do not always present so unambiguously. Each real case presented in this text outlines the patient's history, laboratory findings, disease progress, and outcomes. They are the actual findings in one patient, intended to illustrate in practical terms the real (and often not so clear-cut) presentation, diagnosis, treatment and outcome in an individual with a specific bone or soft tissue lesion.
Gross specimens
The role of the anatomic pathologist is to ensure the proper handling, processing, and diagnosis of the patient's specimen beginning with a fine needle aspiration biopsy, a needle core biopsy or incisional biopsy and commonly ending with a large, complex resection. Although occasionally undervalued, careful gross examination of soft tissue and bone tumors is critical to ensure subsequent accurate diagnosis, to assign the correct tumor grade and to record response to therapy.
Gross assessment of necrosis
Depending on the specific sarcoma type, the amount of pre-treatment necrosis influences grade and prognosis. Percent necrosis is estimated by careful gross review of complete cross sections of tumor. Figure 1.1 is an example of a mixed viable and necrotic tumor.
Value of gross photos
Digital cameras have made photographing gross specimens more convenient. Moreover, some pathology software programs enable integration of photographs into the final report. Gross photos illustrate to the medical team features such as margin status and chemotherapy response. Figures 1.2 through 1.6 compare the gross appearances of multiple tumors.
Tissue banking for research
Collecting fresh tumor (and paired normal tissue, when possible) for biobanking for future research investigations is worthwhile. A small amount (1 cm3, for example) can be frozen and stored for years. At hospitals without sufficient freezer capacity, new solutions are available to store tissue at room temperature while adequately preserving RNA. This banking effort is essential for improving the diagnostic categorization of mesenchymal tumors, identifying biomarkers, and developing targeted therapy. Formalin-fixed, paraffin embedded blocks are also capable of being mined for more limited nucleic acid and protein information.2
zoom view
Figure 1.1: Extraskeletal myxoid chondrosarcoma, gross cut section. This tumor starkly contrasts viable myxoid gelatinous tumor (left) with an area of abrupt necrotic cavitated yellow friable tumor (right). The necrotic, partly cystic area may represent infarction after prior biopsy.
zoom view
Figure 1.2: Lipoblastoma, external surface of tumor, gross photo. The external surface of many tumors exhibits a thin translucent membranous covering.
Ancillary studies to confirm the diagnosis
In addition to harvesting fresh tumor for banking, it may be advisable to conduct cytogenetic analysis, and this will require sending a small sterile sample to the cytogenetic laboratory for karyotyping. In some cases, it is also helpful to prepare touch preps of tumor on unfixed slides for potential fluorescent in situ hybridization (FISH) analysis. These slides can be stored at room temperature.
Very few tumors are subjected to electron microscopy (EM), and therefore tumor aliquots for EM are seldom prepared.
When feasible with bone tumors, it is advisable to submit a portion of tumor without decalcification in order to optimize antigenicity for potential IHC staining (molecular genetic analysis may also be compromised after decalcification). To avoid this harsh pretreatment, a small sample dissected free of bone and cartilage fragments is submitted to the histology department. With heavily calcified or ossified lesions, it may not be possible to obtain a tumor sample that does not require decalcification.
zoom view
Figure 1.3: Lipoblastoma, cut surface, gross photo. On cross sections, a adipocytic tumor may resemble mature adipose tissue with a pale yellow soft, glistening cut surface. Subtle additional clues to the diagnosis may also be present. Lipoblastoma is typically lobulated on microscopic review, and the lobulation can be appreciable also grossly, as in this example.
zoom view
Figure 1.4: Osteosarcoma, gridding post chemotherapy. A band saw cuts thin sections of fresh tumor amenable to careful gridding to assess percent necrosis after neoadjuvant chemotherapy.
John S. J. Brooks, American bone and soft tissue pathologist, suggests asking four questions when dealing with a possible mesenchymal tumor (Brooks, 2010 – questions are paraphrased):
1. Is the mass an actual neoplasm?
Some non-neoplastic lesions can mimic a neoplasm, and the most common of these are myofibroblastic proliferations such as nodular fasciitis and proliferative fasciitis. Other entities that may raise concern for a neoplasm comprise reactive entities such as a maturing fracture callus.3
zoom view
Figure 1.5: Epithelioid hemangioendothelioma, gross photo of cut section. Epithelioid hemangioendothelioma does not reveal its vascular nature on gross inspection. This is true of many intermediate and high grade blood vessel tumors. Notice the multiple discrete nodules, which can be seen in benign and malignant tumors.
zoom view
Figure 1.6: Malignant peripheral nerve sheath tumor, gross photo of cut section. This large, deep seated sarcoma demonstrates aggressive features with multiple, subfascial, coalescing, infiltrating tumor nodules.
To correctly answer whether the lesion is neoplastic, one must have a working knowledge of the histology of non-neoplastic mimics of mesenchymal neoplasms. In histologically ambiguous situations, one must rely upon the contributions of colleagues: clinicians and radiologists who can be counted upon for critical patient history and imaging information.
2. If it is a neoplasm, is it malignant?
Once a lesion is determined to be neoplastic, the next step is to ascertain whether it is benign or malignant. A pathologist begins with general histologic rules of malignancy and benignancy (see below).
3. If this is a malignant neoplasm, is it a sarcoma or could it represent a metastatic carcinoma, melanoma or lymphoma?
For example, not all fascicular spindle cell tumors are mesenchymal. If the lesion is deemed malignant, the origin of the tumor is the next question: primary mesenchymal versus metastatic non-mesenchymal tumor? For example, a spindle cell tumor of the head and neck in an older man is more likely sarcomatoid carcinoma (such as spindle cell squamous carcinoma) or melanoma than sarcoma. Clinical history of melanoma should never be summarily dismissed as irrelevant.
4. Is the diagnosis thoroughly defensible?
In other words, have the other entities in the differential been completely excluded? Painstaking effort should be exerted to be confidently conclusive especially in cases where treatment is dictated by and differs by diagnosis.
This final question is a critical query that is faced by many pathologists who may have limited experience with bone and soft tissue pathology due to the rarity of some of these lesions. If a pathologist cannot be completely confident of the specific diagnosis, the most prudent option is to consult with an acknowledged bone or soft tissue expert pathologist. This has two desirable consequences: optimal diagnosis for the patient and convergence of data pertaining to unusual tumors.
Rules of histology
Benign features of mesenchymal tumors
Table 1.1 lists some of the histologic features that are associated most commonly with benign behavior in mesenchymal tumors. Of course, there are exceptions to this list of features because some malignant tumors may look deceptively benign, while some benign tumors may appear high grade malignant.
Malignant features of mesenchymal tumors
Table 1.2 lists the histologic features of mesenchymal tumors that are commonly associated with malignant (aggressive local or metastatic) behavior. Just as for the benign histologic features listed in Table 1.1, the general histologic features of malignancy in Table 1.2 are not 100% sensitive and specific. Both tables merely provide starting guidelines.
Immunohistochemistry does not supersede histology
The development of consistent, reliable immunohistochemistry (IHC) to detect cell antigens was revolutionary in surgical pathology and dramatically advanced pathology diagnostics. However, IHC cannot replace careful review of the standard histologic characteristics of a tumor. This is because nearly all antigens tested by IHC are not specific, and because quite a few tumors, typically those that are less differentiated, aberrantly express antigens that can obscure the diagnosis. In all instances, the IHC reactivity pattern should be examined in the context of the general overall histologic appearance of the tumor, clinical history and radiologic findings.
Today, thousands of antibodies are available for use in formalin fixed, paraffin embedded tissues. Judicious rather than opulent use of these antibodies is both less costly and less confusing. Table 1.3 provides a broad overall delineation of the more common antibodies useful in bone and soft tissue tumors showing reactivity in the analogous normal tissue, benign lesions, and malignant tumors. Table 1.4 lists tumors that have relatively more specific antibody signatures.4
Table 1.1   Histologic features commonly associated with benign behavior in mesenchymal tumors
Example of low magnification
Example of high magnification
Pushing rather than invasive border
No necrosis
Few if any mitotic figures and no abnormal mitotic figures
Pale uniform nuclei with fine chromatin
Figure 1.7 shows a prominently positive keratin IHC stain in high grade sarcoma. This aberrant epithelial keratin expression illustrates a misleading result, which is especially dangerous if examined in isolation of clinical and radiologic findings, histomorphology and other IHC results. Table 1.5 lists common cell and architectural features found in bone and soft tissue tumors.
Additional figures of characteristic immunohistochemical staining patterns demonstrate that the intracellular location of reactivity can be critical in the diagnostic value of the antibody (Figs 1.8 through 1.10).
Like other neoplasms, benign and malignant mesenchymal tumor cells may contain acquired genetic lesions that result in the activation of oncogenes and/or the inactivation of tumor suppressor genes. Soft tissue and bone tumors are characterized by the acquisition of a series of mutations as they develop. These mutations generally occur in the regulatory or coding regions of genes that are transcribed into mRNAs that are translated into proteins.5
Table 1.2   Histologic features commonly associated with malignant behavior in mesenchymal tumors
Example of low magnification
Example of high magnification
Invasive or permeative border
Nuclear pleomorphism
Primitive blastic (small blue round cell) look
Nuclear hyperchromasia
Coarse and often irregularly distributed chromatin
High mitotic rate
Atypical mitotic figures
The result is either an abnormally high or low concentration of the protein or a protein with aberrant or reduced function. Mutations may also cause the production of abnormal regulatory RNAs. Mutations that may be found in these lesions range from large scale chromosomal rearrangements, detectable by cytogenetic assays to single nucleotide ‘point mutations’ that must be identified by sensitive molecular diagnostic techniques.
Classic cytogenetic, FISH cytogenetic, and molecular diagnostic tests have several basic features in common. A cell or tissue sample from the patient is obtained. Target cellular DNA or RNA is either made readily accessible for DNA hybridization within the cells via enzymatic treatment, or the DNA or RNA is completely isolated and purified from the cells and re-suspended in water. The target DNA is then made single stranded by heating, disrupting the hydrogen bonds that join the complementary Gs and Cs and As and Ts (guanine, cytosine, adenine and thymine respectively). This makes the target DNA accessible to hybridization with a DNA probe specific for a chromosomal region, for a gene, or for a gene region. DNA probes may be synthetic or cloned DNA of known sequence and may range in size from 12 nucleotides to several hundred thousand nucleotides.
When cells are studied with cytogenetic techniques, a karyotype is prepared in which the autosomal pairs of metaphase chromosomes and sex chromosomes are arranged in order from largest to smallest for identification and inspection. Benign and malignant mesenchymal lesions may have normal appearing karyotypes, karyotypes with specific isolated translocations involving the rearrangement of two genes, or bizarre karyotypes with massive chromosomal duplications and deletions and the presence of multiple unidentifiable ‘marker’ chromosomes. Chromosomal abnormalities in mesenchymal lesions may result in the duplication, amplification, or deletion of chromosomes or chromosome regions and the genes contained therein. Alternatively, translocations occurring at specific breakpoints between two chromosomes may alter the regulation of expression of a critical gene or join portions of two genes together resulting in the production of a chimeric mRNA and an encoded protein with an oncogenic function.
Cytogenetic assays allow the study of chromosomes and chromosome regions within cells. A sample from the tumor can be collected under sterile conditions, placed in tissue culture fluid and sent to the laboratory for analysis. Freezing the sample or placing it in fixative makes cytogenetic studies for which the cells must be cultured impossible. In the laboratory, the tumor cells may be cultured for several days before arresting dividing cells in metaphase when the chromatin is condensed, making the chromosomes relatively easy to analyze.
A wide range of molecular genetic technology and clinical techniques have been developed to assist in the detection of acquired mutations in tumor cells. Table 1.6 describes the major clinical genetic assays and techniques used in assessment of bone and soft tissue tumor lesions.
Table 1.7 lists the genetic findings in bone tumors. Table 1.8 lists the genetic findings in soft tissue tumors. These do not capture all genetic abnormalities, as new findings are continuously being reported.7
Table 1.3   Common antibodies useful in mesenchymal tumor diagnosis
Antibody or antibody set
Distribution in normal tissue
Some benign tumors reactive for this antibody
Some malignant tumors reactive for this antibody
Other antibodies uncommonly expressed
Desmin, SMA, MSA, caldesmon, smooth muscle myosin heavy chain
Smooth muscle
S-100 protein
Myogenin, desmin, MyoD, α-sarcomeric actin, CD56
Skeletal muscle
Smooth muscle actin
S-100 protein (rare)
SMA, MSA; desmin less frequently
Pankeratin (fairly common)
S-100 protein (less frequent in high grade tumors)
CDK4 and MDM2 in low grade tumors and dedifferentiated LPS
Adipose tissue
Lipoma (S-100 protein only)
Dedifferentiated LPS may form muscle that will be SMA and/or desmin reactive
S-100 protein, HMB45, Melan-A, MiTF
Melanoma and clear cell sarcoma
Some melanomas are negative for all melanocytic markers
Clear cell sarcoma may express only one marker
S-100 protein
Enchondroma, synovial chondromatosis
No others in common use
S-100 protein
Peripheral schwann cells
Schwannoma, neurofibroma
Malignant peripheral nerve sheath tumor (MPNST) (50% of cases, usually focal but epithelioid variant is usually diffusely positive for S-100 protein)
GFAP +/-.
Keratins may be positive, especially in MPNST.
Desmin and myogenin in Malignant Triton Tumor
EMA, Claudin, GLUT1
Perineurial cells
MPNST with perineurial differentiation
CD34 in sclerosing perineurioma
Neurofilaments, GFAP, NSE, CD56
Axonal cells
MPNST with axonal differentiation
CD56 also stains other tumors
S-100 and CD34 are also positive in neurofibroma
CD34, CD31, FLI-1, ERG (C terminus directed)
Endothelial cells
D2–40 is expressed in normal lymphatics and some angiosarcomas
CD34 is also expressed in nonvascular tumors such as DFSP, GIST and SFT
Keratins may be expressed especially in epithelioid vascular tumors
CD99 (membranous reactivity only), Fli-1
No known normal cell counterpart
No known benign tumor
Ewing sarcoma family of tumors
Pankeratin, S-100 protein
Antibodies in bold font are the main antibodies used for the particular tumor. SMA: smooth muscle actin, MSA: muscle specific actin (HHF-35 actin), MiTF: Microphthalmia-associated transcription factor, EMA: epithelial membrane antigen, GFAP: glial fibrillary acidic protein, NSE: neuron specific enolase, FLI-1: Friend leukemia virus integration 1, GLUT1: Glucose transporter 1, HMB45: Melanosome clone HMB45, CD56: NCAM, neural cell adhesion molecule, Melan-A: Melanocyte A antibody (A103), CDK4: cyclin-dependent kinase 4, MDM2: Mdm2 [murine double minute] p53 binding protein homolog (mouse), DFSP: Dermatofibrosarcoma protuberans, GIST: Gastrointestinal stroma tumor, SFT: Solitary fibrous tumor
Table 1.4   Some tumors with relatively specific antigen reactivity patterns
Antibody reactivity pattern
Additional confirmatory studies
Alveolar soft part sarcoma
TFE3 (nuclear) positive +/-: desmin, S-100 protein, SMA
FISH for translocation
Atypical lipomatous tumor (ALT)/Well differentiated liposarcoma
CDK4 and/or MDM2 (both nuclear) positive
p16: nuclear +/- cytoplasmic positive (must be diffusely positive to be diagnostically useful)
FISH for CDK4/MDM2 amplification
Brachyury, keratins positive +/-: S-100 protein
Imaging may be very helpful
Clear cell sarcoma
S-100, HMB45, Melan-A one or more positive
FISH for translocation
Dermatofibrosarcoma protuberans
CD34 diffusely positive except in areas of fibrosarcomatous transformation
FISH for translocation
Desmoid type fibromatosis
Aberrant nuclear β-catenin expression (70-80%)
Mutation analysis not commonly performed
Desmoplastic small round cell tumor
Odd tetrad reactivity: WT1, NSE, keratin/EMA, dot-like desmin
FISH for translocation
Epithelioid sarcoma
INI-1 loss of expression, EGFR expression, keratins, CD34 (50%)
Gastrointestinal stroma tumor
c-kit (95%), DOG-1 (95%), CD34 (60%), +/−: smooth muscle actin
PCR for c-kit, PDGFRa mutations
Actin and desmin both positive in most cases, caldesmon also commonly positive (~ 70%) and most specific for smooth muscle differentiation
No specific genetic findings
Low grade fibromyxoid sarcoma
MUC4 positive
FISH for translocation
Myogenin, desmin positive
Rarely smooth muscle actin
FISH for translocations in alveolar RMS almost always required by protocol
Synovial sarcoma
EMA, keratins, TLE1 (90%)
+/-: S-100 protein, calponin
FISH for translocation almost always required
Positive = typically positive (85% or more), +/- = 10–20% positive, Rarely positive = 3–5% positive. TFE3 = Transcription factor binding to IGHM enhancer 3, WT1= Wilms tumor 1, INI-1 = Integrase interacter 1 = SMARCB1, DOG-1 = Discovered on GIST-1, MUC4 = Mucin 4, TLE1 = Transducin-like enhancer of split 1, c-kit = CD117, mast/stem cell growth factor receptor (homolog of a feline sarcoma viral oncogene: v-kit)
Table 1.5   Cell shapes, architectural patterns and stromal components in mesenchymal tumors
Low magnification example
High magnification example
Spindle cells
Stellate cells
Epithelioid cells
Small round blue cells
Tumor giant cells
Osteoclastic giant cells
Pleomorphic tumor cells
Myxoid matrix
Fibrous matrix
Bone (mineralized osteoid)
Sheets of cells
Fascicles of cells
Storiforming mats of cells
zoom view
Figure 1.7: Malignant peripheral nerve sheath tumor, pankeratin IHC stain. Sarcomas may strongly and diffusely express epithelial markers such as keratins and EMA. Such reactivity is usually focal but when strong and diffuse can be a confusing finding.
zoom view
Figure 1.8: Ewing sarcoma, high magnification, CD99 IHC stain. Characteristically, the Ewing sarcoma family of tumors expresses CD99 in a relatively specific membranous pattern of reactivity (brown chromogen).
Benign lesions that look malignant
Some benign mesenchymal lesions, by virtue of hypercellularity, high mitotic activity and/or nuclear variability, readily appear malignant. Table 1.9 lists some of these tumors.
Malignant lesions that look benign
By contrast, a few sarcomas may look deceptively benign with minimal cytologic atypia, sparse to absent mitotic activity and no necrosis. Table 1.10 lists these tumors.
zoom view
Figure 1.9: Desmoid-type fibromatosis, high magnification, β-catenin IHC stain, brown chromogen. Approximately 80% of soft tissue desmoid tumors aberrantly express nuclear β-catenin. Nuclear reactivity is a relatively specific feature and helps differentiate the tumor from reactive scar tissue. It may be focal and associated with cytoplasmic and membranous reactivity also present in this example.
zoom view
Figure 1.10: Low grade myofibrosarcoma, high magnification, smooth muscle actin (SMA) IHC stain, brown chromogen. Myofibroblasts often demonstrate a parallel subplasmalemmal pattern of SMA expression (‘tram track’ reactivity). In contrast, smooth muscle cells exhibit diffuse cytoplasmic reactivity for SMA. This is an example of the diagnostic clues provided by the cellular location of antigen expression.
Malignant tumors that look like sarcomas
Sarcoma mimickers encompass non-mesenchymal tumors, and these may be encountered with consternating frequency, especially in the head and neck. Table 1.11 lists these malignant mimics of sarcoma.
Manifold appearance of some bone and soft tissue tumors
A single specific bone or soft tissue tumor may present diverse histologic appearances.12
Table 1.6   Genetic assays used in the clinical assessment of bone and soft tissue tumors
Limit of resolution
Traditional cytogenetics
Chromosomes examined in metaphase
Fresh cells or tissue, not fixed or frozen
Detection of large scale chromosomal duplication, deletions, translocations
5–10 megabases
One molecule of DNA (one chromosome)
Fluorescent in situ hybridization cytogenetics
Examination of specific chromosome regions with fluorescent DNA probes
Fresh or fixed cells or tissue
Detection of finer scale chromosomal duplication, deletions, translocations
100,000 nucleotides
One molecule of DNA (one chromosomal region)
Reverse transcriptase PCR
Polymerase chain reaction amplification of nucleic acids
Fresh cells or tissues
Detection and quantitation of chimeric mRNAs
Several hundred nucleotide target
A few molecules of target mRNA
High resolution melting assays
PCR with controlled melting to detect sequence variation
Fresh or fixed cells and tissue
Detect point mutations in DNA
Single nucleotide mutations
Few percent cells containing mutation in mixture of benign cells
DNA sequencing
Multiple methods for determining DNA sequence of a genomic region
Fresh or fixed cells and tissue
Determine DNA sequence of a region of the genome or the entire genome
Single nucleotide variation
Can begin with one, a few, or many cells
Table 1.7   Representative genetic findings in bone tumors or lesions
Genetic finding
Aneurysmal bone cyst
Benign to intermediate
USP6 (17p13) gene translocations:
t(16;17) = CDH11-USP6
t(1;17) = THRAP3-USP6
t(3;17) = CNBP-USP6
t(9;17) = OMD-USP6
t(17;17) = COL1A1-USP6
Dx but not commonly needed
Intermediate (rarely malignant)
Heterogeneous rearrangements
Not helpful
Chondromyxoid fibroma
6q12–15,6p23–25, 6q23–27 rearrangements
Rare other translocations
Dx but not commonly needed
T (brachyury homolog) and EGFR copy number gains
Dx but not commonly used
Clear cell chondrosarcoma
May have extra copies of 20 or 9p loss or rearrangement
Retinoblastoma pathway commonly altered.
Not helpful
Normal karyotype usually
Not helpful
Conventional chondrosarcoma, low grade
More complex karyotype than most enchondromas
May be helpful
Conventional chondrosarcoma, high grade
Complex aneuploidy
Not helpful
Ewing sarcoma family of tumors
t(11;22)(q24;q12) = EWSR1-FLI1
t(21;22)(q22;q12) = EWSR1-ERG
t(7;22)(p22;q12) = EWSR1-ETV1
t(17;22)(q21;q12) = EWSR1-ETV4
t(2;22)(q33;q12) = EWSR1-FEV
t(16;21)(p11;q22) = FUS-ERG
t(2;16) = FUS-FEV
inv (22) = EWSR1-ZSG
t(2;22)(q31;q12) = EWSR1-SP3
t(20;22)(q13;q12) = EWSR1- NFATC2
t(4;22)(q31;q12) = EWSR1- SMARCA5
Dx (90% are EWSR1FLI1)
A few of the rare translocations have only been described in soft tissue cases
Fibrous dysplasia
GNAS1 gene mutations
Dx but not commonly needed
Giant cell tumor of bone
Telomeric associations
Secondary findings
Dx but not commonly needed
Mesenchymal chondrosarcoma
t(8;8)(q21;q13) = HEY1-NCOA2
Osteosarcoma, low grade
Intermediate to malignant
Normal karyotype
Not helpful
Osteosarcoma, high grade
Complex aneuploidy
Not helpful
Dx = diagnosis. Benign = does not recur (or only rarely), Intermediate = locally aggressive or recurrent or rarely metastasizes (<5%), Malignant = commonly metastasizes and often locally aggressive and recurrent.
Table 1.8   Representative genetic findings in soft tissue tumors
Genetic finding
Deep (aggressive) angiomyxoma
12q15 rearrangements = HMGA2
Dx but not commonly needed
Alveolar soft part sarcoma
t(X;17)(p11;q25) = ASPSCR1(ASPL)-TFE3
Angiomatoid fibrous histiocytoma
t(12;16)(q13;p11) = FUS/TLS-ATF1
t(12;22)(q13;q12) = EWSR1-ATF1
t(2;22)(q33;q12) = EWSR1-CREB1
Complex aneuploidy
Not useful
Clear cell sarcoma
t(12;22)(q13;q12) = EWSR1-ATF1
t(2;22)(q33;q12) = EWSR1-CREB1
Dermatofibrosarcoma protuberans
Ring chromosome composed of 17q21 and 22q10-q31 sequences t(17;22)(q22;q13) = COL1A-PDGFβ
Dx but not commonly needed
Desmoid-type fibromatosis
Benign to intermediate
CTNNB1 mutations Trisomy 8, 20
5q21–22 loss or mutations = APC
Dx but not commonly needed
Desmoplastic fibroblastoma
t(2;11)(q31;q12) = ?
t(11;17)(q12;p11.2) = ?
Dx but not commonly needed
Desmoplastic small round cell tumor
t(11;22)(p13;q12) = EWSR1-WT1
t(21;22)(q22;q12) = EWSR1-ERG
Epithelioid hemangioendothelioma (EHE)
Intermediate to malignant
t(1;3)(p36;q25) = WWTR1-CAMTA1
Epithelioid sarcoma
Complex changes by CGH
22q11 abnormalities = INI1 (SMARCB1)
Not useful
Extraskeletal myxoid chondrosarcoma
t(9;22)(q22;q12) = EWSR1-NR4A3
t(9;17)(q22;q11) = TAF2N-NR4A3
t(9;15)(q22;q21) = TCF12-NR4A3
t(9;22)(q22;q15) = TFG-NR4A3
Fibroma of tendon sheath
t(2;11)(q31-32;q12) = ?
Dx but not commonly needed
c-kit gene mutation (majority)
PDGFRa gene mutation (minority)
Dx but not commonly needed
11q13–21 rearrangements
Dx but not commonly needed
Inflammatory myofibroblastic tumor (IMFT)
ALK1 gene fusions (2p23 translocations): t(1;2)(q25;q23) = TPM3-ALk
t(2;19)(q23;q13) = TPM4-ALK
t(2;17)(q23;q23) = CLTC-ALK
t(2;2)(p23;q13) = RANBP2-ALK
Dx but not commonly needed
Complex aneuploidy
Not useful
Benign to intermediate
8q11–13 rearrangements = PLAG1
Dx but not commonly needed
Lipoma, conventional
12q15 rearrangements = HMGA2
6p21 rearrangements = HMGA1
Dx but not needed
> Myxoid/round cell LPS
Myxoid/RC LPS: t(12;16)(q13;p11) = FUS-DDIT3
t(12;22)(q13;q12) = EWSR1-DDIT3
Dx but not commonly needed
Low grade fibromyxoid sarcoma
t(7;16)(q34;p11) = FUS-CREB3L2
t(11;16)(p11;p11) = FUS-CREB3L1
Malignant peripheral nerve sheath tumor (MPNST)
Complex aneuploidy
22q11.2 abnormalities = INI1 (SMARCB1) in some epithelioid MPNST
Not useful
Myoepithelial carcinoma of soft tissue
t(1;22)(q23;q12) = EWSR1-PBX1
t(19;22)(q13;q12) = EWSR1-ZNF444
t(6;22)(p21;q12) = EWSR1-POU5F1
16p11.2 rearrangement = FUS–?
Complex aneuploidy
Not useful
Myxoinflammatory fibroblastic sarcoma
t(1;10)(p22;q24) = FGF8 and NPM3
Monoallelic or biallelic loss of 17q11.2 = NF1 (neurofibromin)
Not useful
t(7;12)(p22;q13) = ACTB-GLI1
Dx but not commonly needed
10q22–24 deletions rearrangements not ALK1 (sclerosing perineurioma)
22q11.2 alterations in some cases of both types
Dx but not commonly needed
Rhabdoid tumor of ST
Del 22q11.22 = del SMARCB1
A-RMS: t(2;13)(q35;q14) = PAX3-FOX01A
t(1;13)(q36;q14) = PAX7- FOXO1A
t(2;X)(p35;q13) = PAX3-MLLT7
t(2;2)(q35;q23) = PAX3-NCOAI
E-RMS, adult sclerosing/spindle RMS, pleomorphic RMS: complex aneuploidy
Dx (for A-RMS)
22q12 loss = NF2 (Merlin, Schwannomin)
Not useful
Sclerosing epithelioid fibrosarcoma
t(7;16)(q34;p11) = FUS-CREB3L2
t(11;16)(p11;p11) = FUS-CREB3L1
Soft tissue angiofibroma
t(5;8)(p15;q13) = AHRR-NCOA2
Solitary fibrous tumor
NAB2-STAT6 fusion (both genes on 12q13) 12q13–15 rearrangements
May be useful in Dx
Spindle cell lipoma/pleomorphic lipoma
16q13-qter rearrangements
monosomy 13 partial del(13q)
Dx but not commonly needed
Synovial sarcoma
t(X;18)(p11;q11) = SS18-SSX1
t(X;18)(p11;q11) = SS18-SSX2
t(X;18)(p11;q13) = SS18-SSX4
t(X; 20)(p11;q13) = SS18L1- SSX1
Tenosynovial giant cell tumor, diffuse or localized
Benign to intermediate
t(1;2)(p13;q37) = CSF1-COL6A3
Trisomy 5,7
Dx but not commonly needed
Dx = diagnosis, Benign = a tumor that will not recur (or only rarely) and does not metastasize, Intermediate = a tumor that can be locally aggressive and/or recur and/or has a low risk (</= 5%) of metastasis, Malignant = a tumor that has a high risk for metastasis and may be locally aggressive also. This list of genetic abnormalities is necessarily incomplete because new discoveries are ongoing.
Table 1.9   Some benign mesenchymal lesions that appear histologically malignant
Example at low magnification
Example at high magnification
Nodular fasciitis
Myositis ossificans
Pleomorphic hyalinizing angiectatic tumor (PHAT)
Cellular schwannoma
Unstable fracture
Table 1.10   Some malignant mesenchymal tumors that appear histologically benign
Example at low magnification
Example at high magnification
Clear cell sarcoma, sclerosing type
Epithelioid sarcoma
Low grade myofibroblastic sarcoma
Low grade fibromyxoid sarcoma
Parosteal osteosarcoma
Table 1.11   Nonmesenchymal tumors that may histologically mimic sarcoma
Helpful clinical scenario
Useful IHC or molecular tests
Sarcomatoid carcinoma mimicking a spindle cell sarcoma
Head and neck tumor in older patient
Keratins and p63 are positive in sarcomatoid squamous carcinoma
Sarcomatoid breast carcinoma mimicking angiosarcoma
Breast location or history of breast carcinoma
Keratins, MOC31, ER, PR (unlikely to be reactive, however). Lack of reactivity for endothelial markers such as CD34, CD31, ERG (C terminus directed)
Melanoma which can mimic a small round blue cell sarcoma, a sarcoma with epithelioid cells, a spindle cell sarcoma or pleomorphic sarcoma
History of melanoma, even if remote
Melanocytic markers including S-100 protein, HMB45, Melan-A, MiTF, BRAF mutation analysis, c-kit mutation analysis
Follicular dendritic cell sarcoma mimicking a spindle cell sarcoma
May arise in a lymph node or in soft tissue
Clusterin, fascin, CD21, CD23, CD35
True histiocytic sarcoma mimicking a spindle cell or polygonal cell sarcoma
This can easily present as a soft tissue neoplasm, but it is very rare
CD163, CD68, lysozyme positive in tumor cells
This increases the complexity of the diagnosis of bone and soft tissue tumors which otherwise might seem straightforward from the relatively few number of actual entities (WHO 2013 lists 112 soft tissue tumors and 51 bone tumors). Table 1.12 lists some of the mesenchymal tumors that can exhibit a diverse range of histologic appearances.
Some genetic findings are not tumor specific
Mesenchymal tumors, including sarcomas, may be normal diploid in chromosome complement. Finding a normal karyotype can be helpful when a tumor with a known specific translocation is in the diagnostic differential. However, it is important to establish the level of confidence in a diploid karyotype, as it may represent overgrowth of normal cells and thus an artifact of the assay.
Approximately half of sarcomas exhibit a complex aneuploid karyotype without distinguishing, consistently recurrent abnormalities. Another, approximately 50% of sarcomas demonstrate a specific finding, typically a balanced translocation, on genetic analysis. This can be immensely helpful diagnostically and may prove increasingly useful in targeted therapy development. Some translocation specific sarcoma, as this group is termed, also may show secondary genetic abnormalities, and some of these may affect prognosis.
Finally, a few tumors that are discordant histologically and in behavior contain identical translocations (this has been termed ‘tumor infidelity for fusion transcripts’)(Romeo et al. 2010). One paired example is angiomatoid fibrous histiocytoma and clear cell sarcoma which share two translocations: t(12;22) = EWS-ATF1 and t(2;22) = EWS-CREB1. These two tumors look different microscopically, show a different antibody staining profile, and one is usually benign, while the other is malignant. Thus, clinical, radiologic, histologic, antigenic, and even genetic findings must be evaluated in context and in toto to guarantee a conclusive, accurate diagnosis.
Cytogenetic analysis may be unsuccessful
Despite rapidly triaging fresh tumor in a sterile fashion and submitting it quickly to the cytogenetics laboratory for processing, up to 50% of mesenchymal tumors do not subsequently grow in culture, and therefore karyotyping fails. Fortunately, for some specific tumors, an alternate path may be taken to uncover the genetic profile of the tumor. Formalin-fixed, paraffin embedded blocks or touch preps may be used for FISH examination for translocations and other aberrations such as gene amplification. The polymerase chain reaction (PCR) can reveal specific mutational changes. Direct sequencing can provide voluminous information but also can be ‘directed’ to specific genetic sites. The costs of these tests vary, the availability may be limited, and for FISH and PCR, a specific pre-test probability drives the choice of probes: these are not ‘fishing expeditions’. Despite its expense and test complexity, genetic evaluation is becoming increasing important not just diagnostically, but it can also predict behavior and response to therapy.
Still in many cases, it is not yet essential for patient care.
Mesenchymal sarcomas are not always specifically diagnosable
Even an expert soft tissue or bone pathologist will be unable to classify ~ 5–10% of cases in terms of rendering a specific tumor diagnosis. This impairs prognostication and therapy selection. An example of an unclassifiable tumor is provided in Figure 1.11. With continuing refinement of the molecular classification of mesenchymal tumors, the percentage of unclassifiable lesions should fall.
Consulting a bone or soft tissue pathologist in difficult cases is advantageous, as these individuals not only have personal depth of experience but can also access the most advanced diagnostic techniques. Most importantly, the patient's treatment may be optimized as a consequence of subspecialist pathologist review; at a minimum, the patient will be counseled regarding the likely disease course.
Compared with lung, prostate, breast, and colorectal carcinoma, sarcomas afflict only a few. Bone sarcomas are 5–6 fold less common than soft tissue sarcomas.18
Table 1.12   Some mesenchymal tumors that have a diverse array of histologic appearances
Possible histologic appearances
Example 1
Example 2
Low grade myxoid hypocellular tumor ranging to high grade sarcomatous areas with sheets of pleomorphic tumor cells, epithelioid areas, osteoclastic giant cell rich areas
Dedifferentiated liposarcoma
May have well-differentiated areas, metaplastic malignant cartilage, bone, muscle, obscuring inflammation
Synovial sarcoma
Fascicles of crowded plump spindle cells that overlap to hypocellular collagenized calcified areas to areas with true epithelial gland formation to focal myxoid areas
Malignant PEComa
High grade perivascular epithelioid tumor cells to fascicles of lower grade spindle cells
Fascicles of spindled eosinophilic cells with scattered markedly atypical cells to myxoid tumors to epithelioid tumors, to heavily inflamed tumors to tumors with many osteoclastic giant cells
Malignant peripheral nerve sheath MPNST
Fascicles and herringbones of overlapping spindled tumor cells (mimicking classic spindle cell SS) to epithelioid tumors, to low grade hypocellular myxoid tumors to tumors with glands or skeletal muscle differentiation
All of the tumors in this table can exhibit multiple histologic appearances, not just the two patterns shown.
zoom view
Figure 1.11: Sarcoma with epithelioid features unable to further classify, medium magnification. This is an example of a tumor that could not be specifically subclassified. It was reviewed by an expert sarcoma pathologist who employed IHC and FISH to attempt to better define it. The consultant concluded that it was difficult to predict its behavior but would likely at least locally recur. The patient, a woman in her 20s, rapidly developed multiple bone metastases and died of disease within 8 months of presentation despite complete resection and intensive chemotherapy. This case indicates there is still work to be done in mesenchymal tumor classification and prognostication.
A recent review of cases occurring in the Rhone-Alpes region of 6 million inhabitants noted age standardized incidence rates of soft tissue, visceral and bone sarcomas of 3.6, 2.0, and 0.6 per 100,000/year with GIST (18%), unclassified sarcoma (16%), liposarcoma (15%), and leiomyosarcoma (11%) as the most common diagnoses (Ducimetiere et al. 2011). Surveillance Epidemiology and End Results (SEER) U.S. data not subject to central pathology review recorded leiomyosarcoma (many in the uterus), ‘malignant fibrous histiocytoma’, liposarcoma, and dermatofibrosarcoma protuberans as the most common soft tissue and visceral sarcomas (n = 26,758 cases, years 1978–2001) (Toro et al. 2006). For soft tissue sarcomas, international age adjusted incidence rates range from 1.8–5/year (Wibmer et al. 2010). There is some controversy over whether the incidence is increasing, with some studies reporting, and others refuting, such a trend. Median overall survival of metastatic sarcoma improved by 50% over 1987–2006 according to a French Sarcoma Group study (Italiano et al. 2011).
Figure 1.12 depicts in a histogram format the distribution of major soft tissue sarcomas by primary site according to SEER cases abstracted at 13 SEER sites from 1994–2008. Figure 1.13 illustrates the same sarcomas according to stage at presentation. Consistent with expectations, most liposarcomas present as localized lesions, while alveolar soft part sarcoma more commonly is high stage at presentation. There are no clear gender differences in the distribution of the major sarcomas (Fig. 1.14). Median age and age ranges are provided in Table 1.13 (again, SEER data for adults over 20 years from 1994–2008, 11 major sarcoma types). Five year relative survival rates are lowest overall for rhabdomyosarcoma and highest overall for liposarcoma (Fig. 1.15). In general, younger adults survive longer than older adults though the differences may be minor for some types. Tumors categorized as fibrosarcoma would likely benefit from central pathology review as diagnostic criteria and technological advances may permit reclassification of most tumors in this category.
zoom view
Figure 1.12: Distribution of primary site by soft tissue tumor type for adults aged 20 years and older in 13 SEER registries diagnosed between 1994 and 2008.
zoom view
Figure 1.13: Distribution of stage by soft tissue tumor type for adults patients aged 20 years and older in 13 SEER registries diagnosed between 1994 and 2008.
The Surveillance, Epidemiology and End Results (SEER) Program is the data source for the epidemiology sections of this book. SEER was established by the National Cancer Institute in 1973, and it is the primary source for comprehensive, population-based cancer statistics in the US. Currently, the entire program covers 28% of the US population.
The data presented in this book are from cases diagnosed between 1994 and 2008 within 13 SEER Program registries (statewide registries: Connecticut, Hawaii, Iowa, New Mexico, Utah; metropolitan areas: Atlanta, Detroit, Los Angeles, San Francisco-Oakland, San Jose-Monterey, Seattle-Puget Sound; rural areas of Georgia; Alaska Native populations residing in Alaska).20
zoom view
Figure 1.14: Distribution of sex by soft tissue tumor type for adults patients ages 20 years and older in 13 SEER registries diagnosed between 1994 and 2008.
zoom view
Figure 1.15: Five year relative survival rate of soft tissue tumor types by age category for cases diagnosed in 13 SEER registries diagnosed between 1994 and 2008. [*MPNST = malignant peripheral nerve sheath tumor. ** Relative survival could not be calculated for osseous/chondromatous tumors (50–59 years) and alveolar soft part sarcoma tumors (60–69 years and 70+ years)]
Table 1.13   Characterization of age by soft tissue tumor type for cases diagnosed in 13 SEER registries diagnosed between 1994 and 2008
Tumor type
Total number of cases
Percent pediatric cases**
Median age (years)
Age range (years)
Ewing family tumors
Alveolar soft part sarcoma
* MPNST = Malignant peripheral nerve sheath tumors
** Pediatric cases are ages 0 – 19 years
The analyses were restricted to malignant tumors and the histology codes are from the International Classification of Diseases for Oncology – Third Edition (ICDO-3). The ICDO-3 coding system was used for the site of the tumor at diagnosis as well. Codes C40.0-C41.9 were used for tumors of the bone (bones, joints and articular cartilage), and C49.0-C49.9 for tumors of the soft tissue (connective, subcutaneous and other soft tissues).
All SEER incidence rates in the upcoming chapters are per million and are age-adjusted to the 2000 US standard population. SEER obtains the population estimates to calculate these rates from the US Census Bureau, and the mortality data are provided by the National Center for Health Statistics.
The next 10 years should yield major advances in cellular pathobiology that will directly impact the care of patients with bone and soft tissue tumors. Because the human genome has been entirely sequenced, and the ‘omes’ (proteome, metabolome, epigenome and others) are steadily being explored, increasing numbers of specific, cell pathway-based aberrations will be uncovered in mesenchymal tumors. Molecular 21genetic approaches, bioinformatics analyses and mathematical tumor modeling will be required to make and understand these discoveries. Eventually, best practice may entail successively biopsying a patient's sarcoma over time to progressively assess tumor evolution and tumor response to therapy.
In 25 years, with enhanced radiologic and serologic tumor detection, many patients will present with smaller lesions amenable to refined surgical excision. Most sarcomas may then be contained, if not cured, by a combination of surgery and targeted, flexible therapy that shifts as the tumor shifts genetically and molecularly.
Medical students interested in pathology as a career should discuss the field with a pathologist. Shadowing a pathologist would best reveal the vital role this physician plays in patient care. Pathology needs dedicated, inquisitive creative individuals.
Pathology residents particularly fascinated by bone and soft tissue pathology would do well to take an active stance in learning this field by studying as many individual cases as possible including case presentations presented on national pathology websites and actively participating in case sign outs and tumor boards. Few fellowships exist in bone and soft tissue pathology. However, anatomic pathology is edging ever more into the submicroscopic realm, and a combination of a strong traditional surgical pathology background with expertise in genetic molecular analysis of tumors is an optimal preparation for addressing these tumors in one's professional career.
The practice of pathology is not an independent one. The anatomic pathologist depends on the clinical members of the patient care team including the radiologist, medical oncologist, surgeon, and radiation-oncologist. In return, the musculoskeletal pathologist, by typing the tumor, fundamentally determines the patient's therapy. Close collaboration with clinical colleagues impels improved diagnosis and opens opportunities for advancing the field. This collegiality catalyzes the pathologist and champions the patient.
  1. Barretina J, Taylor BS, Banerji S, et al. Subtype specific genome alterations define new targets for soft-tissue sarcoma therapy. Nat Genetics 2010;42:715–723.
  1. Bridge JA. Advances and limitations of cytogenetic, molecular cytogenetic and molecular diagnostic testing in mesenchymal neoplasms. J Orthop Science 2008;13:273–282.
  1. Brooks, SJ. Disorders of soft tissue. In: Mills, SE. Sternberg's Diagnostic Surgical Pathology, 5th ed. Wolters Kluwer/Lippincott Williams and Wilkins:  Baltimore:  2010.
  1. Ducimetiere F, Lurkin A, Ranchere-Vince D, et al. Incidence of sarcoma histotypes and molecular subtypes in a prospective epidemiological study with central pathology review and molecular testing. PLOS One 2011;6:e20294. 1–14.
  1. Fisher C. Soft tissue sarcomas with non-EWS translocations: molecular genetic features and pathologic and clinical correlations. Virch Arch 2010;456:153–166.
  1. Gibault L, Perot G, Chibon F, et al. New insights in sarcoma oncogenesis: a comprehensive analysis of a large series of 160 soft tissue sarcomas with complex genomics. J Pathol 2011;223:64–71.
  1. Guillou L, Aurias A. Soft tissue sarcomas with complex genomic profiles. Virch Arch 2010;456:201–217.
  1. Italiano A, Mathoulin-Pelissier S, Le Cesne A, et al. Trends in survival for patients with metastatic soft tissue sarcoma. Cancer 2011;117:1049–1054.
  1. Jin Y, Moller E, Nord KH, et al. Fusion of the AHRR and NCOA2 genes through a recurrent translocation t(5;8)(p15;q13) in soft tissue angiofibroma results in upregulation of aryl hydrocarbon receptor target genes. Genes Chromosomes Cancer 2012;51:510–520.
  1. Ordonez JL, Osuna D, Garcia-Dominguez DJ, et al. The clinical relevance of molecular genetics in soft tissue sarcomas. Adv Anat Pathol 2010;17:162–181.
  1. Romeo S, Duim RA, Bridge JA, et al. Heterogeneous and complex rearrangements of chromosome arm 6q in chondromyxoid fibroma: delineation of breakpoints and analysis of candidate genes. Am J Pathol 2010;177:1365–76.
  1. Sjogren H, Orndal C, Tingby O, et al. Cytogenetic and spectral karyotype analysis of benign and malignant cartilage tumors. Int J Oncol 2004;24:1385–91.
  1. Toro JR, Travis LB, Wu HJ, et al. Incidence patterns of soft tissue sarcomas, regardless of primary site, in the Surveillance, Epidemiology and End Results program, 1978–2001: an analysis of 26,758 cases. Int J Cancer 2006;119:2922–2930.
  1. Toro JR, Travis LB, Wu HJ, et al. Incidence patterns of soft tissue sarcomas, regardless of primary site, in the Surveillance, Epidemiology, and End Results program, 1978–2001: an analysis of 26,758 cases. Int J Cancer 2006;119:2922–2930.
  1. Verweij J, Baker LH. Future treatment of soft tissue sarcomas will be driven by histological subtype and molecular alterations. Eur J Cancer 2010;46:863–868.
  1. Wang L, Motoi T, Khanin R, et al. Identification of a novel, recurrent HEY1-NCOA2 fusion in mesenchymal chondrosarcoma based on a genome-wide screen of exon-level expression data. Genes Chromosomes Cancer 2012;51:127–139.
  1. Wibmer C, Leithner A, Zielonke N, et al. Increasing incidence rates of soft tissue sarcomas? A population-based epidemiologic study and literature review. Ann Oncol 2010;21(5):1106–11.
  1. Yasuda T, Nishio J, Sumegi J, et al. Aberrations of 6q13 mapped to the COL12A1 locus in chondromyxoid fibroma. Mod Pathol 2009;22:1499–506.
  1. Fletcher CDM, Bridge JA, Hogendoorn P, Mertens F (eds). World Health Organization Classification of Tumours of Soft Tissue and Bone, 4th Ed. IARC Press,  Lyon:  2013.