Although “rhabdomyosarcoma” was once regarded as a single disease, it has become abundantly clear over the past several decades that they are in fact a family of genetically distinct sarcomas, sharing skeletal muscle differentiation, but differing from each other in important ways. This chapter will first cover general aspects of rhabdomyosarcomas, and then discuss specific rhabdomyosarcoma subtypes.

Arthur Purdy Stout was the first to delineate rhabdomyosarcoma as a distinct entity, and Horn and Enterline devised the first rhabdomyosarcoma classification scheme in 1958. During the 1930s and 1940s, a very large percentage of pleomorphic sarcomas in adults were considered to represent pleomorphic rhabdomyosarcoma, and most of the rhabdomyosarcomas reported during this period were of this type. ^, These tumors occurred mainly in the muscles of the lower extremity and affected older patients. They displayed a striking degree of cellular pleomorphism, but cells with cross-striations were typically absent. It subsequently became apparent that virtually all of these tumors were other types of pleomorphic sarcoma, including what we now term undifferentiated pleomorphic sarcoma (UPS).

It also became evident that many childhood sarcomas formerly diagnosed descriptively as “round cell” or “spindle cell” sarcomas were rhabdomyosarcomas of alveolar or embryonal type. Knowledge of these tumors was fostered by the introduction of newer, more effective therapies. Before 1960, childhood rhabdomyosarcoma was an almost uniformly fatal neoplasm that recurred and metastasized in a high percentage of cases. During the last several decades, however, it has been shown that this tumor responds to multimodality therapy—encompassing biopsy or conservative surgery, multiagent chemotherapy, and radiotherapy—and that many children treated by these modalities remain free of recurrent and metastatic disease. The numerous reports of the Intergroup Rhabdomyosarcoma Study (IRS) (now recognized as the Soft Tissue Sarcoma Committee of the Children’s Oncology Group) have contributed greatly to our understanding of childhood rhabdomyosarcomas, especially the effect of the various treatment modalities on the survival of patients with this tumor.

The most recent WHO classification recognizes four major rhabdomyosarcoma subtypes: embryonal , alveolar , pleomorphic , and spindle cell/sclerosing. Spindle cell/sclerosing rhabdomyosarcoma is further divided based on specific genetic events, including MYOD1 -mutant, infantile V GLL2/NCOA2/CITED2 -fused, and fusion-driven tumors ( Box 20.1 ). It is anticipated that the classification of rhabdomyosarcoma will only continue to become more complex, as new, clinically relevant, genetically distinct subtypes are identified. Going forward, molecular genetic features, in particular fusion status, will undoubtedly play a much larger role in risk stratification and therapeutic strategies. ^,

As with other sarcomas, evidence is lacking to suggest that rhabdomyosarcoma actually arises from skeletal muscle cells. In fact, these tumors often arise at sites where striated muscle tissue is normally absent (e.g., common bile duct, urinary bladder), or scant (e.g., nasal cavity, middle ear, vagina). With the notable exception of rhabdomyosarcomas arising in inflammatory rhabdomyoblastic tumor (a skeletal muscle tumor of borderline malignancy), rhabdomyosarcomas are not known to arise from a precursor lesion.

Little is known about the underlying cause of the rhabdomyoblastic proliferations and the stimulus that induces their growth. Genetic factors are implicated by the rare occurrence of the disease in siblings, the occasional presence of the tumor at birth, and the association of the disease with other neoplasms in the same patient. Rhabdomyosarcoma has been described in conjunction with congenital retinoblastoma, familial adenomatous polyposis, multiple lentigines syndrome, type 1 neurofibromatosis, Costello syndrome, Noonan syndrome, and Beckwith–Wiedemann syndrome, among a host of others. Germline mutations in the DICER1 gene predispose affected patients to a broad range of tumors (DICER1 syndrome), including embryonal rhabdomyosarcoma. A 2009 report from the Children’s Oncology Group found an association between first-trimester x-ray exposure and embryonal rhabdomyosarcoma.

Rhabdomyosarcoma is not only the most common soft tissue sarcoma in children under 15 years of age, but also one of the most common soft tissue sarcomas of adolescents and young adults. Rhabdomyosarcoma accounts for an estimated 4.5% of all childhood cancers, with an annual incidence of six cases per 1 million per year. It is rare in persons older than 45 and accounts for an estimated 2%–5% of all adult sarcomas, but it is probably lower than that. There is a bimodal distribution for age at presentation, with the first peak occurring between 2 and 6 years and a second peak between 10 and 18 years. This reflects the peak incidence of embryonal and alveolar rhabdomyosarcomas, respectively.

Each of the rhabdomyosarcoma subtypes occurs in a characteristic age group. For example, embryonal rhabdomyosarcomas affect mainly, but not exclusively, children between birth and 15 years of age. On the other hand, alveolar rhabdomyosarcoma tends to affect older patients, with peak ages of 10–25 years. Rhabdomyosarcomas are uncommon in patients older than 40, with the exception of fusion-driven spindle cell tumors of bone (median patient age 41 years) and soft tissue (median patient age 50 years). MYOD1 -mutant spindle cell rhabdomyosarcomas occur roughly equally in the pediatric and adult populations. There is some correlation between tumor location and age; for example, rhabdomyosarcomas of the urinary bladder, prostate, vagina, and middle ear tend to occur at a younger age (median: 4 years) than those in the paratesticular region or the extremities (median: 14 years for both).

Males are affected more often than females by approximately 1.5:1.0, but the male predominance is less pronounced during adolescence and young adulthood and for rhabdomyosarcomas of the alveolar type. Fusion-driven and MYOD1 -mutant spindle cell rhabdomyosarcomas are somewhat more common in women. ^,

Although rhabdomyosarcomas may arise anywhere in the body, they occur predominantly in three regions: the head and neck, genitourinary tract and retroperitoneum, and upper and lower extremities. Each rhabdomyosarcoma histologic subtype may occur in virtually any location, but each subtype has a site predilection, as discussed in the specific sections.

The head and neck area is the principal location of rhabdomyosarcoma; 970 (26%) of 3717 tumors from IRS-I, IRS-II, and IRS-III occurred in this location ( Table 20.1 ). In the head and neck, parameningeal tumors are the most common. Parameningeal rhabdomyosarcomas should be distinguished from other rhabdomyosarcomas arising in the head and neck because of their potential for intracranial extension and seeding, and therefore less favorable clinical course. For purposes of risk-stratification, parameningeal rhabdomyosarcomas are considered TNM stage 2, whereas tumors occurring in the orbit and other head/neck locations are stage 1.

The orbit is the second most common head and neck site of rhabdomyosarcoma, accounting for 7% of cases from the IRS series. Most rhabdomyosarcomas in this location are of the embryonal subtype. ^, For example, 221 (90%) of 245 orbital tumors from IRS-I through IRS-IV were of the embryonal subtype, although rare botryoid-type embryonal rhabdomyosarcomas and alveolar rhabdomyosarcomas also arise in the orbit. Rhabdomyosarcoma may also involve other head and neck sites, including the nasal cavity and nasopharynx, followed in frequency by the ear and ear canal, paranasal sinuses, soft tissues of the face and neck, and oral cavity (including the tongue, lip, and palate). ^,

After the head and neck, the genitourinary tract is the second most common site for rhabdomyosarcoma. In the IRS series, 650 (17%) of 3717 cases arose in this general region. Histologically, most tumors arising in this location are of the embryonal subtype. The most common location is the paratesticular region, often of the embryonal or spindle cell/sclerosing subtypes. They may also involve the spermatic cord and epididymis, but usually are separate from the testis proper.

The retroperitoneum and pelvis are other sites of involvement. Approximately 45% of tumors in these sites are of the embryonal subtype, but up to 15% are alveolar rhabdomyosarcomas. ^, In general, effective therapy of rhabdomyosarcomas in the retroperitoneum and pelvic region is more difficult than that of paratesticular rhabdomyosarcomas.

Approximately 5% of rhabdomyosarcomas arise in the urinary bladder or prostate. In fact, rhabdomyosarcoma is the most common bladder tumor in children under 10 years of age. Almost all pediatric tumors arising in this location are embryonal or botryoid rhabdomyosarcomas. ^, Those with a botryoid histology typically grow into the lumen of the urinary bladder as a grapelike, richly mucoid, multinodular or polypoid mass, with a broad base that can cause an obstruction of the internal urethral orifice and prostatic urethra. This, in turn, results in incontinence and difficulty with urination. Interestingly, however, adult rhabdomyosarcomas of the urinary bladder are more often of the alveolar type, which can cause morphologic confusion with small cell carcinoma. Rarely, rhabdomyosarcomas arise in other genitourinary or gynecologic sites, including the fallopian tube, uterus, cervix, vagina, labium and vulva, and perineum and perianal region. Tumors in these locations are often (but not always) of the botryoid subtype. Rhabdomyosarcomas that arise in gynecologic organs in adults are morphologically similar to those in pediatric patients, but they seem to behave more aggressively. ^, A significant subset of embryonal rhabdomyosarcomas occurring in the uterine cervix or corpus are of DICER1-mutant type.

Unlike adult soft tissue sarcomas, rhabdomyosarcomas involve the extremities much less often. Only 14.6% of cases from the Armed Forces Institute of Pathology (AFIP) series occurred in this location, with a similar incidence in the upper and lower extremities; alveolar rhabdomyosarcomas outnumbered embryonal rhabdomyosarcomas by 4:3, similar to IRS-I and IRS-II. Most pleomorphic rhabdomyosarcomas arise in the deep soft tissues of the extremities of adults.

Unusual rhabdomyosarcomas arise outside the aforementioned sites. Tumors originating in the hepatobiliary tract usually arise from the submucosa of the common bile duct; most are botryoid type with typical myxoid, grapelike gross and microscopic appearances.

Fusion-driven spindle cell rhabdomyosarcomas in adults segregate into two groups. The first and larger group usually occurs in bone (an otherwise very rare site for rhabdomyosarcomas) and most often harbors FUS/EWSR1::TFCP2 fusions. ^, A smaller group of these tumors occurs in various somatic soft tissue locations and displays significant diversity in fusion type.

Rhabdomyosarcomas display few characteristic features grossly. As with other rapidly growing sarcomas, the appearance of the tumor reflects the degree of cellularity, relative amounts of collagenous or myxoid stroma, and presence and extent of secondary changes (e.g., hemorrhage, necrosis, ulceration). In general, tumors growing into body cavities, such as those in the nasopharynx and urinary bladder, are fairly well-circumscribed, multinodular, or distinctly polypoid. On cross-section, they show a glistening, gelatinous, gray–white surface, with patchy areas of hemorrhage or cyst formation. Deep-seated tumors involving or arising in the musculature are usually less well-defined and almost always infiltrate the surrounding tissues. They are firmer and rubbery, and have a mottled, gray–white to pink-tan, smooth or finely granular, often bulging surface. There are often areas of focal necrosis and cystic degeneration.

Immunohistochemistry in the Diagnosis of Rhabdomyosarcomas

Chapter XX (Immunohistochemistry) covers in depth the various markers used in the diagnosis of rhabdomyosarcomas. Briefly, desmin is the best screening marker for potential rhabdomyosarcomas, in the evaluation of malignant neoplasms with round cell, spindle cell, pleomorphic or epithelioid features ( Fig. 20.1A–B ), although tumors composed predominantly of primitive cells may be only focally positive for this antigen. ^, As discussed in Chapter XX, desmin is by no means a specific marker of myogenic tumors, however, and in the modern era definitive diagnosis of rhabdomyosarcoma requires demonstration of more specific markers, in particular myogenin, MYOD1, and PAX7. Both myogenin and MYOD1 are highly specific markers of rhabdomyosarcoma ( Fig. 20.1C–D ). As is discussed below, there are some interesting differences in myogenin and MYOD1 expression in rhabdomyosarcoma subtypes, with, for example, diffuse expression of myogenin characterizing alveolar rhabdomyosarcoma and diffuse MYOD1 expression often pointing toward MYOD1-mutated spindle cell rhabdomyosarcoma. PAX7 is another highly sensitive marker of skeletal muscle differentiation, although it is not perfectly specific, being expressed in most Ewing sarcomas as well. In general, we use antibodies to actins only when the differential diagnosis includes leiomyosarcoma, keeping in mind that smooth muscle actin expression may be found in up to 13% of rhabdomyosarcomas. There is no role for myoglobin immunohistochemistry at this time, as this is a notably insensitive marker that is often challenging to interpret. ^,

Diffuse, strong immunoreactivity for desmin in embryonal rhabdomyosarcoma ( A ) and in alveolar rhabdomyosarcoma ( B ). In contrast to embryonal rhabdomyosarcoma, which typically shows only patchy expression of myogenin ( C ), alveolar rhabdomyosarcomas are characterized by strong, uniform expression ( D ). This finding can be helpful in the distinction of the “dense” variant of embryonal rhabdomyosarcoma from alveolar rhabdomyosarcoma.

Recent expression profiling studies have established newer markers of rhabdomyosarcoma that are associated with histologic subtype and fusion status. Coexpression of AP-2β and P-cadherin is typically present in FOXO1 -rearranged alveolar rhabdomyosarcomas, whereas coexpression of epidermal growth factor receptor (EGFR) and fibrillin-2 is seen in most embryonal rhabdomyosarcomas. It has been suggested that an immunohistochemical panel consisting of myogenin, AP-2β, NOS-1, and HMGA2 may serve as immunohistochemical surrogates of fusion status in rhabdomyosarcoma, with fusion-positive alveolar rhabdomyosarcomas coexpressing myogenin, AP-2β, and NOS-1, without expression of HMGA2, and fusion-negative rhabdomyosarcomas having the opposite immunophenotype. Although not widely utilized in the United States, these panels may be of some value in settings where molecular genetic testing is not widely available.

During the past 60 years, the prognosis of rhabdomyosarcoma has improved dramatically. Before 1960, the prognosis was extremely poor, and there were few survivors even after radical, often destructive and disfiguring, surgical therapy. For example, an AFIP study in 1969 reported a 5-year mortality rate of 98%.

Since the early 1960s, there has been marked improvement in the survival rates of patients with rhabdomyosarcoma, because of a multidisciplinary therapeutic approach that consists of a biopsy or surgical removal of the neoplasm and multiagent chemotherapy with or without radiotherapy. As a rule, treatment is carried out after a biopsy or resection and careful, comprehensive assessment of tumor stage or tumor group with radiography, computed tomography (CT) scans, magnetic resonance imaging, bone scans, and if necessary, angiograms. Positron emission tomography (PET) CT has emerged as a useful test for staging rhabdomyosarcoma, providing additional information on regional lymph nodes. Recommendations for therapy chiefly depend on the stage or clinical group of the disease, and the site of the tumor following accurate microscopic diagnosis.

The IRS-II study of patients younger than 21 years with a confirmed diagnosis of rhabdomyosarcoma distinguished four clinical groups based on the amount of tumor remaining after initial surgery ( Box 20.2 ). Because this approach is influenced by the variable practices of surgeons, the IRS Committee adopted a modification of the tumor–node–metastasis (TNM) system, which relies on a pretreatment assessment of tumor extent. This system includes evaluation of the site of the primary tumor, maximum diameter of the tumor, determination of tumor invasion into adjacent structures, status of regional lymph nodes, and presence or absence of distant metastases. More recent studies of rhabdomyosarcoma rely on both the IRS clinical grouping system and the TNM stage to determine therapy. Both the IRS clinical group and TNM stage have major prognostic significance ( Tables 20.2 and 20.3 ). Low-risk patients generally have localized embryonal histology tumors. Most of these patients have resected (group I or II) tumors, as well as group III tumors arising in favorable sites (e.g., the orbit). Patients with embryonal tumors that are group III, stage 2 or 3, and all patients with nonmetastatic alveolar tumors are intermediate risk . Patients with metastatic tumors (regardless of subtype) are treated with high-risk protocols. Based on IRS-IV data, overall survival rates were 95%, 75%, and 27% for low-risk, intermediate-risk, and high-risk patients, respectively. Similar results were reported from the European Cooperative Group studies, using a four-tier risk system. ^, As a direct result of advances in risk stratification and therapy secondary to the collaborative group clinical trials, the 5-year failure-free survival rate for low-risk patients is as high as 90%, the 4-year failure-free survival rate for intermediate-risk patients is almost 70%, but little progress has been made on improving survival for high-risk patients.

Age at diagnosis and anatomic site are also important predictors of outcome in patients with rhabdomyosarcoma. ^, Age has its greatest prognostic effect in patients with invasive but nonmetastatic tumors. With regards to site, tumors of the orbit have the best prognosis (92% 5-year survival), followed by tumors of the head and neck and non–bladder/prostate genitourinary tumors (about 80% 5-year survival). ^, A less favorable prognosis is found in patients whose tumors are located in a parameningeal location, bladder and prostate, and the extremities, with approximately 70% 5-year survival for each. The poorest prognosis occurs in patients with tumors at other sites, including the retroperitoneum, biliary tract, and peritoneum. Late detection and large tumor size, difficulties encountered during surgical removal, extension into the meninges (with or without spinal fluid spread), and lymph node metastasis are primarily responsible for the prognostic differences related to anatomic site.

As noted above, the natural histories of the different subtypes of rhabdomyosarcoma are considerably different, and these are best thought of as different tumor types sharing skeletal muscle differentiation, rather than simply morphologic variants of a single disease. Thus, while histologic classification of rhabdomyosarcomas has long been known to be of prognostic significance for patients with this disease, with for example botryoid embryonal rhabdomyosarcomas having an excellent prognosis, conventional embryonal rhabdomyosarcomas having an intermediate prognosis, and alveolar rhabdomyosarcomas having a poor prognosis, these older studies do not account for more recently described, genetically defined rhabdomyosarcoma subtypes, which may have an excellent prognosis (e.g., infantile spindle cell rhabdomyosarcomas with VGLL2 fusions) or an especially poor prognosis (e.g., MYOD1 -mutant rhabdomyosarcomas, EWSR1/FUS::TFCP2 -fused rhabdomyosarcomas). Therefore, molecular genetic subclassification is increasingly important for patients with these diseases.

Among molecular genetic parameters, gene fusion status unquestionably shows the strongest relationship with clinical behavior and patient outcome, as reflected in the current COG rhabdomyosarcoma risk stratification system ( Table 20.3 ). In this context, fusion status refers specifically to the presence or absence of rearrangement of the FOXO1 gene, as seen in alveolar rhabdomyosarcoma. This risk stratification system uses fusion status, rather than histologic subtype (i.e., alveolar vs. embryonal), in part to address difficulties in the distinction of the “dense” form of embryonal rhabdomyosarcoma from true alveolar rhabdomyosarcoma. Historically, it has been claimed that roughly 80% of alveolar rhabdomyosarcoma are FOX01 -rearranged, in contrast to less than 5% of embryonal rhabdomyosarcomas; these figures likely underestimate and overestimate the percentage of positive alveolar and embryonal tumors, respectively. Regardless, it is clear that fusion-negative tumors have a superior event-free survival than fusion-positive tumors. In the largest study to date, evaluating 1727 patients from IRS V and COG ARST studies, only the presence of metastatic disease surpassed FOXO1 fusion status as a predictor of poor outcome.

There are some data to suggest that PAX7::FOXO1 fusion is associated with a better prognosis than PAX3::FOXO1A fusion in patients with alveolar rhabdomyosarcoma. ^, A series of 171 rhabdomyosarcomas, published by Sorensen et al., identified PAX3::FOXO1 and PAX7::FOXO1 fusion transcripts in 55% and 22% of 78 patients with alveolar rhabdomyosarcomas; all embryonal rhabdomyosarcomas were fusion-negative. Although fusion status was not associated with outcome differences in patients with locoregional tumors, patients with metastatic disease whose tumors harbored PAX3::FOXO1 fusions had considerably worse 4-year survival, as compared to patients with PAX7::FOXO1 tumors. Two more recent studies, evaluating 287 patients with rhabdomyosarcoma, found significantly worse 5-year survival for patients with PAX3 fusion as compared to those with PAX7 fusions. ^, It has been suggested, however, that these findings may be confounded by the greater percentage of PAX3 -rearranged tumors occurring in older patients and in larger tumors.

It is important to appreciate that the above analyses apply only to “fusion-positive” tumors harboring canonical PAX3::FOXO1 and PAX7::FOXO1 fusions and not to alveolar rhabdomyosarcomas with extremely rare variant fusions, such as PAX3::FOXO4 , PAX3::NCOA1 , PAX3::INO80D , or FOXO1::FGFR1 ; the clinical significance of these molecular genetic events is unknown. Similarly, infantile spindle cell rhabdomyosarcomas harboring rearrangements of the VGLL2 , TEAD1 , and NCOA2 loci generally have an excellent prognosis and should not be included among the “fusion-positive” tumors for risk assessment purposes. ^, Conversely, the prognosis for some “ FOXO1 -negative” rhabdomyosarcomas is quite poor, in particular MYOD1 -mutant spindle cell/sclerosing rhabdomyosarcomas in children and adults, ^, and adult spindle cell rhabdomyosarcomas harboring EWSR1/FUS::TFCP2 .

A variety of other molecular genetic alterations may also have prognostic significance in patients with rhabdomyosarcoma. For example, amplifications of 2p24 ( MYCN ) and 12q13-q14 ( CDK4 ), found in 10%–15% of alveolar rhabdomyosarcomas, have been associated with worse outcome, although data are limited. ^, TP53 mutations are also associated with poor prognosis in alveolar rhabdomyosarcomas, although they are present in only a very small minority of cases (∼4%), and in embryonal rhabdomyosarcomas.

Poorly differentiated round and spindle cell sarcomas, especially in children or young adults, constitute the most common problem in differential diagnosis. Included in this group are neuroblastoma, Ewing sarcoma, poorly differentiated angiosarcoma, synovial sarcoma, malignant melanoma, melanotic neuroectodermal tumor of infancy, granulocytic sarcoma, and lymphoma. Small cell carcinoma must also be considered when the tumor occurs in a patient older than 45 years. The differential diagnosis requires not only careful evaluation of clinical data, patient age, and tumor location, but also a painstaking examination of multiple sections for specific features (e.g., rhabdomyoblasts, rosettes, biphasic cellular or vascular differentiation, and intracellular pigment). Immunohistochemical assessment with multiple markers and often molecular genetic testing are important ancillary tools (as described earlier). Immunohistochemical analysis using a battery of antibodies is indispensable, including stains for muscle markers such as desmin, muscle-specific actin, and MyoD1 or myogenin. It must also be kept in mind that CD99, although a highly sensitive marker of Ewing sarcoma, is sometimes detected in embryonal or alveolar rhabdomyosarcoma.

Problems in diagnosis may also be caused by benign reactive and neoplastic lesions. They include polypoid cystitis, polyps and pseudosarcomatous myofibroblastic proliferations of the genitourinary tract, infectious granuloma, proliferative myositis, skeletal muscle regeneration, granular cell tumor, and fetal rhabdomyoma. Desmin and more limited expression of myogenin/MYOD1 may be seen in fibroepithelial polyps of the female genital tract, a potentially serious pitfall. Conversely, sparsely cellular, botryoid rhabdomyosarcomas (initially misinterpreted as “myxomas”) may occur. In these cases, consideration of age and location usually allows for the correct diagnosis, because myxomas are virtually nonexistent in children and almost never occur in visceral organs.

Some tumors have heterologous rhabdomyoblastic components . Focal rhabdomyoblastic differentiation occurs in a variety of malignant neoplasms, including those with sarcomatous differentiation only, those with epithelial or germ cell elements, and tumors of neuroectodermal derivation ( Box 20.3 ). Identification of such elements may be obvious by light microscopy alone, but in some cases, immunohistochemistry (IHC) is required to support rhabdomyoblastic differentiation. In addition, sarcomas with a propensity for undergoing dedifferentiation, including chondrosarcomas and liposarcomas, may have areas of divergent rhabdomyoblastic differentiation. Mesenchymal chondrosarcoma typically shows rhabdomyoblastic differentiation in the form of desmin, MyoD1, and myogenin expression, although rhabdomyoblasts are not seen. Similarly, recently described EWSR1::PATZ1 sarcomas commonly express desmin, myogenin and MYOD1 and may even contain rhabdomyoblasts; these rare sarcomas are not currently considered to represent rhabdomyosarcomas for purposes of clinical management.

Epithelial tumors may also exhibit rhabdomyoblastic differentiation, including malignant mixed mesodermal tumors of the uterus, cervix, or ovary; carcinosarcomas of the breast and stomach; pulmonary blastomas; nephroblastomas; and mixed-type hepatoblastomas. The rhabdomyoblastic component may even dominate the microscopic picture. Rhabdomyoblastic differentiation is also encountered in malignant or immature teratomas, but rarely as a major element. In most of these tumors, the rhabdomyoblastic component is accompanied by malignant epithelial and other mesenchymal elements, such as cartilage and bone. Rare ovarian Sertoli–Leydig cell tumors contain heterologous rhabdomyoblastic foci.

Rhabdomyoblastic elements also may be found in various neuroectodermal neoplasms, including malignant peripheral nerve sheath tumor (so-called malignant Triton tumor), ganglioneuroma (ectomesenchymoma), medulloepithelioma, and medulloblastoma. Malignant peripheral nerve sheath tumors with rhabdomyoblastic differentiation chiefly occur in patients older than 30 who have manifestations of type 1 neurofibromatosis. Malignant ectomesenchymoma is primarily a tumor of infants and young children and is not known to be associated with neurofibromatosis; it consists of a mixture of rhabdomyoblastic elements, mature ganglion cells, and neuroma-like structures. Huang et al. found consistent HRAS mutations, and a gene signature that suggested a close relationship to embryonal rhabdomyosarcoma.