Continual improvements in survival have been achieved for children and adolescents with cancer.[1] Between 1975 and 2020, childhood cancer mortality decreased by more than 50%.[1-3] Between 1975 and 2017, the 5-year relative survival rate for patients with rhabdomyosarcoma increased from 53% to 71% for children younger than 15 years and from 30% to 52% for adolescents aged 15 to 19 years.[1,2] In more recent years, improvements in outcome have plateaued.
Childhood and adolescent cancer survivors require close monitoring because side effects of cancer and its therapy may persist or develop months to years later. For specific information about the incidence, type, and monitoring of late effects in childhood and adolescent cancer survivors, see Late Effects of Treatment for Childhood Cancer.
Childhood rhabdomyosarcoma is a soft tissue malignant tumor of mesenchymal origin. It accounts for approximately 2.7% of cancer cases among children aged 0 to 14 years and 1.4% of the cases among adolescents and young adults aged 15 to 19 years.[2] The incidence is 4.6 cases per 1 million children younger than 20 years, which translates into about 350 new cases per year. Fifty percent of these cases are seen in the first decade of life.[2,4]
The 2020 World Health Organization classification distinguishes four histological subtypes of rhabdomyosarcoma, including embryonal, alveolar, spindle cell/sclerosing, and pleomorphic.[5] While these subtypes classify rhabdomyosarcoma into prognostically useful histological categories, FOXO1 gene fusions uniquely occur in alveolar histology tumors; however, not all tumors that have been classified as alveolar histology have a FOXO1 fusion. Molecular characterization has replaced histopathological assessment for treatment risk assignment. Male patients have a higher incidence of embryonal tumors, and Black patients have a slightly higher incidence of alveolar tumors.[4] For more information, see the sections on Cellular Classification for Childhood Rhabdomyosarcoma and Molecular Characteristics of Rhabdomyosarcoma.
Incidence may depend on the histological subtype of rhabdomyosarcoma, as follows:
Rhabdomyosarcoma may occur anywhere in the body. The most common primary sites include the following:[6,7]
Other less common primary sites include the trunk, chest wall, perineal/anal region, and abdomen, including the retroperitoneum and biliary tract.[7]
Most cases of rhabdomyosarcoma occur sporadically, with no recognized predisposing risk factor.
Predisposition factors reported for rhabdomyosarcoma include the following:
The Children's Oncology Group (COG) performed retrospective exome sequencing on germline DNA to determine the prevalence of 63 autosomal dominant cancer-predisposing genes in 615 patients with newly diagnosed rhabdomyosarcoma.[25] They identified germline cancer-predisposition variants in 45 patients with rhabdomyosarcoma (7.3%; all FOXO1 fusion negative) across 15 autosomal dominant genes. Specifically, 73.3% of the predisposition variants were found in predisposition syndrome genes previously associated with pediatric rhabdomyosarcoma risk, such as Li-Fraumeni syndrome (TP53, n = 11) and NF1 (NF1, n = 9). Notably, five patients had well-described oncogenic missense variants in HRAS (p.G12V and p.G12S) associated with Costello syndrome, and two patients each had variants in DICER1 and CBL, respectively. Germline variants were more frequent in patients with embryonal rhabdomyosarcoma than in those with alveolar rhabdomyosarcoma (10% vs. 3%, P = .02), but all of the patients with alveolar rhabdomyosarcoma were FOXO1 negative, and no germline variants were identified in patients with FOXO1 translocations. Although patients with a cancer-predisposition variant tended to be younger at diagnosis (P = .00099), 40% of germline variants were identified in patients older than 3 years.
The COG reviewed the impact of germline variants in cancer predisposition genes on patient outcomes.[26] In this study of 580 individuals with rhabdomyosarcoma, the median age was 5.9 years (range, 0.01–23.23 years), and the male-to-female ratio was 1.5:1 (351 [60.5%] male). For patients with congenital variants in rhabdomyosarcoma-associated cancer-predisposition genes, the event-free survival (EFS) rate was 48.4%, compared with 57.8% for patients without congenital predisposition variants (P = .10). The overall survival (OS) rate was 53.7% for patients with congenital predisposition variants, compared with 65.3% for patients without these variants (P = .06). Analyses were stratified by tumor histology and PAX3::FOXO1 or PAX7::FOXO1 gene fusion status. After adjustment, patients with congenital predisposition variants had significantly worse OS (adjusted hazard ratio [HR], 2.49; 95% confidence interval [CI], 1.39–4.45; P = .002), and patients with embryonal histology did not have better outcomes (EFS: adjusted HR, 2.25; 95% CI, 1.25–4.06; P = .007 and OS: adjusted HR, 2.83; 95% CI, 1.47–5.43; P = .002). These associations were not due to the development of a second malignant neoplasm. In addition, patients with fusion-negative rhabdomyosarcoma who harbored congenital predisposition variants had similarly inferior outcomes as patients with fusion-positive rhabdomyosarcoma who did not have congenital predisposition variants (EFS: adjusted HR, 1.35; 95% CI, 0.71–2.59; P = .37 and OS: adjusted HR, 1.71; 95% CI, 0.84–3.47; P = .14).
The COG reviewed the correlation between anaplastic histology and germline TP53 pathogenic variants in 239 patients with rhabdomyosarcoma. Among the 46 patients with anaplastic rhabdomyosarcoma, 11% (n = 5) carried a germline TP53 pathogenic variant, compared with 1% (n = 2) of the patients without anaplasia (P = .003). The rates of TP53 pathogenic variants in those with diffuse anaplasia and focal anaplasia were 9% (n = 3) and 17% (n = 2), respectively. Among the seven patients with TP53 pathogenic variants, 71% (5 of 7) had tumors with anaplastic histology.[27]
Rhabdomyosarcoma is usually curable in children with localized disease who receive combined-modality therapy, with more than 70% of patients surviving 5 years after diagnosis.[6,7,28] Relapses are uncommon in patients who were alive and event free at 5 years, with a 10-year late-event rate of 9%. Relapses are more common in patients who have unresectable disease, tumor in an unfavorable site at diagnosis, or metastatic disease at diagnosis.[29]
The prognosis for children or adolescents with rhabdomyosarcoma is related to many clinical and biological factors, including the following:
Because treatment and prognosis partly depend on the histology and molecular characterization of the tumor, it is necessary that the tumor tissue be reviewed by expert pathologists with experience in the evaluation and diagnosis of tumors in children. Typically, accurate diagnosis requires additional molecular characterization. The diversity of primary sites, the distinctive surgical and radiation therapy treatments for each primary site, and the subsequent site-specific rehabilitation underscore the importance of treating children with rhabdomyosarcoma in medical centers with appropriate experience in all therapeutic modalities.
Children aged 1 to 9 years have the best prognosis, while those younger than 1 year and older than 10 years fare less well. In Intergroup Rhabdomyosarcoma Study Group (IRSG) and COG trials, the 5-year failure-free survival (FFS) rate was 57% for patients younger than 1 year, 81% for patients aged 1 to 9 years, and 68% for patients older than 10 years. The 5-year OS rates were 76% for patients younger than 1 year, 87% for patients aged 1 to 9 years, and 76% for patients older than 10 years.[30] Historical data show that adults have fared less well than children (5-year OS rates, 27% ± 1.4% vs. 61% ± 1.4%; P < .0001).[31-34]
The 5-year FFS rate was 67% for infants, compared with 81% in a matched group of older patients treated by the COG.[30,38] This inferior FFS rate was largely the result of a relatively high rate of local failure.
In another retrospective study of 126 patients (aged ≤24 months) who were enrolled on the ARST0331 (NCT00075582) and ARST0531 (NCT00354835) trials, the 5-year local failure rate was 24%, the 5-year EFS rate was 68.3%, and the OS rate was 81.9%. Forty-three percent of the patients had an individualized local therapy plan that more frequently omitted radiation therapy. These patients had inferior local control and EFS rates.[38]
Members of the Cooperative Weichteilsarkom Studiengruppe (CWS) reviewed 155 patients with rhabdomyosarcoma presenting from birth to age 12 months; 144 patients had localized disease; 11 patients had metastases; and 32 patients presented with alveolar rhabdomyosarcoma pathology. The following results were reported:[39][Level of evidence C1]
A retrospective analysis of five consecutive studies from the CWS group examined infants and older children with localized rhabdomyosarcoma of the female genitourinary tract.[40] Among 67 patients treated from 1981 to 2019, age of 12 months or younger at diagnosis was the only significant negative prognostic factor that influenced EFS.
The European Paediatric Soft Tissue Sarcoma Study Group (EpSSG) enrolled 490 children younger than 36 months in their prospective RMS2005 study. The study included 110 patients younger than 12 months and 380 patients aged 12 to 36 months. Chemotherapy was given according to the risk group. Radiation therapy (22% received brachytherapy) was administered to 33.6% of the infants and 63.5% of the children aged 12 to 36 months. The 5-year OS rate was 88.4% for the infants, which was significantly better than the 72.5% rate observed in children aged 12 to 36 months. The treatment protocol in this trial, which used an increased application of adequate local therapy, may have contributed to these improved outcomes.[41][Level of evidence B4]
The EpSSG analyzed neonates with congenital rhabdomyosarcoma, which they defined as infants younger than 2 months at diagnosis who were enrolled in EpSSG trials.[42] Twenty-four patients with congenital rhabdomyosarcoma were registered. All patients had favorable histology and localized disease, except for one patient with PAX3::FOXO1 fusion–positive metastatic rhabdomyosarcoma. Three patients had VGLL2::CITED2 or VGLL2::NCOA2 fusions. Complete tumor resection was achieved in ten patients. No radiation therapy was given. Chemotherapy doses were adjusted to age and weight. Only two patients required further dose reduction for toxicity. The 5-year EFS rate was 75.0% (95% CI, 52.6%–87.9%), and the OS rate was 87.3% (95% CI, 65.6%–95.7%).
An international consortium identified 40 infants with spindle cell rhabdomyosarcoma.[43] The 5-year EFS rate for these infants with localized disease was 86% (± 11%; 95% CI), and the OS rate was 91% (± 9%; 95% CI). These outcomes compare favorably with those of all infants with localized rhabdomyosarcoma, for whom the 5-year failure-free survival rates range from 42% to 72% and the 5-year OS rates range from 61% to 88%. This finding suggests that infants with congenital spindle cell rhabdomyosarcoma have a favorable outcome compared with infants with other subtypes of rhabdomyosarcoma.
Two reports from the COG have documented inferior 5-year EFS rates in patients older than 10 years.[37,44] When compared with younger patients, this group of older patients was more likely to present with advanced-stage, large, and invasive alveolar tumors, with nodal involvement arising in the extremity and paratesticular sites. Older patients experienced less myelosuppression and more peripheral nervous system toxicity, suggesting that dose modifications during therapy cannot account for the age-related differences in EFS.
Adolescent and young adult (AYA) patients were more likely to have worse survival outcomes than children.[46]
Prognosis for childhood rhabdomyosarcoma varies according to the primary tumor site (see Table 1).
| Primary Site | Number of Patients | Survival at 5 Years (%) |
|---|---|---|
| aPatients treated on the ARST0331 study.[47] | ||
| bPatients treated on Intergroup Rhabdomyosarcoma Studies III–IV.[48] | ||
| cPooled analysis of European and North American groups.[49] | ||
| dCombined result from the Children's Oncology Group, German Cooperative Soft Tissue Sarcoma Study, Italian Cooperative Group, and International Society of Pediatric Oncology groups.[50] | ||
| ePooled analysis of European and North American groups.[51] | ||
| fPatients treated on Intergroup Rhabdomyosarcoma Study III.[6] | ||
| gPatients treated on Intergroup Rhabdomyosarcoma Studies I–IV.[52] | ||
| hPatients treated on the D9602 and ARST0331 trials.[53] | ||
| Orbita | 82 | 97 |
| Head and neck (nonparameningeal)b | 164 | 83 |
| Cranial parameningealc | 204 | 69.5 |
| Genitourinary (excluding bladder/prostate)b | 158 | 89 |
| Localized bladder/prostated | 322 | 84 |
| Localized extremitye | 643 | 67 |
| Trunk, abdomen, perineum, etc.f | 147 | 67 |
| Biliaryg,h | 25 | 76.5–78 |
Children with tumors 5 cm or smaller have improved survival, compared with children with tumors larger than 5 cm.[6] Both tumor volume and maximum tumor diameter are associated with outcome.[54][Level of evidence C1]
A retrospective review of soft tissue sarcomas in children and adolescents suggests that the 5-cm cutoff used for adults with soft tissue sarcoma may not be ideal for smaller children, especially infants. The review identified an interaction between tumor diameter and BSA.[55] This was not confirmed by a COG study of patients with intermediate-risk rhabdomyosarcoma.[56] This relationship requires prospective study to determine the therapeutic implications of the observation.
The extent of disease after the primary surgical procedure (i.e., the Surgical-pathologic Group, also called the Clinical Group) is correlated with outcome.[6] In the IRS-III study, patients with localized, gross residual disease after initial surgery (Surgical-pathologic Group III) had a 5-year survival rate of approximately 70%, compared with a rate of more than 90% for patients without residual tumor after surgery (Group I) and a rate of approximately 80% for patients with microscopic residual tumor after surgery (Group II).[6,57] Groups I and II represent a minority of patients; approximately 50% of patients have unresectable Group III disease at time of diagnosis.[6]
Resectability without functional impairment is related to the tumor's initial size and site and does not account for the biology of the disease. Outcome is optimized with the use of multimodality therapy. All patients require chemotherapy, and at least 85% of patients also benefit from radiation therapy, with favorable outcomes even for patients with nonresectable disease. In the IRS-IV study, the Group III patients with localized unresectable disease who were treated with chemotherapy and radiation therapy had a 5-year FFS rate of about 75% and a local control rate of 87%.[58] Two intermediate-risk COG rhabdomyosarcoma studies (D9803 and ARST0531 [NCT00354835]) were pooled to assess the benefit of delayed primary excision. In the D9803 study, local control with radiation therapy after either a partial or complete excision was completed at week 12. In the ARST0531 study, radiation was administered upfront at week 4. Patients with bladder or prostate rhabdomyosarcoma who received a delayed primary excision had no difference in survival, whereas patients with extremity rhabdomyosarcoma or nonbladder/nonprostate nonextremity rhabdomyosarcoma had an improved OS with delayed primary excision. Delayed primary excision strategy with a reduction in radiation dose resulted in superior OS for those sites.[59,60]
The alveolar subtype of childhood rhabdomyosarcoma is more prevalent among patients with less favorable clinical features (e.g., younger than 1 year or older than 10 years, extremity and truncal primary tumors, and metastatic disease at diagnosis). It is generally associated with a worse outcome than in similar patients with embryonal rhabdomyosarcoma.
Anaplasia has been observed in 13% of embryonal rhabdomyosarcoma cases, with some studies suggesting the presence of anaplasia adversely influenced clinical outcome in patients with intermediate-risk disease. However, anaplasia has not been shown to be an independent prognostic variable.[65,66]
Approximately 80% of rhabdomyosarcoma cases morphologically defined as alveolar rhabdomyosarcoma express a FOXO1 fusion. FOXO1 gene fusions occur only in alveolar histology tumors.[67] Several retrospective studies found that fusion status is an independent prognostic factor. Patients with translocation-negative alveolar rhabdomyosarcoma have tumors with genetic and molecular profiles and outcomes similar to patients with embryonal rhabdomyosarcoma, and they fare better than patients with fusion-positive alveolar rhabdomyosarcoma.[68,69] Early retrospective studies relied on convenience samples of available tumor tissue.[68,69] A subsequent prospective study from the Soft Tissue Sarcoma Committee of the COG that examined 434 cases of intermediate-risk rhabdomyosarcoma treated on a single intermediate protocol (D9803) confirmed these observations.[70] Analysis of 38 patients enrolled in the COG D9802 (NCT00003955) low-risk study determined that fusion-positive, low-risk patients should be treated as intermediate risk.[71]
The specific fusion partner may have prognostic impact. In a COG study, fusion-positive patients with Stage 2 or 3, Group III, and PAX3-positive tumors had a lower EFS rate (54%) than those with PAX7-positive tumors (65%). Both fusion-positive groups did worse than those with embryonal rhabdomyosarcoma (EFS rate, 77%; P < .001). Patients with alveolar rhabdomyosarcoma and PAX3 fusions had a poorer OS rate (64%) than patients with alveolar rhabdomyosarcoma and PAX7 fusions (87%), patients with alveolar rhabdomyosarcoma who were fusion negative (89%), and patients with embryonal rhabdomyosarcoma (82%; P = .006).[70] Comparable results were observed in the U.K. study; patients with PAX7-positive tumors and patients with fusion-negative tumors had similar outcomes.[72]
Using data from six consecutive COG studies, a retrospective analysis of 1,727 patients with rhabdomyosarcoma refined the risk stratification for childhood rhabdomyosarcoma. The study reported that after metastatic status, FOXO1 status was the most important prognostic factor and improved the risk stratification of patients with localized rhabdomyosarcoma.[69]
The COG performed a retrospective analysis of 269 patients with confirmed FOXO1 fusion–positive rhabdomyosarcoma who were enrolled in three completed clinical trials for localized rhabdomyosarcoma.[73] The estimated 4-year EFS rate was 53% (95% CI, 47%–59%), and the OS rate was 69% (95% CI, 63%–74%). Multivariate analysis identified older age (≥10 years) and larger tumor size (>5 cm) as independent, adverse prognostic factors for EFS within this population. Patients who had both of these adverse features experienced substantially inferior outcomes.
An EpSSG study evaluated the role of clinical factors together with FOXO1 fusion status in patients with nonmetastatic rhabdomyosarcoma, using data from the EpSSG RMS2005 study. The multivariable analysis of 1,661 evaluable patients retained five prognostic variables: age at diagnosis, tumor size, primary site, IRS Group, and FOXO1 status. A nomogram was created, stratifying patients into four risk groups. The 5-year EFS rates were 94.1% for patients in the low-risk group, 78.4% for patients in the intermediate-risk group, 65.2% for patients in the high-risk group, and 52.1% for patients in the very high-risk group.[74]
These studies demonstrated that fusion status was a better predictor of outcome than histology. Similar conclusions were reached in a retrospective study of three consecutive trials in the United Kingdom. Fusion status has now been incorporated into the risk stratification of patients in the current COG ARST1431 (NCT02567435) study for patients with intermediate-risk rhabdomyosarcoma, in subsequent COG trials, and in the new international EpSSG trial.[74] The authors underscored the probable value of treating fusion-negative patients whose tumors have alveolar histology with therapy that is stage appropriate for embryonal histology tumors.[75][Level of evidence C1]
Children with metastatic disease at diagnosis have the worst prognosis.
The prognostic significance of metastatic disease is modified by the following:
The COG performed a retrospective analysis of 179 patients who were diagnosed with rhabdomyosarcoma that was metastatic to the bone marrow. These patients were enrolled in one of four COG rhabdomyosarcoma clinical trials (D9802, D9803, ARST0431, and ARST08P1) between 1997 and 2013.[81] Patients were a median age of 14.8 years and 58% were male. Alveolar histology was the predominant type (76%), the extremity was the most common primary site (32%), and most patients had metastatic disease to additional sites (87%). The 3-year EFS rate was 9.4%, and the 5-year EFS rate was 8.2%. The 3-year OS rate was 26.1%, and the 5-year OS rate was 12.6%.
The COG performed a retrospective review of patients enrolled in high-risk protocols for rhabdomyosarcoma. FOXO1 fusion status correlated with clinical characteristics at diagnosis, including age, stage, histology, and extent of metastatic disease (Oberlin status). Among patients with metastatic disease, PAX::FOXO1 fusion status was not an independent predictor of outcome.[82][Level of evidence B1]
Lymph node involvement at diagnosis is seen in about 23% of patients with rhabdomyosarcoma and is associated with an inferior prognosis.[62,83] Clinical and/or imaging evaluation is performed before treatment and preoperatively. These findings are incorporated into the initial staging and grouping of a patient with rhabdomyosarcoma. The updated TNM staging defines clinical node involvement as larger than 1 cm.[84]
Pathological assessment of nodal disease is determined by biopsy and incorporated in the Surgical/Pathologic Clinical Group classification. Core-needle or open biopsy of clinically enlarged nodes is appropriate to confirm the presence of disease. Approximately 25% of enlarged nodes will be pathologically negative. Suspicious nodes are sampled surgically with open biopsy, preferred to needle aspiration, although needle aspiration may occasionally be appropriate. Pathological evaluation of clinically uninvolved nodes is site specific. In COG studies, it is required for extremity sites and for boys older than 10 years with paratesticular primary tumors.[85] Given the poorer outcomes, pathological node evaluation is required for patients with fusion-positive disease in current European and North American clinical trials.
Data on the frequency of lymph node involvement in various sites are useful for making clinical decisions. For example, up to 40% of patients with rhabdomyosarcoma in genitourinary sites have lymph node involvement, while patients with certain head and neck sites have a much lower likelihood (<10%). Patients with nongenitourinary pelvic sites (e.g., anus/perineum) have an intermediate frequency of lymph node involvement.[86]
In the extremities and select truncal sites, sentinel lymph node evaluation is a more accurate form of diagnosis than random regional lymph node sampling. In clinically negative lymph nodes of the extremity or trunk, sentinel lymph node biopsy is the preferred form of node sampling by the COG. Technical considerations are obtained from surgical experts. Needle or open biopsy of clinically enlarged nodes is appropriate.[87-90] Lymph node removal does not improve outcome, and it is useful for staging but not treatment.
The EpSSG performed a retrospective analysis of 109 patients with rhabdomyosarcoma with extremity primary tumors distal to the elbow or knee who were treated in the EpSSG RMS-2005 (NCT00379457) trial (2005–2016).[91] Thirty-seven of 109 patients (34%) had lymph node metastases at diagnosis. Of the 37 patients, 19 (51%) had in-transit metastases (ITM), especially in lower extremity rhabdomyosarcoma. The 5-year EFS rates were 88.9% for patients with ITM, 21.4% for patients with proximal lymph node involvement, and 20% for combined proximal lymph node involvement and ITM (P = .01). The 5-year OS rates were 100% for patients with ITM, 25.2% for patients with proximal lymph node involvement, and 15% for patients with combined proximal lymph node involvement and ITM (P =. 003). The authors concluded that popliteal and epitrochlear nodes should be considered as true (distal) regional nodes, instead of ITM. The authors recommended biopsy of these nodes, especially for distal extremity rhabdomyosarcoma of the lower limb.
The EpSSG reported a retrospective analysis of 1,294 children with embryonal rhabdomyosarcoma enrolled in the RMS-2005 protocol.[92] Of these patients, 143 had nodal involvement (N1). Patients with N1 disease were older and presented with tumors of unfavorable size, invasiveness, site, and resectability. Unlike alveolar rhabdomyosarcoma, nodal involvement was more frequent in the head and neck area and rare in extremity sites. The 5-year EFS rate was 75.5%, and the OS rate was 86.3% for patients with N0 disease. The 5-year EFS rate was 65.2%, and the OS rate was 70.7% for patients with N1 disease. Nodal involvement and the result of surgery at diagnosis (Intergroup Rhabdomyosarcoma Study group) were independent prognostic factors on multivariate analysis. Investigators concluded that regional nodal involvement is an independent prognostic factor in patients with embryonal rhabdomyosarcoma; therefore, it is appropriate to include this population in the high-risk category.
For more information, see the Molecular Characteristics of Rhabdomyosarcoma section.
It is unlikely that response to induction chemotherapy or best tumor response during therapy, assessed by anatomic imaging, correlates with the likelihood of survival in patients with rhabdomyosarcoma. This finding was based on the IRSG, COG, and International Society of Pediatric Oncology (SIOP) studies that found no association.[93,94]; [95][Level of evidence C2]; [96][Level of evidence C1] However, an Italian study did find that patient response correlated with likelihood of survival.[54][Level of evidence C1] In patients with embryonal rhabdomyosarcoma who had metastases only in the lungs, the CWS assessed the relationship between complete response of the lung metastases at weeks 7 to 10 after chemotherapy and outcome in 53 patients.[97][Level of evidence C1] The 5-year survival rate was 68% for 26 complete responders at weeks 7 to 10 versus 36% for 27 patients who achieved complete responses at later time points (P = .004).
Other studies have investigated response to induction therapy, showing benefit to response. These data are somewhat flawed because therapy is usually tailored on the basis of response. Thus the situation is not as clear as the COG data suggest.[98-103]
Response as judged by sequential functional imaging studies with fluorine F 18-fludeoxyglucose positron emission tomography (18F-FDG PET) may be an early indicator of outcome [104] and is under investigation by several pediatric cooperative groups. A retrospective analysis of 107 patients from a single institution examined PET scans performed at baseline, after induction chemotherapy, and after local therapy.[104] Standardized uptake value measured at baseline predicted PFS and OS, but not local control. A negative scan after induction chemotherapy correlated with statistically significantly better PFS. A positive scan after local therapy predicted worse PFS, OS, and local control. The COG evaluated the relationship between complete metabolic response, as assessed by 18F-FDG PET imaging, and EFS in patients with intermediate- or high-risk rhabdomyosarcoma.[105][Level of evidence B4] The maximum standard uptake values (SUVmax) at study entry did not correlate with EFS for intermediate-risk (P = .32) or high-risk (P = .86) patients. Compared with patients who did not achieve a complete metabolic response, EFS was not superior for intermediate-risk patients who achieved a complete metabolic response at weeks 4 (P = .66) or 15 (P = .46), or for high-risk patients who achieved a complete metabolic response at weeks 6 (P = .75) or 19 (P = .28). Change in SUVmax at weeks 4 (P = .21) or 15 (P = .91) for intermediate-risk patients and at weeks 6 (P = .75) or 19 (P = .61) for high-risk patients did not correlate with EFS.
PET scans have been shown to be useful in understanding patterns of spread, particularly in patients with extremity disease.[106][Level of evidence C2]
A retrospective study of 99 children with rhabdomyosarcoma used reverse transcription–polymerase chain reaction to analyze an 11-gene panel in peripheral blood and bone marrow samples at the time of initial diagnosis.[107] The 5-year EFS rate was 35.5% (95% CI, 17.5%–53.5%) for the 33 patients who were RNA positive, compared with 88.0% (95% CI, 78.9%–97.2%) for the 66 patients who were RNA negative (P < .0001). The predictive power of the assay was maintained in a multivariate analysis, which included the usual clinical characteristics that correlate with prognosis such as the presence of metastatic disease. These investigators also studied the diagnostic potential of ctDNA in 57 patients enrolled in the EpSSG RMS-2005 (NCT00379457) study. ctDNA was detected using both shallow whole-genome sequencing (WGS) and cell-free reduced representation bisulfite sequencing (cfRRBS). Of the 25 samples tested, 21 were correctly classified as embryonal histology by cfRRBS. The presence of methylated RASSF1A correlated with a poor outcome.[108]
The COG analyzed ctDNA in 124 patients with newly diagnosed, intermediate-risk rhabdomyosarcoma from the COG biorepository, which included 75 patients with fusion-negative rhabdomyosarcoma and 49 patients with fusion-positive rhabdomyosarcoma.[109] Ultralow passage WGS was used to detect copy number alterations. Rhabdo-Seq, a new custom sequencing assay, was used to detect rearrangements and single-nucleotide variants (SNVs).
The 5th edition of the World Health Organization (WHO) Classification of Tumors of Soft Tissue and Bone recognizes the following four categories of rhabdomyosarcoma:[1]
The embryonal subtype, which includes classic, dense, and botryoid variants, is the most frequently observed subtype in children, accounting for 70% to 75% of childhood rhabdomyosarcomas.[1,2] Tumors with embryonal histology typically arise in the head and neck region or in the genitourinary tract, although they may occur at any primary site.
Anaplasia has been observed in 13% of embryonal rhabdomyosarcoma cases, with some studies suggesting the presence of anaplasia adversely influenced clinical outcome in patients with intermediate-risk disease. However, anaplasia has not been shown to be an independent prognostic variable.[3,4]
Botryoid tumors, which represent about 10% of all rhabdomyosarcoma cases, are embryonal tumors that arise under the mucosal surface of body orifices such as the vagina, bladder, nasopharynx, and biliary tract. The WHO Classification of Tumors of Soft Tissue and Bone (4th and 5th editions) and the Children's Oncology Group (COG) eliminated botryoid rhabdomyosarcoma as a separate entity, with these cases classified as typical embryonal rhabdomyosarcoma.[1,5]
A COG study of 2,192 children with embryonal rhabdomyosarcoma (including botryoid and spindle cell variants) enrolled in clinical trials showed improved event-free survival (EFS) rates for patients with botryoid tumors (80%; 95% confidence interval [CI], 74%–84%), compared with typical embryonal rhabdomyosarcoma (73%; 95% CI, 71%–75%).[6] However, after adjusting for primary site, resection, and metastatic status, there was no difference in EFS by histological subtype. In this COG report, botryoid tumors accounted for 14% of intermediate-risk patients and 15% of low-risk patients. The botryoid histology retained prognostic significance in only a small proportion of patients with low-risk head and neck tumors, who are known to have excellent outcomes. For these reasons, the COG concluded that the addition of this histological classification of rhabdomyosarcoma has limited clinical utility and endorsed the recommendations of the WHO to remove this subtype from the current COG pathology classification.
One study analyzed the clinical and variant spectrum of 24 pediatric fusion-negative rhabdomyosarcoma tumors with high levels of myogenic differentiation. The analysis revealed that most tumors arose in the head and neck or genitourinary region. The overall survival rate was 100% for these patients (median follow-up, 4.6 years).[7]
Approximately 20% to 25% of children with rhabdomyosarcoma have the alveolar subtype, when histology alone is used to determine subtype.[1] An increased frequency of this subtype is noted in adolescents and in patients with primary sites involving the extremities, trunk, and perineum/perianal region.[2] Eighty percent of patients with alveolar histology tumors will have one of two gene fusions, PAX3 on chromosome 2 or PAX7 on chromosome 1, with the FOXO1 gene on chromosome 13.[8-10] Patients without a fusion have outcomes that are similar to those for patients with embryonal rhabdomyosarcoma.[11-13]
The current trial for intermediate-risk patients from the Soft Tissue Sarcoma Committee of the COG (ARST1431 [NCT02567435]) and all future trials will use fusion status rather than histology to determine eligibility. Fusion-negative patients with alveolar histology will undergo the same treatments as patients with embryonal histology.
The 4th edition of the WHO Classification of Tumors of Soft Tissue and Bone added spindle cell/sclerosing rhabdomyosarcoma as a separate subtype of rhabdomyosarcoma.[5] The 5th edition of the WHO Classification of Tumors of Soft Tissue and Bone continues to identify this separate subtype.[1] The spindle cell variant of embryonal rhabdomyosarcoma is most frequently observed at the paratesticular site.[6,14]
A COG study of 2,192 children with embryonal rhabdomyosarcoma (including botryoid and spindle cell variants) and enrolled in clinical trials showed improved EFS rates for patients with spindle cell rhabdomyosarcoma (83%; 95% CI, 77%–87%) compared with typical embryonal rhabdomyosarcoma (73%; 95% CI, 71%–75%).[6] Patients with spindle cell rhabdomyosarcoma with parameningeal primary tumors (n = 18) were the exception to the overall favorable prognosis for this subtype, with a 5-year EFS rate of 28% (compared with >70% for parameningeal nonspindle cell embryonal rhabdomyosarcoma).
In the WHO classification, sclerosing rhabdomyosarcoma is considered a variant pattern of spindle cell rhabdomyosarcoma, as descriptions note increasing degrees of hyalinization and matrix formation in spindle cell tumors. There are at least two distinct molecular subtypes of spindle cell/sclerosing rhabdomyosarcoma in children:
Pleomorphic rhabdomyosarcoma occurs in adults in their sixth and seventh decades, most commonly involves the extremities, and is associated with a poor prognosis. This histological variant is extremely rare and not well characterized in the pediatric population.[18,19] In children, tumors with extensive pleomorphism are considered anaplastic embryonal rhabdomyosarcoma.[1]
Machine learning of rhabdomyosarcoma histopathology can potentially provide predictive models for identifying the histological subtypes of rhabdomyosarcoma.[20,21] Digital whole-slide hematoxylin and eosin (H&E) images were collected from a cohort of 321 patients with rhabdomyosarcoma enrolled in COG trials from 1998 to 2017. These images were fed into deep learning convolutional neural networks (CNNs) to learn features associated with driver variants and patient outcomes.[22]
The four histological categories recognized in the 5th edition of the World Health Organization (WHO) Classification of Tumors of Soft Tissue and Bone have distinctive genomic alterations and are briefly summarized below.[1,2,23]
The distribution of gene variants and gene amplifications (for CDK4 and MYCN) differs between patients with embryonal histology lacking a PAX::FOXO1 gene fusion (fusion-negative rhabdomyosarcoma) and patients with PAX::FOXO1 gene fusions (fusion-positive rhabdomyosarcoma). See Table 2 below and the text that follows. These frequencies are derived from a combined cohort of the Children's Oncology Group (COG) and United Kingdom rhabdomyosarcoma patients (n = 641).[27]
| Gene | % FN Cases With Gene Alteration | % FP Cases With Gene Alteration |
|---|---|---|
| aAdapted from Shern et al.[27] | ||
| NRAS | 17% | 1% |
| KRAS | 9% | 1% |
| HRAS | 8% | 2% |
| FGFR4 | 13% | 0% |
| NF1 | 15% | 4% |
| BCOR | 15% | 6% |
| TP53 | 13% | 4% |
| CTNNB1 | 6% | 0% |
| CDK4 | 0% | 13% |
| MYCN | 0% | 10% |
Details of the genomic alterations that predominate within each of the WHO histological categories are as follows.
Among the RAS pathway genes, germline variants in NF1 and HRAS predispose to rhabdomyosarcoma. In a study of 615 children with rhabdomyosarcoma, 347 had tumors with embryonal histology. Of these, nine patients had NF1 germline variants, and five patients had HRAS germline variants, representing 2.6% and 1.4% of embryonal histology cases, respectively.[32]
Other genes with recurring variants in fusion-negative rhabdomyosarcoma tumors include FGFR4, PIK3CA, CTNNB1, FBXW7, and BCOR, all of which are present in fewer than 15% of cases.[27,30,31]
TP53 variants: TP53 variants are observed in 10% to 15% of patients with fusion-negative rhabdomyosarcoma and occur less commonly (about 4%) in patients with alveolar rhabdomyosarcoma.[27] In other childhood cancers (e.g., Wilms tumor), TP53 variants are associated with anaplastic histology,[33] and the same is true for embryonal rhabdomyosarcoma. In a study of 146 rhabdomyosarcoma patients with known TP53 status, approximately two-thirds of tumors with TP53 variants showed anaplasia (69%), but only one-quarter of tumors with anaplasia had TP53 variants.[4]
The presence of TP53 variants was associated with reduced EFS in both nonrisk-stratified and risk-stratified analyses for both a COG and a U.K. rhabdomyosarcoma cohort.[27] The poor prognosis associated with TP53 variants was observed for both embryonal and alveolar patients. Based on these results, the COG plans to consider TP53 variant as a high-risk defining characteristic in its upcoming trials.[34]
Rhabdomyosarcoma is one of the childhood cancers associated with Li-Fraumeni syndrome. In a study of 614 pediatric patients with rhabdomyosarcoma, 11 patients (1.7%) had TP53 germline variants. Variants were less common in patients with alveolar histology (0.6%), compared with patients with nonalveolar histologies (2.2%).[32] Rhabdomyosarcoma with nonalveolar anaplastic morphology may be a presenting feature for children with Li-Fraumeni syndrome and germline TP53 variants.[35]
DICER1 variants in embryonal rhabdomyosarcoma: DICER1 variants are observed in a small subset of patients with embryonal rhabdomyosarcoma, most commonly arising in tumors of the female genitourinary tract.[27] More specifically, most cases of cervical embryonal rhabdomyosarcoma,[37-39] which most commonly occurs in adolescents and young adults,[40,41] have DICER1 variants. In contrast, DICER1 variants are rarely observed in patients with vaginal primary sites, an entity occurring primarily in girls younger than 2 or 3 years.[38,40] DICER1 variants are also common in embryonal rhabdomyosarcoma arising in the uterine corpus, but this presentation is primarily observed in adults.[38,42] Cervical rhabdomyosarcoma generally shows a sarcoma botryoides histological pattern, and many cases show areas of cartilaginous differentiation, a feature also observed in other tumor types with DICER1 variants.[40,41,43] In support of the distinctive biology of embryonal rhabdomyosarcoma with DICER1 variants, these cases have a DNA methylation pattern that is distinctive from that of other embryonal rhabdomyosarcoma cases.[39] A diagnosis of cervical rhabdomyosarcoma is an indication for genetic testing for DICER1 syndrome.[38,44]
For the diagnosis of alveolar rhabdomyosarcoma, a FOXO1 gene rearrangement may be detected with good sensitivity and specificity using either fluorescence in situ hybridization or reverse transcription–polymerase chain reaction.[52]
In addition to FOXO1 rearrangements, alveolar tumors are characterized by a lower mutational burden than are fusion-negative tumors, with fewer genes having recurring mutations.[30,31] The most frequently observed alterations in fusion-positive tumors are focal amplification of CDK4 (13%) or MYCN (10%), with small numbers of patients having recurring mutations in other genes (e.g., BCOR, 6%; NF1, 4%; TP53, 4%; and PIK3CA, 2%).[27] TP53 mutations in alveolar rhabdomyosarcoma appear to connote a high risk of treatment failure.[27]
Congenital/infantile spindle cell rhabdomyosarcoma: Several reports have described cases of congenital or infantile spindle cell rhabdomyosarcoma with gene fusions involving VGLL2 and NCOA2 (e.g., VGLL2::CITED2, TEAD1::NCOA2, VGLL2::NCOA2, SRF::NCOA2).[15,54]
MYOD1-altered spindle cell/sclerosing rhabdomyosarcoma: In older children and adults with spindle cell/sclerosing rhabdomyosarcoma, a specific MYOD1 variant (p.L122R) has been observed in a large proportion of patients.[15,56-58] In the combined cohort of COG and U.K. rhabdomyosarcoma patients (n = 641), variants in MYOD1 were found in 3% (17 of 515) of all fusion-negative rhabdomyosarcoma cases and in no fusion-positive cases. The presenting age of patients with MYOD1 variants was 10.8 years.[27] Most cases in this cohort showed spindle or sclerosing features, but cases with densely packed cells that mimicked the dense pattern of embryonal rhabdomyosarcoma were also observed. Most cases in this cohort (15 of 17, 88%) had either head and neck or parameningeal region primary sites. Activating PIK3CA variants are seen in about one-half of cases with MYOD1 variants.[17,27] The presence of the MYOD1 variant is associated with a markedly increased risk of local and distant failure.[15,27,56,57]
Intraosseous spindle cell rhabdomyosarcoma: Primary intraosseous rhabdomyosarcoma is a very uncommon presentation for rhabdomyosarcoma. Most cases present with gene rearrangements involving TFCP2, with either FUS or EWSR1.[59-63] Rhabdomyosarcoma with a FUS::TFCP2 or EWSR1::TFCP2 gene fusion most commonly presents in young adults, although cases in older children and adolescents have been reported.[59,62,63] Craniofacial bones are the most common primary tumor location, and positivity for ALK and cytokeratins by immunohistochemistry is commonly observed. Other characteristics of this entity include a complex genomic profile, with most cases showing deletion of the CDKN2A tumor suppressor gene.[62] Intraosseous spindle cell rhabdomyosarcoma with a FUS::TFCP2 or EWSR1::TFCP2 gene fusion shows an aggressive clinical course. In one study, the median OS was only 8 months.[62]
Recurrent and refractory rhabdomyosarcomas from pediatric (n = 105) and young-adult patients (n = 15) underwent tumor sequencing in the National Cancer Institute–Children's Oncology Group (NCI-COG) Pediatric MATCH trial. Actionable genomic alterations were found in 53 of 120 tumors (44.2%), and patients with these alterations qualified for treatment on MATCH study arms.[64] Variants of MAPK pathway genes (HRAS, KRAS, NRAS, NF1) were most frequent and were reported in 32 of 120 tumors (26.7%). Amplifications of cyclin-dependent kinase genes (CDK4, CDK6) were detected in 15 of 120 tumors (12.5%).
Before a suspected tumor mass is biopsied, imaging studies of the mass and baseline laboratory studies should be obtained. After the patient is diagnosed with rhabdomyosarcoma, an extensive evaluation to determine the extent of the disease should be performed before instituting therapy. This evaluation typically includes the following:
The European Paediatric Soft Tissue Sarcoma Study Group reviewed 367 patients enrolled in the CCLG-EPSSG-RMS-2005 (NCT00379457) study.[1][Level of evidence B4] By prospective study design, patients with indeterminate pulmonary nodules identified on baseline CT scan of the chest (defined as ≤4 pulmonary nodules measuring <5 mm or 1 nodule measuring ≥5 mm and <10 mm) received the same treatment as did patients with no pulmonary nodules identified on baseline CT of the chest. Rates of event-free survival and overall survival for both groups were the same. The authors concluded that indeterminate pulmonary nodules at diagnosis, as defined in this summary, do not affect outcome in patients with localized rhabdomyosarcoma.
Pathological evaluation of normal-appearing regional nodes is currently required for all Soft Tissue Sarcoma Committee of the Children's Oncology Group (COG-STS) study participants with extremity and trunk primary rhabdomyosarcoma. In boys aged 10 years and older with paratesticular rhabdomyosarcoma, retroperitoneal node sampling (ipsilateral nerve sparing) is currently required for normal-appearing lymph nodes because microscopic tumor is often documented, even when the nodes are not enlarged.[8] The International Society of Paediatric Oncology Malignant Mesenchymal Tumour Group has confirmed this is a necessary approach.[9] For more information, see the Regional and in-transit lymph nodes for extremity tumors section.
The efficacy of these imaging studies for identifying involved lymph nodes or other sites of disease is important for staging, and PET imaging is recommended on current COG-STS treatment protocols.
A retrospective study of 1,687 children with rhabdomyosarcoma enrolled in Intergroup Rhabdomyosarcoma Study Group (IRSG) and COG studies from 1991 to 2004 suggests those with localized negative regional lymph nodes, noninvasive embryonal tumors, and Group I alveolar tumors (about one-third of patients) can have limited staging procedures that eliminate bone marrow and bone scan examinations at diagnosis.[13]
Assessing extent of disease of rhabdomyosarcoma is complex. The process includes the following steps:
Prognosis for children with rhabdomyosarcoma depends predominantly on the primary tumor site, tumor size, surgical-pathological Group, presence or absence of nodal disease and distant metastasis, and fusion status. Favorable prognostic groups were identified in previous IRSG studies, and treatment plans were designed on the basis of patient assignment to different treatment protocols according to prognosis.
Current COG-STS protocols for rhabdomyosarcoma use the TNM-based pretreatment staging system that incorporates the primary tumor site, presence or absence of tumor invasion of surrounding tissues, tumor size, clinical (imaging) assessment of regional lymph node status, and the presence or absence of metastases. This staging system is described in Table 4 below.[14-16]
Terms defining the TNM criteria are described in Table 3.
| Term | Definition |
|---|---|
| CSF = cerebrospinal fluid; CT = computed tomography; MRI = magnetic resonance imaging. | |
| aAdapted from Crane et al.[16] | |
| Favorable site | Orbit; head and neck (excluding parameningeal); genitourinary tract (nonbladder/nonprostate). |
| Unfavorable site | Any site other than a favorable site. |
| T1 | Tumor confined to anatomical site of origin. |
| T2 | Extension and/or fixative to surrounding tissue. |
| a | Tumor ≤5 cm in longest diameter. |
| b | Tumor >5 cm in longest diameter. |
| N0 | Regional nodes not clinically involved. |
| N1 | Regional nodes clinically involved as defined as >1 cm measured in short axis on CT or MRI. |
| NX | Clinical status of regional nodes unknown (especially sites that preclude lymph node evaluation). |
| M0 | No distant metastases. |
| M1 | Distant metastases present (Note: the presence of positive cytology in pleural fluid, abdominal fluid, or CSF and the presence of pleural or peritoneal implants are considered evidence of metastases). |
| Stage | Sites of Primary Tumor | Tumor Sizec | Regional Lymph Nodesd | Distant Metastasisd |
|---|---|---|---|---|
| cTumor size: (a) <5 cm in longest diameter; (b) >5 cm in longest diameter. | ||||
| dFor definitions of the TNM criteria, see Table 3. | ||||
| 1 | Favorable sites | a or b | N0 or N1 or NX | M0 |
| 2 | Unfavorable sites | a | N0 or NX | M0 |
| 3 | Unfavorable sites | a | N1 | M0 |
| b | N0 or N1 or NX | |||
| 4 | Any site | a or b | N0 or N1 or NX | M1 |
The IRS-I, IRS-II, IRS-III, and IRS-IV studies prescribed treatment plans on the basis of the surgical-pathological Group system. In this system, Groups are defined by the extent of disease and by the completeness or extent of initial surgical resection after pathological review of the tumor specimen(s). The definitions for these Groups are shown in Table 5 below.[16-19]
| Group | Incidence | Definition |
|---|---|---|
| CSF = cerebrospinal fluid. | ||
| aAdapted from Crane et al.[16] | ||
| I | Approximately 15% | Localized disease, completely resected (regional lymph nodes not involved). |
| II | Approximately 16% | Localized disease, grossly resected with microscopic residual disease or regional disease, grossly resected with or without microscopic residual disease. (a) Localized disease, grossly resected tumor with microscopic residual disease, regional nodes not involved. (b) Regional disease with involved nodes, completely resected with no microscopic residual disease (including most distal node is histologically negative). (c) Regional disease with involved nodes, grossly resected with evidence of microscopic residual and/or histological involvement of the most distal regional node in the dissection. |
| III | Approximately 50% | Localized or regional disease, biopsy only or incomplete resection with gross residual disease. |
| IV | Approximately 20% | Distant metastatic disease present at onset. Although not limited to these, the following are considered evidence of metastatic disease: (a) presence of positive cytology in CSF, (b) positive cytology in pleural or abdominal fluids, (c) presence of implants on pleural or peritoneal surfaces. (Note: Regional lymph node involvement and adjacent organ infiltration are not considered metastatic disease. Presence of a pleural effusion or ascites, without positive cytological evaluation, is not considered evidence of metastatic disease.) |
After patients are categorized by Stage and surgical-pathological Group, a Risk Group is assigned on the basis of the Stage, Group, and FOXO1 fusion status. The planned COG low-risk study will also use TP53 and MYOD1 variant status to assign risk group. Patients are classified for protocol purposes as having a low risk, intermediate risk, or high risk of disease recurrence.[20-22] Treatment assignment is based on Risk Group, as shown in Table 6.
| Risk Group | Fusion Status/Molecular Profile | Stage | Group |
|---|---|---|---|
| Very low risk | Fusion negative: MYOD1 wild-type, TP53 wild-type | 1 | I |
| Low risk | Fusion negative: MYOD1 wild-type, TP53 wild-type | 1 | II, III (orbit only) |
| 2 | I, II | ||
| Intermediate risk | Fusion negative | 1 | III (nonorbit) |
| 2, 3 | III | ||
| 3 | I, II | ||
| 4 | IV (age <10 years) | ||
| Fusion positive | 1, 2, 3 | I, II, III | |
| High risk | Fusion positive | 4 | IV |
| Fusion negative | 4 | IV (age ≥10 years) | |
| aAdapted from Crane et al.[16] | |||
The most recent COG protocols use fusion status and molecular findings, as opposed to histology, to define Risk Groups.
All children with rhabdomyosarcoma require multimodality therapy with systemic chemotherapy, in conjunction with either surgery, radiation therapy (RT), or both modalities to maximize local tumor control.[1-3] Surgical resection is performed before chemotherapy if it will not result in disfigurement, functional compromise, or organ dysfunction. If this is not possible, only an initial biopsy is performed.
Low-risk Group I (complete tumor resection, about 15% of patients) patients are treated with multiagent chemotherapy after surgical resection. Group II patients typically require chemotherapy and local tumor bed irradiation (about 20% of patients). Most patients (about 50%) have Group III (gross residual) disease.[4] After initial chemotherapy, Group III patients receive definitive RT for local control of the primary tumor. Some patients with initially unresected tumors may undergo delayed primary excision after induction chemotherapy to remove residual tumor before the initiation of RT. This is appropriate only if the delayed excision is deemed feasible with acceptable functional and cosmetic outcome and if a grossly complete resection is anticipated. If a delayed primary excision results in complete resection or microscopic residual disease, a modest (15%–30%) reduction in RT could be utilized.[5] Patients with Group IV disease (about 15%) receive chemotherapy and RT to the primary tumor and metastatic disease sites when feasible.
RT is given to clinically suspicious lymph nodes (detected by palpation or imaging) unless the suspicious lymph nodes are biopsied and shown to be free of rhabdomyosarcoma. RT is also administered to lymph node basins where a sentinel lymph node biopsy has identified microscopic disease.[5]
The discussion of treatment options for children with rhabdomyosarcoma is divided into the following sections:
Rhabdomyosarcoma treatment options used by the Children's Oncology Group (COG) and by groups in Europe (as exemplified by trials from the Soft Tissue Sarcoma Committee of the COG [COG-STS], the Intergroup Rhabdomyosarcoma Study Group [IRSG], the International Society of Pediatric Oncology Malignant Mesenchymal Tumor [MMT] Group, and the European Paediatric Soft Tissue Sarcoma Study Group [EpSSG]) differ in management and overall treatment philosophy, as noted below:[2]
The MMT Group approach led to an overall survival (OS) rate of 71% in the European MMT89 study, compared with an OS rate of 84% in the IRS-IV study. Similarly, EFS rates at 5 years were 57% in the MMT89 study versus 78% in the IRS-IV study. Differences in outcomes were most striking for patients with extremity and head and neck nonparameningeal tumors. Failure-free survival was lower for patients with bladder/prostate primary tumors who did not receive RT as part of their initial treatment, but there was no difference in OS between the two strategies for these patients.[6] The overall impression is that survival for most patient subsets is superior with the use of early local therapy, including RT.[1-3]
Cancer in children and adolescents is rare, although the overall incidence has slowly increased 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 pediatric specialists and others to ensure that children receive treatment, supportive care, and rehabilitation to achieve optimal survival and quality of life:
For specific information about supportive care for children and adolescents with cancer, see the summaries on Supportive and Palliative Care.
The American Academy of Pediatrics has outlined guidelines for pediatric cancer centers and their role in the treatment of children and adolescents with cancer.[2] At these 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 diagnosed with cancer are generally designed to compare potentially better therapy with current standard therapy. Other types of clinical trials test novel therapies when there is no standard therapy for a cancer diagnosis. Most of the progress in identifying curative therapies for childhood cancers has been achieved through clinical trials. Information about ongoing clinical trials is available from the NCI website.
Optimizing care for patients with rhabdomyosarcoma requires a multidisciplinary team approach. All patients require chemotherapy and effective local tumor control. Because rhabdomyosarcoma can arise from multiple sites, surgical care decisions and radiotherapeutic options must be tailored to the specific aspects of each site and should be discussed with a multidisciplinary team, including representatives of those specialties and pediatric oncologists. These multidisciplinary discussions ideally occur at the time of diagnosis, either before or after the diagnostic biopsy and before the initiation of therapy.
Local control remains a significant problem in children with rhabdomyosarcoma. The predominant site of treatment failure in patients with initially localized rhabdomyosarcoma has been local recurrence. In the Intergroup Rhabdomyosarcoma Study Group (IRS)-II trial, of patients who achieved a complete remission with chemotherapy and surgery, almost 20% of patients with Groups I to III disease relapsed locally or regionally, and 30% of patients with Group IV disease relapsed locally or regionally. Local or regional relapses accounted for 70% to 80% of all relapses in children with Groups I to III disease and 46% of all relapses in patients with Group IV disease.[1]
Both surgery and radiation therapy (RT) are procedures primarily focused on local tumor control, but each treatment has risks and benefits.
For more information about surgical and radiotherapeutic management of the more common primary sites, see the Surgery and RT by Primary Site of Disease (Local Control Management) section.
Treatment options for childhood rhabdomyosarcoma include the following:
Surgical removal of the entire tumor should be considered initially, but only if functional and cosmetic impairment will not result.[2] With that stipulation, complete gross resection of the primary tumor, with a surrounding margin of normal tissue, and biopsy are recommended by the authors of one study. For some tumor sites, sampling of regional draining lymph nodes is necessary. Children's Oncology Group (COG) protocols require regional draining node sampling in extremity tumors and paratesticular tumors in patients older than 10 years. Important exceptions to achieving an R0 resection (negative margins) are in tumors of the orbit and the genitourinary region.[3,4] Additionally, the principle of wide and complete resection of the primary tumor is less applicable for patients known to have metastatic disease at the initial operation, but it is an appropriate approach if easily accomplished without loss of form (cosmesis) and function.
Patients with microscopic residual tumor after their initial surgery appear to have improved prognoses if a second operation (primary re-excision) to resect the primary tumor bed before beginning chemotherapy can completely remove the tumor without loss of form and function.[5]
There is no evidence that debulking surgery (i.e., surgery that is expected to leave macroscopic residual tumor) improves outcomes, compared with biopsy alone; therefore, debulking surgery is not recommended for patients with rhabdomyosarcoma.[6][Level of evidence B4] Rather than debulking a tumor at the time of initial biopsy, it is preferable to delay definitive surgery until after induction chemotherapy (delayed primary excision). In a retrospective study of 73 selected patients, delayed primary excision allowed for the identification of viable tumor that remained after initial chemotherapy. Of the 73 patients, 65 also received RT. Patients with viable tumor had shorter event-free survival (EFS) rates than did patients without viable tumor, but there was no effect on overall survival (OS).[7] There is also no evidence that performing surgical resection on residual masses detected by imaging at completion of all planned therapy improves outcomes.[8] Thus, residual masses can be monitored without therapeutic intervention.
For children with low-risk rhabdomyosarcoma, local control was not diminished with reduced doses of RT after surgical resection.[9] Subsequently, delayed primary excision was evaluated by the Soft Tissue Sarcoma Committee of the COG (COG-STS) in the D9602 and D9803 studies.[8] Delayed primary excision at week 12 after induction chemotherapy was completed in 45% to 54% of patients with Group III rhabdomyosarcoma tumors when appropriate (anticipated complete resection with no loss of form or function at select sites such as bladder, prostrate, extremity, trunk, retroperitoneum, intrathoracic, perineum, or perianal). Of these patients, 81% to 84% were eligible for modest RT dose reduction. Approximately 50% of these patients had an R0 resection (negative margins) and received a reduced RT dose of 36 Gy, and 30% of patients had an R1 resection (margins were microscopically involved) and received a reduced RT dose of 41.4 Gy (from the standard 50.4-Gy dose). Local control and survival outcomes were similar to those of patients who received full-dose RT alone in the IRS-IV study.[7]
A retrospective analysis compared patients with clinical Group III rhabdomyosarcoma treated on consecutive COG protocols D9803 (encouraged delayed primary excision) and ARST0531 (NCT00354835) (discouraged delayed primary excision).[10] Among 369 patients in an adjusted-regression analysis, the risk of death (hazard ratio [HR], 0.71; 95% confidence interval [CI], 0.43–1.16) was similar for patients who did or did not undergo delayed primary excision. A subset of patients who had tumors of the trunk and retroperitoneum did have a reduced risk of death with delayed primary excision (HR, 0.44; 95% CI, 0.20–0.97).
RT is an effective method for achieving local control of the tumor for patients with microscopic or gross residual disease after biopsy, initial surgical resection, or chemotherapy.
A study of Group I patients with alveolar rhabdomyosarcoma and undifferentiated soft tissue sarcoma found that omission of RT was followed by decreased local control.[12] A subsequent review of patients with only alveolar rhabdomyosarcoma found that the improvement in outcome with RT did not reach statistical significance for patients with Stage 1 and Stage 2 tumors. There were very few patients (n = 4) with large tumors (Stage 3, >5 cm) who did not receive RT, but their outcome was poor.[13][Level of evidence C2] COG recommends the use of RT for all patients with FOXO1 fusion–positive disease (previously called alveolar rhabdomyosarcoma).
The German Cooperative Weichteilsarkom Studiengruppe (CWS) conducted a review of European trials between 1981 and 1998, in which RT was omitted for some Group II patients. This review demonstrated a benefit to using RT as a component of local tumor control for all Group II patient subsets, as defined by tumor histology, tumor size, and tumor site.[15]
The CWS performed a retrospective analysis of 395 children with parameningeal rhabdomyosarcoma. Patients had IRS Groups II (n = 15) and III (n = 380) disease. Delayed resection was performed in 88 of 395 patients (22%), and RT was also given to 79 of the 88 patients (90%) who underwent resections. RT was the predominant local treatment for 355 of 395 patients (90%), which included hyperfractionated accelerated photon RT (HART) (n = 77), conventionally fractionated photon RT (n = 91) or proton-beam RT (n = 126), brachytherapy (n = 4), and heavy ions (n = 1). Details of the RT received were not available for 56 patients.[17]
Investigators performed a retrospective analysis of 1,470 patients (aged 21 years or younger) with localized rhabdomyosarcoma. These patients were enrolled in the CWS-96, CWS-2002P, and Soft Tissue Sarcoma Registry (SoTiSaR) trials. The study analyzed and compared the indications, doses, and application methods of RT and their influence on prognosis.[18] The authors concluded the following:
As with the surgical management of patients with rhabdomyosarcoma, recommendations for RT depend on the following:
For optimal care of pediatric patients undergoing radiation treatments, it is imperative that radiation oncologists, radiation therapists, and nurses who are experienced in treating children are available. An anesthesiologist may be necessary to sedate young patients. Computerized treatment planning with a 3-dimensional planning system is essential. Techniques to deliver radiation specifically to the tumor while sparing normal tissue (e.g., conformal radiation therapy, intensity-modulated radiation therapy [IMRT], volumetrical modulated arc therapy, proton-beam therapy [charged-particle radiation therapy], or brachytherapy) are appropriate.[19-24]
Dosimetric comparison of proton-beam RT and photon IMRT treatment plans has shown that proton-beam treatment plans may spare more normal tissue adjacent to the targeted volume than IMRT plans, but with no difference in local control using photon RT. Late effects data are lacking.[25,26]
Evidence (radiation delivery techniques):
The radiation doses according to Group, histology, and disease site for children with rhabdomyosarcoma are described in Table 7:
| Group | Treatment |
|---|---|
| N = regional lymph node. | |
| Group I | |
| Fusion negative (embryonal) | No RT required. |
| FOXO1 fusion positive | 36 Gy to involved (prechemotherapy) site. |
| Group II | |
| N0 (microscopic residual disease after surgery) | 36 Gy to involved (prechemotherapy) site. |
| N1 (resected regional lymph node involvement) | 36 Gy to involved (prechemotherapy) site and 41.4 Gy to nodes. |
| Group III | |
| Orbital and nonorbital tumors | 45 Gy for orbital tumors with complete response to chemotherapy. For other sites and orbital tumors in partial remission, 50.4 Gy with volume reduction after 36 Gy if excellent response to chemotherapy (or complete remission after delayed re-excision) and noninvasive pushing tumors; no volume reduction for invasive tumors. 59.4 Gy boost to residual disease at 9 weeks for tumors >5 cm at diagnosis (if enrolled on the COG ARST1431 [NCT02567435]) protocol. |
| N1 with gross residual disease after surgery/chemotherapy | 50.4 Gy |
| Group IV | |
| As for other Groups and including all metastatic sites, if safe and possible. Exception: lung (pulmonary metastases) treated with 12 Gy to 15 Gy depending on age. | |
In the COG ARST1431 (NCT02567435) study, risk group is in part determined by fusion status. The recommended dose of RT depends on the amount of residual disease, if any, after the initial primary surgical procedure and fusion status. For patients with fusion-positive rhabdomyosarcoma who have had an initial complete resection (Group I), radiation therapy with 36 Gy is recommended.
Select COG subgroups with Group III disease received somewhat reduced radiation doses of 36 Gy after delayed gross-total resection with negative margins (R0 resection), and 41.4 Gy if the margins were microscopically involved (R1 resection) or the nodes were positive. In the COG-D9602 study, a limited number of low-risk patients had a greater than 85% likelihood of local control with 36 Gy.[9] Similarly, the intermediate-risk studies for patients with Group III disease investigated the paradigm of delayed resection in amenable patients (anticipated complete resection with no loss of form or function at select sites such as bladder, prostate, extremity, trunk, peritoneum, intrathoracic, perineum, and perianal), with a subsequent reduced dose of RT (36 Gy for R0 resections or 41.4 Gy for R1 resections). The study demonstrated that patients who received reduced doses of RT had outcomes equivalent to patients who were treated with full-dose RT of 50.4 Gy.[8,10]
In the D9803 study of patients with intermediate-risk rhabdomyosarcoma, local control was 90% in 41 patients with Groups I and II alveolar rhabdomyosarcoma but lower in 280 patients with Group III embryonal (80%) and alveolar (83%) rhabdomyosarcoma. Histology, regional lymph node status, and primary site were not related to the likelihood of local failure; however, the local failure rate for 47 patients with retroperitoneal tumors was 33% (probably caused by tumors ≥5 cm in diameter), compared with 14% to 19% for patients with bladder/prostate, extremity, and parameningeal tumors. Tumor size was the strongest predictor of local failure (10% for patients with primary tumors <5 cm vs. 25% for larger tumors; P = .0004).[36][Level of evidence C2]
The treated radiation volume should be determined by the extent of tumor at diagnosis before surgical resection and before chemotherapy, including clinically involved regional lymph nodes. With conformal plans and image-guided RT, a margin of 1 cm to 1.3 cm to a clinical target volume or planning target volume may be used.[12] This clinical tumor volume can be modified on the basis of anatomical constraints, especially in situations where the tumor was pushing, rather than invading, the adjacent normal tissues, or when adjacent normal tissues are functionally critical (e.g., head and neck rhabdomyosarcoma). Thus, while the volume irradiated may be modified because of considerations for normal tissue tolerance, gross residual disease at the time of RT should receive full-dose RT. A reduction in volume after 36 Gy is appropriate in chemoresponsive disease for patients with noninvasive displacement (T1) that has regressed in size, but not for invasive tumors (T2). Gross residual disease still receives the full RT dose (50.4–59.4 Gy, the higher dose if >5 cm at diagnosis).
For involved nodal sites, the treated volume is defined as the extent of nodal involvement at diagnosis, factoring in changes in anatomy, plus a 3-cm margin superiorly and inferiorly in the direction of lymphatic drainage, or inclusion of the entire nodal chain where there is uncertainty.
For metastatic disease, the treated volume is the extent of metastases at diagnosis, with the exception of the lung or extensive brain metastases where the whole organ is irradiated, or diffuse peritoneal metastases where the entire peritoneal cavity is included. The use of novel techniques, such as stereotactic body RT to appropriate sites (e.g., bone or small volume soft tissue metastases), can be considered.
The timing of RT generally allows for chemotherapy to be given for up to 3 months before RT is initiated. RT is usually administered over 5 to 6 weeks (e.g., 1.8 Gy once per day, 5 days per week), during which time chemotherapy is usually modified to avoid the radiosensitizing agents dactinomycin, doxorubicin, and temsirolimus. Another consideration is the administration of RT before a planned second surgical excision that will be R0 or R1, particularly if RT might facilitate surgical resection to decrease the chances of loss of form or function. This approach is protocol dependent.
For metastatic sites, RT is usually given after 16 to 20 weeks of chemotherapy or, rarely, as consolidation at the completion of planned chemotherapy.
Thus, conventional RT remains the standard for treating patients who have rhabdomyosarcoma with gross residual disease.[38]
Brachytherapy, using either intracavitary or interstitial implants, is another method of local control that has been used in selected situations for children with rhabdomyosarcoma, especially for patients with primary tumors at a vaginal site [39-44] and selected bladder/prostate sites.[45][Level of evidence C1] This technique requires specialized technical skill and expertise and is limited to only a few institutions. In small series from one or two institutions, this treatment approach was associated with a high survival rate and retention of a functional organ or tissue in most patients.[40,46]; [47][Level of evidence C2] Other sites, especially head and neck, have also been treated with brachytherapy.[48]
Local control for pediatric solid tumors, such as rhabdomyosarcoma, often requires high doses of EBRT, which can cause unwanted damage to the normal tissues surrounding the tumor. After maximal tumor resection, delivering some or all of the radiation dose to the sites of highest recurrence risk intraoperatively could mitigate this issue, especially in relapsed patients who have received previous EBRT or very young children. This procedure is called intraoperative radiation therapy (IORT), and it can be used in challenging cases where standard full-dose EBRT is contraindicated. During IORT, a single large dose of radiation is administered during surgical exploration with direct visualization of the tumor bed and radiation field. IORT has been deemed safe to use for malignancies in pediatric patients, with minimal long-term toxicities.[49]
A single-institution retrospective study examined IORT (108 applications) in 96 pediatric patients with solid tumors (42 with rhabdomyosarcoma) who were treated from 1995 to 2022. The median age at time of IORT was 8 years (range, 0.8–20.9 years). The median follow-up was 16 months for all patients and 3 years for surviving patients. About one-half of patients (54%) were treated with upfront IORT to the primary tumor because of difficult circumstances, such as very young age or challenging anatomy. The median IORT dose was 12 Gy (range, 4–18 Gy). The cumulative incidence rate of local failure was 17% at 2 years and 23% at 5 years. A total of 15 patients (16%) experienced postoperative complications likely related to IORT.[50]
While IORT may be advantageous in the treatment of certain high-risk patients, there are important disadvantages. IORT is only available at certain institutions. In addition, while IORT minimizes the radiation dose to surrounding tissues by delivering one large fraction, any healthy tissue that is exposed is at risk of long-term treatment effects later in life.[50]
Very young children (aged ≤36 months) diagnosed with rhabdomyosarcoma pose a therapeutic challenge because of their increased risk of treatment-related morbidity.[9] Reduced radiation doses have been used when delayed surgery can provide negative margins. However, for most patients and those in whom surgical resection is not appropriate, higher doses of RT are given.[51] Radiation techniques are designed to maximize normal tissue sparing and should include conformal approaches, often with intensity-modulation or protons. When radiation is omitted, even in those with Stage 1 disease, there is a high risk of recurrence, with local recurrence being the most common, confirming the need for RT.[52-54]
Delayed primary excision may allow for a radiation dose reduction and has been studied in select patients.[8] However, the youngest patients frequently do not get appropriate RT because of concerns about normal tissue toxicity, and these are the best patients for whom surgical resection by delayed primary excision is a particularly important consideration. Local control can be achieved by both RT and surgery. Both treatments are optimal, but at least one approach is necessary in addition to chemotherapy. Local control rates from delayed primary excision and reduced-dose RT are equivalent to that from RT alone.[8]
In studies of infants younger than 1 or 2 years, 77 patients with nonmetastatic rhabdomyosarcoma were included. These studies showed 5-year failure-free survival (FFS) rates of 57% to 68% and OS rates of 76% to 82%.[55] Most failures were local, often because RT was withheld in violation of protocol guidelines. In contrast, for infants treated according to guidelines, both FFS and OS were clearly superior.[56] This experience has been confirmed for children up to age 2 years.[55] Consequently, the COG recommends treating children aged 2 years or younger with the same guidelines as recommended for children older than 2 years.
Local control of primary disease in rhabdomyosarcoma has evolved with the use of more effective chemotherapy protocols, improved surgical approaches and techniques, and improvements in RT, including better definition of therapy fields, tailored dosing, and new techniques such as IMRT, brachytherapy, and proton therapy. Data are predominantly derived from retrospective reviews of primary tumor sites from cooperative group studies, including the IRSG, COG, EpSSG, CWS, Gesellschaft für Pädiatrische Onkologie und Hämatologie, International Society for Pediatric Oncology (SIOP) Malignant Mesenchymal Tumour (MMT), and the Associazione Italiana di Ematologia e Oncologia Pediatrica Soft Tissue Sarcoma Committee. These groups created the International Soft Tissue Sarcoma Consortium (INSTRuCT) and agreed to form a single data commons by merging multiple cooperative group databases. Leaders of INSTRuCT have initiated efforts to define international consensus statements for approaches to several primary tumor sites, predominantly through their expert review of published data, sometimes enhanced with new analyses of merged data.
Primary sites for childhood rhabdomyosarcoma within the head and neck include the orbit; nonorbital head and neck and cranial parameningeal; and nonparameningeal, nonorbital head and neck. Specific considerations for the surgical and radiotherapeutic management of tumors arising at each of these sites are discussed below.
For patients with head and neck primary tumors that are considered unresectable, chemotherapy and RT with organ preservation are the mainstay of primary management.[57-62] Several studies have reported excellent local control in patients with rhabdomyosarcoma of the head and neck treated with IMRT, fractionated stereotactic radiation therapy, or proton RT, and chemotherapy. Further study is needed, but the use of IMRT and chemotherapy in patients with head and neck rhabdomyosarcoma may result in less-severe late effects.[63-65]; [66][Level of evidence C1]
Rhabdomyosarcomas of the orbit should not undergo exenteration, but biopsy is needed for diagnosis.[67,68] Biopsy is followed by chemotherapy and RT, with orbital exenteration reserved for the small number of patients with locally persistent or recurrent disease.[59,69] RT and chemotherapy are the standard of care, with survival rates exceeding 90% to 95%. When RT is omitted, there is risk of local relapse. For patients with orbital tumors, precaution should be taken to limit the RT dose to the lens, conjunctiva, and cornea.
The COG investigators have shown that patients with embryonal rhabdomyosarcoma of the orbit who achieve a complete response to induction chemotherapy have improved local control after radiation therapy of 45 Gy, compared with patients who fail to achieve a complete response.[70][Level of evidence B4] For patients in whom a complete response has not been achieved with induction chemotherapy, 50.4 Gy of RT is recommended by the investigators.
The COG studied a lower dose of cyclophosphamide to reduce the risk of infertility. In the COG ARST0331 (NCT00075582) trial, only four cycles of therapy contained cyclophosphamide, for a total cyclophosphamide exposure of 4.8 g/m2. Sixty-two patients with Group III orbital embryonal rhabdomyosarcoma were treated. None of the 15 patients with radiographic complete response (CR) had local recurrences, compared with 6 of the 38 patients who had less than a CR after 12 weeks of vincristine, dactinomycin, and cyclophosphamide (VAC) chemotherapy (P = .11). The authors concluded that for patients with Group III orbital embryonal rhabdomyosarcoma achieving a CR after VAC chemotherapy that includes modest-dose cyclophosphamide, 45 Gy of RT may be sufficient for durable FFS. However, for patients with less than a CR who were treated with the ARST0331 systemic therapy, a radiation dose of 50.4 Gy or a higher dose of cyclophosphamide may be needed to achieve the control rate reported in the IRS-IV trial.[70][Level of evidence B4]
Long-term outcomes were evaluated in 218 patients with orbital rhabdomyosarcoma enrolled in COG clinical trials between 1997 and 2013. The 192 patients with low-risk orbital rhabdomyosarcoma (clinical groups I–III with mostly embryonal histology treated on the low-risk D9602 and ARST0331 studies) had 10-year EFS and OS rates of 85.5% (95% CI, 77.0%–94.0%) and 95.6% (95% CI, 90.8%–100.0%), respectively. The 26 patients with non–low-risk orbital rhabdomyosarcoma (mostly tumors with alveolar histology that were treated with more intensive intermediate-risk protocols [D9802, D9803 and ARST0531]), had 5-year EFS and OS rates of 88.5% (95% CI, 75.6%–100.0%) and 95.8% (95% CI, 87.7%–100.0%), respectively. Twenty-eight patients experienced a recurrence, including 25 who were treated in low-risk trials (6 patients did not receive radiation therapy during initial therapy). Twenty-seven recurrences were local. One metastatic recurrence occurred in a patient with Group III, PAX3::FOXO1 fusion–positive alveolar rhabdomyosarcoma. Patients with recurrent orbital rhabdomyosarcoma had a 10-year OS rate of 69.4% (95% CI, 50.0%–88.8%) from time of recurrence, showing that a significant number of patients with recurrent orbital rhabdomyosarcoma may achieve long-term survival.[71]
If the tumors are nonorbital and cranial parameningeal (arising in the middle ear/mastoid, nasopharynx/nasal cavity, paranasal sinus, parapharyngeal region, or pterygopalatine/infratemporal fossa), a magnetic resonance imaging (MRI) scan with contrast of the primary site and brain should be obtained to check for presence of base-of-skull erosion and possible extension onto or through the dura.[60,72,73] If skull erosion and/or transdural extension is equivocal, a computed tomography (CT) scan with contrast of the same regions is indicated. Also, if there is any suspicion of extension down the spinal cord, an MRI scan with contrast of the entire cord should be obtained. The cerebrospinal fluid (CSF) should be examined for malignant cells in patients with high-risk parameningeal tumors. Because complete removal of these tumors is not feasible, owing to their location, the initial surgical procedure for these patients is usually only a biopsy for diagnosis.
Nonorbital head and neck rhabdomyosarcomas, including cranial parameningeal tumors, are optimally managed by conformal RT and chemotherapy. Patients with parameningeal disease with intracranial extension bordering the primary tumor and/or signs of meningeal impingement (i.e., cranial base bone erosion and/or cranial nerve palsy) do not require whole-brain irradiation or intrathecal therapy, unless tumor cells are present in the CSF at diagnosis.[72] Patients should receive RT to the site of primary tumor with a 1.5-cm margin to include the meninges adjacent to the primary tumor and the region of intracranial extension, if present, with a 1.5-cm margin.[73]
Evidence (timing of RT for nonorbital and cranial parameningeal tumors):
Children who present with tumor cells in the CSF (Stage 4) may or may not have other evidence of diffuse meningeal disease and/or distant metastases. In a review of experience from IRSG protocols II though IV, eight patients had tumor cells in the CSF at diagnosis. Three of four patients without other distant metastases were alive at 6 to 16 years after diagnosis, as was one of the four patients who had concomitant metastases elsewhere.[80]
Patients may also have multiple intraparenchymal brain metastases from a distant primary tumor. They may be treated with central nervous system–directed RT in addition to treatment with chemotherapy and RT for the primary tumor. Craniospinal axis RT may also be indicated.[81,82]
For nonparameningeal, nonorbital head and neck tumors, wide excision of the primary tumor (when feasible without functional impairment) and ipsilateral neck lymph node sampling of clinically involved nodes may be appropriate but requires postoperative RT if margins or nodes are positive.[83]; [84][Level of evidence C1] Narrow resection margins (<1 mm) are acceptable because of anatomical restrictions. Cosmetic and functional factors should always be considered, but with modern techniques, complete resection in patients with superficial tumors is consistent with good cosmetic and functional results.
The EpSSG RMS-2005 (NCT00379457) study prospectively enrolled 165 patients with localized head and neck, nonparameningeal rhabdomyosarcoma. Local therapy included surgery (58%) and/or RT (72%). Chemotherapy was given according to the patient’s risk group. Low-risk patients received vincristine and dactinomycin (VA) therapy. High-risk patients were randomly assigned to receive either neoadjuvant therapy with ifosfamide, vincristine, and dactinomycin (IVA) or IVA and doxorubicin for four courses followed by five courses of IVA. The 5-year EFS rate was 75% (95% CI, 67.3%–81.2%), and the OS rate was 84.9% (95% CI, 77.5%–89.7%). Favorable histology was associated with a better EFS rate (82.3% vs. 64.6%, P = .02), and nodal spread was associated with a worse OS rate (88.6% vs. 76.1%, P = .04). Locoregional relapse/progression was the main tumor failure event (84% of events).[85][Level of evidence B4]
Specialized, multidisciplinary surgical teams also have performed resections of anterior skull-based tumors in areas previously considered inaccessible to definitive surgical management, including the nasal areas, paranasal sinuses, and temporal fossa. However, these procedures should be considered only in children with recurrent locoregional disease or residual disease after chemotherapy and RT.
A pooled analysis of 642 patients from four international cooperative groups in Europe and North America was performed to identify prognostic factors in patients with localized extremity rhabdomyosarcoma. Regional lymph node involvement was approximately 2.5 times higher with alveolar rhabdomyosarcoma than with embryonal rhabdomyosarcoma. The 5-year OS rate was 67%. Multivariate analysis showed that decreased OS was correlated with age older than 3 years, T2 invasive disease and N1 nodal status, incomplete initial surgery, treatment before 1995, and treatment by European groups. This analysis also suggested that duration of chemotherapy might have an impact on outcome in these patients.[86]
Primary re-excision before initiation of chemotherapy (i.e., not delayed) may be appropriate in patients whose initial surgical procedure leaves microscopic residual disease that is deemed resectable by a second procedure without loss of cosmesis or function.[5] Chemotherapy or delayed primary excision does not improve outcome over chemotherapy and RT.[8]
Delayed primary excision has been studied in patients with extremity tumors enrolled in the COG intermediate-risk rhabdomyosarcoma trials. Two COG studies (D9803 and ARST0531 [NCT00354835]) were pooled to assess the benefit of delayed primary excision. In the D9803 study, local control with RT after a partial or complete excision was completed at week 12. In the ARST0531 study, RT was done upfront at week 4. Patients with bladder or prostate rhabdomyosarcoma who received a delayed primary excision had no difference in survival, whereas patients with extremity rhabdomyosarcoma had an improved OS with delayed primary excision. The delayed primary excision strategy with a reduction in RT dose resulted in superior OS for those sites.[8,10] Delayed primary excision may be most appropriate for infants because the late effects of RT are more severe than in older patients; thus, even a moderate reduction in radiation dose is desirable. For more information, see the Surgery (Local Control Management) section.
IMRT can be used to spare the bone yet provide optimal soft tissue coverage in extremity rhabdomyosarcoma. Complete primary tumor removal from the hand or foot is not feasible in most cases because of functional impairment.[87][Level of evidence C1] For children presenting with a primary tumor of the hands or feet, COG studies have shown a 10-year local control rate of 100% using RT along with chemotherapy, avoiding amputation in these children.[88][Level of evidence C1] Definitive RT and chemotherapy for Group III tumors resulted in a local control rate of 90% to 95% in the IRS-IV trial.[37]
Because of the significant incidence of regional nodal spread in patients with extremity primary tumors (often without clinical evidence of involvement) and because of the prognostic and therapeutic implications of nodal involvement, extensive pretreatment assessment of regional and in-transit nodes is warranted.[89-93]; [94][Level of evidence C2] In-transit nodes are defined as epitrochlear and brachial for upper-extremity tumors and popliteal for lower-extremity tumors. Regional lymph nodes are defined as axillary/infraclavicular nodes for upper-extremity tumors and inguinal/femoral nodes for lower-extremity tumors.
Positron emission tomography (PET) scanning is recommended for evaluation and staging of extremity primary tumors before initiation of therapy [94] and is useful in RT treatment planning.[95]
For patients enrolled in clinical trials, the COG-STS recommends biopsy of all enlarged or clinically suspicious lymph nodes, if possible, without delay in therapy or adverse functional outcome. If biopsy is not feasible, clinically abnormal nodes need to be included in the RT treatment plan.
In the trunk and extremity, if no enlarged lymph nodes are identified in the draining nodal basin, a sentinel lymph node biopsy is recommended. This type of biopsy is a more accurate way of assessing regional lymph nodes than random lymph node sampling. Techniques for sentinel lymph node biopsy are standardized and should be completed by an experienced surgeon.[92,96-102]
In a single-institution study of 28 patients aged 6 months to 32 years with soft tissue sarcomas, but not confined to rhabdomyosarcoma, sentinel lymph node biopsy was prospectively compared with PET-CT scan for detection of lymph node metastases. Forty-three percent of patients (3 of 7) with proven malignant sentinel lymph nodes had negative cross-sectional and functional imaging (PET-CT). Also, PET-CT suggested nodal involvement in 14 patients, whereas only 4 of those were proven to have metastatic disease. The study does not address relapse rate or follow-up in these patients. Therefore, the use of PET-CT staging to diagnose lymph node disease in soft tissue sarcomas is of uncertain utility.[103]
Primary sites for childhood rhabdomyosarcoma within the trunk include the chest wall or abdominal wall, intrathoracic or intra-abdominal area, biliary tree, and perineum or anus. Specific considerations for the surgical and radiotherapeutic management of tumors arising at each of these sites are discussed below.
The surgical management of patients with lesions of the chest wall or abdominal wall follows the same guidelines as those used for lesions of the extremities (i.e., wide local excision and an attempt to achieve negative microscopic margins if cosmetic and functional outcomes are acceptable).[104] These resections may require use of prosthetic materials for subsequent reconstruction.
Initial primary resection is performed if there is a realistic expectation of achieving negative margins (R0 resection). However, most patients who present with large tumors in these sites have localized disease that is unresectable at diagnosis but may become amenable to resection with negative margins after preoperative chemoradiation therapy. These patients may have excellent long-term survival.[104-107]
Chest wall rhabdomyosarcoma, which is usually Group III, does not require R0 resection (no microresidual disease) at delayed primary resection. The COG data show equivalent survival for R0 and R1 (microresidual disease at the margin) resections in chest wall rhabdomyosarcoma, likely because of the addition of postoperative RT.[107] Aggressive resections at diagnosis before chemotherapy are not necessary because rhabdomyosarcoma is chemosensitive and radiosensitive.
Intrathoracic or intra-abdominal sarcomas may not be resectable at diagnosis because of the massive size of the tumor and extension into vital organs or vessels.[108]
For patients with initially unresectable retroperitoneal/pelvic tumors, complete surgical removal after induction chemotherapy, with or without RT, offers a significant survival advantage (73% vs. 34%–44% without removal).[108]
Evidence (chemotherapy with or without RT followed by surgery):
With rhabdomyosarcoma of the biliary tree, total resection at diagnosis is rarely feasible. The standard treatment includes chemotherapy and RT. Outcomes for patients with this primary tumor site were considered favorable despite residual disease after surgery;[112] however, an analysis of COG low-risk studies found that patients with this site had suboptimal outcomes.[113] The CWS also reported poorer outcomes,[114] confirmed by a systematic review and meta-analysis.[115] The COG now recommends that this site be classified as unfavorable. External biliary drains significantly increase the risk of postoperative infectious complications. Thus, external biliary drainage is not warranted.[112]
Evidence (chemotherapy, surgery, and RT):
Patients with rhabdomyosarcoma arising from tissue around the perineum or anus often present with advanced disease. These patients have a high likelihood of regional lymph node involvement, and about half of the tumors have alveolar histology.[116] The high frequency of nodal involvement and the prognostic association between nodal involvement and poorer outcome support the recommendation to sample the regional lymph nodes.[117] When feasible and without unacceptable morbidity, removing all gross tumor before chemotherapy may improve the likelihood of cure; however, chemotherapy and RT remain the standard of care. With the goal of organ preservation, patients with tumors of the perineum or anus are preferentially managed with chemotherapy and RT without aggressive surgery, as aggressive surgery may result in the loss of sphincter control. Very aggressive surgery is not indicated because of multiple critical structures that limit the ability to achieve negative margins near the anus and urethra.[117]
Primary sites for childhood rhabdomyosarcoma within the genitourinary system include the paratesticular area, bladder, prostate, kidney, vulva, vagina, and uterus. Specific considerations for the surgical and radiotherapeutic management of tumors arising at each of these sites are discussed below.[120]
Recommendations for paratesticular primary tumors are primarily based on the results from cooperative group trials and a recent INSTRuCT consensus opinion.[121]
Lesions occurring adjacent to the testis or spermatic cord and up to the internal inguinal ring should be removed by orchiectomy with resection of the spermatic cord, using an inguinal incision with proximal vascular control (i.e., radical orchiectomy).[122] Resection of hemiscrotal skin is required when there is tumor fixation or invasion.
Hemiscrotectomy had been recommended by the COG, German groups, and Italian groups when a previous transscrotal biopsy had been performed. A retrospective German CWS study of 28 patients with embryonal rhabdomyosarcoma found a 5-year EFS rate of 91.7% in 12 patients with an initial transscrotal excision followed by hemiscrotectomy, while the 5-year EFS rate was 93.8% in 16 patients without subsequent hemiscrotectomy. All of these patients also received chemotherapy with vincristine, dactinomycin, an alkylating agent, and other agents.[123][Level of evidence C2]
A retrospective study examined 842 patients with localized paratesticular rhabdomyosarcoma who were enrolled in COG, CWS, EpSSG, Italian Cooperative Group, and MMT studies from 1988 to 2013. Of all patients, 7.7% had a transscrotal resection; however, this surgical factor did not contribute to an inferior EFS in stratified univariable and multivariable analysis.[124] A COG study evaluated 279 patients with paratesticular rhabdomyosarcoma. The study also found that hemiscrotectomy did not improve outcome after transscrotal violation.[125][Level of evidence C1] These findings support the consensus statement from INSTRuCT that hemiscrotectomy is no longer recommended for scrotal violation.[121]
The EpSSG RMS-2005 (NCT00379457) study enrolled 237 patients with paratesticular tumors. Seventy-five patients (32%) had an inappropriate first surgery, defined as tumorectomy without orchidectomy, transscrotal orchidectomy without an inguinal approach, or biopsy in a resectable tumor. These patients required intensified therapy to maintain excellent OS and EFS. Ten patients required additional local surgery and intensified chemotherapy.[126]
For patients with incompletely removed paratesticular tumors that require RT, temporarily repositioning the contralateral testicle into the adjacent thigh before scrotal radiation may preserve hormone production; however, more data are needed.[127][Level of evidence C1] A retrospective review of 49 patients with paratestis rhabdomyosarcoma referred to Memorial Sloan Kettering Cancer Center found that 20 patients had scrotal violation as a part of their original surgery. Fifteen of these patients underwent salvage surgery or RT. Eleven of these patients had continuous PFS, whereas four of the five patients who were treated without a salvage procedure developed recurrent disease.[128][Level of evidence C2]
Paratesticular tumors have a relatively high incidence of lymphatic spread (26% in IRS-I and IRS-II).[89] All patients with paratesticular primary tumors should have thin-cut abdominal and pelvic CT scans with intravenous contrast to evaluate nodal involvement. For patients who have Group I disease, are younger than 10 years, and in whom CT scans show no evidence of lymph node enlargement, retroperitoneal node biopsy/sampling is unnecessary, but a repeat CT scan every 3 months is recommended.[129,130] For patients with suggestive or positive CT scans, retroperitoneal, ipsilateral, infra-renal vein lymph node sampling of 10 to 12 nodes (but not formal node dissection) is recommended, and treatment is based on the findings of this procedure.[4,38,131] Patients with suspicious or documented retroperitoneal/pelvic lymph nodes require nodal RT.
In patients aged 10 years and older, only 9% will have clinical or radiological evidence of retroperitoneal node enlargement. However, pathological evaluation has shown that imaging alone will miss 50% of nodal disease. Therefore, patients aged 10 years and older should have an ipsilateral, nerve-sparing retroperitoneal node dissection, regardless of imaging findings.[132] Staging ipsilateral retroperitoneal lymph node sampling is currently required for all children aged 10 years and older with paratesticular rhabdomyosarcoma on COG-STS studies.
Many European investigators relied on radiographic, rather than surgical-pathological assessment, for retroperitoneal lymph node involvement.[122,129] European studies, as well as an international pooled data analysis, demonstrated worse outcomes in this patient population when surgical lymph node evaluation was not performed.[124,126,133] On the basis of these results and with the high relapse rate and worse EFS in Stage N0 patients, investigators from SIOP, EpSSG, and COG recommended surgical resection, in the form of ipsilateral retroperitoneal lymph node sampling of clinically normal nodes (not enlarged by CT or MRI), in patients aged 10 years and older with paratesticular rhabdomyosarcoma.[124] A consensus document regarding paratesticular rhabdomyosarcoma from all North American and European cooperative groups concurred that all patients aged 10 years or older should undergo ipsilateral, infra-renal vein, retroperitoneal surgical lymph node evaluation by sampling 7 to 12 nodes or a nerve-sparing dissection.[121]
Evidence (lymph node sampling):
RT should be considered for patients whose nodes are biopsy positive.
Bladder preservation is a major goal of therapy for patients with tumors arising in the bladder and/or prostate. Two reviews provide information about the historical, current, and future treatment approaches for patients with bladder and prostate rhabdomyosarcomas.[135,136]
The initial surgical procedure in most patients consists of a biopsy, which often can be performed using ultrasound guidance or cystoscopy, or by a direct-vision transanal route.[137]
In rare cases, the tumor is confined to the dome of the bladder and can be completely resected, leaving a functional bladder intact. Otherwise, to preserve a functional bladder in patients with gross residual disease, chemotherapy and RT have been used in North America and some parts of Europe to reduce tumor bulk.[138,139] This is sometimes followed, when necessary, by a more limited surgical procedure such as partial cystectomy.[140] Early experience with this approach was disappointing, with only 20% to 40% of patients with bladder/prostate tumors alive and with functional bladders 3 years after diagnosis (3-year OS rate was 70% in IRS-II).[140,141] The later experience from the IRS-III and IRS-IV studies, which used more intensive chemotherapy and RT and had a greater emphasis on bladder preservation, showed 55% of patients alive with functional bladders at 3 years postdiagnosis, with 3-year OS rates exceeding 80%.[139,142,143]
In a prospective registry study of 19 patients (median age, 1.8 years at diagnosis; range, 0.5–5.0 years) who were treated with proton therapy, the 5-year OS and PFS rates were 76%. The 5-year local-control rate was 76%. Tumor size predicted the local-control rate, with 5-year local-control rates of 43% for patients whose tumors were larger than 5 cm versus 100% for patients whose tumors were 5 cm or smaller (P = .006). The four patients who had a relapse all died.[144]
Patients with a primary tumor of the bladder or prostate who present with a large pelvic mass, resulting from a distended bladder caused by outlet obstruction at diagnosis, receive RT. The RT volume is defined by imaging studies after initial chemotherapy to relieve outlet obstruction. This approach to therapy remains generally accepted, with the belief that more effective chemotherapy and RT will continue to increase the frequency of bladder salvage.
For patients with biopsy-proven, residual malignant tumor after chemotherapy and RT, appropriate surgical management may include partial cystectomy, prostatectomy, or exenteration (usually approached anteriorly with preservation of the rectum). Very few studies report objective long-term assessment of bladder function. Urodynamic studies can accurately evaluate bladder function.[145]
An alternative strategy, used in European SIOP protocols, has been to avoid major radical surgery when possible and omit EBRT if complete disappearance of tumor can be achieved by chemotherapy and conservative surgical procedures. The goal is to preserve a functional bladder and prostate without incurring the late effects of RT or having to perform a total cystectomy/prostatectomy. From 1984 to 2003, 172 patients with nonmetastatic bladder and/or bladder/prostate rhabdomyosarcoma were enrolled in a SIOP-MMT study. Of the 119 survivors, 50% had no significant local therapy, and only 26% received RT. The 5-year OS rate was 77%.[146][Level of evidence C1]
Another alternative strategy in highly selected patients is to perform conservative surgery, followed by brachytherapy at a specialized center.[147]; [148][Level of evidence C2]; [149][Level of evidence C1] A prospective, nonrandomized analysis of this strategy reported the outcomes of 100 children. The 5-year disease-free survival rate was 84%, and the OS rate was 91%. At last follow-up, most survivors presented with only mild-to-moderate genitourinary sequelae and a normal diurnal urinary continence. Five patients required a secondary total cystectomy, three patients for a nonfunctional bladder and two patients for relapse. In another series, bladder-conserving surgery plus brachytherapy for boys with prostate or bladder-neck rhabdomyosarcoma led to excellent survival rates, bladder preservation, and short-term functional results.[45][Level of evidence C1]
In patients who have been treated with chemotherapy and RT for rhabdomyosarcoma arising in the bladder or prostate region, the presence of well-differentiated rhabdomyoblasts in surgical specimens or biopsies obtained after treatment does not appear to be associated with a high risk of recurrence and is not an indication for a major surgical procedure such as total cystectomy.[142,150,151] One study suggested that in patients with residual bladder tumors with histological evidence of maturation, additional courses of chemotherapy should be given before cystectomy is considered.[142] Surgery should be considered if malignant tumor cells do not disappear over time after initial chemotherapy and RT. Because of limited data, it is unclear whether this situation is analogous for patients with rhabdomyosarcoma arising in other parts of the body.
The kidney is rarely the primary site for sarcoma. Ten patients were identified among 5,746 eligible patients enrolled in IRSG protocols, including six with embryonal rhabdomyosarcoma and four with undifferentiated sarcoma. The tumors were large (mean widest diameter, 12.7 cm), and anaplasia was present in four (67%) patients. Of the patients with embryonal rhabdomyosarcoma, three Group I and Group II patients survived, one Group III patient died of infection, and two Group IV patients died of recurrent disease; these children were aged 5.8 and 6.1 years at diagnosis. This limited experience concluded that the kidney is an unfavorable site for primary sarcoma.[152]
For patients with genitourinary primary tumors of the vulva, vagina, or uterus, the initial surgical procedure is usually a vulvar or transvaginal biopsy. Initial radical surgery is not indicated for rhabdomyosarcoma of the vulva, vagina, or uterus.[4] Conservative surgical intervention for vaginal rhabdomyosarcoma, with primary chemotherapy and radiation (external beam or brachytherapy) for Group II or III disease results in excellent 5-year survival rates.[52,153,154][Level of evidence C1]
In the COG-ARST0331 study, there was an unacceptably high rate of local recurrences in girls with Group III vaginal tumors who did not receive RT.[52][Level of evidence C2] In 21 girls with genitourinary tract disease who were not treated with RT (mostly Group III vaginal primary tumors), the 3-year FFS rate was 57%, compared with 77% in the other 45 patients with non–female genitourinary primary tumors (P = .02).[53][Level of evidence B4] Therefore, the COG-STS recommended that RT be administered to patients with residual viable vaginal tumor, beginning at week 12.[54][Level of evidence C1]
Because of the small number of patients with uterine rhabdomyosarcoma, it is difficult to make a definitive treatment decision, but chemotherapy with or without RT is effective.[153,155] Twelve of 14 girls with primary cervical embryonal (mainly botryoid) rhabdomyosarcoma were disease-free after VAC chemotherapy and conservative surgery. Of note, two girls also had a pleuropulmonary blastoma and another had a Sertoli-Leydig cell tumor.[156] Exenteration is usually not required for primary tumors at these sites, but may be done if needed, with rectal preservation possible in most cases.
Four cooperative groups in the United States and Europe evaluated patients with localized vaginal or uterine tumors (N = 427). Some patients received initial RT for local control of residual disease after induction chemotherapy, while others had it later, or not at all if no demonstrable disease was found. The 10-year EFS rate was 74%, and the 10-year OS rate was 92%. Unfavorable factors were positive lymph node disease and uterine corpus primary site. There was no statistical difference in outcomes between patients who received early RT and patients who received later RT. About one-half of these patients were cured without radical surgery or systematic RT.[44][Level of evidence C1]
A study of five CWS trials (and one registry) included 67 patients with localized vaginal or uterine rhabdomyosarcoma diagnosed at a median age of 2.89 years (0.09–18.08). Multimodality treatment consisted of chemotherapy (n = 66), secondary surgery (n = 32), and RT (n = 11). The study reported the following results:[157][Level of evidence C1]
The INSTRuCT group summarized its consensus expert opinion about local treatment of female genital tract tumors as follows:[158]
For girls with genitourinary primary tumors who will receive pelvic irradiation, ovarian transposition (oophoropexy) before radiation therapy should be considered unless dose estimations suggest that ovarian function is likely to be preserved.[159] Alternatively, ovarian tissue preservation is under investigation and can be considered.[160]
Rhabdomyosarcoma occasionally arises in sites other than those previously discussed.
Patients with localized primary rhabdomyosarcoma of the brain can occasionally be cured using a combination of tumor excision, RT, and chemotherapy.[161][Level of evidence C2]
Patients with laryngeal rhabdomyosarcoma will usually be treated with chemotherapy and RT after biopsy in an attempt to preserve the larynx.[162]
Patients with diaphragmatic tumors often have locally advanced disease that is not grossly resectable initially because of fixation to adjacent vital structures such as the lung, great vessels, pericardium, and/or liver. In such circumstances, chemotherapy and RT should be initiated after diagnostic biopsy. Removal of residual tumor at a later date if clinically indicated could be considered.[163]
Two well-documented cases of primary ovarian rhabdomyosarcoma (one Stage III and one Stage IV) have been reported to supplement the eight previously reported patients. These two patients were alive at 20 and 8 months after diagnosis. Six of the previously reported eight patients had died of their disease.[164][Level of evidence C2] Treatment with combination chemotherapy, followed by removal of the residual mass or masses, can sometimes be successful.[164]
The EpSSG reported a retrospective analysis of ten patients with rhabdomyosarcoma and unknown primary sites, most of whom were adolescents (median age, 15.8 years; range, 4.6–20.4 years).[165] Nine patients had fusion-positive alveolar rhabdomyosarcoma. Seven patients had multiple organ involvement, two patients had only bone marrow disease, and one patient had only leptomeningeal dissemination. All patients received chemotherapy, four received radiation therapy, and none underwent surgery. Three patients underwent allogeneic bone marrow transplant. At the time of this analysis, only two patients were alive in complete remission: one who was treated with radiation therapy, and one who was treated with a bone marrow transplant.
Primary resection of metastatic disease at diagnosis (Stage 4, M1, Group IV) is rarely indicated. A site of gross disease is rarely cured with chemotherapy alone; thus, the COG recommends RT to sites of gross disease.
In the COG protocols, resection of the primary tumor in patients with metastatic disease may be considered before initiating chemotherapy if a complete resection is anticipated without the loss of form or function. After induction chemotherapy, delayed resection can be performed, with the same caveat regarding complete resection without loss of form or function, followed by RT of the primary tumor. The paradigm of aggressive local control of primary tumors in patients with metastases is supported by a European evaluation of 101 patients treated from 1998 to 2011 using MMT protocols. OS rates were best when both surgical resection and RT were combined (44%) versus surgical resection alone (19%) or RT alone (16%) (P < .006).[166][Level of evidence C1] Outcome also correlated with completeness of the surgical resection (R0, 41%; R1, 56%; R2, 20%; P < .03). Primary resection of metastatic disease at diagnosis (Stage 4, M1, Group IV) is rarely indicated. Treatment of metastatic disease occurs near the end of therapy using RT and, rarely, resection or other ablative techniques. The primary treatment for bony metastatic disease is RT.
Members of the EpSSG evaluated the role of indeterminate pulmonary nodules at diagnosis in patients with rhabdomyosarcoma. The criteria for indeterminate pulmonary nodules were one to four nodules smaller than 5 mm or one nodule measuring 5 mm to 10 mm. Of 316 patients, 67 patients had nodules and 249 patients did not have nodules. At a median follow-up of 75 months, the 5-year EFS rate was 77% for patients with nodules and 73.2% for patients without nodules (P = .68). The 5-year OS rate was 82% for patients with nodules and 80.8% for patients without nodules (P = .76). The authors concluded that there was no need to perform a biopsy on or upstage the patients with indeterminate pulmonary nodules at diagnosis.[167][Level of evidence C1]
Evidence (treatment of lung-only metastatic disease):
All children with rhabdomyosarcoma should receive chemotherapy. The intensity and duration of the chemotherapy are dependent on the Risk Group assignment.[170] For more information about Risk Groups, see Table 6.
Adolescents treated with chemotherapy for rhabdomyosarcoma experience less hematologic toxicity and more peripheral nerve toxicity than do younger patients.[171]
Cooperative group studies have defined low-risk patient populations who have better outcomes. The specific definition of the low-risk group is protocol dependent, and while outcomes have typically been excellent, some subgroups of low-risk patients have received relatively aggressive therapy. In the COG D9602 and ARST0331 studies, low-risk patients had localized (nonmetastatic) embryonal histology tumors in favorable sites that were grossly resected (Groups I and II), embryonal tumors in the orbit that were not completely resected (Group III), and localized tumors in unfavorable sites that were grossly resected (Groups I and II). Approximately 25% of newly diagnosed patients are low risk. For more information, see Table 5 in the Stage Information for Childhood Rhabdomyosarcoma section.
COG and EpSSG studies have evaluated two- and three-drug chemotherapy schedules with varying intensity of alkylator therapy and variations in length of therapy. The goals are to maximize cure rates while attempting to mitigate late effects of chemotherapy. These cooperative groups have evaluated different approaches in different patient subsets.
Evidence (chemotherapy for low-risk Group patients):
| Subset | Tumor Site | Tumor Size | Surgical-Pathological Group | Nodes |
|---|---|---|---|---|
| N0 = absence of nodal spread; N1 = presence of regional nodal spread beyond the primary site. | ||||
| A | Favorable | Any | I, IIA | N0 |
| Orbital | Any | I, II, III | N0 | |
| Unfavorable | ≤5 cm | I | N0 | |
| B | Favorable (orbital or nonorbital) | Any | IIB, IIC, III | N0, N1 |
| Unfavorable | <5 cm | II | N0 | |
| Unfavorable | >5 cm | I, II | N0, N1 | |
| Subset | Tumor Site | Tumor Size | Surgical-Pathological Group | Nodes |
|---|---|---|---|---|
| N0 = absence of nodal spread; N1 = presence of regional nodal spread beyond the primary site. | ||||
| 1 | Favorable | Any | I | N0 |
| II | N0, N1 | |||
| Orbital | Any | III | N0 | |
| Unfavorable | <5 cm | I, II | N0 | |
| 2 | Favorable (nonorbital) | Any | III | N0, N1 |
| Unfavorable | >5 cm | I, II | N0, N1 | |
Approximately 50% of newly diagnosed patients are in the intermediate-risk category. In North America, VAC is the standard multiagent chemotherapy regimen used for intermediate-risk patients. In Europe, ifosfamide is typically used in place of cyclophosphamide. COG studies for intermediate-risk rhabdomyosarcoma use VAC plus vincristine and irinotecan (VI).
Evidence (chemotherapy for intermediate-risk Group patients):
The patients classified as high risk by the EpSSG had:
These patients would be classified as intermediate risk by the COG.
Patients received initial treatment with cycles of IVA—ifosfamide (6 g/m2), dactinomycin (1.5 mg/m2), and vincristine (1.5 mg/m2)—for 7 weeks, followed by randomization to continue IVA or IVA with doxorubicin (60 mg/m2). IVA represents a lower alkylating agent dose than the cyclophosphamide dose of 2.2 g/m2 used in COG rhabdomyosarcoma studies. Patients assessed to be in complete remission at the end of initial therapy were randomly assigned to either observation or the addition of six 4-week cycles of maintenance chemotherapy with vinorelbine (25 mg/m2) on days 1, 8, and 15 of each cycle with continuous daily cyclophosphamide (25 mg/m2/day).
Approximately 20% of Group III patients have a residual mass at the completion of therapy. The presence of a residual mass had no adverse prognostic significance.[183,191] Aggressive alternative therapy is not warranted for patients with rhabdomyosarcoma who have a residual mass at the end of planned therapy unless it has biopsy-proven residual malignant disease. A 2009 analysis by the COG reported that for Group III patients, best response (complete resolution versus partial response or no response) to initial chemotherapy had no impact on overall outcome.[191] In 2020, the COG reported a retrospective analysis of 601 patients with clinical Group III disease. The patients were enrolled in two COG studies (ARST0531 [n = 285] and D9803 [n = 316]) and completed all protocol therapy without developing progressive disease.[192] Response was defined radiographically: 393 patients had complete resolution (65.4%), and 208 patients had partial response/no response (34.6%). The overall 5-year FFS rate was 75% for patients who achieved complete resolution and 66.5% for those who had a partial response/no response (adjusted [adj.] P = .094). Radiographic response was not associated with OS at any site of disease (adj. P = .21). Resection of the end-of-therapy mass did not improve FFS (P = .12) or OS (P = .37). Patients with parameningeal primary sites who achieved complete resolution had significantly improved FFS (adj. P = .037), while those with nonparameningeal primary sites had similar outcomes (adj. P = .47). In conclusion, complete resolution status at the end of protocol therapy in patients with parameningeal clinical Group III rhabdomyosarcoma was associated with improved FFS but not OS.
While induction chemotherapy is commonly administered for 9 to 12 weeks, 2.2% of patients with intermediate-risk rhabdomyosarcoma in the IRS-IV and D9803 studies were found to have early disease progression and did not receive their planned local control therapy.[186][Level of evidence A1]
High-risk patients have metastatic disease in one or more sites at diagnosis (Stage IV, Group IV). These patients continue to have a relatively poor prognosis with current therapy (5-year survival rate of ≤50%), and new approaches to treatment are needed to improve survival in this group.[169,193,194] Two retrospective studies have examined patients who present with metastases limited to the lungs;[168,169] results are summarized in the Metastatic disease section of this summary.
The standard systemic therapy for children with metastatic rhabdomyosarcoma is the three-drug combination of VAC.
Evidence (chemotherapy for high-risk Group patients):
The analysis identified several adverse prognostic factors (Oberlin risk factors):
The EFS rate at 3 years depended on the number of adverse prognostic factors:[195][Level of evidence C1]
Many clinical trials have tried to improve outcomes by adding additional agents to standard VAC chemotherapy or substituting new agents for one or more components of VAC chemotherapy. To date, no chemotherapy regimens have been shown to be more effective than VAC, including the following:
The following results were observed:
The following results were observed:
Information about NCI-supported clinical trials can be found on the NCI website. For information about clinical trials sponsored by other organizations, see the ClinicalTrials.gov website.
The following is an example of a national and/or institutional clinical trial that is currently being conducted:
| Subset | Fusion status | Tumor Site | Tumor Size | Surgical-pathological Group | MYOD1 or TP53 Status |
|---|---|---|---|---|---|
| Very Low Risk | Negative | Favorable | Any | I | Wild-type |
| Low Risk | Negative | Favorable | Any | II | Wild-type |
| Unfavorable | >5 cm | I, II | |||
| Orbit | Any | III |
Use our advanced clinical trial search to find NCI-supported cancer clinical trials that are now enrolling patients. The search can be narrowed by location of the trial, type of treatment, name of the drug, and other criteria. General information about clinical trials is also available.
Although patients with progressive or recurrent rhabdomyosarcoma sometimes achieve complete remission with secondary therapy, the long-term prognosis is usually poor.[1,2] Rhabdomyosarcoma may relapse locally or in the lung, bone, or bone marrow. Less commonly, the site of first recurrence can be the breast in adolescent females or the liver.[3]
The following studies reported on the prognostic factors associated with progressive or recurrent disease:
The selection of additional treatment depends on many factors, including the site(s) of progression or recurrence, previous treatment, and individual patient considerations.
Treatment options for progressive or recurrent childhood rhabdomyosarcoma include the following:
The following chemotherapy regimens have been used to treat progressive or recurrent rhabdomyosarcoma:
Very intensive chemotherapy followed by autologous bone marrow reinfusion is also under investigation for patients with recurrent rhabdomyosarcoma. However, a review of the published data did not determine a significant benefit for patients who underwent this salvage treatment approach.[42-44]
Patients or families who desire additional disease-directed therapy should consider entering trials of novel therapeutic approaches because no standard agents have demonstrated clinically significant activity.
Regardless of whether a decision is made to pursue disease-directed therapy at the time of progression, palliative care remains a central focus of management. This ensures that quality of life is maximized while attempting to reduce symptoms and stress related to the terminal illness.
Palliation of painful lesions in children with recurrent or progressive disease can be achieved using a short course (10 or fewer fractions) of radiation therapy. In a retrospective study of 213 children with various malignancies, who were treated with short course radiation therapy, 85% of patients had complete or partial pain relief, with low levels of toxicity.[45]
Information about National Cancer Institute (NCI)–supported clinical trials can be found on the NCI website. For information about clinical trials sponsored by other organizations, see the ClinicalTrials.gov website.
The following are examples of national and/or institutional clinical trials that are currently being conducted:
Patients with tumors that have molecular variants addressed by open treatment arms in the trial may be enrolled in treatment on Pediatric MATCH. Additional information can be obtained on the NCI website and ClinicalTrials.gov website.
Use our advanced clinical trial search to find NCI-supported cancer clinical trials that are now enrolling patients. The search can be narrowed by location of the trial, type of treatment, name of the drug, and other criteria. General information about clinical trials is also available.
The PDQ cancer information summaries are reviewed regularly and updated as new information becomes available. This section describes the latest changes made to this summary as of the date above.
General Information About Childhood Rhabdomyosarcoma
Added text about the results of a Children's Oncology Group (COG) study that evaluated the impact of germline variants in cancer predisposition genes on patient outcomes in 580 individuals with rhabdomyosarcoma (cited Martin-Giacalone et al. as reference 26).
Treatment of Childhood Rhabdomyosarcoma
Added Intraoperative radiation therapy (IORT) as a new subsection.
Added text about the results of a randomized, open-label, phase III COG trial that evaluated the addition of temsirolimus to chemotherapy in pediatric patients with intermediate-risk rhabdomyosarcoma (cited Gupta et al. as reference 190).
This summary is written and maintained by the PDQ Pediatric Treatment Editorial Board, which is editorially independent of NCI. The summary reflects an independent review of the literature and does not represent a policy statement of NCI or NIH. More information about summary policies and the role of the PDQ Editorial Boards in maintaining the PDQ summaries can be found on the About This PDQ Summary and PDQ® Cancer Information for Health Professionals pages.
This PDQ cancer information summary for health professionals provides comprehensive, peer-reviewed, evidence-based information about the treatment of childhood rhabdomyosarcoma. It is intended as a resource to inform and assist clinicians in the care of their patients. It does not provide formal guidelines or recommendations for making health care decisions.
This summary is reviewed regularly and updated as necessary by the PDQ Pediatric Treatment Editorial Board, which is editorially independent of the National Cancer Institute (NCI). The summary reflects an independent review of the literature and does not represent a policy statement of NCI or the National Institutes of Health (NIH).
Board members review recently published articles each month to determine whether an article should:
Changes to the summaries are made through a consensus process in which Board members evaluate the strength of the evidence in the published articles and determine how the article should be included in the summary.
The lead reviewers for Childhood Rhabdomyosarcoma Treatment are:
Any comments or questions about the summary content should be submitted to Cancer.gov through the NCI website's Email Us. Do not contact the individual Board Members with questions or comments about the summaries. Board members will not respond to individual inquiries.
Some of the reference citations in this summary are accompanied by a level-of-evidence designation. These designations are intended to help readers assess the strength of the evidence supporting the use of specific interventions or approaches. The PDQ Pediatric Treatment Editorial Board uses a formal evidence ranking system in developing its level-of-evidence designations.
PDQ is a registered trademark. Although the content of PDQ documents can be used freely as text, it cannot be identified as an NCI PDQ cancer information summary unless it is presented in its entirety and is regularly updated. However, an author would be permitted to write a sentence such as “NCI’s PDQ cancer information summary about breast cancer prevention states the risks succinctly: [include excerpt from the summary].”
The preferred citation for this PDQ summary is:
PDQ® Pediatric Treatment Editorial Board. PDQ Childhood Rhabdomyosarcoma Treatment. Bethesda, MD: National Cancer Institute. Updated <MM/DD/YYYY>. Available at: https://www.cancer.gov/types/soft-tissue-sarcoma/hp/rhabdomyosarcoma-treatment-pdq. Accessed <MM/DD/YYYY>. [PMID: 26389243]
Images in this summary are used with permission of the author(s), artist, and/or publisher for use within the PDQ summaries only. Permission to use images outside the context of PDQ information must be obtained from the owner(s) and cannot be granted by the National Cancer Institute. Information about using the illustrations in this summary, along with many other cancer-related images, is available in Visuals Online, a collection of over 2,000 scientific images.
Based on the strength of the available evidence, treatment options may be described as either “standard” or “under clinical evaluation.” These classifications should not be used as a basis for insurance reimbursement determinations. More information on insurance coverage is available on Cancer.gov on the Managing Cancer Care page.
More information about contacting us or receiving help with the Cancer.gov website can be found on our Contact Us for Help page. Questions can also be submitted to Cancer.gov through the website’s Email Us.