1. Departments of Oncology and Surgery, The Children’s Hospital at Westmead, New South Wales.
2. Department of Radiotherapy, Westmead Hospital, New South Wales.
3. Department of Radiation Oncology, Westmead Hospital, New South Wales.
Rhabdomyosarcoma is the most common soft tissue sarcoma in childhood, accounting for approximately 6% of paediatric tumours. These tumours arise in various different locations, at all age groups in childhood and adolescence, and with various different histologic subtypes. As a result, there are a number of important considerations that affect the choice of local control measures. Our aim here is to explore these issues. Local control of rhabdomyosarcoma can rarely be achieved by chemotherapy and so local treatment in the form of surgery and/or radiotherapy is essential in most circumstances. Risk stratification which incorporates clinical group stage, TNM stage, histology and tumour location is an important consideration in determining the modality of local control. Other important considerations include resectability and age of the patient. The timing of local treatment is dependant on the tumour location, feasibility of complete resection without unacceptable loss of function or cosmesis, and the presence of an acute emergency such as spinal cord compression.
Rhabdomyosarcoma is a chemotherapy-responsive tumour, and all patients are treated with chemotherapy because this is a systemic disease.1 The principle North American chemotherapy regimen is vincristine, actinomycin-D and cyclophosphamide (VAC). In European studies, ifosfamide has been substituted for cyclophosphamide (IVA). The role of anthracycline drugs remains controversial; in randomised studies, the addition of doxorubicin or epirubicin did not lead to improved outcomes. However, these drugs do have activity against rhabdomyosarcoma, and some units routinely incorporate them into treatment protocols. A current European study is re-examining the role of doxorubicin, while a recent rhabdomyosarcoma trial within the Children’s Oncology Group for patients with ‘high risk disease’ incorporated doxorubicin in a single arm trial. This latter trial also added irinotecan, ifosfamide and etoposide; early outcome data appear promising.2 Irinotecan had earlier been shown to be active in a phase II trial given with vincristine,3 and is being studied in randomised fashion within an ongoing Children’s Oncology Group trial. Other active agents include carboplatin, topotecan, and melphalan.
The use of irinotecan as a radio-sensitiser to improve local control in patients with intermediate risk disease is being evaluated in the most recent Children’s Oncology Group study. Doxorubicin is a potent radiosensitiser, however anthracyclines are usually used only at the beginning of a radiotherapy regimen. The administration of anthracyclines concurrent with radiotherapy, or in the immediate post-radiation period, should be avoided as it often results in unacceptable augmentation of normal tissue radiation toxicity and a radiation recall phenomenon. Other novel radiosensitising agents, including idoxuridine, razoxane and ifosfamide have also been evaluated.4
Within North American studies, local therapy, comprising surgery and/or radiation therapy, has been applied systematically.
A different philosophy was adopted in European trials. Because of the long-term morbidities often associated with local therapies, a strategy of evaluating chemotherapy response prior to local therapy was adopted. Those patients, who responded promptly, and completely, did not undergo radiation therapy. The expectation was that there would be patients who had a recurrence, but that a second course of treatment, including a local therapy, might be able to achieve a durable second remission. Indeed, this is what the data showed, and so these European trials have lower event free survival than the North American trials, but equivalent overall survival rates confirmed that certain relapsing patients were indeed salvaged.5 It is clear that local treatment is essential for local control at least in certain settings.
Patients with rhabdomyosarcoma are allocated a pre-treatment clinical stage and a post surgical clinical group, both of which carry prognostic significance.
Staging is a modification of the UICC-TNM staging system and is based on site, size, clinical regional nodal status and distance spread. ‘Staging’ is clinical and should be performed by the responsible surgeon based on pre-operative imaging and physical findings (table 1). Intraoperative and/or pathologic results do not affect the stage. Site designation alters stage, with certain sites considered favourable and others unfavourable. Careful evaluation of clinical and/or imaging finding should precede multi-disciplinary site assignment.
The surgical-pathologic (clinical) ‘group’ is based on intraoperative findings and postoperative pathologic status, and includes final pathologic verification of margins, residual tumour, node involvement, and cytological examination of pleural, peritoneal and cerebrospinal fluid when applicable.
Final risk stratification within recent North American studies has combined group and stage with histological sub-type. Low, intermediate and high-risk groups are therefore defined (table 2). Similarly complex stratification takes place within European trials.
When possible and reasonable, a wide and complete resection of the primary tumour, with a surrounding envelope of normal tissue, should be performed as an initial and/or subsequent operation. This may be possible with extremity or trunk primaries, but is often not possible with head and neck, orbital and some genitourinary sites. Procedures which would lead to an unacceptable loss of function or cosmesis are not recommended.
Approximately half of all patients have unresectable tumours (clinical group 3) at presentation and 15% have metastatic disease (group 4). Microscopically complete (group 1) and incomplete (group 2) resections are achieved at diagnosis in 16% and 20% respectively.
Patients with unresectable tumours undergo biopsy. Adequate tissue needs to be obtained to facilitate immuno-histochemical and molecular studies. Fine needle biopsies are generally inadequate.
Selected sites require further surgical staging. Patients with extremity tumours should have aggressive sampling of relevant lymph nodes. Lymphoscintography may guide this nodal sampling, especially in truncal tumours; radical lymph node dissection is not performed. Patients over the age of 10 years with para-testicular rhabdomyosarcoma routinely undergo selective ipsilateral retroperitoneal lymph node dissection in North American trials.
The remainder of the staging studies include a computerised tomography scan of the chest, bone scan or positron emission tomography scan, bilateral bone marrow trephines and lumbar puncture in patients with para-meningeal tumours.
Sites considered to be parameningeal:
Surgery and external beam radiation therapy are the principle modalities. In general, surgery is preferred where wide resection can be obtained without unacceptable loss of function. Radiation therapy is effective for local control, however is generally associated with growth, developmental and cosmetic abnormalities, as well as a small risk of secondary malignant neoplasia, in the order of 2% long term.6
As discussed above, an initial wide resection of the primary tumour at diagnosis is optimal. Primary re-excision of a tumour is defined as a second attempt at complete resection before the initiation of any other forms of therapy. This should be encouraged when an initial excision results in positive margins and further resection can be accomplished without significant functional or cosmetic morbidity. This strategy has been shown to improve survival in selected tumours. In patients with microscopic or macroscopic residual disease, North American investigators have systematically applied external beam radiation therapy following a phase of induction chemotherapy.
More recently, the Children’s Oncology Group studied the role of a delayed, or “second look” operation to resect residual tumour at selected sites after induction chemotherapy, and before radiation therapy. The goals of second look surgery are to remove residual tumour and to determine pathological response. The rationale was that the tumour may become amenable to resection following chemotherapy, and that the second look procedure would improve local control and/or allow the use of a lower dose of radiation therapy. The radiotherapy dose was adjusted according to the completeness of the delayed resection – patients with gross residual disease received 50.4 Gy, microscopic residual received 41.4 Gy and those with a complete resection received 36 Gy. Seventy three patients with tumours at the selected primary sites (bladder/ extremity/trunk) underwent second look surgery and 84% of these achieved removal of all gross disease and were eligible for a reduced dose of radiation therapy. The authors concluded that a second look operation was feasible, and was able to be performed in approximately half of the patients with tumours at the selected sites. A majority of patients who underwent induction chemotherapy and delayed surgery were then eligible for radiotherapy dose reduction. Long-term follow-up of disease control is awaited.7
Radiation treatment, whether definitive or post-operative, may be delivered by external beam or brachytherapy. Brachytherapy involves the insertion of a radioactive source directly into the tumour or tumour bed, concentrating the radiation dose here rather than scattering radiation dose to surrounding structures. This technique is suitable when the area to treat is small and accessible to implantation or is in proximity to a body cavity.
External beam radiation is delivered by a linear accelerator on a daily outpatient basis and may require the use of general anaesthesia to ensure immobilisation of a younger child. Three dimensional conformal radiation and intensity modulated radiation therapy are technologies in current practice designed to conform the radiation treatment to the target volume as concisely as possible.
Rhabdomyosarcomas occur at multiple different locations throughout the body. Typical rhabdomyosarcoma clinical trial protocols give comprehensive recommendations for local control measures at the various sites. Within this review, we will limit discussion to a few general points at key anatomical sites:
As discussed above, a wide local resection should be performed at initial diagnosis, or as pre-treatment re-excision following an initial biopsy, if possible with acceptable morbidity. Resection at a second look operation may be performed following induction chemotherapy; post-operative radiotherapy has been employed in this context within American studies.
The timing of radiotherapy varies. It is given at the start of treatment for those patients with parameningeal tumours with intracranial extension, or if there is an acute emergency such as spinal cord compression. Otherwise radiotherapy is generally given following approximately 12 weeks of induction chemotherapy.
A research question within an ongoing trial of the Children’s Oncology Group is whether the delivery of radiotherapy earlier in treatment, at week four of chemotherapy, may improve local control.
When radiotherapy is to be applied to metastatic sites, treatment generally follows a longer phase of chemotherapy, for instance being given at week 20.
Special considerations are required for the very young patient, particularly those under the age of 24 months. The long-term sequelae of radiation therapy given to such young patients may make that modality of therapy unacceptable. Clinical trial protocols acknowledge this, and often allow for the clinical team to deviate from those local control guidelines employed in older children.
Very young children with parameningeal tumours and intracranial extension should still undergo radiation therapy early in treatment, as cure cannot be achieved without radiotherapy.
The prognostic significance of tumour histology is well known. Patients with group I embryonal tumours do very well with multi-agent chemotherapy alone, hence radiotherapy is not recommended. However, those with group I alveolar or undifferentiated tumours achieve superior outcomes when radiotherapy is administered.14 Conversely, in group III tumours, histology did not correlate with the risk of relapse.15
Depending on the primary tumour location, consequences of local treatment of rhabdomyosarcoma may include growth disturbances, pituitary failure, cataract formation, hearing loss and dentition malformations. Early referral to a paediatric dentist, endocrinologist, facio-maxillary surgeon or orthopaedic surgeon needs to be co-ordinated through a multidisciplinary clinic.
Failure of local control remains the major cause of treatment failure in rhabdomyosarcoma. Data analysed from the Third Intergroup Rhabdomyosarcoma Study for Group III patients showed that the risk of relapse was 33%, and 71% of relapsing patients had local relapse with or without distant relapse.15 Radiotherapy as a modality for local control was less frequently used in frontline treatment in European trials; in those studies local recurrence accounted for 85% of treatment failures.
Myriad factors impact on decisions regarding local control. These decisions are best made in the context of a multidisciplinary team, incorporating the sarcoma surgeon, radiation oncologist, paediatric oncologist, radiologist and pathologist.
1. Miller RW, Young JL Jr, Novakovic B; Childhood cancer. Cancer. 1995;75(Suppl 1):395–405.
2. Weigel B, Lyden E, Anderson JR, Galster A, Arndt CA, Michalski J, et al. Early Results from Children’s Oncology Group (COG) ARST 0431: Intensive multidrug therapy for patients with metastic rhabdomyosarcoma (RMS). J Clin Oncol 2010 ASCO Annual Meeting Proceedings (Post-Meeting Edition) 28(15 Suppl May 20):9503.
3. Pappo AS, Lyden E, Breitfeld P, Donaldson SS, Wiener E, Parham D. Two consecutive phase II window trials of irinotecan alone or in combination with vincristine for the treatment of metastatic rhabdomyosarcoma: the Children’s Oncology Group. J Clin Oncol. 2007;25(4):362-9.
4. Pisters P, Ballo M, Patel S. Preoperative Chemoradiation Treatment Strategies for Localised Sarcoma. Ann Surg Oncol. 2002;9(6);535- 542.
5. Flamant F, Rodary C, Rey A, Praquin MT, Sommelet D, Quintana E, et al. Treatment of non-metastatic rhabdomyosarcomas in childhood and adolescence. Results of the second study of the international society of paediatric oncology: mmt84. Eur J Cancer. 1998;34(7):1050-62.
6. Raney RB, Asmar L, Vassilopoulou – Sellin R, Klein MJ, Donaldson SS, Green J, Heyn R, Wharam M, Glicksman AS, Gehan EA, Anderson J, Maurer HM. Late complications of therapy in 213 children with localized, nonorbital soft-tissue sarcoma of the head and neck: A descriptive report from the Intergroup Rhabdomyosarcoma Studies (IRS)-II and III. IRS Group of the Children’s Cancer Group and the Pediatric Oncology Group. Med Pedaitr Oncol; 1999; 33(4):362-71.
7. Rodenberg DA, Wharam MD, Lyden E, Stoner JA, Brown KL, Wolden SL, et al. Second-look operation with subsequent modification of radiotherapy dose for intermediate-risk rhabdomyosarcoma (RMS): A Report from the Children’s Oncology Group (COG); J Clin Oncol ASCO Annual Meeting Proceedings (Post-Meeting Edition). 28(15 Suppl May 20):9504.
8. Raney RB, Chintagumpala M, Anderson J, Pappo A, Qualman S, Wharam M,et al. Soft-Tissue Sarcoma Committee of the Children’s Oncology Group, Arcadia, California; Results of treatment of patients with superficial facial rhabdomyosarcomas on protocols of the Intergroup Rhabdomyosarcoma Study Group (IRSG), 1984-1997. Pediatr Blood Cancer. 2008;50(5):958-64.
9. Tefft M, Fernandez C, Donaldson M, Newton W, Moon TE. Incidence of meningeal involvement by rhabdomyosarcoma of the head and neck in children: A report of the Intergroup Rhabdomyosarcoma Study (IRS). Cancer. 1978;42:253–258.
10. Defachelles AS, Rey A, Oberlin O, Spooner D. Stevens MC. Treatment of Nonmetastatic Cranial Parameningeal Rhabdomyosarcoma in Children Younger Than 3 Years Old: Results From International Society of Pediatric Oncology Studies MMT 89 and 95. J Clin Oncol. 2009;27(8);1310-1315.
11. Stewart RJ, Martelli H, Oberlin O, Rey A, Bouvet N, Spicer RD, Godzinski J, et al. International Society of Pediatric Oncology. Treatment of children with nonmetastatic paratesticular rhabdomyosarcoma: results of the Malignant Mesenchymal Tumors studies (MMT 84 and MMT 89) of the International Society of Pediatric Oncology. J Clin Oncol. 2003;21(5):793-8.
12. Martelli H, Oberlin O, Rey A, Godzinski J, Spicer RD, Bouvet N, et al. Conservative treatment for girls with nonmetastatic rhabdomyosarcoma of the genital tract: A report from the Study Committee of the International Society of Pediatric Oncology. J Clin Oncol. 1999;17(7):2117-22.
13. Martelli H, Haie-Meder C, Branchereau S, Franchi-Abella S, Ghigna MR, Dumas I. Conservative surgery plus brachytherapy treatment for boys with prostate and/or bladder neck rhabdomyosarcoma: a single team experience. J Pediatr Surg. 2009;44(1):190-6.
14. Wolden S, Anderson J, Crist W, Breneman J, Wharam M, Wiener E, et al. Indications for Radiotherapy and Chemotherapy After Complete Resection in Rhabdomyosarcoma: A Report From the Intergrooup Rhabdomyosarcoma Studies I to III. J Clin Oncol.1999;17:3468- 3475.
15. Wharam MD, Meza J, Anderson J, Breneman JC, Donaldson SS, Fitzgerald TJ, et al. Failure Pattern and Factors Predictive of Local Failure in Rhabdomyosarcoma: A Report of Group III patients on the Third Intergroup Rhabdomyosarcoma Study. J Clin Oncol. 2004;22(10)1902-1908.