Treatment of Nonmetastatic Rhabdomyosarcoma in Childhood and Adolescence: Third Study of the International Society of Paediatric Oncology—SI
http://www.100md.com
《临床肿瘤学》
the University of Bristol, Bristol
United Kingdom Children's Cancer Study Group Data Centre, Leicester
Royal Manchester Children's Hospital, Pendlebury
University Hospital, Birmingham, United Kingdom
Institut Gustave Roussy, Villejuif
Hpital Bicetre, Le Kremlin-Bicetre, France
Hospital Val d'Hebron, Barcelona, Spain
Jeroen Bosch Ziekenhuis, 's-Hertogenbosch, the Netherlands
ABSTRACT
PURPOSE: To improve outcome for children with nonmetastatic rhabdomyosarcoma and to reduce systematic use of local therapy.
PATIENTS AND METHODS: Five hundred three previously untreated patients aged from birth to 18 years, recruited between 1989 and 1995, were allocated to one of six treatment schedules by site and stage.
RESULTS: Five-year overall survival (OS) and event-free survival (EFS) were 71% and 57%, respectively. Primary site, T-stage, and pathologic subtype were independent factors in predicting OS by multivariate analysis. Differences between EFS and OS reflected local treatment strategy and successful re-treatment for some patients after relapse. Patients with genitourinary nonbladder prostate tumors had the most favorable outcome (5-year OS, 94%): the majority were boys with paratesticular tumors treated successfully without alkylating agents. Patients with stage III disease treated with a novel six-drug combination showed improved survival compared with the Malignant Mesenchymal Tumor 84 study (MMT 84; 5-year OS, 60% v 42%, respectively). OS was not significantly better than that achieved in the previous MMT 84 study, but 49% of survivors were cured without significant local therapy.
CONCLUSION: Selective avoidance of local therapy is justified in some patients, though further work is required to prospectively identify those for whom this is most applicable. Exclusion of alkylating agents is justified for the most favorable subset of patients. The value of the new six-drug chemotherapy combination is being evaluated further in a randomized study (MMT 95).
INTRODUCTION
Rhabdomyosarcoma is the most common form of soft tissue sarcoma in the first two decades of life, with a peak age incidence in very young children. Survival rates for patients with nonmetastatic disease have improved substantially throughout the last 30 years.1,2 All strategies incorporate multiagent chemotherapy, and most include the delivery of systematic local therapy. Although surgical resection, either as a primary procedure or a secondary strategy (after initial chemotherapy), is important for local control, complete resection is not feasible at some sites of disease. Radiotherapy (RT) has therefore assumed a major role in the management of many patients,3 though late sequelae can be significant. Clinical trials organized by the Malignant Mesenchymal Tumor committee of the International Society of Pediatric Oncology (SIOP MMT) have explored strategies by which the use of local therapy can be modified or avoided on the basis of response to chemotherapy and conservative surgery.4,5
The overall objectives of the MMT 89 protocol were to improve treatment outcome for children with rhabdomyosarcoma and to reduce, where possible, the consequences of local therapy. In standard and high-risk patients, specific aims were to (1) improve outcome by evaluating early tumor response and modifying chemotherapy in poor responders; (2) explore the value of an increased dose intensity of ifosfamide (9 g/m2 per course compared with 6 g/m2 per course in the previous MMT 84 study5); and (3) assess the impact of an intensified chemotherapy combination in selected poor-prognosis patients. Systematic use of RT was avoided in patients who achieved complete local tumor control with chemotherapy with or without surgery, except in children 3 years of age with parameningeal tumors. In good-prognosis patients, the aim was to maintain excellent survival without the use of alkylating agents.
PATIENTS AND METHODS
Eligibility and Consent
Patients aged from birth to 18 years with a diagnosis of rhabdomyosarcoma were eligible provided they had had no prior treatment except surgery, and started chemotherapy within 6 weeks of the date of initial biopsy or other surgery. Informed consent was obtained from parent or patient, or both, according to the research ethics requirements of the individual institution.
Staging
Primary disease was evaluated by standard clinical and radiologic techniques. Routine lymph node sampling was not required, but fine-needle aspiration or biopsy was recommended for accessible, clinically suspicious regional lymph nodes. Abdominal and pelvic computed tomography (CT) scans were required for all lower limb tumors. Abdominal CT and ultrasound examination of the para-aortic area were required for patients with paratesticular tumors. CSF examination was required for all head and neck tumors (except patients with localized orbital disease).
Chest x-ray and technetium bone scan (with plain x-rays of abnormal sites) were required for detection of distant metastases. CT chest was not mandatory if chest x-ray was normal. A bone-marrow aspirate was performed in all cases, and bilateral bone marrow trephines obtained when metastatic disease was identified elsewhere.
Open surgical biopsy was recommended in order to obtain adequate material for diagnosis and central pathology review. All diagnoses were reviewed by an international panel of pathologists and classified according to the new international classification.6
Clinical staging was based on the SIOP–International Union Against Cancer TNM classification of pretreatment disease.7
Treatment Schedule
The overall schema for each treatment schedule used are shown in Figures 1, 2, and 3, and the allocation of patients to treatment schedules defined by site, stage, and age are given in Table 1. Details of the chemotherapy doses and schedules are given in Table 2.
A primary attempt at surgical resection was encouraged only if it was thought possible without residual disease and with the avoidance of important functional or cosmetic consequences. Chemotherapy doses were reduced for infants younger than 1 year.
Treatment schema 89.1. Patients with localized completely excised stage I tumors (stage I pT1) at any site received only two courses of vincristine and actinomycin D (VA) delivered over 10 weeks (Fig 1).
Treatment schema 89.2. Patients with localized but incompletely excised (stage I, pT3ab) tumors and locally invasive but completely resected (stage I, pT2) tumors at favorable sites (paratesticular, vagina, orbit, nonparameningeal head and neck, bladder, and prostate) received chemotherapy with the intention of administering four courses of ifosfamide and VA (IVA; Fig 1). Patients with a poor response (defined as < 50% partial response) after the second course were considered for surgery. If surgical resection was impossible or incomplete, treatment was changed to a second-line chemotherapy combination known as "Vincaepi" (VCE; vincristine, carboplatin, and epidophyllotoxin). Patients with a good response to IVA ( 50% partial response) and those achieving a complete surgical resection continued with IVA to a total of four courses. Local treatment was required at the end of chemotherapy for those in whom complete remission (CR) had not been achieved.
Treatment schema 89.3. Patients with locally invasive but completely resected (stage II pT2) tumors and with localized but incompletely resected (stage I and II pT3abc) tumors at nonfavorable sites (excluding parameningeal tumors) were treated in this group with an initial intention to administer six courses of IVA (Fig 2). Assessment and treatment adjustment was similar to that for patients in group 89.2. RT was required at the end of treatment for all patients in whom CR could not be confirmed by or achieved with surgery.
Treatment schema 89.4. Patients aged 3 years with stage I and II parameningeal tumors all received radiation therapy, except for those without evidence of erosion of the skull base or cranial nerve palsy who were considered to be at low risk of meningeal extension. RT was delivered after the third course of IVA (or after two courses of IVA + two courses of Vincaepi). Chemotherapy with ifosfamide and vincristine was continued during RT. No intrathecal chemotherapy was given (Fig 3).
Treatment schema 89.5. Patients with lymph node involvement at any site received intensified treatment utilizing a six-drug chemotherapy combination (IVA–carboplatin, epirubicin, and vincristine [CEV]–ifosfamide, avincristine, and etoposide [IVE]). Surgical evaluation of involved nodes was required at the end of the second cycle. If pathologically negative, local treatment to the nodes was not required. Local treatment to the primary tumor depended on response to chemotherapy (Fig 3). Persistently positive nodes were irradiated if they were not surgically resected.
Treatment schema 89.6. Patients with parameningeal disease who were younger than 3 years did not undergo systematic irradiation and received intensified chemotherapy. Ultimately, the need for local therapy was determined by the achievement of CR with chemotherapy alone (Fig 3).
Cytokine (granulocyte colony-stimulating factor) support was not recommended with any of the chemotherapy combinations utilized in this study.
The prescribed dose for conventional radiation treatment was 45 Gy given by single daily fractions of 1.5 to 2.0 Gy. A 5-Gy boost was permitted if appropriate for the age of the patient, treatment volume, and extent of residual disease. Patients with tumors at all sites except the orbit were eligible for inclusion in a pilot study of accelerated hyperfractionated RT (HFRT) as determined by individual institutional policy. In this case, the dose was 40 Gy delivered in twice-daily fractions of 1.5 Gy given at least 6 hours apart. A 5-Gy boost was also permitted if appropriate for the age of the patient, treatment volume, and extent of residual disease. Treatment was given to the initial tumor volume with a 2- to 3-cm margin.
Treatment for relapse was directed by the protocol, and definitive local therapy (surgery ± RT) was required in all cases. Chemotherapy depended on the drugs previously used. The Vincaepi combination was used for patients relapsing after IVA. Patients who relapsed after previous exposure to VCE or the six-drug schedule were eligible for local phase II studies according to physician choice. A maximum of 6 months' therapy was given after relapse.
Quality assurance procedures involved central review of chemotherapy dose and scheduling; review of original surgical reports for all procedures other than biopsy; and review of planning films and prescribed dose for patients relapsing locally after RT.
Statistical Methods
Analyses were performed on the whole study population and on subsets, according to potential prognostic factors.
CR was defined as either the absence of any residual radiologic abnormalities, assessed pathologically when possible, or the presence of residual abnormalities that remained stable for 6 months after the end of therapy.
The starting point for survival analysis was the start of the first course of chemotherapy. Outcome was defined both by overall survival (OS) and event-free survival (EFS). For EFS, events were defined as relapse after CR, death from any cause, and nonresponse (taken as date of change to "off-protocol" therapy). OS and EFS curves were calculated using the Kaplan and Meier method.8 The statistical significance of each variable was first tested by the log-rank test (univariate procedure).9 Multivariate analysis was then performed with the Cox proportional hazard model (BMDP program)10 for OS. The multivariate analysis models were determined by the likelihood ratio test. A stepwise procedure was used to identify the major prognostic events among clinical and pathological factors. All eligible patients were included in the analysis, regardless of their compliance with the protocol.
RESULTS
Characteristics of the Patient Population
Five hundred three patients with rhabdomyosarcoma were enrolled at 83 institutions from 10 countries between January 1989 and June 1995. All had central pathology review.
Median age at diagnosis was 4.7 years; 81% of the patients were younger than 10 years; 29%, younger than 3 years; and 6.5%, younger than 1 year at diagnosis. Sixty-one percent were boys.
The pathology review committee classified 18% of tumors as botryoid or leiomyomatoid, 67% as embryonal, and 15% as alveolar.
Distribution of patients according to tumor site, clinical and postsurgical stage, size, median age, and treatment group is presented in Table 1. Of the 135 patients with parameningeal tumors, 97 (72%) had evidence of erosion of the base of the skull and/or cranial nerve palsy, and were considered high risk (for meningeal involvement). Twenty-seven patients were younger than 3 years at diagnosis.
Median follow-up for survivors was 87 months (7.2 years), ranging from 3 months to 10.9 years. Only six patients were lost to follow-up before 3 years, and 95% of surviving patients (334 of 353) were followed up for at least 5 years.
Median survival beyond the end of therapy was 5 months in patients who did not achieve CR.
CR Rates and Intensity of Primary Treatment
CR was achieved in 93% of patients (467 of 503). Treatment to achieve CR was not always the same as total treatment received, as, for example, patients aged 3 years with high-risk parameningeal disease were systematically irradiated regardless of initial response to chemotherapy.
Overall Outcome
Five-year EFS and OS rates were 57% (SE = 2) and 71% (SE = 2), respectively (Table 3, Fig 4). Thirty-four patients died of disease without achieving control of their tumor, and 185 subsequently relapsed. Four patients (< 1%) died of treatment-related toxicity; two, of second malignancy; and two, of causes unrelated to tumor or treatment.
Outcome by Prognostic Factors
Table 3 presents outcome by clinical stage (Fig. 5 and 6), pathologic subtype (Fig 7 and 8), and primary site. Children younger than 3 years with parameningeal tumors (group 89.6) achieved 5-year OS and EFS rates of 59% and 33% compared with 65% and 62% in those aged 3 years who received systematic RT (group 89.4; Fig. 9 and 10).
Univariate analysis (Table 3) showed that T-stage, tumor size, pathology, primary site, and node involvement were statistically related to OS. T-stage, histology, and primary site were independently prognostic for OS in multivariate analysis (Table 4). Clinical stage and the size of the primary tumor were not selected in the Cox model because these variables were strongly dependent on T stage.
Relapse: Patterns and Outcome
Median time between diagnosis and first relapse was 12 months (range, 3 to 102 months), and 80% of relapses occurred within 24 months of diagnosis. The first event was metastatic in 30 patients (6%) and locoregional in 155 patients (34%); 15 patients had combined local and metastatic relapse. Among those with isolated locoregional relapse, 105 (75%) occurred at the primary site, 24 (17%) in nodes, and 11 (8%) at both.
Alveolar histology was associated with a significantly higher risk of relapse (P = .01), and with a very much higher risk of metastases (13 of 78 v 17 of 425; P > .0001).
Five-year survival was 44% after isolated local relapse and 18% after metastatic relapse. All patients surviving after relapse remained under active follow-up at the time of the analysis.
Toxicity
Four patients (< 1%) died of treatment-related toxicity—one each from sepsis, pneumonitis, veno-occlusive disease (one infant received a larger dose of actinomycin than recommended for their age), and neurological complications possibly associated with ifosfamide. Three patients developed four second malignancies. Acute myeloid leukemia occurred in two patients (one after treatment for relapse and the other after extended treatment for failure to achieve primary disease control), and a third patient with Li Fraumeni syndrome developed osteosarcoma and then oligodendroglioma.
There were 294 episodes of grade 3 or 4 infection in 143 patients. The risk was related to the intensity of chemotherapy—55% in those receiving the six-drug schedule compared with 26% with IVA and less than 1% with VA. Nephrotoxicity (glomerular filtration rate < 60 mL/min/1.73 m2 or symptomatic renal tubular leak) was seen in 28 patients (5.5%). Symptomatic cardiotoxicity was reported in three patients.
Information on surviving patients continues to be collected, and evaluation of late sequelae of treatment is ongoing.
Total Burden of Therapy
To define the total burden of therapy received by survivors, it was necessary to consider not only primary therapy but also that required to retrieve patients who relapsed. At the final analysis, 345 patients were alive: 276 were in first CR with at least 24 months follow-up from diagnosis, and 69 were in second or subsequent CR 24 months from their last event. These patients were considered cured, as their median follow-up was 81 months from their last event. Excluding 49 patients with parameningeal tumors who systematically received RT as part of primary treatment, 169 (58%) of the remainder (n = 294) avoided "significant" local therapy (Table 5). Significant local therapy was defined either as "large" (ie, major surgery but without predictably important long-term functional or cosmetic consequences and/or external-beam RT) or "radical" (major surgery with predictable functional or cosmetic consequences; eg, total cystectomy, orbital exenteration, amputation). Within this group, 54 patients had had limited chemotherapy after initial complete surgery, 46 had received chemotherapy after incomplete initial surgery or biopsy without further local treatment, and 69 had had chemotherapy and "conservative" local therapy. Conservative local therapy was defined as functionally and cosmetically unimportant surgery, or brachytherapy. Thirty-three percent of the survivors had had incomplete initial surgery but avoided subsequent significant local therapy, and, overall, 49% of all survivors were treated without significant local therapy. Table 5 also provides comparative data from the previous MMT 84 study.5
DISCUSSION
Failure of local control remains the major cause of treatment failure in rhabdomyosarcoma. Assessment for initial surgical resection followed by chemotherapy and local RT, with or without further surgery, achieves local control in more than 80% of patients. Such good results are not achieved at all sites (eg, trunk and limbs) or in patients with large (> 10 cm) primary tumors, and failure rates of 11% to 18% are still reported at more favorable sites.11 In survivors, all forms of therapy can contribute to major late morbidity.12-15 Experience in the first SIOP study (RMS 75) showed that patients who received initial chemotherapy followed by delayed local therapy achieved the same overall survival with less extensive local therapy.4,16
The second SIOP study (MMT 84) attempted to assess whether local treatment could be decided according to response to initial chemotherapy. It was recognized that this strategy could result in a higher local relapse rate, but a secondary objective was to determine whether patients initially treated by chemotherapy without definitive local therapy could be salvaged by local treatment (and further chemotherapy) at the time of relapse.5 Particular subsets of patients appeared more likely to benefit from this strategy; for example, 38% of patients with orbital tumors were treated successfully without RT or radical surgery, with 5-year OS of 88%.5,17 A subsequent international meta-analysis of children with orbital RMS showed that this approach did not compromise survival, though patients treated after relapse were also exposed to further chemotherapy, a factor to be considered in the context of the total burden of therapy borne by survivors.18 Similar data have emerged from a more recent meta-analysis of children treated for bladder prostate rhabdomyosarcoma, where the omission of RT as part of primary therapy had no adverse impact on OS.19
For very good prognosis patients, the intention was to maintain excellent survival but reduce treatment burden by avoiding alkylating agents. These patients had the same outcome using VA (5-year OS, 87%) as those in MMT 84 using three courses of IVA (89%), though EFS was less satisfactory (67% v 85%). The majority of these patients were boys with paratesticular tumors.20 Improved results (5-year OS, 95%) have been reported using similar chemotherapy, but of longer duration.21 These data also confirm the MMT group's approach to retroperitoneal lymph node dissection.22 However, evidence is accumulating that boys older than 10 years, particularly those with tumors 5 cm, carry a higher risk of relapse, and such limited therapy would not be appropriate.20,21,23
Despite an increase in ifosfamide dose intensity and the early introduction of alternative drugs for those with an inadequate initial response, the overall outcome for patients in MMT 89 showed no statistical improvement over MMT 84 (5-year OS, 71% v 68% and EFS 57% v 53%). The large difference observed between EFS and OS in both studies illustrates the possibility of salvage after relapse and implies that the main criterion for a comparison of this strategy with other studies should be OS rather than EFS.
Prolonged follow-up was required to provide a valid final report, as the evaluation of a strategy that limits local therapy must incorporate outcome of treatment for relapse and the total burden of therapy to which survivors were exposed. This report is based on analysis at a median follow-up more than 7 years from diagnosis. Although EFS in this study does not compare favorably with rates reported by others, the OS compared with a 5-year OS of 71% in IRS III24 and 69% in the German cooperative group study (CWS 86).25 A 3-year OS of 86% has been reported for the IRS IV study.26
Age and sex had no impact on OS, despite a trend for an increased relative risk in children aged 10 years. The prognostic impact of age has emerged in other studies,26 and there is a strong interrelationship between age, site, and histology. There was also a relative under-representation of older patients (81% aged < 10 years), which reduced the power of the analysis in this study. Site, T-stage, and histology were the most important independent prognostic variables for OS. The adverse importance of alveolar pathology was also apparent in the results of IRS IV.26
The value of chemotherapy in contributing to local control was evaluated by an assessment of the number of patients receiving local therapy. The data are consistent with those from MMT 84 (Table 5): approximately 50% of survivors could be treated without significant local therapy, and 33% of those who start treatment with residual disease (SIOP TNM pT3 or IRS Group II and III) can avoid significant local therapy. These findings establish the principle that some children with rhabdomyosarcoma may benefit from selective avoidance of local therapy; the challenge for the future is to predict at the outset which individual patients have a realistic chance of cure with this approach.
Systematic local therapy is more important for some sites of disease than for others, particularly at parameningeal sites.27 Only three of 27 patients younger than 3 years with parameningeal disease were ultimately cured without RT, but further exploration of strategies to delay RT in young parameningeal patients seems justified in view of the lack of statistical difference in their OS as compared with the older parameningeal patients (59% v 65%).
The rationale for using an intensified six-drug chemotherapy strategy28 for patients with node-positive disease was derived from the very poor results seen for this group of patients treated with IVA in MMT 84. Although valid statistical comparison could not be made, no difference was identified between the two groups of patients in terms of age, site, or extent of local therapy, and the results obtained in MMT 89 suggested benefit (5-year OS, 60% v 42%; EFS, 51% v 42%). This is being explored as a randomized study in MMT 95.29
Exploration of the value of HFRT was intended, but many participating centers were unable to participate, and only limited nonrandomized experience was acquired. A randomized evaluation as part of IRS IV suggested no benefit for HFRT.30
Sequelae from the use of chemotherapy cannot be ignored, particularly in relation to risks of nephrotoxicity, cardiotoxicity, and second malignancy. The use of ifosfamide has been controversial. Data from IRS IV show no advantage of ifosfamide over cyclophosphamide,26 and there is concern about the renal toxicity of ifosfamide. Significant nephrotoxicity was very low, consistent with experience with cumulative doses less than 60 g/m2.31,32 Ifosfamide at 9 g/m2 did not require routine use of granulocyte colony-stimulating factor in contrast to the intensified cyclophosphamide dose utilized in IRS IV. Data are not yet available to determine whether ifosfamide is less damaging to gonadal function.
Appendix
Patients were entered onto the study from treatment centers in the following cities. We are grateful to the clinicians concerned for their support. Argentina: Buenos Aires; Belgium: Bruxelles (Fabiola), Bruxelles (St Joseph), Bruxelles (St Luc), Leuven, Montegnee; Denmark: Aarhus, Copenhagen; France: Angers, Brest, Besancon, Bordeaux (Centre Hospitalier Universitaire [CHU]), Bordeaux (CAC), Caen (CHU), Caen (CAC), Clermont Ferrand, Colmar, Grenoble, Lille (CAC), Lille (Calmette), Lille (St Antoine), Limoges, Lyon, Marseille, Montpellier, Nancy, Nantes, Nice, Paris (Curie), Paris (Trousseau), Poitiers, Reims, Rennes, Rouen (CHU), Rouen (CAC), St Etienne, Strasbourg (CHU), Toulouse (CHU), Tours, Villejuif; Republic of Ireland: Dublin; The Netherlands: Amsterdam (AMC), Amsterdam (VU), Nijmegen; Poland: Wroclaw; Spain: Badalona, Barcelona, Compostella, Las Palmas, Madrid (La Paz), Madrid (Nino Jesus), Madrid (12 October), Madrid (Zarzuela), Oviedo, Pamplona, Valencia (La Fe), Viscaya, Zaragoza; Switzerland: Basel, Lausanne, Lucerne, Zurich; United Kingdom: Belfast, Birmingham, Bristol, Cambridge, Cardiff, Edinburgh, Glasgow, Leeds, Leicester, Liverpool, London (Great Ormond Street), London (Middlesex), London (St Bartholomew's), Manchester, Newcastle, Nottingham, Oxford, Sheffield, Southampton, Sutton (Royal Marsden).
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
NOTES
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Raney RB, Anderson JR, Barr FG, et al: Rhabdomyosarcoma and undifferentiated sarcoma in the first two decades of life: A selective review of Intergroup rhabdomyosarcoma study group experience and rationale for Intergroup Rhabdomyosarcoma Study V. J Pediatr Hematol Oncol 23:215-220, 2001
Stiller CA, Stevens MCG, Magnani C, et al: Survival of children with soft tissue sarcoma in Europe since 1978: Results from the EUROCARE study. Eur J Cancer 37:767-774, 2001
Wolden SL, Anderson JR, Crist WM, et al: Indications for radiotherapy and chemotherapy after complete resection in rhabdomyosarcoma: A report from the Intergroup Rhabdomyosarcoma Studies I to III. J Clin Oncol 17:3468-3475, 1999
Flamant F, Rodary C, Voute PA, et al: Primary chemotherapy in the treatment of rhabdomyosarcoma in children: Trial of the International Society of Pediatric Oncology (SIOP)—Preliminary results. Radiother Oncol 3:227-236, 1985
Flamant F, Rodary C, Rey A, et al: Treatment of non metastatic rhabdomyosarcomas in childhood and adolescence: Results of the second study of the International Society of Paediatric Oncology: MMT 84. Eur J Cancer 34:1050-1062, 1998
Newton WA, Gehan EA, Webber BL, et al: Classification of rhabdomyosarcoma and related sarcomas: Pathologic aspects and proposals for a new classification—An Intergroup Rhabdomyosarcoma Study. Cancer 76:1073-1085, 1995
Rodary C, Flamant F, Donaldson SS, et al: An attempt to use a common staging system in rhabdomyosarcoma: A report of an international workshop initiated by the International Society of Pediatric Oncology (SIOP). Med Pediatr Oncol 17:210-215, 1989
Kaplan ES, Meier P: Non-parametric estimation from incomplete observation. J Am Stat Assoc 53:457-480, 1958
Peto R, Pike MC, Armitage P, et al: Design and analysis of randomised clinical trials requiring prolonged observation of each patient. Br J Cancer 35:1-39, 1977
Cox DR: Regression models and life table. J R Stat Soc (B) 34:187-220, 1972
Wharam MD, Hanfelt JJ, Tefft MC, et al: Radiation therapy for rhabdomyosarcoma: Local failure risk for Clinical Group III patients on Intergroup Rhabdomyosarcoma Study II. Int J Radiat Oncol Biol Phys 38:797-804, 1997
Heyn R, Raney RB, Hays DM, et al: Late effects of therapy in patients with paratesticular rhabdomyosarcoma. J Clin Oncol 10:614-623, 1992
Raney RB, Asmar L, Vassilopoulou-Sellin R, et al: Late complications of therapy in 213 children with localized, non orbital soft tissue sarcoma of the head and neck: A descriptive report from the Intergroup Rhabdomyosarcoma Studies (IRS)-II and -III. Med Pediatr Oncol 33:362-371, 1999
Raney RB, Anderson JR, Kollath J, et al: Late effects of therapy in 94 patients with localized rhabdomyosarcoma of the orbit: Report from the Intergroup Rhabdomyosarcoma Study (IRS)-III, 1984-1991. Med Pediatr Oncol 34:413-420, 2000
Heyn R, Haeberlen V, Newton WA, et al: Second malignant neoplasms in children treated for rhabdomyosarcoma. J Clin Oncol 11:262-270, 1993
Rodary C, Rey A, Olive D, et al: Prognostic factors in 281 children with non metastatic rhabdomyosarcoma (RMS) at diagnosis. Med Pediatr Oncol 16:71-77, 1988
Rousseau P, Flamant F, Quintana E, et al: Primary chemotherapy in rhabdomyosarcomas and other malignant mesenchymal tumours of the orbit: Results of the International Society of Pediatric Oncology MMT 84 Study. J Clin Oncol 12:516-521, 1994
Oberlin O, Rey A, Anderson J, et al: Treatment of orbital rhabdomyosarcoma: Survival and late effects of treatment—Results of an international workshop. J Clin Oncol 19:197-204, 2001
Anderson J, Raney RB, Carli M, et al: International study of characteristics and outcome of patients with primary rhabdomyosarcoma of the bladder/prostate. Med Pediatr Oncol 37:181, 2001
Stewart RJ, Martelli H, Oberlin O, et al: Treatment of children with non metastatic paratesticular rhabdomyosarcoma: Results of the MMT 84 and MMT 89 studies of the International Society of Paediatric Oncology. J Clin Oncol 21:793-795, 2003
Ferrari A, Bisogno G, Casanova M, et al: Paratesticular rhabdomyosarcoma: Report from the Italian and German Cooperative Group. J Clin Oncol 20:449-455, 2002
Rodary C, Flamant F, Maurer H, et al: Initial lymphadenectomy is not necessary in localised completely resected paratesticular rhabdomyosarcoma. Med Pediatr Oncol 20:430, 1992
Wiener ES, Anderson JR, Ojimba JI, et al: Controversies in the management of paratesticular rhabdomyosarcoma: Is staging retroperitoneal lymph node dissection necessary for adolescents with resected paratesticular rhabdomyosarcoma Semin Pediatr Surg 10:146-152, 2001
Crist W, Gehan EA, Ragab AH, et al: The third Intergroup rhabdomyosarcoma study. J Clin Oncol 13:610-630, 1995
Koscielniak E, Harms D, Henze G, et al: Results of treatment for soft tissue sarcoma in childhood and adolescence: A final report of the German Cooperative Soft Tissue Sarcoma Study CWS-86. J Clin Oncol 17:3706-3719, 1999
Crist WM, Anderson JR, Meza JL, et al: Intergroup rhabdomyosarcoma study-IV: Results for patients with non metastatic disease. J Clin Oncol 19:3091-4102, 2001
Benk V, Rodary C, Flamant F, et al: Parameningeal rhabdomyosarcoma: Results of an international workshop. Int J Radiat Oncol Biol Phys 36:533-540, 1996
Carli M, Colombatti R, Oberlin O, et al: High dose melphalan with autologous stem-cell rescue in metastatic rhabdomyosarcoma. J Clin Oncol 17:2796-2803, 1999
Oberlin O, Rey A, Praquin MT, et al: Non metastatic rhabdomyosarcoma (RMS): Outlines for the SIOP MMT 95 study. Med Pediatr Oncol 27:265, 1996
Donaldson SS, Meza J, Breneman JC, et al: Children's Oncology Group Soft Tissue Sarcoma Committee (formerly Intergroup Rhabdomyosarcoma Group) representing the Children's Oncology Group and the Quality Assurance Review Center: Results from the IRS-IV randomized trial of hyperfractionated radiotherapy in children with rhabdomyosarcoma—A report from the IRSG. Int J Radiat Oncol Biol Phys 51:718-728, 2001
Suarez A, Mc Dowell H, Niaudet P, et al: Long term follow-up of Ifosfamide renal toxicity in children treated for malignant mesenchymal tumours: An International Society of Pediatric Oncology report. J Clin Oncol 9:2177-2182, 1991
Skinner R, Cotterill SJ, Stevens MCG: Risk factors for nephrotoxicity after ifosfamide treatment in children: A UKCCSG Late Effects Group Study. Br J Cancer 82:1636-1645, 2000(Michael C.G. Stevens, Ann)
United Kingdom Children's Cancer Study Group Data Centre, Leicester
Royal Manchester Children's Hospital, Pendlebury
University Hospital, Birmingham, United Kingdom
Institut Gustave Roussy, Villejuif
Hpital Bicetre, Le Kremlin-Bicetre, France
Hospital Val d'Hebron, Barcelona, Spain
Jeroen Bosch Ziekenhuis, 's-Hertogenbosch, the Netherlands
ABSTRACT
PURPOSE: To improve outcome for children with nonmetastatic rhabdomyosarcoma and to reduce systematic use of local therapy.
PATIENTS AND METHODS: Five hundred three previously untreated patients aged from birth to 18 years, recruited between 1989 and 1995, were allocated to one of six treatment schedules by site and stage.
RESULTS: Five-year overall survival (OS) and event-free survival (EFS) were 71% and 57%, respectively. Primary site, T-stage, and pathologic subtype were independent factors in predicting OS by multivariate analysis. Differences between EFS and OS reflected local treatment strategy and successful re-treatment for some patients after relapse. Patients with genitourinary nonbladder prostate tumors had the most favorable outcome (5-year OS, 94%): the majority were boys with paratesticular tumors treated successfully without alkylating agents. Patients with stage III disease treated with a novel six-drug combination showed improved survival compared with the Malignant Mesenchymal Tumor 84 study (MMT 84; 5-year OS, 60% v 42%, respectively). OS was not significantly better than that achieved in the previous MMT 84 study, but 49% of survivors were cured without significant local therapy.
CONCLUSION: Selective avoidance of local therapy is justified in some patients, though further work is required to prospectively identify those for whom this is most applicable. Exclusion of alkylating agents is justified for the most favorable subset of patients. The value of the new six-drug chemotherapy combination is being evaluated further in a randomized study (MMT 95).
INTRODUCTION
Rhabdomyosarcoma is the most common form of soft tissue sarcoma in the first two decades of life, with a peak age incidence in very young children. Survival rates for patients with nonmetastatic disease have improved substantially throughout the last 30 years.1,2 All strategies incorporate multiagent chemotherapy, and most include the delivery of systematic local therapy. Although surgical resection, either as a primary procedure or a secondary strategy (after initial chemotherapy), is important for local control, complete resection is not feasible at some sites of disease. Radiotherapy (RT) has therefore assumed a major role in the management of many patients,3 though late sequelae can be significant. Clinical trials organized by the Malignant Mesenchymal Tumor committee of the International Society of Pediatric Oncology (SIOP MMT) have explored strategies by which the use of local therapy can be modified or avoided on the basis of response to chemotherapy and conservative surgery.4,5
The overall objectives of the MMT 89 protocol were to improve treatment outcome for children with rhabdomyosarcoma and to reduce, where possible, the consequences of local therapy. In standard and high-risk patients, specific aims were to (1) improve outcome by evaluating early tumor response and modifying chemotherapy in poor responders; (2) explore the value of an increased dose intensity of ifosfamide (9 g/m2 per course compared with 6 g/m2 per course in the previous MMT 84 study5); and (3) assess the impact of an intensified chemotherapy combination in selected poor-prognosis patients. Systematic use of RT was avoided in patients who achieved complete local tumor control with chemotherapy with or without surgery, except in children 3 years of age with parameningeal tumors. In good-prognosis patients, the aim was to maintain excellent survival without the use of alkylating agents.
PATIENTS AND METHODS
Eligibility and Consent
Patients aged from birth to 18 years with a diagnosis of rhabdomyosarcoma were eligible provided they had had no prior treatment except surgery, and started chemotherapy within 6 weeks of the date of initial biopsy or other surgery. Informed consent was obtained from parent or patient, or both, according to the research ethics requirements of the individual institution.
Staging
Primary disease was evaluated by standard clinical and radiologic techniques. Routine lymph node sampling was not required, but fine-needle aspiration or biopsy was recommended for accessible, clinically suspicious regional lymph nodes. Abdominal and pelvic computed tomography (CT) scans were required for all lower limb tumors. Abdominal CT and ultrasound examination of the para-aortic area were required for patients with paratesticular tumors. CSF examination was required for all head and neck tumors (except patients with localized orbital disease).
Chest x-ray and technetium bone scan (with plain x-rays of abnormal sites) were required for detection of distant metastases. CT chest was not mandatory if chest x-ray was normal. A bone-marrow aspirate was performed in all cases, and bilateral bone marrow trephines obtained when metastatic disease was identified elsewhere.
Open surgical biopsy was recommended in order to obtain adequate material for diagnosis and central pathology review. All diagnoses were reviewed by an international panel of pathologists and classified according to the new international classification.6
Clinical staging was based on the SIOP–International Union Against Cancer TNM classification of pretreatment disease.7
Treatment Schedule
The overall schema for each treatment schedule used are shown in Figures 1, 2, and 3, and the allocation of patients to treatment schedules defined by site, stage, and age are given in Table 1. Details of the chemotherapy doses and schedules are given in Table 2.
A primary attempt at surgical resection was encouraged only if it was thought possible without residual disease and with the avoidance of important functional or cosmetic consequences. Chemotherapy doses were reduced for infants younger than 1 year.
Treatment schema 89.1. Patients with localized completely excised stage I tumors (stage I pT1) at any site received only two courses of vincristine and actinomycin D (VA) delivered over 10 weeks (Fig 1).
Treatment schema 89.2. Patients with localized but incompletely excised (stage I, pT3ab) tumors and locally invasive but completely resected (stage I, pT2) tumors at favorable sites (paratesticular, vagina, orbit, nonparameningeal head and neck, bladder, and prostate) received chemotherapy with the intention of administering four courses of ifosfamide and VA (IVA; Fig 1). Patients with a poor response (defined as < 50% partial response) after the second course were considered for surgery. If surgical resection was impossible or incomplete, treatment was changed to a second-line chemotherapy combination known as "Vincaepi" (VCE; vincristine, carboplatin, and epidophyllotoxin). Patients with a good response to IVA ( 50% partial response) and those achieving a complete surgical resection continued with IVA to a total of four courses. Local treatment was required at the end of chemotherapy for those in whom complete remission (CR) had not been achieved.
Treatment schema 89.3. Patients with locally invasive but completely resected (stage II pT2) tumors and with localized but incompletely resected (stage I and II pT3abc) tumors at nonfavorable sites (excluding parameningeal tumors) were treated in this group with an initial intention to administer six courses of IVA (Fig 2). Assessment and treatment adjustment was similar to that for patients in group 89.2. RT was required at the end of treatment for all patients in whom CR could not be confirmed by or achieved with surgery.
Treatment schema 89.4. Patients aged 3 years with stage I and II parameningeal tumors all received radiation therapy, except for those without evidence of erosion of the skull base or cranial nerve palsy who were considered to be at low risk of meningeal extension. RT was delivered after the third course of IVA (or after two courses of IVA + two courses of Vincaepi). Chemotherapy with ifosfamide and vincristine was continued during RT. No intrathecal chemotherapy was given (Fig 3).
Treatment schema 89.5. Patients with lymph node involvement at any site received intensified treatment utilizing a six-drug chemotherapy combination (IVA–carboplatin, epirubicin, and vincristine [CEV]–ifosfamide, avincristine, and etoposide [IVE]). Surgical evaluation of involved nodes was required at the end of the second cycle. If pathologically negative, local treatment to the nodes was not required. Local treatment to the primary tumor depended on response to chemotherapy (Fig 3). Persistently positive nodes were irradiated if they were not surgically resected.
Treatment schema 89.6. Patients with parameningeal disease who were younger than 3 years did not undergo systematic irradiation and received intensified chemotherapy. Ultimately, the need for local therapy was determined by the achievement of CR with chemotherapy alone (Fig 3).
Cytokine (granulocyte colony-stimulating factor) support was not recommended with any of the chemotherapy combinations utilized in this study.
The prescribed dose for conventional radiation treatment was 45 Gy given by single daily fractions of 1.5 to 2.0 Gy. A 5-Gy boost was permitted if appropriate for the age of the patient, treatment volume, and extent of residual disease. Patients with tumors at all sites except the orbit were eligible for inclusion in a pilot study of accelerated hyperfractionated RT (HFRT) as determined by individual institutional policy. In this case, the dose was 40 Gy delivered in twice-daily fractions of 1.5 Gy given at least 6 hours apart. A 5-Gy boost was also permitted if appropriate for the age of the patient, treatment volume, and extent of residual disease. Treatment was given to the initial tumor volume with a 2- to 3-cm margin.
Treatment for relapse was directed by the protocol, and definitive local therapy (surgery ± RT) was required in all cases. Chemotherapy depended on the drugs previously used. The Vincaepi combination was used for patients relapsing after IVA. Patients who relapsed after previous exposure to VCE or the six-drug schedule were eligible for local phase II studies according to physician choice. A maximum of 6 months' therapy was given after relapse.
Quality assurance procedures involved central review of chemotherapy dose and scheduling; review of original surgical reports for all procedures other than biopsy; and review of planning films and prescribed dose for patients relapsing locally after RT.
Statistical Methods
Analyses were performed on the whole study population and on subsets, according to potential prognostic factors.
CR was defined as either the absence of any residual radiologic abnormalities, assessed pathologically when possible, or the presence of residual abnormalities that remained stable for 6 months after the end of therapy.
The starting point for survival analysis was the start of the first course of chemotherapy. Outcome was defined both by overall survival (OS) and event-free survival (EFS). For EFS, events were defined as relapse after CR, death from any cause, and nonresponse (taken as date of change to "off-protocol" therapy). OS and EFS curves were calculated using the Kaplan and Meier method.8 The statistical significance of each variable was first tested by the log-rank test (univariate procedure).9 Multivariate analysis was then performed with the Cox proportional hazard model (BMDP program)10 for OS. The multivariate analysis models were determined by the likelihood ratio test. A stepwise procedure was used to identify the major prognostic events among clinical and pathological factors. All eligible patients were included in the analysis, regardless of their compliance with the protocol.
RESULTS
Characteristics of the Patient Population
Five hundred three patients with rhabdomyosarcoma were enrolled at 83 institutions from 10 countries between January 1989 and June 1995. All had central pathology review.
Median age at diagnosis was 4.7 years; 81% of the patients were younger than 10 years; 29%, younger than 3 years; and 6.5%, younger than 1 year at diagnosis. Sixty-one percent were boys.
The pathology review committee classified 18% of tumors as botryoid or leiomyomatoid, 67% as embryonal, and 15% as alveolar.
Distribution of patients according to tumor site, clinical and postsurgical stage, size, median age, and treatment group is presented in Table 1. Of the 135 patients with parameningeal tumors, 97 (72%) had evidence of erosion of the base of the skull and/or cranial nerve palsy, and were considered high risk (for meningeal involvement). Twenty-seven patients were younger than 3 years at diagnosis.
Median follow-up for survivors was 87 months (7.2 years), ranging from 3 months to 10.9 years. Only six patients were lost to follow-up before 3 years, and 95% of surviving patients (334 of 353) were followed up for at least 5 years.
Median survival beyond the end of therapy was 5 months in patients who did not achieve CR.
CR Rates and Intensity of Primary Treatment
CR was achieved in 93% of patients (467 of 503). Treatment to achieve CR was not always the same as total treatment received, as, for example, patients aged 3 years with high-risk parameningeal disease were systematically irradiated regardless of initial response to chemotherapy.
Overall Outcome
Five-year EFS and OS rates were 57% (SE = 2) and 71% (SE = 2), respectively (Table 3, Fig 4). Thirty-four patients died of disease without achieving control of their tumor, and 185 subsequently relapsed. Four patients (< 1%) died of treatment-related toxicity; two, of second malignancy; and two, of causes unrelated to tumor or treatment.
Outcome by Prognostic Factors
Table 3 presents outcome by clinical stage (Fig. 5 and 6), pathologic subtype (Fig 7 and 8), and primary site. Children younger than 3 years with parameningeal tumors (group 89.6) achieved 5-year OS and EFS rates of 59% and 33% compared with 65% and 62% in those aged 3 years who received systematic RT (group 89.4; Fig. 9 and 10).
Univariate analysis (Table 3) showed that T-stage, tumor size, pathology, primary site, and node involvement were statistically related to OS. T-stage, histology, and primary site were independently prognostic for OS in multivariate analysis (Table 4). Clinical stage and the size of the primary tumor were not selected in the Cox model because these variables were strongly dependent on T stage.
Relapse: Patterns and Outcome
Median time between diagnosis and first relapse was 12 months (range, 3 to 102 months), and 80% of relapses occurred within 24 months of diagnosis. The first event was metastatic in 30 patients (6%) and locoregional in 155 patients (34%); 15 patients had combined local and metastatic relapse. Among those with isolated locoregional relapse, 105 (75%) occurred at the primary site, 24 (17%) in nodes, and 11 (8%) at both.
Alveolar histology was associated with a significantly higher risk of relapse (P = .01), and with a very much higher risk of metastases (13 of 78 v 17 of 425; P > .0001).
Five-year survival was 44% after isolated local relapse and 18% after metastatic relapse. All patients surviving after relapse remained under active follow-up at the time of the analysis.
Toxicity
Four patients (< 1%) died of treatment-related toxicity—one each from sepsis, pneumonitis, veno-occlusive disease (one infant received a larger dose of actinomycin than recommended for their age), and neurological complications possibly associated with ifosfamide. Three patients developed four second malignancies. Acute myeloid leukemia occurred in two patients (one after treatment for relapse and the other after extended treatment for failure to achieve primary disease control), and a third patient with Li Fraumeni syndrome developed osteosarcoma and then oligodendroglioma.
There were 294 episodes of grade 3 or 4 infection in 143 patients. The risk was related to the intensity of chemotherapy—55% in those receiving the six-drug schedule compared with 26% with IVA and less than 1% with VA. Nephrotoxicity (glomerular filtration rate < 60 mL/min/1.73 m2 or symptomatic renal tubular leak) was seen in 28 patients (5.5%). Symptomatic cardiotoxicity was reported in three patients.
Information on surviving patients continues to be collected, and evaluation of late sequelae of treatment is ongoing.
Total Burden of Therapy
To define the total burden of therapy received by survivors, it was necessary to consider not only primary therapy but also that required to retrieve patients who relapsed. At the final analysis, 345 patients were alive: 276 were in first CR with at least 24 months follow-up from diagnosis, and 69 were in second or subsequent CR 24 months from their last event. These patients were considered cured, as their median follow-up was 81 months from their last event. Excluding 49 patients with parameningeal tumors who systematically received RT as part of primary treatment, 169 (58%) of the remainder (n = 294) avoided "significant" local therapy (Table 5). Significant local therapy was defined either as "large" (ie, major surgery but without predictably important long-term functional or cosmetic consequences and/or external-beam RT) or "radical" (major surgery with predictable functional or cosmetic consequences; eg, total cystectomy, orbital exenteration, amputation). Within this group, 54 patients had had limited chemotherapy after initial complete surgery, 46 had received chemotherapy after incomplete initial surgery or biopsy without further local treatment, and 69 had had chemotherapy and "conservative" local therapy. Conservative local therapy was defined as functionally and cosmetically unimportant surgery, or brachytherapy. Thirty-three percent of the survivors had had incomplete initial surgery but avoided subsequent significant local therapy, and, overall, 49% of all survivors were treated without significant local therapy. Table 5 also provides comparative data from the previous MMT 84 study.5
DISCUSSION
Failure of local control remains the major cause of treatment failure in rhabdomyosarcoma. Assessment for initial surgical resection followed by chemotherapy and local RT, with or without further surgery, achieves local control in more than 80% of patients. Such good results are not achieved at all sites (eg, trunk and limbs) or in patients with large (> 10 cm) primary tumors, and failure rates of 11% to 18% are still reported at more favorable sites.11 In survivors, all forms of therapy can contribute to major late morbidity.12-15 Experience in the first SIOP study (RMS 75) showed that patients who received initial chemotherapy followed by delayed local therapy achieved the same overall survival with less extensive local therapy.4,16
The second SIOP study (MMT 84) attempted to assess whether local treatment could be decided according to response to initial chemotherapy. It was recognized that this strategy could result in a higher local relapse rate, but a secondary objective was to determine whether patients initially treated by chemotherapy without definitive local therapy could be salvaged by local treatment (and further chemotherapy) at the time of relapse.5 Particular subsets of patients appeared more likely to benefit from this strategy; for example, 38% of patients with orbital tumors were treated successfully without RT or radical surgery, with 5-year OS of 88%.5,17 A subsequent international meta-analysis of children with orbital RMS showed that this approach did not compromise survival, though patients treated after relapse were also exposed to further chemotherapy, a factor to be considered in the context of the total burden of therapy borne by survivors.18 Similar data have emerged from a more recent meta-analysis of children treated for bladder prostate rhabdomyosarcoma, where the omission of RT as part of primary therapy had no adverse impact on OS.19
For very good prognosis patients, the intention was to maintain excellent survival but reduce treatment burden by avoiding alkylating agents. These patients had the same outcome using VA (5-year OS, 87%) as those in MMT 84 using three courses of IVA (89%), though EFS was less satisfactory (67% v 85%). The majority of these patients were boys with paratesticular tumors.20 Improved results (5-year OS, 95%) have been reported using similar chemotherapy, but of longer duration.21 These data also confirm the MMT group's approach to retroperitoneal lymph node dissection.22 However, evidence is accumulating that boys older than 10 years, particularly those with tumors 5 cm, carry a higher risk of relapse, and such limited therapy would not be appropriate.20,21,23
Despite an increase in ifosfamide dose intensity and the early introduction of alternative drugs for those with an inadequate initial response, the overall outcome for patients in MMT 89 showed no statistical improvement over MMT 84 (5-year OS, 71% v 68% and EFS 57% v 53%). The large difference observed between EFS and OS in both studies illustrates the possibility of salvage after relapse and implies that the main criterion for a comparison of this strategy with other studies should be OS rather than EFS.
Prolonged follow-up was required to provide a valid final report, as the evaluation of a strategy that limits local therapy must incorporate outcome of treatment for relapse and the total burden of therapy to which survivors were exposed. This report is based on analysis at a median follow-up more than 7 years from diagnosis. Although EFS in this study does not compare favorably with rates reported by others, the OS compared with a 5-year OS of 71% in IRS III24 and 69% in the German cooperative group study (CWS 86).25 A 3-year OS of 86% has been reported for the IRS IV study.26
Age and sex had no impact on OS, despite a trend for an increased relative risk in children aged 10 years. The prognostic impact of age has emerged in other studies,26 and there is a strong interrelationship between age, site, and histology. There was also a relative under-representation of older patients (81% aged < 10 years), which reduced the power of the analysis in this study. Site, T-stage, and histology were the most important independent prognostic variables for OS. The adverse importance of alveolar pathology was also apparent in the results of IRS IV.26
The value of chemotherapy in contributing to local control was evaluated by an assessment of the number of patients receiving local therapy. The data are consistent with those from MMT 84 (Table 5): approximately 50% of survivors could be treated without significant local therapy, and 33% of those who start treatment with residual disease (SIOP TNM pT3 or IRS Group II and III) can avoid significant local therapy. These findings establish the principle that some children with rhabdomyosarcoma may benefit from selective avoidance of local therapy; the challenge for the future is to predict at the outset which individual patients have a realistic chance of cure with this approach.
Systematic local therapy is more important for some sites of disease than for others, particularly at parameningeal sites.27 Only three of 27 patients younger than 3 years with parameningeal disease were ultimately cured without RT, but further exploration of strategies to delay RT in young parameningeal patients seems justified in view of the lack of statistical difference in their OS as compared with the older parameningeal patients (59% v 65%).
The rationale for using an intensified six-drug chemotherapy strategy28 for patients with node-positive disease was derived from the very poor results seen for this group of patients treated with IVA in MMT 84. Although valid statistical comparison could not be made, no difference was identified between the two groups of patients in terms of age, site, or extent of local therapy, and the results obtained in MMT 89 suggested benefit (5-year OS, 60% v 42%; EFS, 51% v 42%). This is being explored as a randomized study in MMT 95.29
Exploration of the value of HFRT was intended, but many participating centers were unable to participate, and only limited nonrandomized experience was acquired. A randomized evaluation as part of IRS IV suggested no benefit for HFRT.30
Sequelae from the use of chemotherapy cannot be ignored, particularly in relation to risks of nephrotoxicity, cardiotoxicity, and second malignancy. The use of ifosfamide has been controversial. Data from IRS IV show no advantage of ifosfamide over cyclophosphamide,26 and there is concern about the renal toxicity of ifosfamide. Significant nephrotoxicity was very low, consistent with experience with cumulative doses less than 60 g/m2.31,32 Ifosfamide at 9 g/m2 did not require routine use of granulocyte colony-stimulating factor in contrast to the intensified cyclophosphamide dose utilized in IRS IV. Data are not yet available to determine whether ifosfamide is less damaging to gonadal function.
Appendix
Patients were entered onto the study from treatment centers in the following cities. We are grateful to the clinicians concerned for their support. Argentina: Buenos Aires; Belgium: Bruxelles (Fabiola), Bruxelles (St Joseph), Bruxelles (St Luc), Leuven, Montegnee; Denmark: Aarhus, Copenhagen; France: Angers, Brest, Besancon, Bordeaux (Centre Hospitalier Universitaire [CHU]), Bordeaux (CAC), Caen (CHU), Caen (CAC), Clermont Ferrand, Colmar, Grenoble, Lille (CAC), Lille (Calmette), Lille (St Antoine), Limoges, Lyon, Marseille, Montpellier, Nancy, Nantes, Nice, Paris (Curie), Paris (Trousseau), Poitiers, Reims, Rennes, Rouen (CHU), Rouen (CAC), St Etienne, Strasbourg (CHU), Toulouse (CHU), Tours, Villejuif; Republic of Ireland: Dublin; The Netherlands: Amsterdam (AMC), Amsterdam (VU), Nijmegen; Poland: Wroclaw; Spain: Badalona, Barcelona, Compostella, Las Palmas, Madrid (La Paz), Madrid (Nino Jesus), Madrid (12 October), Madrid (Zarzuela), Oviedo, Pamplona, Valencia (La Fe), Viscaya, Zaragoza; Switzerland: Basel, Lausanne, Lucerne, Zurich; United Kingdom: Belfast, Birmingham, Bristol, Cambridge, Cardiff, Edinburgh, Glasgow, Leeds, Leicester, Liverpool, London (Great Ormond Street), London (Middlesex), London (St Bartholomew's), Manchester, Newcastle, Nottingham, Oxford, Sheffield, Southampton, Sutton (Royal Marsden).
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
NOTES
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Raney RB, Anderson JR, Barr FG, et al: Rhabdomyosarcoma and undifferentiated sarcoma in the first two decades of life: A selective review of Intergroup rhabdomyosarcoma study group experience and rationale for Intergroup Rhabdomyosarcoma Study V. J Pediatr Hematol Oncol 23:215-220, 2001
Stiller CA, Stevens MCG, Magnani C, et al: Survival of children with soft tissue sarcoma in Europe since 1978: Results from the EUROCARE study. Eur J Cancer 37:767-774, 2001
Wolden SL, Anderson JR, Crist WM, et al: Indications for radiotherapy and chemotherapy after complete resection in rhabdomyosarcoma: A report from the Intergroup Rhabdomyosarcoma Studies I to III. J Clin Oncol 17:3468-3475, 1999
Flamant F, Rodary C, Voute PA, et al: Primary chemotherapy in the treatment of rhabdomyosarcoma in children: Trial of the International Society of Pediatric Oncology (SIOP)—Preliminary results. Radiother Oncol 3:227-236, 1985
Flamant F, Rodary C, Rey A, et al: Treatment of non metastatic rhabdomyosarcomas in childhood and adolescence: Results of the second study of the International Society of Paediatric Oncology: MMT 84. Eur J Cancer 34:1050-1062, 1998
Newton WA, Gehan EA, Webber BL, et al: Classification of rhabdomyosarcoma and related sarcomas: Pathologic aspects and proposals for a new classification—An Intergroup Rhabdomyosarcoma Study. Cancer 76:1073-1085, 1995
Rodary C, Flamant F, Donaldson SS, et al: An attempt to use a common staging system in rhabdomyosarcoma: A report of an international workshop initiated by the International Society of Pediatric Oncology (SIOP). Med Pediatr Oncol 17:210-215, 1989
Kaplan ES, Meier P: Non-parametric estimation from incomplete observation. J Am Stat Assoc 53:457-480, 1958
Peto R, Pike MC, Armitage P, et al: Design and analysis of randomised clinical trials requiring prolonged observation of each patient. Br J Cancer 35:1-39, 1977
Cox DR: Regression models and life table. J R Stat Soc (B) 34:187-220, 1972
Wharam MD, Hanfelt JJ, Tefft MC, et al: Radiation therapy for rhabdomyosarcoma: Local failure risk for Clinical Group III patients on Intergroup Rhabdomyosarcoma Study II. Int J Radiat Oncol Biol Phys 38:797-804, 1997
Heyn R, Raney RB, Hays DM, et al: Late effects of therapy in patients with paratesticular rhabdomyosarcoma. J Clin Oncol 10:614-623, 1992
Raney RB, Asmar L, Vassilopoulou-Sellin R, et al: Late complications of therapy in 213 children with localized, non orbital soft tissue sarcoma of the head and neck: A descriptive report from the Intergroup Rhabdomyosarcoma Studies (IRS)-II and -III. Med Pediatr Oncol 33:362-371, 1999
Raney RB, Anderson JR, Kollath J, et al: Late effects of therapy in 94 patients with localized rhabdomyosarcoma of the orbit: Report from the Intergroup Rhabdomyosarcoma Study (IRS)-III, 1984-1991. Med Pediatr Oncol 34:413-420, 2000
Heyn R, Haeberlen V, Newton WA, et al: Second malignant neoplasms in children treated for rhabdomyosarcoma. J Clin Oncol 11:262-270, 1993
Rodary C, Rey A, Olive D, et al: Prognostic factors in 281 children with non metastatic rhabdomyosarcoma (RMS) at diagnosis. Med Pediatr Oncol 16:71-77, 1988
Rousseau P, Flamant F, Quintana E, et al: Primary chemotherapy in rhabdomyosarcomas and other malignant mesenchymal tumours of the orbit: Results of the International Society of Pediatric Oncology MMT 84 Study. J Clin Oncol 12:516-521, 1994
Oberlin O, Rey A, Anderson J, et al: Treatment of orbital rhabdomyosarcoma: Survival and late effects of treatment—Results of an international workshop. J Clin Oncol 19:197-204, 2001
Anderson J, Raney RB, Carli M, et al: International study of characteristics and outcome of patients with primary rhabdomyosarcoma of the bladder/prostate. Med Pediatr Oncol 37:181, 2001
Stewart RJ, Martelli H, Oberlin O, et al: Treatment of children with non metastatic paratesticular rhabdomyosarcoma: Results of the MMT 84 and MMT 89 studies of the International Society of Paediatric Oncology. J Clin Oncol 21:793-795, 2003
Ferrari A, Bisogno G, Casanova M, et al: Paratesticular rhabdomyosarcoma: Report from the Italian and German Cooperative Group. J Clin Oncol 20:449-455, 2002
Rodary C, Flamant F, Maurer H, et al: Initial lymphadenectomy is not necessary in localised completely resected paratesticular rhabdomyosarcoma. Med Pediatr Oncol 20:430, 1992
Wiener ES, Anderson JR, Ojimba JI, et al: Controversies in the management of paratesticular rhabdomyosarcoma: Is staging retroperitoneal lymph node dissection necessary for adolescents with resected paratesticular rhabdomyosarcoma Semin Pediatr Surg 10:146-152, 2001
Crist W, Gehan EA, Ragab AH, et al: The third Intergroup rhabdomyosarcoma study. J Clin Oncol 13:610-630, 1995
Koscielniak E, Harms D, Henze G, et al: Results of treatment for soft tissue sarcoma in childhood and adolescence: A final report of the German Cooperative Soft Tissue Sarcoma Study CWS-86. J Clin Oncol 17:3706-3719, 1999
Crist WM, Anderson JR, Meza JL, et al: Intergroup rhabdomyosarcoma study-IV: Results for patients with non metastatic disease. J Clin Oncol 19:3091-4102, 2001
Benk V, Rodary C, Flamant F, et al: Parameningeal rhabdomyosarcoma: Results of an international workshop. Int J Radiat Oncol Biol Phys 36:533-540, 1996
Carli M, Colombatti R, Oberlin O, et al: High dose melphalan with autologous stem-cell rescue in metastatic rhabdomyosarcoma. J Clin Oncol 17:2796-2803, 1999
Oberlin O, Rey A, Praquin MT, et al: Non metastatic rhabdomyosarcoma (RMS): Outlines for the SIOP MMT 95 study. Med Pediatr Oncol 27:265, 1996
Donaldson SS, Meza J, Breneman JC, et al: Children's Oncology Group Soft Tissue Sarcoma Committee (formerly Intergroup Rhabdomyosarcoma Group) representing the Children's Oncology Group and the Quality Assurance Review Center: Results from the IRS-IV randomized trial of hyperfractionated radiotherapy in children with rhabdomyosarcoma—A report from the IRSG. Int J Radiat Oncol Biol Phys 51:718-728, 2001
Suarez A, Mc Dowell H, Niaudet P, et al: Long term follow-up of Ifosfamide renal toxicity in children treated for malignant mesenchymal tumours: An International Society of Pediatric Oncology report. J Clin Oncol 9:2177-2182, 1991
Skinner R, Cotterill SJ, Stevens MCG: Risk factors for nephrotoxicity after ifosfamide treatment in children: A UKCCSG Late Effects Group Study. Br J Cancer 82:1636-1645, 2000(Michael C.G. Stevens, Ann)