THE OUTCOME OF treatment for children with metastatic

Transcription

1 High-Dose Melphalan With Autologous Stem-Cell Rescue in Metastatic Rhabdomyosarcoma By M. Carli, R. Colombatti, O. Oberlin, M. Stevens, L. Masiero, E. Frascella, E. Koscielniak, J. Treuner, and C.R. Pinkerton Purpose: The European Collaborative MMT4-91 trial was conducted as a prospective nonrandomized study to evaluate the potential benefit of high-dose melphalan as consolidation of first complete remission in children with stage IV rhabdomyosarcoma. Patients and Methods: Fifty-two patients in complete remission after six courses of chemotherapy received megatherapy : 42 received melphalan alone, whereas 10 received melphalan in combination with etoposide, carboplatin/etoposide, or thiotepa/busulfan and etoposide. The outcome of this group of patients was compared with that observed in 44 patients who were also in complete remission after six courses of identical chemotherapy (plus surgery or radiotherapy) but went on to receive a total of up to 12 courses of conventional chemotherapy (four cycles). No differences were found between the two groups regarding clinical characteristics, chemotherapy received before complete remission, or response to chemotherapy. In particular, there was no significant difference between the groups for site of primary tumor, histologic subtype, age at presentation, presence of bone or bone marrow metastases, or number of metastases. THE OUTCOME OF treatment for children with metastatic rhabdomyosarcoma (RMS) remains disappointing. Despite intensification of conventional chemotherapy, there has been little evidence of any significant improvement in overall survival, which has remained at approximately 20% to 30%. 1,2 A preliminary analysis of prognostic factors in 175 children on the European Collaborative Studies MMT4-89 and MMT4-91 has shown the following factors to have a significant adverse impact on prognosis: presence of bone and bone marrow metastases; number of metastases; From the Department of Pediatrics, Oncology/Hematology Division, University of Padova, Padova, Italy; Institut Gustave Roussy, Villejuif, France; The Birmingham Children s Hospital, Birmingham, and The Royal Marsden Hospital, Children s Department, London, United Kingdom; and the Department of Oncology/Hematology, Pediatric Center Olgahospital, Stuttgart, Germany. Submitted February 5, 1999; accepted May 27, Supported in part by the Consiglio Nazionale delle Ricerche Progetto Finalizzato (CNR ACRO n PF39) and MURST. Address reprint requests to Modesto Carli, MD, Department of Pediatrics. Oncology/Hematology Division, University of Padova, Via Giustiniani 3, Padova, Italy; carli@child.pedi.unipd.it by American Society of Clinical Oncology X/99/ Results: The 3-year event-free survival (EFS) and overall survival (OS) rates were 29.7% and 40%, respectively, for those receiving high-dose melphalan or other multiagent high-dose regimens and 19.2% and 27.7%, respectively, for those receiving standard chemotherapy. The difference was not statistically significant (P.3 and P.2 for EFS and OS, respectively). There was a significant prolongation in the time from the last day of high-dose chemotherapy or the end of chemotherapy cycle 4 to the time of relapse in those receiving megatherapy (168 days for patients receiving megatherapy v 104 days for those receiving standard therapy; P.05). Conclusion: The addition of a high-dose alkylating agent to consolidation therapy may have prolonged progression-free survival in this poor-risk patient group, but it did not significantly improve the ultimate outcome. J Clin Oncol 17: by American Society of Clinical Oncology. limbs, parameningeal head and neck, and other sites as primary sites; and age older than 10 years at presentation. In particular, on multivariate analysis, age, site of primary tumor, and bone marrow metastases each achieved statistical significance for prognostic impact. 3 Similar observations have been confirmed in recent analyses of the Intergroup Rhabdomyosarcoma Study (IRS) data. 4 High-dose chemotherapy with autologous stem-cell rescue has been investigated over several years as a method of eliminating micrometastatic residual disease in patients who achieve a clinical complete remission (CR) after conventional chemotherapy. This strategy has been applied in the treatment of a number of pediatric cancers, in particular neuroblastoma, rhabdomyosarcoma, Ewing s sarcoma, and Wilm s tumor. 5 The majority of published studies demonstrate the feasibility of this approach in small series of patients, but few if any conclusions can be drawn regarding efficacy. Melphalan has been shown to be an active agent in RMS in the xenograft model. 6 Although response rates were low in a phase II study using conventional doses of melphalan, 7 the drug was clearly active in a window study of untreated patients Journal of Clinical Oncology, Vol 17, No 9 (September), 1999: pp

2 HIGH-DOSE MELPHALAN FOR RHABDOMYOSARCOMA 2797 In early studies of megatherapy, melphalan was selected because its major dose-limiting toxicity is myelosuppression, and the short half-life of the drug (related to hepatic metabolism) allowed the reinfusion of noncryopreserved bone marrow within 12 to 24 hours, thus avoiding the need for cryopreservation. At doses greater than 240 mg/m 2, dose-limiting oral mucositis and intestinal toxicity occur. The phase II studies by McElwain et al 9 demonstrated that high-dose melphalan was an active agent in RMS. Response rates with high-dose administration seemed to be higher than those achieved with the conventional dose, which is consistent with in vitro studies showing a clear dose-response relationship. The innovative rapid vincristine, doxorubicin, and cyclophosphamide (VAC)/melphalan protocol from the Royal Marsden Hospital comprised six courses of standard-dose VAC delivered over an 8-week period followed by a single course of high-dose melphalan, with unpurged autologous bone marrow rescue. Irradiation was given to sites at which CR had not been achieved. In patients who were without metastatic disease, this short, high-dose regimen produced results comparable to those achieved with the more conventional 18 months to 2 years VAC regimen in use at that time. However, no obvious improvement in survival was apparent in children with metastatic disease. 10 Subsequently, a wide range of high-dose chemotherapy regimens have been used in patients with poor-risk soft tissue sarcoma on the basis of results achieved with highdose melphalan, multiagent chemotherapy, or total-body irradiation (TBI). Data from the European Bone Marrow Transplantation Group have indicated little, if any, impact on patients with relapsed RMS, but it has been difficult to draw any firm conclusions about the potential impact when such regimens are used as consolidation therapy for high-risk patients who are in first remission. 11 In 1989, the European Intergroup Study (comprising the International Society of Pediatric Oncology [SIOP], the Associazione Italiana Ematologia ed Oncologia Pediatrica, and the German Society for Pediatric Oncology and Hematology) designed a very intensive six-drug multiagent regimen (MMT4-89) for patients with metastatic soft tissue sarcoma. This regimen included most of the drugs active in RMS at doses which, in combination, were close to the maximum-tolerated doses. 12 In 1991, the protocol was modified (MMT4-91) to evaluate the potential of high-dose melphalan as consolidation therapy of first CR. All patients in clinical CR before the third cycle of six-drug chemotherapy were eligible for high-dose chemotherapy. In this analysis, the outcome of this group was compared with that of a similar group of patients treated in the two studies but who were not given high-dose chemotherapy, either as directed by protocol MMT4-89 or as a result of participating center decisions in MMT4-91. PATIENTS AND METHODS One hundred seventy-five eligible RMS patients, ranging in age from 3 months to 18 years (median, 7.7 years), were enrolled onto the Intergroup European Studies MMT4-89 and MMT4-91. The treatment plan is shown in Fig 1. Each cycle consisted of a combination of the following drugs administered every 21 days: course 1, epirubicin 150 mg/m 2 administered intravenously (IV) on day 1, carboplatin 500 mg/m 2 IV on day 1, vincristine 1.5 mg/m 2 IV on days 1 and 8; course 2, ifosfamide 3 g/m 2 IV on days 1 through 3, actinomycin 1.5 mg/m 2 IV on Fig 1. Treatment schema. Abbreviations: HCR, histologic complete remission; CCR, clinical complete remission; PR, partial remission; BMT, bone marrow transplantation; MEGATH, megatherapy.

3 2798 CARLI ET AL day 1, vincristine 1.5 mg/m 2 IV on days 1 and 8; and course 3, ifosfamide g/m 2 IV on days 1 through 3, etoposide 200 mg/m 2 IV on days 1 through 3, vincristine 1.5 mg/m 2 IV on days 1 and 8. In the MMT4-89 protocol, treatment consisted of four identical 9-week cycles; in the MMT4-91 protocol, the fourth cycle was replaced with high-dose melphalan (200 mg/m 2 IV) with autologous bone marrow or peripheral-blood stem-cell (PBSC) rescue. Bone marrow harvest or PBSC collection was performed after the third cycle in patients who had bone marrow metastatic involvement at diagnosis (assuming bone marrow CR had been achieved) and after the second cycle for those who did not have metastatic involvement of the bone marrow. Second-look surgery was planned, if feasible, after the second cycle. Radiotherapy (40 Gy) was recommended for patients with microscopic residual disease or uncertain margins after surgical resection involving either primary tumor or metastases. Second-look surgery was not electively performed for patients with sites of metastatic disease at which there was a CR after chemotherapy. Ninety-seven of 175 patients, in CR before the third cycle, were considered eligible for this analysis, and 96 were assessable. One patient was excluded from the analysis because of missing information regarding the type of consolidation therapy used. Statistical Analysis Data collected by the dbase-iv program (Borland Inprise Corp, Scotts Valley, CA) were analyzed using the SAS for Windows package (SAS Institute Inc., Cary, NC). In comparing the two treatment protocols, the 2 test was used for determining the heterogeneity of all clinical patient characteristics considered. 13 The differences in the distributions of time interval from diagnosis to last chemotherapy cycle or to megatherapy and from the end of therapy to relapse in the two patient groups were analyzed with t test and distribution-free tests, such as the Wilcoxon nonparametric score test. Survival curves were calculated according to the Kaplan-Meier method. 14 The overall survival (OS) curve was calculated considering the time interval between the date of the diagnosis and the date of last follow-up or death. For event-free survival (EFS), the interval between diagnosis and the date of last follow-up, relapse, or death from any causes was considered. All data were obtained through June 30, The log-rank test was adopted to assess differences in univariate analysis. 15 Clinical CR was defined as the disappearance of all signs of tumor based on clinical and imaging evidence. Histologic CR was defined as CR with histologically confirmed absence of disease, and partial remission was defined as a greater than 50% decrease in the sum of the products of the maximum perpendicular diameters of all measurable lesions. Table 1. Clinical Characteristics of the Patient Population Variable Megatherapy (no. of patients) Standard Chemotherapy (no. of patients) Total no. of patients Age, years Site Extremity PM Other GU HN 13 9 Histology Alveolar Embryonal NOS Bone metastases Yes No Bone marrow metastases Yes No Lung metastases Yes No Lymph node metastases Yes No No. of metastases Invasiveness T T Regional lymph nodes N N NX 8 9 Size, cm Metastatic site Single Multiple Abbreviations: PM, parameningeal head and neck; GU, genitourinary tract; HN, head and neck (nonparameningeal); NOS, not otherwise specified. RESULTS Ninety-six assessable patients were in CR before the third cycle of chemotherapy: 55 were in CR at week 18 after the second cycle, whereas the remainder achieved CR with excision of the residual tumor, which was accomplished via second-look surgery and/or radiation therapy. Fifty-two of 96 children received megatherapy, whereas 44 continued with a fourth cycle of the same chemotherapy. The decision not to use megatherapy in MMT4-91 was largely centerbased and did not depend on presenting features or response to initial chemotherapy in individual cases. Of 52 children included in the megatherapy group, 42 received melphalan alone according to the protocol guideline, whereas 10 children received melphalan in combination with etoposide, carboplatin/etoposide, thiotepa/busulfan, and etoposide. The clinical characteristics of the two groups of patients are listed in Table 1. In the megatherapy group, there was a slightly higher number of children who were older than 10 years and had alveolar histology, but overall there were no P

4 HIGH-DOSE MELPHALAN FOR RHABDOMYOSARCOMA 2799 Variable Table 2. Therapeutic Variables Megatherapy (no. of patients) Standard Chemotherapy (no. of patients) Total no. of patients Week 9 response Complete response Partial response No response 0 2 Missing data Week 18 response Complete response Partial response No response 1 3 Surgery Yes No Radiotherapy Yes No Missing data 2 significant differences between the two groups for any of the characteristics with unfavorable prognostic impact (bone, bone marrow involvement, alveolar histology, unfavorable sites of primary tumor, and age older than 10 years). A significant difference was observed in the incidence of lymph node metastases, which were more frequent among megatherapy patients. Details of therapy are shown in Table 2. No differences were observed between the two groups either in response to chemotherapy as evaluated after the first cycle (week 9) or P the second cycle (week 18), the number of patients who had surgery for residual tumor, or the use of radiation therapy (mostly for microresidual disease). The median time from diagnosis to high-dose melphalan administration was 264 days (range, 145 to 420 days), whereas the median time to the start of the fourth chemotherapy cycle was 259 days (range, 191 to 321 days). Median follow-up of the surviving patients was 42.8 months. The median follow-up durations for standard chemotherapy and megatherapy groups were 45.1 and 34.7 months, respectively (P.9). The 3-year EFS rates were 29.7% for patients who received high-dose chemotherapy versus 19.2% for those who received standard chemotherapy (P.3; Fig 2); 3-year OS rates were 40% for patients who received high-dose chemotherapy versus 27.7% for those who received standard chemotherapy (P.2; Fig 3). The median time from the last days of high-dose chemotherapy or the end of the fourth cycle to relapse was 168 days for patients in the high-dose group and 104 days for patients in the standard chemotherapy group. This reaches statistical significance (P.05, Wilcoxon test). Within the limitation of the sample size, no difference was noted in the 3-year EFS rate between patients who received consolidation with high-dose melphalan alone and those who received melphalan combined with a range of other chemotherapy agents (27 of 42 patients v seven of 10 patients, respectively; alive event-free P.5). The tumor recurred in the majority of patients at sites of previous disease. Fig 2. EFS of patients who received megatherapy versus those who received standard chemotherapy.

5 2800 CARLI ET AL Fig 3. OS of patients who received megatherapy versus those who received standard chemotherapy. The 3-year EFS and OS rates for the 175 patients enrolled on the study were 20% (95% confidence interval [CI], 14% to 27%) and 28% (95% CI, 21% to 35%), respectively. The 3-year EFS rate of the 78 children who did not achieve CR before the third cycle was 10.8%, compared with 27.8% for the 97 children who were in CR (P.0001; Fig 4). The corresponding 3-year OS rates were 14% and 41% (P.0001), respectively (Fig 5). The toxicity of high-dose melphalan was comparable to that previously reported in detail in other tumor types, particularly neuroblastoma. The median time for neutrophil recovery (polymorphonuclear neutrophils 500/µL) was 16 days; median time for platelet recovery ( 50,000/µL) was 41 days. There was one toxic death as a result of sepsis associated with megatherapy. There were no documented cases of acute hepatotoxicity compatible with venoocclusive disease. One death due to anthracycline-related cardiotoxicity was observed in the standard chemotherapy group. DISCUSSION Due to the relative rarity of metastatic RMS, it has proved difficult to mount a prospective randomized trial of megatherapy, even at a European collaborative level. One attempt by the SIOP group to evaluate the role of high-dose consolidation chemotherapy in young children with parameningeal RMS was unsuccessful due to poor recruitment. 16 Because of the relatively short time period over which the current trial was run and the lack of selection in metastatic patients entered on the MMT4-89 and MMT4-91 trial, a historical comparison provides useful information regarding the role of high-dose melphalan in this setting. It is apparent from this study that a single course of high-dose melphalan or melphalan combined with a range of other chemotherapy agents has a limited role as consolidation of clinical CR in metastatic RMS. The relatively small numbers of patients assigned to the two arms of this study and the lack of complete standardization of high-dose regimens represent limitations of the study. The study would have been incapable of demonstrating a significant difference of 10% to 15% in EFS between the two groups. However, it is apparent that the outcome in the group of children who received an intensive induction regimen and then megatherapy is little different from that which was previously reported in unselected children with metastatic RMS, for example, by the IRS group. 1 There is a suggestion that there was some effect from megatherapy in that progression-free survival and OS at 3 years were higher, but this was not statistically significant. Only the time to relapse was significantly extended. A similar observation was seen in a study conducted by the European Neuroblast Study Group I, in which high-dose melphalan seemed to shift the relapse-free survival curve to the right, but this had little impact on overall survival. 17

6 HIGH-DOSE MELPHALAN FOR RHABDOMYOSARCOMA 2801 Fig 4. EFS of patients in CR before the third chemotherapy cycle versus those who were not in CR. Although the comparison group seems similar to the group receiving high-dose consolidation therapy, the slight overrepresentation of children younger than 10 years of age with embryonal tumors, who in the IRS group did much better than all others, 4 might have reduced any difference in EFS and OS. Moreover, the duration of treatment in the MMT4-91 regimen was relatively short, and it is possible that a similar marginal benefit could have been achieved by more prolonged chemotherapy as used by the IRS. In the new MMT4-98 study, the use of VAC for a further 6 months after an intensive induction regimen is to be evaluated. Analysis of these data emphasizes the importance of achieving CR before week 18 (third cycle). In fact, there is a statistically significant difference (P.001) both in the EFS and OS between the patients who were in CR before the third cycle and those who were not. Is it likely that a more intensive high-dose regimen would be any more effective? A number of studies of multiagent regimens have been reported. Some investigators have added TBI to the regimen, but it seems unlikely that a dose of 10 to 14 Gy would be sufficient in RMS, when at least 45 Gy is administered with conventional radiotherapy. The dose equivalence between single dose and fractionated dose remains uncertain. The small studies that have described combinations of cyclophosphamide/tbi, melphalan/tbi, or carboplatin/etoposide/tbi have included too few patients with previously untreated metastatic RMS to draw any real conclusions about efficacy. Similarly, the addition of other alkylating agents to melphalan, such as busulfan, cyclophosphamide, or carmustine, or the combination with carboplatin and etoposide, have only been reported in pilot studies. 24,25 These combinations inevitably increase toxicity, both in the short and long term. The two largest studies using megatherapy in first remission reported to date are from the German/Austrian Pediatric Bone Marrow Transplantation Group and the Memorial Sloan-Kettering (MSK) Cancer Center. Koscielniak et al 26 reported the outcome in 27 patients with primary metastatic disease. Although the induction therapy was relatively standardized, there were considerable variations in the megatherapy used. Most were based on the combination of melphalan, etoposide, and carboplatin, but one half of patients in the study group also received TBI. Moreover, five patients were allografted and several received postgraft interleukin-2. Only five of 27 patients were reported to be disease-free at median follow-up (postgraft) of 27 months. Boulad et al 27 described a series of 26 children with poor-risk sarcoma, including 14 with primary metastatic disease (two with soft tissue Ewing s sarcoma) who received a dose-intensive regimen with planned megatherapy as consolidation of first remission; 19 eventually received consolidation, of whom nine originally had metastatic disease. Five of nine patients with metastatic sarcoma were alive at last follow-up (2-year EFS of 50%).

7 2802 CARLI ET AL Fig 5. OS of patients in CR before third chemotherapy cycle versus those who were not in CR. The MSK regimen is similar to the MMT4-89/MMT4-91 induction chemotherapy regimen, but it does not contain carboplatin and applies a higher dose of etoposide. It is unclear which component of the MSK regimen contributed to their encouraging results, as both intensive induction and consolidation were used. The study reported by Horowitz et al 20 combined 91 high-risk Ewing s sarcoma and RMS patients, of whom 49 presented with metastases. Patients received megatherapy in remission using doxorubicin/vac/tbi regimen. In these children, the EFS at 6 years was 19%. For the group as a whole (including nonmetastatic cases), there was no difference in outcome for poor-risk RMS and Ewing s sarcoma patients; OS rates were 28% and 34%, respectively. The MMT4-91 study is, to date, the largest series of children with primary metastatic RMS treated with a standardized induction regimen who then received a melphalanbased, non-tbi regimen. There are many reasons why a single course of high-dose chemotherapy given 4 to 6 months after initial presentation might not be expected to be effective. 28 At this stage, residual tumor is likely to have developed a range of drug resistance mechanisms, and in view of the extensive prior exposure to alkylating agents such as ifosfamide or cyclophosphamide, it would not be unexpected for there to be a relative resistance to high-dose melphalan. Unlike leukemia and lymphoma, the anticipated cell kill from a single course of high-dose therapy is unlikely to be sufficient to overcome such resistance. There is little support from studies of neuroblastoma that double high-dose procedures are any more effective at this time. 29,30 An alternative strategy may be to begin dose escalation earlier in treatment in an attempt to avoid drug resistance, and a current SIOP pilot study of stage IV malignant mesenchymal tumor has been designed to evaluate the potential value of sequential high-dose therapy, starting within 4 to 6 weeks of initial presentation. 31 Because of the need to perform PBSC harvest and the likelihood that many of these patients may have bone marrow involvement at diagnosis, earlier introduction of dose escalation is impractical. Repeated courses of high-dose cyclophosphamide, high-dose etoposide, and high-dose carboplatin are relatively well tolerated, and PBSC rescue is only necessary after high-dose carboplatin. There is also interest in alternative agents, such as high-dose thiotepa, for which there have been encouraging phase II data. 32 Attempts have been made to purge reinfused marrow or PBSC using mafosfamide, but this was not used in the present study nor will it be used in the new study. The relevance of potentially reinfusing tumor cells remains contentious, although recent gene-marking studies have shed some light on the subject. 33 The intention has been to delay harvesting until a stage at which there has been in vivo cytoreduction, although this is more difficult when highdose therapy and PBSC rescue are given earlier in treatment. In conclusion, although this was not a randomized comparison of two treatment strategies, each group was

8 HIGH-DOSE MELPHALAN FOR RHABDOMYOSARCOMA 2803 similar with respect to clinical characteristics at diagnosis and therapy they received to achieve CR. There may have been some delay to relapse, but there was no clear evidence of any survival benefit from the use of high-dose melphalan as consolidation therapy in patients with metastatic disease who achieve CR with conventional chemotherapy. We would conclude that on the basis of these data, the routine use of high-dose melphalan alone is not justified. Alternative strategies are required to improve outcome for this group of patients with poor-prognosis disease. 1. Crist W, Gehan EA, Ragab AH, et al: The Third Intergroup Rhabdomyosarcoma Study. J Clin Oncol 13: , Koscielniak E, Rodary C, Flamant F, et al: Metastatic rhabdomyosarcoma and histologically similar tumours in childhood: A restrospective European multicentre analysis. Med Pediatr Oncol 20: , Carli M, Colombatti R, Pinkerton R, et al: Prognostic factors in children with metastatic rhabdomiosarcoma: Results of the European Intergroup Studies (EIS) MMT 89 and MMT 91. Proc SIOP Med Pediatr Oncol 29:324, 1997 (abstr 0-32) 4. Anderson JR, Ruby E, Link M, et al: Identification of a favorable subset of patients (PTS) with metastatic (MET) rhabdomyosarcoma (RMS): A report from the Intergroup Rhabdomyosarcoma Study Group (IRSG). Proc Am Soc Clin Oncol 16:510, 1997 (abstr 1836) 5. Atra A, Pinkerton CR: Autologous stem cell transplantation in solid tumours of childhood. Ann Med 28: , Houghton JA, Cook RL, Lutz PJ: Melphalan: A potential new agent in the treatment of childhood rhabdomyosarcoma. Cancer Treat Rep 69:91-96, Belasco JB, Mitchell CD, Rohrbaugh T, et al: IV Melphalan in children. Cancer Treat Rep 1: , Horowitz M, Etcubanas E, Christesed M, et al: Phase II testing of melphalan in children with newly diagnosed rhabdomyosarcoma: A model for anticancer drug development. J Clin Oncol 6: , Bagnulo S, Perez DJ, Barrett A, et al: High dose melphalan and autologous bone marrow transplantation for solid tumours of childhood. Eur Paediatr Haematol Oncol 2: , Pinkerton CR, Groot-Loonen, Barrett J, et al: Rapid VAC high dose melphalan regimen: A novel chemotherapy approach for childhood soft tissue sarcomas. Br J Cancer 64: , Pinkerton CR: Megatherapy for soft tissue sarcomas: EBMT experience. Bone Marrow Transplant 7: , Carli M, Pinkerton P, Frascella E, et al: Intensive chemotherapy for metastatic sarcoma in children. SIOP European Intergroup Study MM89. Proc Am Soc Clin Oncol 12:410, 1993 (abstr 1404) 13. Pearson ES, Hartley HO: Biometrika Tables for Statisticians. Cambridge, England, Cambridge University Press, Kaplan EL, Meier P: Nonparametric estimation from incomplete observation. J Am Stat Assoc 53: , Cox DR, Oakes D: Analysis of Survival Data. London, England, Chapman and Hall, 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 MMT84. Eur J Cancer 34: , Pinkerton CR: ENSGI randomised study on high dose melphalan in neuroblastoma. Bone Marrow Transplant 7: , 1991 (suppl 3) 18. Dumontet C, Biron P, Bouffet E, et al: High dose chemotherapy with ABMT in soft tissue sarcoma: A report of 22 cases. Bone Marrow Transplant 20: , Blay JY, Bouhour D, Brunat-Mentigny M, et al: High dose chemotherapy (VIC) and bone marrow support in advanced sarcomas. Proc Am Clin Oncol 13:479, 1994 (abstr 1672) REFERENCES 20. Horowitz ME, Kinsella TJ, Wexler LH, et al: Total-body irradiation and autologous bone marrow transplant in the treatment of high-risk Ewing s sarcoma and rhabdomyosarcoma. J Clin Oncol 11: , Emminger W, Emminger-Schmidmeier W, Peters C, et al: Is treatment intensification by adding etoposide and carboplatin to fractionated total body irradiation and melphalan acceptable in children with solid tumors with respect to toxicity? Bone Marrow Transplant 8: , Pinkerton R, Philip T, Barrett A, et al: Massive therapy with total body irradiation and bone marrow transplantation in children with relapsed or advanced rhabdomyosarcoma. Proc SIOP XVII Meeting, Venice, Italy, September 30-October 4, Hoover ML, Walterhouse DO, Morgan EM, et al: High dose chemotherapy with peripheral blood stem cell rescue (HDC/PBSCR) as consolidation for stage IV rhabdomyosarcoma (RMS). Proc Am Clin Oncol 15:351, 1996 (abstr 1031) 24. Hartmann O, Benhampou E, Beaujean F, et al: High-dose busulfan and cyclophosphamide with autologous bone marrow transplantation support in advanced malignancies in children: A phase II study. J Clin Oncol 4: , Ozkaynak MF, Matthay K, Cairo M, et al: A non TBI regimen for autologous hematopoietic stem cell transplantation in pediatric solid tumors. Proc Am Clin Oncol 16:514, 1997 (abstr 1848) 26. Koscielniak E, Klingebiel TH, Peters C, et al: Do patients with metastatic and recurrent rhabdomyosarcoma benefit from high-dose therapy with hematopoietic rescue? Report of the German/Austrian Pediatric Bone Marrow Transplantation Group. Bone Marrow Transplant 19: , Boulad F, Kernan NA, LaQuaglia MP, et al: High dose induction chemoradiotherapy followed by autologous bone marrow transplantation as consolidation therapy in rhabdomyosarcoma, extraosseous Ewing s sarcoma, and undifferentiated sarcoma. J Clin Oncol 16: , Pinkerton CR: Massive chemotherapy followed by bone marrow graft in paediatric oncology: Arguments against. Bull Cancer 82:42-45, Grupp SA, Diller J, Gorlin J, et al: Feasibility of double peripheral blood progenitor cell (PBPC) transplant in children with advanced neuroblastoma and sarcomas. Proc Am Soc Clin Oncol 15:465, 1996 (abstr 1465) 30. Hartmann O, Benhamou E, Beaujean F, et al: Repeated high dose chemotherapy followed by purged autologous bone marrow transplantation as consolidation therapy in metastatic neuroblastoma. J Clin Oncol 5: , Foot ABM, Pinkerton CR, Stevens M, et al: Sequential rapid high dose single agent consolidation therapy for metastatic sarcoma in children. Proc SIOP Med Pediatr Oncol 29:409, 1997 (abstr P-89) 32. Valteau-Couanet D, Lucidarme N, Pein F, et al: High dose thiotepa with hematopoietic stem cell support in children with solid tumors: A phase II study. Proc Am Soc Clin Oncol 15:464, 1996 (abstr 1458) 33. Shapall EJ, Cagnoni PJ, Gehling U, et al: Bone marrow purging, in Armitage JO, Antman KH (eds): High Dose Cancer Therapy. Baltimore, MD, Williams and Wilkins, 1995, pp