|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 1999 American Society for Clinical Oncology Indications for Radiotherapy and Chemotherapy After Complete Resection in Rhabdomyosarcoma: A Report From the Intergroup Rhabdomyosarcoma Studies I to IIIFrom the Department of Radiation Oncology, Memorial Sloan-Kettering Cancer Center, New York, NY, Department of Biostatistics, University of Nebraska Medical Center, Omaha, NE; Department of Pediatric and Adolescent Medicine, Mayo Clinic, Rochester, MN; Department of Radiation Oncology, University Hospital of Cincinnati, Cincinnati, and Department of Laboratory Medicine, Columbus Children's Hospital, Columbus, OH; Department of Radiation Oncology, Johns Hopkins Oncology Center, Baltimore, MD; Department of Surgery, Children's Hospital of Pittsburgh, Pittsburgh, PA; and Department of Radiation Oncology, Stanford University Medical Center, Stanford, CA. Address reprint requests to Suzanne Wolden, MD, Department of Radiation Oncology, Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY 10021.
PURPOSE: To evaluate the outcome of patients with rhabdomyosarcoma (RMS) treated with complete surgical resection and multiagent chemotherapy, with or without local radiotherapy (RT). PATIENTS AND METHODS: Four hundred thirty-nine patients with completely resected (ie, group I) RMS were further treated with chemotherapy (vincristine and actinomycin D ± cyclophosphamide, doxorubicin, and cisplatin) on Intergroup Rhabdomyosarcoma Studies (IRS) I to III between 1972 and 1991. Eighty-three patients (19%) also received local RT as a component of initial treatment. RESULTS: Eighty-six patients relapsed (10-year failure-free survival [FFS] 79%, overall survival 89%). Six percent of failure sites were local, 6% were regional, and 7% were distant. Poor prognostic factors were tumor size greater than 5 cm, alveolar or undifferentiated histology, primary tumor sites other than genitourinary, and treatment on IRS-I or II. For patients with embryonal RMS who were treated with RT, there was a trend for improved FFS but no difference in overall survival. On IRS-I and II, patients with alveolar or undifferentiated sarcoma who received RT compared with those who did not receive RT had greater 10-year FFS rates (73% v 44%, respectively; P = .03) and overall survival rates (82% v 52%, respectively; (P = .02). Such patients who received RT on IRS III also benefited more than those who did not receive RT (10-year FFS, 95% v 69%; P = .01; overall survival, 95% v 86%; P = .23). CONCLUSION: Patients with group I embryonal RMS have an excellent prognosis when treated with adjuvant multiagent chemotherapy without RT. Patients with alveolar RMS or undifferentiated sarcoma fare worse; however, FFS and overall survival are substantially improved when RT is added to multiagent chemotherapy (IRS-I and II). The best outcome occurred in IRS-III, when RT was used in conjunction with intensified chemotherapy.
RHABDOMYOSARCOMA (RMS) is a highly malignant childhood cancer thought to arise from striated muscle progenitor cells. Remarkable progress has occurred in both biology and therapy, and the majority of children are now cured with combined modality treatment.1 Since 1972, the Intergroup Rhabdomyosarcoma Study Group (IRSG) has conducted a series of prospective randomized trials with the aim of improving cure rates, while minimizing treatment-related morbidity, using multidisciplinary management.2-4 The benefit of multiagent systemic chemotherapy for all patients with RMS has been clearly demonstrated. Localized tumors that can be completely resected are categorized as group I according to the IRSG clinical grouping system.5 The use of radiotherapy (RT) has been investigated for patients with group I (ie, completely excised) tumors collectively, but this report is the first to look at the value of RT within subsets of patients with pathologically different tumors (ie, alveolar RMS and undifferentiated sarcoma v embryonal RMS). Patients with alveolar RMS or undifferentiated sarcoma have been shown not to fare as well on similar therapy compared with those with embryonal tumors, including botryoid, and spindle cell variants.6,7 We present mature data from the Intergroup Rhabdomyosarcoma Studies (IRS) I to III that define prognostic factors, patterns of failure, outcome, and optimal management for patients with group I disease.
Patients Our analysis included previously untreated patients with biopsy proven RMS, extraosseous Ewing's sarcoma, or undifferentiated sarcoma, who were entered onto IRSG studies IRS-I through IRS-III with clinical-pathologic group I disease. All patients were under 21 years old. Patients were classified as having group I disease if they had complete removal of a localized tumor, leaving microscopically uninvolved surgical margins, and no clinical or radiographic evidence for nodal or distant metastases. If regional lymph nodes were sampled, they had to be pathologically negative.
Staging and Treatment Patients treated in IRS-I with group I tumors were scheduled to receive 2 years of vincristine (V), actinomycin D (A), and cyclophosphamide (C) (VAC) and to be randomized to either receive or not receive concurrent RT.2 Children who underwent amputation were assigned to VAC chemotherapy without RT. When RT was used, volumes treated included the tumor bed but not the regional nodal areas. The IRS-I recommended radiation doses were 50 to 60 Gy in once-daily fractions of 1.5 to 2.0 Gy over 5 to 6 weeks beginning immediately after surgery. Doses for children under 3 years old were limited to a maximum of 40 Gy. The IRS-II protocol specified no RT for patients with group I tumors.3 They were randomized to receive VAC for 2 years versus VA for 1 year. Patients with extremity alveolar RMS were treated with 2 years of repetitive pulse VAC without RT.8 IRS-III treatment regimens separated patients with group I tumors by histology.4 Patients with embryonal RMS and undifferentiated tumors were treated with VA chemotherapy for 1 year without RT. Patients with alveolar histology were entered onto a single regimen of pulse VAC and doxorubicin plus cisplatin, with local RT to the preoperative tumor volume plus a 2-cm margin beginning at week 6. The radiation dose was 41.4 Gy in standard 1.8 Gy daily fractions over 4 to 5 weeks. For this analysis, the records of all patients with group I disease that were coded as having any event (relapse, death, or second cancer) were reviewed to confirm the nature of the event(s). Special attention was given to the site(s) of initial relapse for the patterns of failure analysis. Charts were also scrutinized to determine whether patients actually received the RT to which they were randomized or assigned.
Definition of End Points
Statistical Methods Patients and disease characteristics predictive of FFS were investigated using the Cox proportional hazards model.10 Because histologic subtype has been known to be a strong predictor of outcome for these patients, multivariate analyses were performed separately for embryonal and alveolar or undifferentiated tumors. Factors investigated for their predictive value included age, sex, disease site, study, and receipt of RT as part of treatment. Tumor size could not be evaluated because this information was not available for many patients.
A total of 439 eligible patients with group I disease were enrolled onto IRS-I (1972 to 1978, n = 109), IRS-II (1978 to 1984, n = 126), and IRS-III (1984 to 1991, n = 204). Patients with group I RMS presented with localized tumors in a wide variety of anatomical sites. Tumor histology and primary site distributions are displayed in Table 1. The distribution of tumor sites was similar for patients who did and did not receive RT among each histologic group (ie, embryonal and alveolar/undifferentiated sarcomas). The male to female ratio was 2/1 because most patients with paratesticular primaries had group I disease. Sixty-four percent of patients were between less than 1 year and 9 years old.
Eighty-five of the 439 patients in IRS I to III received RT as part of initial treatment; 49 patients had alveolar and undifferentiated histology, 27 had embryonal histology, and nine had other histologies. Fifty-nine of 85 patients received RT according to appropriate protocol assignment, 32 patients on IRS-I and 27 on IRS-III. The other 26 patients were treated with RT because of inappropriate group assignment by the treating institution or protocol violation. Twenty-nine of the 110 patients assigned to regimens that included RT did not receive RT. At the time of this analysis, median follow-up for surviving patients was 12.9 years for IRS-I, 9.1 years for IRS-II, and 4.6 years for IRS-III. Eighty-six of the 439 patients with group I disease relapsed, resulting in an estimated 10-year FFS rate of 79% (95% confidence interval [CI], 75% to 83%) and estimated 10-year overall survival rate of 89% (95% CI, 86% to 92%). Forty-eight percent of relapses were apparent within 1 year and 93% within 3 years of diagnosis. Two patients failed after 10 years of follow-up. Patterns of initial relapse were nearly equally divided, as one third of the recurrences were local (6%), one third were regional (6%), and an equal fraction were distant failures (7%). Three patients with alveolar histology suffered stump recurrences after amputation. Univariate analysis of potential prognostic factors revealed that certain characteristics were associated with a significantly poorer FFS. These included tumors larger than 5 cm versus less than or equal to 5 cm (P < .004), alveolar or undifferentiated histology versus embryonal or other histology (P < .001), and tumors arising in primary sites other than genitourinary and paratesticular (GU) locations versus GU sites (P < .001). GU primary site conferred an advantage in FFS and overall survival for patients with embryonal RMS (P < .002) but not for patients with alveolar or undifferentiated sarcoma. Among patients with group I tumors, no sites could be identified that correlated with a better prognosis for patients with alveolar or undifferentiated RMS. Five-year FFS rates for patients improved significantly in IRS-III (86%) compared with IRS-I (77%) and IRS-II (74%) (P = .04, Fig 1).
Multivariate analyses were performed separately for patients with embryonal and alveolar or undifferentiated tumors because histologic subtype is known to be a strong predictor of outcome. For embryonal RMS, only favorable tumor site was shown to be predictive of FFS (P = .003). No differences in outcome were observed in these patients by age, sex, study, or by whether or not the patient received RT. For alveolar and undifferentiated RMS, multivariate analysis demonstrated that both treatment on the more recent IRS study (IRS-III v IRS-I or II, P = .04) and the addition of RT compared with no RT (IRS-I to II, P = .04; IRS-III, P = .03) were associated with improved FFS. No statistically significant differences in FFS were observed by age or sex for these patients. The effect of a favorable primary site could not be assessed in this group because only four patients had tumors in such sites.
Analysis by Intent to Treat In IRS-II, it was not intended that patients with group I RMS receive RT. Excluding patients with alveolar extremity lesions, no significant difference in patient prognosis was found between the chemotherapy regimens tested (standard dose VAC v VA).3 However, those patients with alveolar and undifferentiated RMS were found to have a worse prognosis irrespective of chemotherapy regimen. Ten-year FFS was estimated to be 81% for patients with embryonal tumors (n = 70) compared with 47% for those with alveolar and undifferentiated tumors (n = 17). There was a trend toward increased FFS for patients with alveolar extremity lesions treated with more intensive VAC chemotherapy when compared with historical controls (69% v 43% at 3 years, P = .06).8 In IRS-III, patients with group I tumors, excluding those with alveolar histology, were treated on a single regimen with postoperative VA chemotherapy but no RT. Overall, they experienced an excellent 10-year FFS (88%, n = 132). However, the subset of patients with undifferentiated RMS (n = 14) had an estimated 10-year FFS of only 57%. Patients with alveolar histology tumors were to receive more intensive chemotherapy (VAC plus doxorubicin and cisplatin) plus local RT. Estimated 10-year FFS for these patients was estimated to be 90%, compared with 60% in IRS-I, 57% in IRS-II, and 57% for patients with undifferentiated RMS who received standard chemotherapy and no RT in IRS-III (P = .003).
Analysis by Treatment Received
Table 2 lists the distribution of patients with alveolar and undifferentiated RMS according to treatment regimen and RT status in IRS-I to III. Patients treated with radiation and chemotherapy experienced fewer distant metastases as well as fewer local and regional failures compared with those who received adjuvant chemotherapy alone (Fig 3A and 3B). In IRS-I and II, 10-year FFS for patients with alveolar or undifferentiated sarcomas who received RT (n = 22) was superior to that observed for patients who did not receive RT (n = 45) (73% v 44%, respectively; P = .03; Fig 4A); 10-year overall survival was also substantially better for RT patients compared with patients who received no RT (82% v 52%, respectively; P = .02; Fig 4B). In IRS-III, children who received RT (n = 27) had a better 10-year FFS than those who did not (n = 29) (95% v 69%, respectively; P = .01; Fig 4A); also, there was a trend for improved 10-year overall survival for patients who received RT compared with those who did not (95% v 86%, respectively; P = .23; Fig 4B).
The most important finding in this analysis was that patients with alveolar and undifferentiated RMS who were treated with intensive chemotherapy, surgery, and RT fared better than similar patients with these tumor subtypes who did not receive RT. Of 67 patients with alveolar RMS and undifferentiated tumors treated on IRS-I and II, 22 patients received RT and 45 did not. Patients who received RT had a significantly improved FFS and overall survival compared with those who did not. No obvious difference in the proportion of patients receiving different chemotherapy regimens was noted between the patient groups who received or did not receive RT. However, this study was not a randomized comparison of patients who received or did not receive RT. Therefore, we cannot be certain that bias does not explain the apparent advantage of RT for this patient subset. In IRS-III, patients with alveolar or undifferentiated tumors were analyzed together. There was, again, a significant improvement in FFS and overall survival for the 27 patients with group I alveolar or undifferentiated tumors treated with RT compared with the 29 similar children who did not receive RT. Potentially important differences in patient characteristics exist between the RT and no-RT groups. Nearly all (26 of 27) patients in the RT group had alveolar histology tumors, whereas histology in the no-RT group was more evenly distributed between alveolar (16 patients) and undifferentiated (13 patients) histology. In addition, 26 of the 27 patients in the RT group received intensive chemotherapy, whereas only 15 of 29 patients in the no-RT group had intensive systemic treatment. The remaining patients in both groups were given standard VA chemotherapy. This presents a significant confounding variable in this study. Multivariate analysis of the entire patient cohort revealed that receipt of RT and use of a more contemporary therapy protocol (ie, IRS-III v IRS-I or II) were associated with better FFS. Considered altogether, these data seem to indicate that there is a benefit from intensified therapy plus RT for this patient group. Patients with group I embryonal RMS have enjoyed excellent FFS and overall survival in IRS-I to III. As treatment strategies have evolved, group I patients with alveolar RMS or undifferentiated RMS have experienced variable long-term FFS (IRS-I 60%, IRS-II 47%, and IRS-III 90%), which seemed to improve with use of more intensified chemotherapy for these patients in IRS-III. Given that outcome was known to be histologically dependent in the context of the therapy used in IRS-I to III, the effect of RT was examined according to histologic classification. There was a trend toward improved FFS but no difference in overall survival for patients with embryonal RMS who did or did not receive RT. These patients fare extremely well, regardless of protocol used (approximately 90% cure rate). Although compliance with IRSG protocols is generally quite good, there were significant numbers of patients who did not receive RT despite protocol assignment to RT. Noncompliance was most pronounced for patients with alveolar tumors in IRS-III, where only 26 of 41 subjects received the intended RT (Table 2). The patients who were not irradiated were not younger than those who did receive RT, and none had experienced a failure before the scheduled initiation of RT. Thus, there were no identifiable differences among the patients with alveolar RMS who did or did not receive RT in IRS-III. A number of patients were also treated with RT inappropriately because they were incorrectly assigned to be group II or because physicians violated protocol specifications. Findings in this study underscore the prognostic significance of tumor histology. When treatment decisions are based on histologic subclassification, tissue should be evaluated by a pathologist with extensive experience in RMS. IRSG central pathology review frequently disagreed with the tumor subtype assigned by participating institutions, and, thus, central review remains an important feature of multi-institutional protocols.7,11 Cytogenetic analysis has led to identification of histologically associated translocations (t[2;13][q35;q14], 50% of cases; t[1;13][q36;q14], 10% of cases) in alveolar RMS that aid in confirming the diagnosis.12 Primary tumor location is strongly associated with outcome for patients with group I embryonal tumors. Patients with GU (including paratesticular) primary tumors have improved outcome compared with other patients. Such differences were not apparent for patients with alveolar or undifferentiated sarcomas. Alveolar tumors tend to occur primarily in trunk and extremity sites, whereas embryonal tumors are more often found in head and neck or GU sites.13 In addition, tumors in certain locations are unlikely to be completely resected.14 Patients on IRS-I to III were treated according to clinical group, which is often affected by the surgical procedure used. This approach has been criticized because it places an emphasis on the surgical procedure rather than on tumor biology and clinical presentation.14-17 The use of a preoperative tumor-node-metastasis staging system has been studied in a prospective fashion for patients entered onto IRS IV but was not used for IRS I to III; thus, we cannot analyze by stage.15,18 However, data on tumor size was available for some patients, enabling a retrospective subset analysis. In this series of group I patients, large tumor size (> 5 cm) was found to confer a worse prognosis. The 5-year FFS for those with large tumors was 75%, whereas, for those with smaller tumors, it was 91% (P < .004). The use of RT for group I RMS may depend on tumor-node-metastasis stage as well as tumor histology, but this issue could not be addressed in this analysis. The IRSG continues to discourage disfiguring surgery when RT can be used to achieve limb preservation and organ salvage.19 Amputation does not always prevent local relapse. Three patients in this series experienced a stump recurrence after amputation with generous surgical margins. All three cases were alveolar tumors, and none received postoperative RT. The probability of salvage for patients with relapsed RMS has recently been reported by Pappo et al.20 The actuarial 5-year survival after relapse for patients with initial group I alveolar or undifferentiated histology was 40% (n = 16) in IRS-III and IV. Despite the reasonable success rate for this small cohort of patients, every effort should be made to cure patients with initial therapy. The findings of this analysis have influenced the design of IRS-V, in which all patients with group I alveolar and undifferentiated RMS will receive local RT.
Members of the Intergroup Rhabdomyosarcoma Study Group of the Children's Cancer Group and the Pediatric Oncology Group James R. Anderson, PhD; Richard J. Andrassy, MD; Carola Arndt, MD; Scott Baker, MD; Frederic G. Barr, MD; Archie Bleyer, MD; Philip Breitfield, MD; John C. Breneman, MD; Julia Bridge, MD; Ken Brown, MD; William M. Crist, MD; Sarah S. Donaldson, MD; Holcombe E. Grier, MD; Douglas Hawkins, MD; Peter J. Houghton, PhD; Michael Link, MD; Thom E. Lobe, MD; Harold M. Maurer, MD; William H. Meyer, MD; Jeff Michalski, MD; Sharon Murphy, MD; Charles N. Paidas, MD; Alberto S. Pappo, MD; David M. Parham, MD; Stephen J. Qualman, MD; R. Beverly Raney, MD; Leslie Robison, PhD; Eric Sandler, MD; Stephen Skapek, MD; Lynn Smith, MD; Paul H.B. Sorenson, MD, PhD; Sheri Spunt, MD; Lisa Teot, MD; Timothy Triche, MD, PhD; Teresa J. Vietti, MD; David Walterhouse, MD; Moody Wharam, MD; Eugene Wiener, MD; Suzanne L. Wolden, MD; and Richard Womer, MD.
Supported in part by National Institutes of Health/National Cancer Institute grants no. CA24507-22 and CA72989-02.
Presented in part at the American Society of Therapeutic Radiology and Oncology Meeting, Phoenix, AZ, October 28, 1998.
1. Pappo AS, Shapiro DN, Crist WM, et al: Biology and therapy of pediatric rhabdomyosarcoma. J Clin Oncol13:2123-2139, 1995 2. Maurer HM, Beltangady M, Gehan EA, et al: The Intergroup Rhabdomyosarcoma Study-I: A final report. Cancer61:209-220, 1988[Medline] 3. Maurer HM, Gehan EA, Beltangady M, et al: The Intergroup Rhabdomyosarcoma Study-II. Cancer71:1904-1922, 1993[Medline]
4.
Crist W, Gehan EA, Ragab AH, et al: The Third Intergroup Rhabdomyosarcoma Study. J Clin Oncol13:610-630, 1995 5. Maurer HM: The Intergroup Rhabdomyosarcoma Study: Objectives and clinical staging classification. J Pediatr Surg10:977-978, 1975 6. Crist WM, Garnsey L, Beltangady MS, et al: Prognosis in children with rhabdomyosarcoma: A report of the Intergroup Rhabdomyosarcoma Studies I and II. J Clin Oncol8:443-452, 1990[Abstract] 7. Newton WA, Soule EH, Hamoudi AB, et al: Histopathology of childhood sarcomas: Intergroup Rhabdomyosarcoma Studies I and II—Clinicopathologic correlation. J Clin Oncol6:67-75, 1988[Abstract] 8. Heyn R, Beltangady M, Hays D, et al: Results of intensive therapy in children with localized alveolar extremity rhabdomyosarcoma: A report from the Intergroup Rhabdomyosarcoma Study. J Clin Oncol7:200-207, 1989[Abstract] 9. Wharam MD, Hanfelt JJ, Tefft MC, et al: Radiation therapy for rhabdomyosarcoma: Local failure risk for clinical group II patients on the Intergroup Rhabdomyosarcoma Study II. Int J Radiat Oncol Biol Phys38:797-804, 1997[Medline] 10. Cox DR: Regression models and life-tables. J R Stat Soc B334:187-202, 1972 11. Qualman SJ, Coffin CM, Newton WA, et al: Intergroup Rhabdomyosarcoma Study: Update for pathologists. Ped Develop Path1:550-561, 1989
12.
Shapiro DN, Sublett JE, Li B, et al: Fusion of PAX3 to a member of the forkhead family of transcription factors in human alveolar rhabdomyosarcoma. Cancer Res53:5108-5112, 1993 13. Crist WM, Kun LE: Common solid tumors of childhood. N Engl J Med324:461-471, 1991[Medline]
14.
Donaldson SS, Anderson J: Factors that influence treatment decisions in childhood rhabdomyosarcoma. Radiology203:17-22, 1997 15. Lawrence W, Gehan EA, Hays DM, et al: Prognostic significance of staging factors of the UICC staging system in childhood rhabdomyosarcoma: A report form the Intergroup Rhabdomyosarcoma Study (IRS-II). J Clin Oncol5:46-54, 1987[Abstract] 16. Rodary C, Rey A, Olive D, et al: Prognostic factors in 281 children with nonmetastatic rhabdomyosarcoma (RMS) at diagnosis. Med Pediatr Oncol16:71-77, 1988[Medline] 17. Pedrick TJ, Donaldson SS, Cox RS: Rhabdomyosarcoma: The Stanford experience using a TNM staging system. J Clin Oncol3:370-378, 1986 18. Donaldson SS, Belli JA: A rational clinical staging system for childhood rhabdomyosarcoma. J Clin Oncol2:135-139, 1984[Abstract] 19. Raney B Jr, Gehan EA, Hays DM, et al: Primary chemotherapy with or without radiation therapy and/or surgery for children with localized sarcoma of the bladder, prostate, vagina, uterus, and cervix. Cancer66:2072-2081, 1990[Medline] 20. Pappo AS, Anderson JR, Grier HE, et al: Survival after relapse in children with rhabdomyosarcoma: A report from the Intergroup Rhabdomyosarcoma Study Group (IRSG). Proc Am Soc Clin Oncol 18:555a, 1999 (abstr 2142) Submitted January 5, 1999; accepted June 24, 1999.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 1999 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
