Pii: s0360-3016(99)00063-2

Int. J. Radiation Oncology Biol. Phys., Vol. 44, No. 3, pp. 569 –577, 1999 Copyright 1999 Elsevier Science Inc.
PII S0360-3016(99)00063-2
RADIATION THERAPY FOR CONSOLIDATION OF METASTATIC OR
RECURRENT SARCOMAS IN CHILDREN TREATED WITH INTENSIVE
CHEMOTHERAPY AND STEM CELL RESCUE. A FEASIBILITY STUDY
E. ANN DUNPHY CZYZEWSKI, M.D.,* STEWART GOLDMAN, M.D.,† ARNO J. MUNDT, M.D.,‡ JAMES NACHMAN, M.D.,† CHARLES RUBIN, M.D.,† AND DENNIS E. HALLAHAN, M.D.§ *Department of Cancer Biology, University of Chicago, Chicago, IL; †Department of Pediatric Hematology/Oncology, University of Chicago, Chicago, IL; ‡Department of Radiation and Cellular Oncology, University of Chicago,Chicago, IL; and §Vanderbilt University Purpose: To assess the role of consolidative radiation therapy (CRT) in conjunction with myeloablative therapy
with or without total body irradiation (TBI) in children and young adults with metastatic or recurrent sarcoma.
Methods and Materials: Twenty-one pediatric sarcoma patients with metastatic (10) or recurrent (11) disease
were entered on a prospective feasibility study of intensive myeloablative therapy with or without TBI. Median
patient age was 17.8 years (range, 9.4 –24.7 years). Primary histologies included Ewing’s (12), PNET (3), and
other soft tissue sarcomas (6). Twenty patients received induction chemotherapy. Myeloablative therapy con-
sisted of TBI in 11 patients with either high dose melphalan/etoposide (9) or high dose cytoxan/thiotepa (2). TBI
consisted of 12 Gy in 2 Gy fractions delivered twice daily over 3 days. Ten patients received high dose
chemotherapy alone, either with thiotepa/carboplatinum/etoposide (8) or cytoxan/carboplatinum (2). Myeloab-
lative therapy was followed by autologous stem cell rescue (ASCR) 24 to 48 hours after completing chemother-
apy. Fourteen patients (67%) received CRT either prior to (5) or following (9) myeloablative therapy. Median
CRT dose was 37.2 Gy (range, 20 – 60). Fifty-one disease sites were present prior to myeloablative therapy.
Twelve (24%) were bulky (> 8 cm) and 18 (35%) underwent surgical debulking. The median follow-up of
surviving patients was 15 months (range, 8 –20) with 25% of patients having been followed for more than 20
months.
Results: The 3-year actuarial disease-free (DFS) and overall survival (OS) rates for the entire group were 36%
and 27%, respectively. Following myeloablative treatment, responses were: 11 complete, 6 partial, 1 stable, and
3 progressive disease. Sixteen patients (71%) have relapsed. The most common site of relapse was the lung (13).
Of the 51 disease sites present prior to myeloablative therapy, 36 sites (71%) were amenable to CRT. Non-
amenable sites were: multiple lung metastases (13) and bone marrow (2). Twenty-six amenable sites (51%)
received CRT either prior to (14) or following (12) ASCR. Amenable sites treated with CRT had a better 3-year
actuarial local control (80 vs 37%) (p ؍ 0.0065) than amenable sites not treated with CRT. Factors associated
with improved disease-free survival (DFS) in univariate analysis were induction chemotherapy response (p ؍
0.002) and extent of surgical resection (p ؍ 0.045). There was a trend toward improved DFS on univariate
analysis with the use of TBI as part of myeloablative therapy (p ؍ 0.07). The one factor associated with improved
OS on univariate analysis was induction chemotherapy response (p ؍ 0.007). Multivariate analysis revealed that
induction chemotherapy response is the only factor that remains significant for DFS (p ؍ 0.032) as well as for
OS (p ؍ 0.017). Patients with complete response to induction therapy had 40% probability of survival versus all
other patients who had 10% survival (p ؍ 0.05).
Conclusion: Consolidative radiotherapy is feasible in primary metastatic or recurrent pediatric sarcoma patients
treated with myeloablative therapy with or without TBI. CRT to sites amenable to irradiation provided an
improved 3-year actuarial local control than that seen in sites amenable to CRT that did not undergo
radiotherapy. There was a trend for improved DFS with the use of TBI. Improved DFS and OS can be predicted
by response to induction therapy. This intensive regimen may improve the cure rate of advanced pediatric
sarcomas in select patients.
1999 Elsevier Science Inc.
Sarcoma, Autologous bone marrow transplant, Radiation therapy.
Reprint requests to: Dennis Hallahan, M.D., Chairman, Depart- Lane, Ann Hosewitz, Sharon Bolster, RN, for data management ment of Radiation Oncology, 1301 22nd Avenue South, B-902, Vanderbilt Clinic, Nashville, TN 37232-5671. Tel 615-343-9244; Accepted for publication 2 February 1999.
Fax 615-343-3075; Email: [email protected].
Presented at the 40th Annual Meeting of the American Society for Therapeutic Radiology and Oncology, Phoenix, AZ, October, Acknowledgments—We thank Carmalita Chotipradit, RN, Debra I. J. Radiation Oncology ● Biology ● Physics INTRODUCTION
chemotherapy response. In addition, 13 patients had lungmetastases on initial presentation. Despite this, the 3-year Large single or multi-institution studies for high-risk or DFS was 36% and OS was 27%, comparable to the outcome metastatic Ewing’s sarcoma report 3-year event-free surviv- of large prospective, randomized trials reported in the liter- als of approximately 30% whether treated with myeloabla- ature. The presence of bulky disease did not have an impact tive regimens (1– 4) or conventional multimodality therapy on local control. This may be due to the use of CRT for (4 – 6). Similar poor results have been reported for meta- amenable sites, which was found to have a significant im- static peripheral primitive neuroectodermal (PPNET) tu- pact on DFS. The response to induction chemotherapy pre- mors (7,8) and other metastatic pediatric sarcomas (3,8 – dicted for improved DFS and OS in those with a complete 13). A poor prognosis is also expected for recurrent response compared to those with other responses. The pres- pediatric sarcoma when presenting early (within 2 years ence of pulmonary metastases did not predict outcome and from initial treatment), or if multiple bony or pulmonary 6 of the 13 patients who had lung disease at initial presen- sites are involved (14 –16). It is uncertain whether either tation are alive: 3 who are relapse-free and another 3 who local radiotherapy or total body irradiation (TBI) is tolerable are alive following lung relapse. The fact that the presence or effective in the treatment of metastatic sarcoma. The of pulmonary mestastases on initial presentation did not purpose of the present study was to determine the feasibility predict for a worse outcome may indicate that these patients and efficacy of using radiation therapy as consolidation benefit from this aggressive treatment regimen.
treatment in pediatric patients with metastatic sarcomas In addition, we compared the outcome of patients receiv- treated with myeloablative chemoradiotherapy.
ing TBI to those receiving myeloablative chemotherapy Myeloablative therapy may benefit select patients with alone. We found that there was a trend for improved DFS metastatic or recurrent disease. These include patients with with the use of TBI, but TBI did not significantly impact on low tumor burden or high tumor burden, which is amenable overall survival (OS). Likewise, although CRT did have a to consolidative radiation therapy (CRT). For example, an significant impact on DFS, CRT also had no impact on improved 6-year event-free survival (EFS) of 45% was overall survival. However, only an appropriately designed achieved in select metastatic or recurrent Ewing’s sarcoma prospective, randomized trial would be able to determine patients who received high dose chemotherapy (HDC) and the true impact of TBI and CRT. But this may not be 12 Gy TBI compared to 2% EFS in a historic control group possible as a large number of patients and a long follow-up treated with conventional therapy (17). These select poor period would be required, which has been shown to be risk patients could not have bulky disease or progression needed to detect any survival advantage when radiation during induction chemotherapy. One study that reported provides local control (24,25). Also, we still do not have a superior results for bulk disease with myeloablative therapy prospective, randomized trial that compares conventional and CRT was a retrospective review by the University of multimodality therapy to myeloablative regimens in pediat- Florida (18). Patients had primarily localized disease and ric patients with poor-risk disease.
2-year disease-free survival (DFS) was 80%.
The present study analyzes the feasibility and efficacy of The University of Florida report supports the use of CRT using CRT in pediatric/young adult patients with poor-risk to reduce bulky disease such that HDC would also be as disease treated with myeloablative therapy with or without successful in patients with high tumor burden; however, TBI. The impact of both tumor and treatment factors on these were patients with primarily localized disease. Pa- prognosis were also retrospectively assessed.
tients with metastasis have been shown to have a worseoutcome (5,15). In a study conducted at the National Cancer METHODS AND MATERIALS
Institute (1), 31 patients with metastatic Ewing’s sarcomaand rhabdomyosarcoma who had a complete response (CR) Eligibility criteria for patients entered into this high dose to induction therapy and irradiation to the primary and chemoradiotherapy protocol included children and young metastatic sites received high dose chemotherapy and TBI adults with metastatic or recurrent sarcomas referred to the with autologous bone marrow infusion. Patients with met- University of Chicago Children’s Hospital. Patients that astatic disease had a 19% 6-year EFS, significantly worse were excluded from the study included adults and all pa- compared to the 45% 6-year EFS reported by Burdach et al. tients with osteosarcomas. Twenty-three patients were eli- gible for this study and 21 patients with a median age of The most common site of metastatic disease on initial 17.8 years (9.4 –24.7 years) were treated with myeloablative presentation is the lung (1,5,19) and this site is not amenable therapy with ASCR between June 1987 and April 1997.
to curative irradiation. Aggressive multimodality therapy to Two of the 23 original patients had rapidly progressive the lung is limited by our inability to resect all disease disease during induction chemotherapy and did not undergo and/or irradiate to a high enough dose due to risk of organ myeloablative therapy or consolidative radiation therapy injury. Likewise, the lung is the most common site of failure and were not involved in this analysis. Patients were staged both in other studies and in the present study (13 of 21 total with plain films of the primary site, a bone scan, CXR, patients relapsed in the lung). In this study, patients were computed tomography (CT) of the lungs, CT or magnetic eligible despite presence of bulky diseases or induction resonance imaging (MRI) scan of the primary site and bone Radiation therapy for metastatic pediatric sarcomas ● E. A. DUNPHY CZYZEWSKI et al. Table 1. Patient and tumor characteristics tients all had individualized chemotherapy according toprevious chemotherapy treatment.
Surgical resections included resection of the primary and thoracotomy for resection of lung metastases. Primary tu- mor resections included the tumor and surrounding com- partment. The adjacent bone was resected when local inva- sion into bone was noted on the MRI scan. Metastatic lesions to the lung were resected by thoracotomy if residual lesions were found after chemotherapy induction. In one patient who underwent thoracotomy for lung metastases, resection of pericardial disease was also performed.
Fifteen patients underwent surgical resection to a total of 17 sites; 7 of these sites were diffuse lung disease. The procedure used in the treatment of 7 diffuse lung sites included thoracotomies with wedge resections and all nod- ules were completely resected in 3 (patients 3, 14, and 15), but incompletely resected in 4 patients (patients 12, 16, 20, and 21). Patients 3, 12, and 15 received whole lung CRT following myeloablative therapy (details to follow). The 11 remaining sites underwent surgical resection with CRT (4 patients) or surgery alone (7 patients). The 6 sites that rib, scapula, chest wall/lung, flank, breast, spine, underwent surgery alone had complete resection in 4 (pari- etal bone, breast, pericardium, humerus), microscopic resid- ual in 1 (breast/axilla) and gross residual in 1 (breast). The 4 sites that underwent surgery and CRT consisted of pleura (2), pelvis/thigh (1), and thigh (1) masses; these were all completely resected after induction therapy but prior to myeloablative therapy and CRT except in one patient who had a complete resection of pubic ramus 6 weeks after CRT and stem cell reinfusion (patient 8).
Surgical resection was performed prior to myeloablative therapy in all patients except one; this patient had resectionof pubic ramus 6 weeks after stem cell reinfusion.
marrow biopsy. The clinical characteristics are shown in Table 1. In the 10 patients with primary metastatic disease, Stem cells were obtained by bone marrow aspiration (11 the metastatic sites were: lung only (5 patients), a solitary patients) or peripheral stem cell harvesting (10 patients).
bone (1 patient), and multiple sites (4 patients). In the 11 High dose cytoxan was used to mobilize peripheral stem patients with recurrent disease, the relapse sites were: pri- cells. Myeloablative therapy consisted of high dose chemo- mary site only (2 patients), lung only (6 patients), multiple therapy alone in 10 patients, either with thiotepa/carboplati- bones (2 patients), and multiple sites (1 patient).
num/etoposide (8 patients) or cytoxan/carboplatinum (2 pa-tients). Eleven patients underwent TBI with either high dose melphalan/etoposide (9 patients) or high dose cytoxan/thio- Twenty patients were treated with induction chemother- tepa (2 patients). TBI consisted of 12 Gy in 2 Gy fractions apy. The one patient who was not treated with induction delivered twice daily over 3 days using techniques as de- chemotherapy presented with lung relapse and was treated scribed below and previously (20,21). These fractions were with thoracotomy alone prior to high dose chemotherapy given a minimum of 4 h apart. One patient received an and ASCR. All 11 recurrent disease patients had been alternate regimen of 15 Gy in 1.25 Gy fractions delivered treated with previous chemotherapy regimens prior to re- three times daily over 4 days. TBI was delivered with 24 mV photons, right and left lateral opposed fields, at an Induction chemotherapy regimens consisted of ifospha- extended source to patient midplane distance of 410 cm at a mide and etoposide in 5 patients; vincristine, adriamycin dose rate of approximately 0.12 to 0.15 Gy/min. Dose and cytoxan (VAC) in 2 patients; VAC and MESNA in 4 homogeneity was 97 to 105% along the patient midplane.
patients, VAC and ifosphamide in 1 patient; VAC and The dose to the lung was kept below 12 Gy with use of ifosphamide/etoposide in 2 patients. The remaining 6 pa- customized partial transmission lung blocks. Patients were I. J. Radiation Oncology ● Biology ● Physics treated in the supine position with head and extremity bolusand a 1.2 cm polycarbonate beam spoiler as previouslydescribed (18). Myeloablative therapy was followed bybone marrow or autologous stem cell rescue 24 to 48 hoursafter completing high dose chemotherapy.
Two patients had consecutive transplants; one patient due to progressive disease during initial transplant and the sec-ond due to extensive disease at presentation consisting ofmultiple sites including bone marrow involvement.
Consolidative radiation therapy (CRT) was used to treat the original site of the primary tumor (6 patients) and/ormetastatic lesions (12 patients). The median dose to the Fig. 1. Survival of 21 patients receiving myeloablative chemora- primary was 50 Gy (range 42.8 to 60 Gy) in patients with diotherapy and consolidation radiotherapy for metastatic and re- Ewing’s sarcoma or PPNET. One patient with rhabdomyo- current sarcomas. Kaplan-Meyer analysis was used to generatesurvival these data.
sarcoma received 47 Gy and one patient with synovialsarcoma received 60 Gy to the primary site. The radiationtherapy dose for metastatic lesions other than whole lung was given to liver metastases and in patient 18 there were ranged from 20 to 50 Gy (median dose 31.5 Gy) for a total diffuse bony metastases and not all rib metastases could of 19 sites. Lung metastases treated with whole lung irra- feasibly be irradiated or resected. The sites that received diation consisted of 12.8 in 1.6 Gy fractions twice daily CRT, 26 of 36 amenable sites and 3 unamenable (whole (bid) (1 patient) and 15 Gy in 1.5 Gy fractions delivered 1.5 lung) sites, were chosen for CRT based on the discretion of Gy daily (2 patients). In 6 patients with multiple lung the treating physician. The 11 amenable sites, which under- metastases with Ewing’s sarcoma or PPNET, the lungs went surgery, as described in “Surgical Resection.” received irradiation via TBI consisting of 12 Gy in 2 Gy On completion of curative therapy, patients were fol- fractions bid (not considered CRT). For twice daily irradi- lowed initially at 2 to 3 month intervals and subsequently ation, the interfraction interval was at least 4 hours apart.
biannually, then yearly. Follow-up studies included clinical None of these CRT doses include the TBI dose.
examination, CXR or CT scan of the chest and CT or MRI The radiation volume included the primary tumor, tumor scans of the primary site. The median follow-up of surviv- bed (including the scar) or metastatic site with a margin of ing patients was 15 months (range, 8 –20) with 25% of 2 cm. Complete circumferential irradiation of extremities patients having been followed for more than 20 months.
was not done to prevent development of lymphedema.
Treated sites received external beam radiation with mega- voltage in 1.8 to 2.0 Gy fractions. Exceptions consisted of All statistical analyses were performed using statistical 2.5 Gy to 31.5 Gy to a paraspinal mass and use of 9 MeV software. Local control, DFS and OS actuarial analyses electrons to frontal bone and to a thigh scar (both in 2 Gy were plotted according to the method of Kaplan-Meier and fractions to 50 Gy). The majority of sites were treated with compared by the log rank test. Comparison of proportions opposed fields, usually AP/PA. Abdominal sites usually was performed using the Chi-square test. All intervals were were treated with 3 fields, which were individualized, to determined from the date of the bone marrow transplant or spare critical structures. Patients received appropriate im- mobilization and customized blocking.
Fourteen patients (67%) underwent CRT either prior to (5 patients) or following (9 patients) myeloablative therapy.
Sites were divided into amenable and unamenable sites. An amenable site was defined as metastasis in an organ or tissue Twenty-three patients were eligible for myeloablative that could be treated to a curative dose without exceeding therapy for metastatic or recurrent pediatric sarcoma. Two the dose or volume that is tolerated by that tissue. The 51 of these patients had rapidly progressive disease during disease sites consisted of 36 amenable and 15 unamenable induction chemotherapy and were not treated with myeloa- to CRT sites (13 diffuse lung and 2 bone marrow sites). Of blative therapy or consolidative radiation therapy and were the 36 amenable sites, 3 received no local therapy, 22 not involved in this analysis. Of the 21 patients analyzed, 6 received CRT, 4 received surgery and CRT, and 7 received are disease-free and 3 are alive with disease; 11 patients surgery alone. Three amenable sites did not receive any have died from progressive disease and 1 patient died from local therapy under the discretion of the treating physician: a pulmonary embolus 2 months after ASCR. The 3-year in patient 11 it was not feasible to irradiate all 6 involved actuarial and OS rates are 36% and 27%, respectively (Figs.
sites, therefore, an L3/L4 was not irradiated and surgical resection was not feasible; in patient 13 no local therapy Of the 21 patients treated with myeloablative therapy, 20 Radiation therapy for metastatic pediatric sarcomas ● E. A. DUNPHY CZYZEWSKI et al. Stem cell engraftment was defined as ANC Ͼ1000 and platelets Ͼ100,000. The average time to engraftment was16 days for neutrophils (21 patients) and 43 days for plate-lets (14 patients who attained engraftment). Three patientscontinued to have thrombocytopenia Ͼ 3 months followingbone marrow or stem cell reinfusion. Two of these threepatients received TBI and 2 had CRT to Ͼ 2 bone sites. Twopatients died before there was sufficient time for plateletengraftment.
Toxicities following autologous bone marrow or stem Fig. 2. Disease-free survival 21 patients receiving myeloablative cell reinfusion included grade 2 mucositis (4 patients), grade chemoradiotherapy and consolidation radiotherapy for metastatic 2 esophagitis (2 patients), prolonged thrombocytopenia with and recurrent sarcomas. Kaplan-Meyer analysis was used to gen- platelet dependence (5 patients), neutropenic fever (5 pa- tients), sepsis (1 patients), infectious diarrhea (3 patients),pneumonia (3 patients), skin infection (4 patients), andpulmonary embolus (1 patients). Treatment related fatalities were treated with induction chemotherapy and 1 patient occurred in one patient with pulmonary embolus. A partial underwent thoracotomy alone for lung relapse prior to my- sagittal sinus thrombosis developed in one patient following eloablative therapy. The response to induction therapy is surgery. Diminished range of motion occurred in one pa- shown in Table 2: 8 complete, 6 partial, 2 stable, and 4 tient who required a prosthesis after surgical resection of the progressive disease. Induction therapy included surgical proximal femur. One patient developed chronic enteritis debulking in 11 patients and CRT in 5 patients. Following requiring donnatal following abdominal irradiation.
engraftment, 9 patients underwent CRT and surgical exci-sion was performed in one patient resulting in a CR.
Responses following myeloablative treatment were: 11 The following tumor and treatment factors were analyzed complete, 6 partial, 1 stable, and 3 progressive disease. Six for their impact on DFS and OS: induction chemotherapy patients are disease-free (NED) and all had either a PR/CR response, TBI, CRT, extent of surgical resection, histology, with induction therapy and a CR with myeloablative ther- age at diagnosis, stage, presence of pulmonary metastases, apy. Table 2 lists the disease sites and type of local therapy and prior treatment. Factors associated with improved DFS if given. In the 6 NED patients, 5 unamenable sites were in univariate analysis were induction chemotherapy re- controlled, consisting of 4 lung and 1 bone marrow. All of sponse (p ϭ 0.002) and extent of surgical resection (p ϭ these patients had TBI. When comparing myeloablative 0.045). There was a trend for improved DFS on univariate regimens, we found that 5 of 10 patients treated with TBI analysis with the use of TBI as part of myeloablative ther- and 1 of 11 patients treated with high dose chemotherapy apy (p ϭ 0.07). The one factor associated with improved OS alone are presently free of disease (p Ͻ 0.07).
on univariate analysis was induction chemotherapy re-sponse (p ϭ 0.007). There was a trend for improved OS on univariate analysis if there was a complete surgical resec- Fifty-one disease sites were present prior to myeloabla- tion (p ϭ 0.07). Multivariate analysis with respect to DFS tive therapy. Twelve sites (24%) were bulky, defined as Ͼ and OS revealed that induction chemotherapy response is 8 cm in size. Eighteen (35%) underwent surgical debulking.
the only factor that remains significant (p ϭ 0.032 for DFS Of the 51 disease sites present prior to myeloablative ther- and p ϭ 0.017 for OS). Patients with complete response to apy, 36 sites (71%) were amenable to CRT. Non-amenable induction therapy had 40% probability of survival versus all sites were: multiple lung metastases (13) and bone marrow other patients who had 10% survival (p ϭ 0.05).
(2). Twenty-six amenable sites (51%) received CRT eitherprior to (14) or following (12) ASCR. Amenable sites treated with CRT had a better 3-year actuarial local control Fifteen patients (71%) have relapsed. The most common (80 vs. 37%) (p ϭ 0.0065) than amenable sites not treated sites of relapse were lung (13) and/or bone marrow (2).
with CRT. Median CRT dose was 37.2 Gy (range, 20 – 60).
Recurrence within the field of CRT occurred in 6 patients: A higher rate of local control was achieved with the dose in 3 unamenable sites consisting of all 3 patients who range 31 to 60 Gy compared to 20 to 30 Gy. There were 4 received whole lung CRT, and in 3 amenable sites consist- bulky disease sites in the NED group, all were controlled ing of 1 patient who received 27 Gy to a femur lesion, 1 with CRT. The 5 other amenable sites were controlled with patient with a frontal bone lesion treated with electrons to either surgery or CRT and one site was controlled with no 50 Gy and in 1 patient with extensive vertebral disease from T3 to S1 treated to 43.2 Gy. There were no recurrences at I. J. Radiation Oncology ● Biology ● Physics A, amenable; UL, unamenable/lung; UB, unamenable/bone marrow; B, bulky; S, surgery; R, CRT; 0, no local therapy.
the margin of irradiation. Recurrence outside the field of therapy in four patients. The three salvaged patients who consolidation therapy occurred in eight patients. Out of field remain alive were salvaged with lung resections.
CRT recurrences included the lung in 6 patients, lung andbone marrow in 1 patient, and lung, liver, and sacrum in 1patient.
DISCUSSION
We retrospectively analyzed the impact of local irradia- tion for consolidation with respect to local control in pedi- Ten patients were treated with salvage therapy. Seven of atric patients with metastatic or recurrent sarcomas treated these patients did not survive and their salvage therapy with high dose chemo/radiotherapy and stem cell rescue as consisted of a second course of myeloablative therapy in part of a feasibility study. The patients in this study received one patient who had progressive disease during high dose myeloablative therapy regardless of their response to induc- chemotherapy; etoposide followed by tumor vaccine with tion chemotherapy unless they were rapidly progressing and interleukin-2 in one patient; high dose methotrexate fol- regardless of the initial extent of disease. The present study lowed by alpha interferon in one patient; palliative radiation included 2 patients with no response and 3 patients who Radiation therapy for metastatic pediatric sarcomas ● E. A. DUNPHY CZYZEWSKI et al. progressed during induction chemotherapy, 10 patients with (3). Those without TBI achieved a better EFS rate at 34% bulky disease (Ͼ 8 cm), and 8 patients with multiple sites compared to 19% in the group with TBI. Also, 8 Gy TBI (two (Ͼ 3 sites). Consolidative radiation therapy was able to 4 Gy fractions) and HDC did not improve outcome in the 31 provide 80% 3-year actuarial local control to those sites patients treated at the NCI with metastatic Ewing’s sarcoma irradiated. This is significantly improved compared to the and rhabdomyosarcoma who had a complete response (CR) to other amenable sites of disease that did not receive radiation induction therapy and CRT (1). Seventy percent of patients which had a 37% 3-year actuarial local control (p ϭ with metastatic disease at diagnosis failed distantly and 70% of these distant failures were either in the lung and/or bone The use of CRT may have contributed to our comparable marrow. It was concluded that either the induction therapy results despite many patients with poor induction chemother- delivered 4 to 6 months prior to TBI caused radioresistance or apy response and bulk disease. CRT is useful in additionally that the amount of cell kill from 2 fractions of 4 Gy was reducing local tumor that is not eradicated by induction che- insufficient to eliminate occult disease. The Austrian/German motherapy either due to (a) tumor bulk or (b) chemoresistance.
experience found that after 8 Gy fractionated TBI in patients CRT could potentially convert patients into a “select” category with multifocal Ewing’s sarcoma, the lung was the predomi- by eliminating disease at bulky sites and at sites of progression nant site of relapse after BMT (19). We propose that 8 Gy TBI during induction chemotherapy. In the present study, it is is an insufficient dose and we are continuing to accrue patients difficult to determine the CRT dose that is required to achieve to a regimen consisting of 12 Gy given as 2 Gy twice daily.
tumor control. There were a total of 26 amenable sites treated The presence of pulmonary metastases on initial presen- with CRT and only 3 of these sites failed following treatment.
tation did not predict outcome and 6 of the 13 patients who These sites were a femur treated to 27 Gy, spine T3-S1 treated had lung disease at initial presentation are alive: 3 who are to 43.2 Gy, and a frontal bone treated to 50 Gy. All other sites relapse-free and another 3 who are alive following lung of disease were controlled, consisting of 7 sites that received relapse. The 3 patients who are relapse-free all had TBI and 20 –30 Gy and 16 sites that received 31– 60 Gy. The NCI 1 had lung wedge resections; in those 3 patients who are reported a 19% EFS in 31 patients with metastatic Ewing’s alive following lung relapse, 2 patients had both whole lung sarcoma or rhabdomyosarcoma who received induction che- irradiation and lung wedge resections at time of myeloabla- motherapy and CRT (50 to 60 Gy) to local disease (1). In tive therapy. There were a total of 3 patients who received another study, a median CRT dose of 35 Gy achieved excellent whole lung irradiation, all have relapsed in the lung. The 2 local control (6). Patients received chemotherapy followed by patients who had bone marrow disease at initial presentation surgery and/or CRT. At almost 4 years, 10 patients were remain relapse-free. Both of these patients received TBI. It is difficult to make any definite conclusions with regards to In the present study, bulky disease was not associated the impact of lung irradiation/TBI in these patients present- with a worse prognosis. Local control was achieved in 10 of ing with pulmonary metastases and/or bone marrow in- the 12 bulky sites: 7 with CRT alone, 2 with surgery and volvement on initial presentation. However, it is promising CRT, and in 1 site with surgery alone. The two bulky sites that TBI did show a trend for improved disease-free survival that failed consisted of chest/axilla (patient 14) treated with on univariate analysis. Likewise, the fact that the presence surgery alone and spine T3-S1 (patient 18) treated with of pulmonary metastases on initial presentation did not CRT to a dose of 43.2 Gy. Bulky disease has been shown to predict for a worse outcome may indicate that these patients be a poor prognostic factor in previous studies (15,22). The benefit from this aggressive treatment regimen.
NCI reported on their experience from 1968 –1980 for both Certain patients with lung metastases or bone marrow dis- local and primary metastatic disease for Ewing’s sarcoma of ease can be cured. It appears that our present myeloablative the bone. Poor prognostic factors were found to be meta- chemotherapy regimens and TBI radiation doses are most static disease at presentation, high LDH (Ͼ350 IU) in effective if lung or bone marrow tumor burden is below a localized disease patients and central primary tumors in certain threshold and if the tumor has demonstrated a response localized disease patients (15). All of these factors were to chemotherapy. This rationale would support giving myeloa- considered to be associated with tumor bulk. At the Uni- blative therapy only to those patients who have had a CR or PR versity of Florida between 1966 and 1981, Ewing’s sarcoma from induction chemotherapy. Those patients who had only a patients with metastatic or primary tumor size of Ͼ 8 cm PR would benefit from myeloablative therapy if they could be who were treated for cure had a worse outcome (22).
rendered a CR with CRT or surgery prior to myeloablative Therefore, patients with bulky and or metastatic disease are therapy. The role of myeloblative therapy and TBI would then often treated more aggressively with myeloablative therapy.
be to eliminate any residual micrometastases. The above treat- We found that patients receiving 12 Gy TBI had an im- ment rationale is supported by the fact that in the 6 patients proved disease-free survival as compared to patients receiving who remain NED in the present study, all had either a CR or chemotherapy alone for myeloablation. This finding was sup- PR from induction chemotherapy and all amenable sites except ported by the study by Burdach et al. (17) that showed superior one (a lumbar spine metastasis) had local therapy with either results using TBI. In contrast, the European BMT Tumor Registry experience with myeloablative therapy in Ewing’s Another concern these patients are the high relapse rate in sarcoma patients compared regimens with and without TBI the lungs, illustrated by the fact that in the present study, 13 I. J. Radiation Oncology ● Biology ● Physics patients failed in the lung. Another approach to possibly In conclusion, consolidative radiotherapy is feasible in improve outcome would be increasing the radiation dose to patients with metastatic or recurrent pediatric sarcomas the lungs. A retrospective analysis by Dunst et al. (23) treated with myeloablative therapy with or without TBI.
showed a dose-response relationship in 22 Ewing’s sarcoma Overall, this aggressive treatment regimen was tolerable patients with pulmonary metastases at diagnosis treated with the primary toxicity from myeloablative therapy with bilateral lung irradiation.Four of 10 patients treated being prolonged thrombocytopenia. Only one patient de- with 12 to 16 Gy vs. 5 of 6 patients receiving 18 to 21 Gy veloped a long-term complication from CRT consisting lung irradiation were in remission. The results suggested of chronic enteritis. DFS is improved with CRT and there that 18 to 21 Gy might control 75% of pulmonary disease in is a trend for improved DFS with TBI. Both DFS and OS patients with a complete radiographic remission following are improved in those with a CR following induction chemotherapy. No severe late lung toxicities were reported.
chemotherapy compared to those with other responses.
We are presently treating the lungs with a total dose of 18 The 3-year DFS of 36% and OS of 27% are comparable Gy in patients with biopsy-proven lung metastases.
to the outcome of large prospective, randomized trials The toxicity associated with CRT was acceptable consisting reported in the literature, including trials that did not use of a single patient with chronic enteritis requiring donnatal myeloablative therapy. It is not known whether myeloa- following treatment to the abdomen. Myeloablative therapy blative therapy adds any additional benefit in patients consisted of the expected severe acute toxicities to the marrow, who have already achieved a CR prior to myeloablative mucosa, and gastrointestinal tract. There was one death from a therapy. Only a prospective, randomized trial would be pulmonary embolus soon after myeloablative therapy.
REFERENCES
1. Horowitz ME, Kinsella TJ, Wexler LH, Belasco J, Triche T, 10. Crist W, Gehan EA, Ragab AH, et al. The Third Intergroup Tsokos M, Steinberg SM, McClure L, Longo DL, Steis RG, Rhabdomyosarcoma study. J Clin Oncol 1995;13:610 – 630.
Glatstein E, Pizzo P, Miser JS. Total-body irradiation and 11. Emminger W, Emminger-Schmidmeier W, Hawliczek R, Pe- autologous bone marrow transplant in the treatment of high- ters C, Hocker P, Gadner H. High-dose melphalan, etopo- risk Ewing’s sarcoma and rhabdomyosarcoma. J Clin Oncol side Ϯ carboplatin (MEC) combined with 12-gray fractionated total-body irradiation in children with generalized solid tu- 2. Ladenstein R, Gadner H, Hartmann O, Pico J, Biron P, Thierry mors. Pediatr Hematol Oncol 1991;8:13–22.
P. The European experience with megadose therapy and au- 12. Koscielniak E, Rodary C, Flamant F, Carli M, Treuner J, tologous bone marrow transplantation in solid tumors with Pinkerton CR, Grotto P. Metastatic rhabdomyosarcoma and poor prognosis Ewing sarcoma, germ cell tumors and brain histologically similar tumors in childhood: A retrospective tumors. Wiener Medizinische Wochenschrift 1995;145:55– European multi-center analysis. Med Pediatr Oncol 1992;20: 3. Ladenstein R, Hartmann O, Pinkerton CR. The role of mega- 13. Seeger RC, Reynolds CP. Treatment of high-risk solid therapy with autologous bone marrow rescue in solid tumours tumors of childhood with intensive therapy and autologous of childhood. Ann Oncol 1993;1:45–58.
bone marrow transplantation. Pediatr Clin N Am 1991;38: 4. Miser JS, Kinsella TJ, Triche TJ, Tsokos M, Forquer R, Wesley R, Horvath K, Belasco J, Longo DL, Steis R, Glatstein 14. Hayes FA, Thompson EI, Kumar M, Hustu HO. Long-term E, Pizzo PA. Preliminary results of treatment of Ewing’s survival in patients with Ewing’s sarcoma relapsing after sarcoma of bone in children and young adults: Six months of completing therapy. Med Pediatr Oncol 1987;15:254 –256.
intensive combined modality therapy without maintenance.
15. Kinsella TJ, Miser JS, Waller B, Venzon D, Glatstein E, Weaver-McClure L, Horowitz ME. Long-term follow-up of 5. Cangir A, Vietti TJ, Gehan EA, Burgert E Jr, Thomas P, Ewing’s sarcoma of bone treated with combined modality Tefft M, Nesbit ME, Kissane J, Pritchard D. Ewing’s sar- therapy. Int J Radiat Oncol Biol Phys 1991;20:389 –395.
coma metastatic at diagnosis. Results and comparisons of 16. Klingebiel TH, Bode U, et al. Treatment of relapse in soft two intergroup Ewing’s sarcoma studies. Cancer 1990;66: tissue and Ewing’s sarcoma patients. A phase II trial (CESS/ CWS REZ 91). Med Pediatr Oncol 1993;21:573 (Abst.).
6. Hayes FA, Thompson EI, Parvey L, Rao B, Kun L, Parham D, 17. Burdach S, Jurgens H, Peters C, Nurnberger W, Mauz-Kor- Hustu HO. Metastatic Ewing’s sarcoma: Remission induction holz C, Korholz D, Paulussen M, Pape H, Dilloo D, Ko- and survival. J Clin Oncol 1987;5:1199 –1204.
scielniak E, Gadner H, Goebel U. Myeloablative radiochemo- 7. Bader JL, Horowitz ME, Dewan R, Watkins E, Triche TJ, therapy and hematopoietic stem-cell rescue in poor-prognosis Tsokos M, Kinsella TJ, Miser JS, Steinberg SM, Glatstein E.
Ewing’s sarcoma. J Clin Oncol 1993; 11:1482–1488.
Intensive combined modality therapy of small round cell and 18. Marcus RB, Graham-Pole JR, Springfield DS, Fort JA, Gross undifferentiated sarcomas in children and young adults: local S, Mendenhall NP, Elfenbein GJ, Weiner RS, Enneking WF, control and patterns of failure. Radiother Oncol 1989;16:189 – Million RR. High-risk Ewing’s sarcoma: End-intensification using autologous bone marrow transplantation. Int J Radiat 8. Jurgens H, Bier V, Harms D, et al. Malignant peripheral neuroectodermal tumors. A retrospective analysis of 42 pa- 19. Pape H, Wurm R, Burdach S, Nuernberger W, Bannach B, Schmitt G. Die lunge als hauptort der rezidive nach knochen- 9. Carli M, Pinkerton R, et al. Risk group analysis in metastatic mark-transplantation multifokaler Ewing-Sarkome: Sollte die ra- soft tissue sarcomas in children. European Intergroup Study diotherapie intensiviert werden? Zbl Radiol 1993;147:948 –949.
20. Bradley J, Reft C, Goldman S, Rubin C, Nachman J, Larson R, Radiation therapy for metastatic pediatric sarcomas ● E. A. DUNPHY CZYZEWSKI et al. Hallahan DE. High energy total body irradiation in the prep- sults of the CESS-studies. Strahlentherapie und Onkologie aration for bone marrow transplantation in leukemia patients.
Int J Radiat Oncol Biol Phys 1998;40:391–396.
24. Ragaz J, Jackson SM, Le N, Plenderleith IH, Spinelli JJ, 21. Sibley G, Mundt A, Hallahan D, Goldman S, Nachman J, Reft BascoVE, Wilson KS, Knowling MA, Coppin CM, Paradis M, C, Weichselbaum R, Hallahan D, Johnson L. Pattern of fail- Coldman AJ, Olivotto IA. Adjuvant radiotherapy and chemo- ures following total body irradiation and bone marrow trans- therapy in node-positive premenopausal women with breast plantation Ϯ local boost for advanced Neuroblastoma. Int J cancer. N Engl J Med 1997; 337(14):956 –962.
Radiat Oncol Biol Phys 1995;32:1127–1135.
25. Overgaard M, Hansen PS, Overgaard J, Rose C, Andersson 22. Marcus RB Jr, Million RR. The effect of primary tumor size M, Bach F, Kjaer M, Gadeberg CC, Mouridsen HT, Jensen on the prognosis of Ewing’s sarcoma. Int J Radiat Oncol Biol MB, Zedeler K. Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive ad- 23. Dunst J, Paulussen M, Jurgens H. Lung irradiation for Ew- juvant chemotherapy. Danish Breast Cancer Cooperative ing’s sarcoma with pulmonary metastases at diagnosis: Re- Group 82b Trial. N Engl J Med 1997;337(14):949 –955.

Source: http://radonc.ucsd.edu/research/publications/Documents/Sarcoma_Paper03.pdf