Abstract

Background:

Many papers have reported the results achieved with combined therapy for Ewing’s tumors, but little is known about the treatment and outcome of those 30–40% of patients who relapse.

Patients and methods:

In a retrospective study, we evaluated 195 patients with Ewing’s tumors treated at our institution from 1979 to 1997 with chemotherapy, radiotherapy, surgery or combined therapies after recurrence.

Results:

A second complete remission was achieved in only 26 patients (13.3%); 12 relapsed again and died of the tumor. The 5-year post-relapse event-free survival and overall survival were 9.7% and 13.8%, respectively; both of which were significantly better for patients who had relapsed ≥2 years after the beginning of the first treatment (14.3% versus 2.5%; P <0.001) and for patients who relapsed with only lung metastases (14.5% versus 0.9%; P <0.0005). In terms of treatment, patients treated with surgery or radiotherapy, alone or in combination with chemotherapy, had better survival rates than patients treated with chemotherapy alone (15.4% versus 0.9%; P <0.0001).

Conclusions:

The outcome of Ewing’s tumor patients who relapse after combined treatment is very poor. However, these patients may be divided into two groups: those that can be cured with traditional treatments (late relapse and/or only lung metastases), and a second group of patients (early relapses with metastases in lungs and/or other sites) who gain no benefit from traditional therapies. For the latter group, multicenter studies are needed to evaluate new strategies of treatment.

Received 12 March 2003; revised 14 April 2003; accepted 11 August 2003

Introduction

Due to a multimodal approach combining surgery and/or radiotherapy with systemic chemotherapy, the prognosis for patients with Ewing’s tumors of the bone which are non-metastatic at presentation has dramatically improved over the last 30 years. However, in spite of these results, metastases and/or local recurrence still develop in ∼30–40% of cases [17]. In papers reporting the results of adjuvant or neoadjuvant studies of Ewing’s tumors of the bone, little attention is given to the treatment and outcome of post-relapse patients. This is probably due to the many different types of relapse that occur, and, as a consequence, treatments are tailored almost on an individual basis. This lack of information in the literature makes evaluation of outcome and the analysis of factors influencing prognosis extremely difficult.

The aim of this study was to retrospectively report on post-relapse treatment and outcome of 195 patients with non-metastatic Ewing’s tumors at presentation, treated at a single institution from 1979 to 1997, who relapsed after adjuvant or neoadjuvant treatment.

Patients and methods

Patient selection

Patients who experienced a recurrence of an Ewing’s tumor (local, distant or combined) after treatment with any of the four consecutive protocols (REA-2, REN-1, REN-2 and REN-3) of the Rizzoli Orthopaedic Institute (Bologna, Italy) from 1979 to 1997 were eligible to enter this study. Patients with metastases at presentation were excluded.

When first admitted to our institution, all patients fulfilled the following criteria: histological diagnosis of Ewing’s sarcoma of the bone; age <40 years; absence of metastases at diagnosis; no previous treatment and a less than 4-week interval between biopsy and the beginning of treatment. Of a total of 429 patients who entered our study, 206 relapsed. One-hundred and ninety-five of them were included in the present review, for whom it was possible to collect exact information on the date and type of first recurrence (local and/or systemic), on treatment performed after relapse and on final outcome. Patient characteristics are listed in Table 1.

There were 138 male and 57 female patients. The median age at the time of diagnosis was 18.1 years (range 1.5–40) and the median volume of tumor was 214 ml (range 1.5–2008). The primary tumor was located in the extremity in 112 patients, the pelvis in 50 patients and at other sites in 33 patients. The pelvis (49 patients), femur (45) and tibia (22) were the most common sites of primary tumor. The serum level of lactate dehydrogenase (LDH) was normal in 103 patients and elevated in 92 patients.

Initial diagnosis and frontline treatment

Diagnosis of Ewing’s sarcoma was made on representative specimens obtained from an open biopsy. The primary tumor was staged using X-rays, Tc99 bone scan, computed tomography (CT) and, in more recent cases, magnetic resonance imaging (MRI). The presence of metastases was investigated using bone scintigraphy and CT of the lungs.

Chemotherapy was given according to four consecutive protocols, sequentially activated at the Rizzoli Orthopaedic Institute, which are described in detail elsewhere [810].

Local treatment consisted of surgery and/or radiotherapy. The choice of local treatment was based on the patient’s age, the site and size of the tumor and the presence of pathological fractures. Its aim was to achieve complete local control, while attempting to save the best function of the tumor-affected site. Strategies for local control have changed over the years.

After surgery, all specimens were carefully observed and surface labeled histological sections were taken. Surgical margins were evaluated according to the method proposed by Enneking et al. [11]. The response to chemotherapy was determined by a thorough histological examination of an entire coronal section of the tumor as previously described [12] and patients were classified as ‘poor responders’ (evidence of macroscopic foci of viable tumor cells) or ‘good responders’ (only isolated microscopic nodules of viable tumor cells or no viable tumor cells).

A retrospective medical chart and database review was carried out to identify Ewing’s sarcoma patient characteristics and disease features both at the time of diagnosis and at the time of recurrence; treatment received and outcome after recurrence were also evaluated.

Statistics

We evaluated whether patients in the study reached a second complete remission after treatment; the 5-year post-relapse event-free survival (PREFS) and the 5-year post-relapse overall survival (PROS) were calculated.

The duration of PREFS was defined as the time interval between the start of successful treatment of the recurrence and the appearance of a new recurrence, or to the most recent follow-up examination. The duration of PROS was defined as the time interval between the date of the first recurrence and death of any cause, or the most recent follow-up examination.

For each patient the following parameters were investigated: site, size and treatment of the primary tumor; patient age and gender; serum level of LDH at presentation; relapse-free interval from the beginning of treatment to first relapse; pattern of relapse (local, systemic, or both) including the site of first metastasis; treatment performed at recurrence and its result; relapse-free interval from treatment of recurrence to eventual second recurrence.

The probability of having complete remission, PREFS and PROS were estimated using the Kaplan–Meyer method. Demographic data, treatment factors and tumor characteristics were examined with univariate Cox regression models to investigate their relationships with the probability of second remission, PREFS and PROS.

Results

In this study, 30 patients were treated with the REA-1 protocol, 61 with REN-1, 39 with REN-2 and 65 with REN-3. Local treatment was radiotherapy in 87 patients, surgery in 57 patients, and surgery followed by radiotherapy in 50 patients. One patient (a 3-year-old boy with Ewing’s tumor in the proximal femur, the parents of which refused amputation) received only chemotherapy (Table 1). In the operated patients, the surgical margins were adequate (radical or wide) in 166 patients (85%) and inadequate (marginal or intralesional) in 29. In the 89 patients treated with neoadjuvant protocols and locally by surgery, the histological response to chemotherapy was good in 35 patients (39%) and poor in 54 (61%).

Time and type of recurrence

Median time between the start of frontline treatment and first recurrence was 27.3 months (range 2–222). Seventy-seven first recurrences (38.9%) occurred after 2 years and 25 (12.8%) after 5 years.

There were 138 (70.8%) distant and 57 (29.2%) local recurrences. Local recurrence was isolated in only one case, while in the other 56 patients it was associated with distant recurrences. In 29 patients, local and distant recurrence were contemporary, in 18 patients local recurrence was before distant metastases at 4–12 months, and in nine patients metastases appeared 3–14 months before local recurrence. The first site of distant relapse was the lung in 83 patients (42.6%), bone in 77 patients (39.5%), lung and bone in 29 patients (14.9%) and other sites in five patients (four at the central nervous system and one in the skin).

Treatment after relapse

Treatment after relapse was not homogeneous and varied according to the type of relapse (systemic, local, or combined), the site of distant relapse (lung, bone, lung and bone, other sites), the number of pulmonary metastases and the type of first-line local and systemic treatment performed.

More specifically, the patient who only relapsed locally (local recurrence of a Ewing’s tumor of the tibia initially treated by local radiotherapy) was treated with surgery alone (amputation). The 194 patients who relapsed with metastases and/or local recurrence received treatment as follows: (i) chemotherapy, 110 patients (56.4%); (ii) radiotherapy, 33 patients (17.%); (iii) surgery, 12 patients (5.6%); (iv) surgery followed by chemotherapy, 15 patients (7.7%); (v) radiotherapy followed by chemotherapy, five patients (2.6%). Palliative treatment was performed in only 19 patients, who were treated with chemotherapy alone, due to the extension and location of the recurrent disease, i.e. site, number and size of tumors that were too numerous and/or too large to be treated with surgery and/or radiotherapy.

Of a total of 57 patients with local recurrence, 10 patients were treated with surgery (amputation in seven cases and resections in three), 35 with radiotherapy (61.4%) and nine with chemotherapy alone (15.7%). In three patients no treatment was performed. The metastases surgically treated were located in the lung (25 patients) and in the bone (one patient); surgery consisted of wedge resection in 20 patients, lobectomy in five patients and resection of the rib in one patient. Radiotherapy was applied at different doses ranging from 16.5 to 51.8 Gy. Chemotherapy was used in 97 patients at conventional doses using different combinations of drugs (cisplatinum, melphalan and ifosfamide). In 35 patients, chemotherapy was performed with high doses of melphalan and busulfan with peripheral stem-cell transplantation (five cases) or autologous bone marrow transplantation (30 cases). Only in recent years have these high-dose therapies been used to treat patients.

Results of salvage treatment

Among the 195 relapsed patients, 26 entered remission (13.3%). According to the type of relapse, the rate of remission was 100% (1/1) for the patient who only relapsed locally, 7.1% for the 56 who had local and systemic recurrence and 15.2% for the 130 patients with distant recurrence: differences that are statistically significant (P <0.01).

As shown in Table 2, patients who relapsed with distant metastases located only in the lung (83 patients) had a rate of remission significantly higher than those (111 patients) with single metastases located both inside and outside the lung (228.9% versus 0.9%; P = 0.0001). The rate of remission was 29% for the 65 patients who relapsed with only lung metastases, 23% for 17 patients who relapsed with local recurrence and lung metastases, and only 4% for those 53 patients who relapsed with bone metastases. None of the patients who relapsed with local recurrence and bone metastases (25 patients), local recurrences and bone-and-lung metastases (14), lung-and-bone metastases (15) or metastases outside lung and bone (five) achieved remission of the disease.

The mean time to first relapse for all patients was 27.3 months (range 2–222). The time to first relapse was significantly longer for the 26 patients who reached second remission after treatment, than for the 169 who were never free of the disease [52 months (range 17–169) versus 23.5 months (range 2–222); P <0.0001].

According to therapy after relapse, the rate of patients who entered remission was 86.6% for 15 patients treated with surgery plus chemotherapy, 33.3% for patients treated with radiotherapy plus chemotherapy, 66.7% for patients treated by surgery alone, 6% for the 33 patients treated by radiotherapy alone and 0.9% for the 110 patients treated by chemotherapy alone. These differences are highly significant (P <0.0001). The relationship between second remission and type of treatment at first relapse should be considered with caution. In fact, treatment for patients with single lung metastases is surgery, whereas patients with multiple unresectable metastatic locations in the lung and/or in other sites are treated with chemotherapy alone. None of the 15 patients treated with high-dose chemotherapy alone reached remission, while of the 96 patients treated with standard-dose chemotherapy, one achieved disease-free status. With regard to patients treated with chemotherapy combined with surgery or radiotherapy, remission was achieved by one of two (50%) patients treated with standard-dose chemotherapy and by 13 of 18 (72.2%) patients treated with high-dose chemotherapy.

Final outcome

Of the 169 patients who never achieved disease-free status, 164 died of widespread tumors, and five were alive with uncontrolled disease 6–18 months after their last treatment. For the patients who died, the interval between start of first treatment and death was 33.6 months (range 4–222), while the interval from relapse to death was 11.7 months (range 2–16). Among the 26 patients who entered remission, 12 relapsed again 9–50 months (mean 16.8 months) after the last treatment. None of them reached a new (third) remission. Ten died of widespread tumors 16–56 months after remission (mean 26.7 months), while two were alive with uncontrolled disease 6 and 13 months after last treatment, respectively. The remaining 14 patients were alive and apparently free of disease 1.5–17 years after their last treatment (mean 7.9 years). The primary site of tumor in these patients was the extremity in 12 cases and the pelvis in two. Relapses were one local recurrence (treated by amputation), 12 lung metastases (treated by surgery plus chemotherapy in five cases, by surgery alone in five and by radiotherapy plus chemotherapy in one) and one local recurrence followed by lung metastases. In this last patient, an Ewing’s tumor of the humerus was initially treated by radiotherapy; there followed a local recurrence after 38 months which was treated by surgery alone (resection and prosthesis). Three years later, he had a lung metastasis that was resected. The patient is alive and free of disease 14 years after this last treatment.

With a mean follow-up of 9.5 years (range 1.5–19), the 5-year PREFS and PROS were 9.7% and 13.8%, respectively. The 5-year PREFS was significantly related to the following: time of first relapse (5% for patients who relapsed within 2 years versus 19% for those who relapsed later; P <0.003); type of relapse (2% for patients who relapsed with local recurrence and metastases versus 14% for patients who had only distant recurrence; P <0.02); site of first metastases (4% for patients who had metastases outside the lung, or more sites involved, versus 19% for patients who had only lung metastases); type of treatment after relapse (2.7% for patients treated with chemotherapy, 13% for patients treated by radiotherapy and/or chemotherapy and 46% for patients treated by surgery and/or chemotherapy). The statistical significance, when comparing therapies, are as follows: chemotherapy versus surgery, P <0.0001; chemotherapy versus radiotherapy, P <0.04; radiotherapy versus surgery, P <0.006. Since relapsed patients reached a second remission, the differences concerning the PROS can essentially be superimposed.

In summary, although our study confirmed that Ewing’s tumor patients who relapse after combined treatment have a very poor prognosis, there are significant differences in survival according to different sub-groups of patients with specific features. Patients who experienced recurrence ≥2 years after the first treatment had a significantly better prognosis than patients who relapsed early. The 5-year post-relapse survival in the two groups were 14% and 7%, respectively (P <0.001). These results confirm the observations reported by other authors [2, 3, 7, 13, 14]. Another favorable prognostic factor is the site of the first metastasis. Patients who relapsed with only lung metastases had a significantly better outcome in comparison with patients who relapsed with metastases other than in the lung: 5-year PREFS were 14% and 0.9%, respectively (P <0.0001). It is important to stress that in these patients outcome does not seem to be related to the number of pulmonary lesions.

If we compare the 14 patients who relapsed with only lung metastases ≥2 years after the beginning of treatment with the other 181 patients, the prognostic importance of time of relapse and site of relapse is even more evident. In fact, the 5-year PREFS in the two groups were 57% and 6%, respectively (P <0.0001).

Discussion

Although improved local control therapy and the development of new and more intensive chemotherapy combinations have reduced the frequency of relapse, a large proportion of patients with Ewing’s tumors non-metastatic at presentation still experience recurrence. While there are many studies about treatment and outcomes of patients presenting with non-metastatic Ewing’s tumors [110], very few papers described the strategies used to cure the 30% of patients who relapse after first-line treatments [13, 14].

Even if this population is not homogeneous for time and type of relapse and for type of local and/or systemic salvage treatment performed, some interesting data resulted from this retrospective review. First, we were able to confirm that the prognosis of patients with Ewing’s tumors who relapse after combined treatment is very poor. In our present series, only 13% of patients who relapsed achieved a second remission. Surprisingly, in these cases the remission was maintained in more than half of the cases. This seems to indicate that the major issue for patients is to reach a second remission.

The 5-year PROS of our study (13.8%) is significantly worse than the one reported by Rodriguez-Galindo et al. [13]; these authors, in a study concerning 71 patients, observed a 5-year post-relapse survival of 23.7% for patients with localized disease at presentation relapsing after treatment with one of the three consecutive institutional protocols applied. On the other hand, our data are similar to the ones reported for 64 patients in the UKCCSG ET-2 study [5]. These authors evaluated the post-relapse outcome of 64 patients with localized Ewing’s sarcoma, treated according to the ET-2 protocol of the United Kingdom Children’s Cancer Study Group (UKCCSG), who had relapsed. Treatment at relapse was different, including chemotherapy, surgery, radiotherapy and high-dose therapy or megatherapy in various combinations. The overall actuarial median survival from relapse for all patients was 14 months. In multivariate analysis the duration of first remission was the only factor associated with longer survival after relapse.

In contrast with the results of Rodriguez-Galindo et al. [13], we found that patients with local recurrence did not have a better prognosis than patients who relapse only with metastases. In only one of the 57 patients who experienced local recurrence did the disease remain isolated. In all other cases, local recurrence was associated with systemic relapse and all of these patients died of the tumor.

From our study we also found that the type of treatment after relapse influenced outcome. The rate of patients entering remission, as well as the rate of 5-year PREFS, was significantly higher for patients treated with surgery or radiotherapy (with or without chemotherapy) than for patients treated by chemotherapy alone. No patients treated with chemotherapy alone were cured. Of course this result must be treated with caution. In fact, patients treated with surgery or radiotherapy alone were highly selected, i.e. they were those patients with a better expected prognosis, while the treatment with chemotherapy alone was restricted to those patients with a poorer prognosis because of unresectable lesions and/or no indication for radiotherapy.

For these reasons our study does not allow a ‘standard’ definition of best treatment for relapsing ET patients. Nonetheless, our results indicate that it is possible to clearly separate relapsing ET patients into two groups with different prognoses. The first group (long disease-free interval, resectable metastases located only in the lung with a 5-year EFS after relapse of 57%) may benefit from conventional treatments (surgery, radiotherapy, chemotherapy with drugs commonly used in this tumor). The prognosis in the second group of patients (short disease-free interval, local and distant recurrence, unresectable metastases outside the lung) is not improved by any of the standard treatments. We believe international cooperative studies are needed to conceive new strategies in order to give this last group of patients a chance of cure.

+

Correspondence to: Dr G. Bacci, Sezione di Chemioterapia, Istituto Ortopedico Rizzoli, via Pupilli 1, 40136 Bologna, Italy. Tel: +39-051-6366400; Fax: +39-051-6366277; E-mail: gaetano.bacci@ior.it

Table 1.

Second remission, 5-year EFS and overall survival after relapse according to several variables related to primary tumor and treatment

Variable No. of patients Rate of patients who enter second remission (%) P 5-year EFS after relapse (%) P Survival after relapse (months) P 
Gender        
 Male 138 12.3 0.51  7.2 0.51 18.6  0.85 
 Female  57 15.9   7  17.7  
Age, years        
 ≤14  70 17.1 0.34 10 0.39 18.2  0.96 
 >14 125 11.2   5.6  18.5  
Site        
 Extremity 112 17.9 0.09 10.7 0.06 21  0.24 
 Pelvis  50  6.1   4.1  14.7  
 Other sites   33  9.1   0   15.2 
Serum LDH        
 Normal 103 18.4 0.04 10.7 0.08 22.3  0.6 
 Elevated  92  7.6   3.3  14.0  
Therapy for primary tumora        
 Radiotherapy  87  9.2 0.09  5.7 0.09 15.7  0.32 
 Surgery  57 22.8  14  22.8  
 Surgery + radiotherapy  50 10   2  17.9  
Histological response to chemotherapy        
 Poor  54  9.3 0.21  4.1 0.90 18.4  0.56 
 Good  35 20   5.7  15.2  
Variable No. of patients Rate of patients who enter second remission (%) P 5-year EFS after relapse (%) P Survival after relapse (months) P 
Gender        
 Male 138 12.3 0.51  7.2 0.51 18.6  0.85 
 Female  57 15.9   7  17.7  
Age, years        
 ≤14  70 17.1 0.34 10 0.39 18.2  0.96 
 >14 125 11.2   5.6  18.5  
Site        
 Extremity 112 17.9 0.09 10.7 0.06 21  0.24 
 Pelvis  50  6.1   4.1  14.7  
 Other sites   33  9.1   0   15.2 
Serum LDH        
 Normal 103 18.4 0.04 10.7 0.08 22.3  0.6 
 Elevated  92  7.6   3.3  14.0  
Therapy for primary tumora        
 Radiotherapy  87  9.2 0.09  5.7 0.09 15.7  0.32 
 Surgery  57 22.8  14  22.8  
 Surgery + radiotherapy  50 10   2  17.9  
Histological response to chemotherapy        
 Poor  54  9.3 0.21  4.1 0.90 18.4  0.56 
 Good  35 20   5.7  15.2  

aOne patient did not receive any local treatment.

EFS, event-free survival; LDH, lactate dehydrogenase.

Table 2.

Remission, 5-year EFS and overall survival after relapse according to time and type of relapse, first site of metastases and second-line treatment

Variable No. of patients Rate of patients who enter second remission (%) P 5-year EFS after relapse (%) P Survival after relapse (months) P 
Time to relapse, months        
 ≤24 118  6.8 0.001  2.5 0.02 12.5 0.0009 
 >24  77 23.4  14.3  27  
Type of relapse        
 Only distant 138 15.2 0.01  8.6 0.0004 19.3 0.20 
 Local ± distant  57  8.7   3.5  15.2  
Site of first distant relapse        
 Lung  83 28.9 0.0001 14.5 0.0001 27.1 0.0001 
 Bone  77  1.3   1.3  12.7  
 Lung and bone  29  0   0   8.2  
 Other sites  5  0   0   6.0  
Treatmenta        
 Chemotherapy 111  0.9 0.0001  0 0.0001 11.4 0.002 
 Radiotherapy  33  6.1   3.0  18.7  
 Surgery  12 63.4  45.1  78.9  
 Chemo + surgery  15 86.7  33.3  30.5  
 Chemo + radiotherapy  5 33.3  33.3  20.3  
 Other treatments  19  0   0  11.0  
Type of chemotherapy        
 Standard dose  98  2.0 0.0001  0 0.0001 11.1 0.0001 
 High dose  33 39  21.2  27.1  
Variable No. of patients Rate of patients who enter second remission (%) P 5-year EFS after relapse (%) P Survival after relapse (months) P 
Time to relapse, months        
 ≤24 118  6.8 0.001  2.5 0.02 12.5 0.0009 
 >24  77 23.4  14.3  27  
Type of relapse        
 Only distant 138 15.2 0.01  8.6 0.0004 19.3 0.20 
 Local ± distant  57  8.7   3.5  15.2  
Site of first distant relapse        
 Lung  83 28.9 0.0001 14.5 0.0001 27.1 0.0001 
 Bone  77  1.3   1.3  12.7  
 Lung and bone  29  0   0   8.2  
 Other sites  5  0   0   6.0  
Treatmenta        
 Chemotherapy 111  0.9 0.0001  0 0.0001 11.4 0.002 
 Radiotherapy  33  6.1   3.0  18.7  
 Surgery  12 63.4  45.1  78.9  
 Chemo + surgery  15 86.7  33.3  30.5  
 Chemo + radiotherapy  5 33.3  33.3  20.3  
 Other treatments  19  0   0  11.0  
Type of chemotherapy        
 Standard dose  98  2.0 0.0001  0 0.0001 11.1 0.0001 
 High dose  33 39  21.2  27.1  

aThe patient who relapsed with only local recurrence was treated with amputation.

EFS, event-free survival.

References

1.
Burgert EO, Nesbit ME, Garnsey LA et al. Multimodal therapy for the management of nonpelvic, localized Ewing’s sarcoma of bone: Intergroup Study IESS-II.
J Clin Oncol
 
1990
;
8
:
1514
–1524.
2.
Nesbit ME, Gehan EA, Burgert EO et al. Multimodal therapy for the management of primary, nonmetastatic Ewing’s sarcoma of bone: a long-term follow up of the first intergroup study.
J Clin Oncol
 
1990
;
8
:
1664
–1674.
3.
Oberlin O, Patte C, Demeocq F et al. The response to initial chemotherapy as a prognostic factor in localized Ewing’s sarcoma.
Eur J Cancer Clin Oncol
 
1985
;
21
:
463
–467.
4.
Nilbert M, Saeter G, Elomaa I et al. Ewing’s sarcoma treatment in Scandinavia 1984–1990: ten-year results of the Scandinavian Sarcoma Group protocol SSGIV.
Acta Oncol
 
1998
;
37
:
375
–378.
5.
Craft A, Cotterill S, Malcolm A et al. Ifosfamide-containing chemotherapy in Ewing’s sarcoma: the second United Kingdom Children’s Cancer Study Group and the Medical Research Council Ewing’s Tumor Study.
J Clin Oncol
 
1998
;
16
:
3628
–3633.
6.
Oberlin O, Deley MC, Bui BN et al. Prognostic factors in localized Ewing’s tumours and peripheral neuroectodermal tumours: the third study of the French Society of Paediatric Oncology (EW88 study).
Br J Cancer
 
2001
;
85
:
1646
–1654.
7.
Paulussen M, Ahrens S, Dunst J et al. Localized Ewing’s tumor of bone: final results of the Cooperative Ewing’s Sarcoma Study CESS 86.
J Clin Oncol
 
2001
;
19
:
1818
–1829.
8.
Bacci G, Mercuri M, Longhi A et al. Neoadjuvant chemotherapy for Ewing’s tumour of bone: recent experience at the Rizzoli Orthopaedic Institute.
Eur J Cancer
 
2002
;
38
:
2243
–2251.
9.
Bacci G, Picci P, Ferrari S et al. Neoadjuvant chemotherapy for Ewing’s sarcoma of bone: no benefit observed after adding ifosfamide and etoposide to vincristine, actinomycin, cyclophosphamide, and doxorubicin in the maintenance phase: results of two sequential studies.
Cancer
 
1998
;
82
:
1174
–1183.
10.
Bacci G, Toni A, Avella M et al. Long-term results in 144 localized Ewing’s sarcoma patients treated with combined therapy.
Cancer
 
1989
;
63
:
1477
–1486.
11.
Enneking W, Dunham W, Gebhardt M et al. A system for the classification of skeletal resections.
Chir Organi Mov
 
1990
;
75
:
217
–240.
12.
Picci P, Bohling T, Bacci G et al. Chemotherapy-induced tumor necrosis as a prognostic factor in localized Ewing’s sarcoma of the extremities.
J Clin Oncol
 
1997
;
15
:
1553
–1559.
13.
Rodriguez-Galindo C, Billups CA, Kun LE et al. Survival after recurrence of Ewing’s tumors: the St Jude Children’s Research Hospital experience.
Cancer
 
2002
;
94
:
561
–569.
14.
Schuck A, Ahrens S, Paulussen M et al. Local therapy in localized Ewing’s tumors: results of 1058 patients treated in the CESS 81, CESS 86, and EICESS 92 trials.
Int J Radiol Biol Phys
 
2003
;
55
:
168
–177.

Author notes

1Department of Chemotherapy, 2Department of Thoracic Surgery, Istituto Ortopedico Rizzoli, Bologna; 3Department of Radiotherapy, Policlinico S.Orsola/Malpighi, Bologna, Italy