Allogeneic and autologous stem-cell transplantation in advanced Ewing tumors

Annals of Oncology 11: 1451-1462, 2000. O 2000 Kluwer Academic Publishers. Primed in the Netherlands

Original article Allogeneic and autologous stem-cell transplantation in advanced Ewing tumors An update after long-term follow-up from two centers of the European Intergroup Study EICESS S. Burdach,1 B. van Kaick,2 H. I Laws,2 S. Ahrens,3 R. Haase,1 D. Korholz,4 H. Pape,2 J. Dunst,1 T. Kahn,4 R.Willers,2 B. Engel,5 U. Dirksen,2 C. Kramm,2 W. Niirnberger,2 A. Heyll,2 R. Ladenstein,6 H. Gadner,6 H. Jiirgens3 & U. Gobel2 for the Stem-Cell Transplant Programs at Diisseldorf University Medical Center, Germany and St. Anna Kinderspital, Vienna, Austria Martin Luther University Halle-Wittenberg; 2Heinrich Heine University Diisseldorf; 3 Wilhelms-University Munster; 4University of Leipzig, Germany; i Children's Hospital, University of Southern California, Los Angeles, USA; 6St. Anna Kinderspital, Vienna, Austria

Introduction Ewing tumor (FT) is a systemic malignancy probably derived from a common embryonic progenitor of neuroectodermal and hematopoetic tissue. Myeloablative consolidation and hematopoetic stem-cell transplantation have become in widespread use as part of intensified treatment protocols in advanced Ewing tumors (AET) [1—6]. The role of autologous vs. allogeneic grafting is still unclear in most solid tumors. In some hematological malignancies allogeneic transplantation has been proven to be more efficacious than autologous transplantation [7-13]. There may be two reasons for a potential advantage

of allogeneic transplantation in Ewing tumors. Firstly, as evidenced by gene marking studies in patients with pediatric malignancy autologous transplants may contain tumor cells that contribute to relapse [14], whereas allografts are free of malignant cells. Secondly graftversus-tumor (GVT) effects have been observed in various kinds of hematological malignancies; in some solid tumors the possible existence of a GvT effect is illustrated by case reports [15-26]. In contrast, increased toxicity after allogeneic transplantation seems to hamper overall improvement in event-free survival (EFS) as evidenced by lack of increased cure rates after allogeneic grafts in Hodgkin's lymphoma [15, 27] and neuroblastoma [28].

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 5, 2016

comitant lung disease, age at time of diagnosis, pelvic involvement, involved compartment radiation, histopathological diBackground; An update of results from the High Risk Protocol agnosis. of the Meta-EICESS Study, conducted at the Pediatric StemResults: EFS for the 36 patients was 0.24 (0.21) ± 0.07. Cell Transplant Centers of DOsseldorf and Vienna. In order to Eighteen of thirty-six patients suffered relapse or died of disevaluate a possible therapeutic benefit after allogeneic SCT in ease, nine of thirty-six died of treatment related toxicity (DOC). patients with advanced Ewing tumors (AET), we compared Nine of thirty-six patients are alive in CR. Age ^ 17 years at outcome after autologous and allogeneic stem-cell transplan- initial diagnosis (P < 0.005) significantly deteriorated outcome. tation (SCT). According to the type of graft, EFS was 0.25 ± 0.08 after Patients and methods: We analyzed 36 patients treated with autologous and 0.20 ± 0.13 after allogeneic SCT. Incidence of the myeloablative Hyper-ME protocol (hyperfractionated total DOC was more than twice as high after allogeneic (40%) body irradiation, melphalan, etoposide ± carboplatin) between compared to autologous (19%) SCT, even though the difference November 1986 and December 1994. Minimal follow-up for did not reach significance (P = 0.08, Fisher's exact test). all patients wasfiveyears. All patients underwent remission Conclusions: Because of the rather short observation period, induction chemotherapy and local treatment before myelo- secondary malignant neoplasm (SMN) may complicate the ablative therapy. Seventeen of thirty-six patients had multifocal future clinical course of some of our patients who are currently primary Ewing's tumor, eighteen of thirty-six had early, multi- viewed as event-free survivors. EFS in AET is not improved by ple or multifocal relapse, one of thirty-six patients had unifocal allogeneic SCT due to a higher complication rate. The patient late relapse. Twenty-six of thirty-six were treated with autolo- group was to small to analyze for a possible graft-versus-tumor gous and ten of thirty-six with allogeneic hematopoetic stem effect. cells. We analyzed the following risk factors, that could possibly influence the event-free survival (EFS): number of involved Key words: advanced Ewing tumors, allogeneic stem-cell bones, degree of remission at time of SCT, type of graft, transplantation, autologous stem-cell transplantation, IL-2 indication for SCT, bone marrow infiltration, bone with con- therapy Summary

1452 To our knowledge, there are no published studies comparing allogeneic and autologous hematopoietic stem-cell transplantation in Ewing tumors. In order to evaluate a possible therapeutic benefit after allogeneic transplantation we analyzed the available data from 12years experience with the EICESS Hyper-ME protocol concerning EFS, death of complication (DOC) and death of disease (DOD) in patients with AET.

Study population Type of graft The type of graft was autologous BM in 5 of 36 (14%), allogeneic BM in 10 of 36 (28%) and autologous unselected peripheral blood stem cells (PBSC) in 21 of 36 (58%). Two patients (Tables 1 and 2, patient nos. 10 and 32) received unselected PBSC and autologous BM and are registered in the patient group with autologous BM grafts. Local stage and extent of disease

Patients and methods Patients Patient characteristics

Histopathological diagnosis Diagnosis was established by the reference pathologists in Kiel (D. Harms, D. Schmidt) or Vienna (M. Salzer-Kuntschik, G. Amann). According to the histologic definition of Ewing's tumors proposed by Schmidt et al., Schmidt, Harms and Burdach [29, 50] 31 of 36 (86%) patients were diagnosed as Ewing's sarcoma and 5 of 36 (14%) as malignant peripheral neuroectodermal tumor. Staging The extent of metastatic disease was evaluated either by Tc bone scan alone (n = 22) or by Tc bone scan plus partial body magnetic resonance imaging (PB MRI, n = 5) or by Tc bone scan plus total body MRI (TB MRI, n = 9). Lung disease was assessed by thoracic computed tomography (CT). Inclusion criteria The EICESS group had reached the following agreement on the indications for high-dose therapy (HDT) in advanced Ewing tumors. Patients with multifocal disease at either initial or relapse diagnosis, patients with early relapse (less than two years after diagnosis) or patients with multiple relapses have the worst prognosis and require HDT[1]. Patients who showed a clinical response to induction chemotherapy became eligible for myeloablative intensification irrespective of tumor volume or histological response. Clinical response was judged as amelioration of fever, anorexia, malaise or reduction in tumor mass. The allocation criteria for patients receiving allogeneic bone marrow transplantation was the availability of an HLA-identical sibling. All donors were completely matched for HLA class I and II by sero typing. Institutional review This protocol has been approved by the Ethics Committees of DQsseldorf and Vienna. Informed consent was obtained from each patient or if appropriate, from the patient's guardian.

Indicationfor stem-cell transplantation (SCT) Seventeen of thirty-six (47%) patients were transplanted because of primary multifocal disease. In 19 of 36 (53%) patients, the diagnosis that made them eligible for transplantation was relapse. Seven of nineteen (37%) patients had early relapse ( < 2 4 months after diagnosis of primary tumor), three of nineteen (16%) patients had late but multifocal relapse and two of nineteen patients had two relapses before SCT. In 1 of 19 (5%) relapse patients there was unifocal late relapse 36 months after localized primary bone disease (Tables 1 and 2, patient no. 14). Six of nineteen (32%) additional relapse patients had primary multifocal disease followed by early (n = 4) or late (n = 2) relapse before SCT.

Study protocol The current time schedule of diagnostic and therapeutic procedures for treatment of AET is depicted in Figure 1. Induction protocol After diagnosis, all patients underwent remission induction chemotherapy. In the patient group with primary multifocal disease (n = 17) 7 of 17 (41%) patients were treated according to the CESS 86 or CWS 86 protocol with vincristine, adriamycin, ifosfamide and actinomycin D (VAIA regime) or vincristine, adriamycin, carboplatin and actinomycin D (VACA regime) [30]. Ten of seventeen (59%) patients were randomized for EVAIA or VAIA cycles of chemotherapy according to the CESS 91 or EICESS 92 protocol [I, 6]. EVAIA contains etoposide, vincristine, ifosfamide and adriamycine alternating with actinomycine D. In the patient cohort with relapsed Ewing tumors (n = 19) 16 of 19 (84%) patients received ifosfamide and etoposide alone or in combination with carboplatin, cisplatin or cyclophosphamide. Two of nineteen (11%) patients received EVAIA and one of nineteen (5%) patients received VAIA or VACA induction chemotherapy with actinomycine D instead of adriamycine. Harvest of grafts After the third and/or the fourth cycle of EVAIA, autologous bone marrow (BM, n = 5) or autologous peripheral blood stem cells (PBSC, n = 21) were harvested 8-14 days after start of chemotherapy. Priming of peripheral blood progenitor cells with 250 ug/m 2 /d E. colt derived non-glycosylated granulocyte colony stimulating factor (G-CSF) or granulocyte macrophage-colony stimulating factor (GM-CSF) was started 24 hours after the last dose of chemotherapy and was discontinued after completion of harvest. A CD34+ cell count greater than

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 5, 2016

Between November 1986 and December 1994, 36 patients were treated for advanced ETat the transplant centers of the University of Dusseldorf (n = 24) and of the hospital of Vienna (n = 12). The median age at initial diagnosis was 15.6 ± 5 years (range 4—30). Twelve of thirty-six (33%) patients were above seventeen years old at initial diagnosis. Ratio of males to females was approximately 1.1. Minimal follow-up was 5 years with a range from 5 to 15 years. Median follow up was 7.4 years from diagnosis and 6.7 years from transplantation. Detailed patient characteristics are listed in Tables 1 and 2.

Twenty-seven of thirty-six (75%) patients presented with bone metastases at diagnosis, which led to transplantation. To avoid any bias from resolution of the imaging technique, the number of involved bones was calculated from Tc bone scans for all patients analyzed. Ten of thirtysix (28%) patients presented with lung disease at diagnosis before transplantation. Five of nineteen (26%) patients presented with relapse in the lung without bone or soft tissue metastases Fourteen of thirtysix (39%) patients had bone marrow involvement, eighteen of thirty-six (50%) patients had pelvic involvement. For more details see Table 3.

1453 Table 1. Patient data, relapse (early, late or multiple). Patient no

Age at initial diagnosis (years)

Date Site of initial diagnosis

Therapy of primary disease

Date of

1

22

4-89

Humerus. CESS-86/ Th 12, Rad skull 28-48.2 Gy

3-90

2

19

1-87

Fibula

8-88

12

8-78

Fibula left

Relapse therapy

Remis- Date of sion graft status at transplant

Pelvis, femur

Carbo, Ifo/ Rad 42 Gy

2 PR

Lung, pleura, chest wall, 4 rib left, mamma left

relapse

CESS-86 Sur/Rad

CyC,VCR, ADR/Rad 30 Gy

Mode Ablative of therapy graft

Cytokines Outcome 3/98

Date of death or relapse

5-6-90

PBSC Mel 180 mg/m 2 , GM-CSF Eto 60 mg/kg, 12GyTBI

DOD

2-9-90

Ifo, cto/Sur/ 2.CR Rad 35.2 Gy

18-4-89

PBSC Mel 160 mg/m 2 , GM-CSF Eto 60 mg/kg, 12GyTBI

DOD

3-8-89

12-88 Chest wall, lung

CESS81/ Sur/Rad 50 Gy

16-5-90

Auto BM

12-89 Mediastinum, lung

CESS86/Sur

4 6 G y

3

Relapse site

3.CR

Mel 180 mg/m 2 , GM-CSF Eto 60 mg/kg, 12GyTBI

CR

7

6-86

Scapula left

CESS-86, Sur/Rad 44.8 Gy

2-88

Pelvis, BM

Ifo, Eto/Rad 2.CR 44.8 Gy

17-10-88 PBSC M e l l 2 0 m g / m 2 , Eto 60 mg/kg, 12GyTBI

5

8

4-87

Femur

CESS-86, Sur/Rad 45 Gy

8-88

Lung

Ifo, Eto/Sur

2.PR

27-12-88 PBSC Mel 140 mg/m 2 , GM-CSF, CR Eto 60 mg/kg, IL-2 12GyTBI

6

17

12-86 Os ischium, left

CESS-86, Rad 60.8 Gy

5-88 Chest wall, 12-88 scapula

Ifo, Eto, carboplatin/ Rad 30 Gy

3.CR

27-3-89

PBSC Mell40mg/m 2 , Eto 60 mg/kg, 12GyTBI

GM-CSF

DOD

24-10-89

7

8

12-87

Fibula

CESS-86, Sur/Rad 46 Gy

10-89 Lung

VCR, ActoD, Ifo, Eto/Rad 10 Gy/Sur

2.CR

24-7-90

PBSC Mel 160 mg/m 2 , GM-CSF, Eto 60 mg/kg, IL-2 12GyTBI

DOD

21-*-92

8

6

5-86

Femur

CESS-86, Sur/Rad 46 Gy

1-88

lung

Cispl, Ifo, VP16/Sur

2.CR

10-5-88

Auto BM

Mel 120 mg/m 2 , GM-CSF Eto 60 mg/kg, 12GyTBI

DOD

15-3-89

9

23

8-86

Osmetatarsale

CESS-86, Sur/Rad 46 Gy

9-87

Orbita, ribs, scull

Cispl, Ifo, VP16/Rad 50 Gy

2.PR

14-6-88

PBSC Mel 120 mg/m 2 , Eto 40 mg/kg, !2GyTBI

DOD

27-10-88

10

12

12-83 Chest CESS-81/ wall, ribs Sur/Rad 39 6 Gy

5-86

Lung

Cispl + Ifo + 2.CR VP16

27-11-86 Auto M e l l 8 0 m g / m 2 , BM + 12GyTBI PBSC

CR

11

7

11-84

Humerus

CESS81, Sur/70Gy

7-87

Scull, lung

Cispl, Ifo, VP16, Rad 55 Gy

2.PR

3-11-87

CR

12

15

9-89

Tibialeft

CESS-86/ Sur/Rad 44.8 Gy

6-92

C3-C6

CWSRez91/Sur/ Rad32Gy

2.CR

15-12-92 PBSC M e l l 2 0 m g / m 2 , Eto 40 mg/kg, Carbo 800 mg/ m, 12GyTBI

13

23

2-90

Ribs, spine

CESS-86/ Sur/Rad 44 Gy

1-91

Humerus

CESS-Rez 91/Rad 50 Gy

3.CR

22-12-92 allo BM

4-92

Th 10

VIP Rez.

15

Allo BM

GM-CSF

Eto 60 mg/kg, 12GyTBI

G-CSF

Mel 120mg/m 2 , G-CSF Eto 40 me /kg

CR

CR

DOD

15-8-93

Carbo 500 mg/ m 2 , 12GyTBI

16

4-88

Tibia

CESS-86/ Sur

4-91

Lung

Ifo + Carbo + VP16+ Cispl/Sur

2.CR

3-9-91

Allo BM

Mel 120 mg/m 2 , Eto 60 mg/kg, 12GyTBI

DOC

9-7-92

27

1-93

Femur, Orbita, BM

EICESS 92

10-93 Scull, sternummandible, spine, pelvis, femura, humeri, tibia, ribs

EVAIA/ Rad32Gy/ Sur

2.PR

5-7-94

Allo BM

Mel 120 mg/m 2 , G-CSF Etol800mg/ kg,12GyTBI

DOD

5-11-94

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 5, 2016

4

1454 Tablet. (Continued). Patient Age at initial no. diagnosis (years)

Date

Site

of

initial diagnosis

Therapy of Date of primary relapse disease

Relapse site

Relapse therapy

Remis- Date of sion graft status

Mode Ablative therapy graft

Cytokines

of

Out- Date of come death or 3/98 relapse

at

transplant Right femur, CESS-Rez/ chest wall 54 Gy right (suspected)

2.CR

19-7-94

PBSC Mel 120 mg/m 2 , Eto 1800 mg/kg, 12GyTBI

G-CSF, CR IL-2

Carbo/Eto, Rad 32 Gy

2.PR

9-8-94

PBSC Mell20mg/m 2 , Eto 1800 mg/kg, !2GyTBI

G-CSF, DOD 15-3-95 IL-2

27-9-94

Allo BM

17

5-91

Right femur

17

24

7-92

Os ilium EICESS 16-12-93 Pelvis, Th5-L2, 92/Sur/ nght, Os Rad 55 Gy femura, rib, pubis humen, right tibia clavicula scapula, BM thalus

18

20

8-91

Left pelvis, L1-L5

CESS-91/ 15-3-94 Sur/Rad 44-54.5 Gy

L2-L3, BM CESS-CWS, 2 PR Rez 91

19

18

4-91

Left femur

CESS 86,

Pelvis, spine, EVAIA/ sternum, 44.8 Gy humerus, radius

CESS91/ Sur

11-93

13-5-94

Sur

2.CR

22-1 -94 Allo BM

G-CSF DOC 21-10-94 Mel 120 mg/m 2 , Eto 1800 mg/kg, 12 GyTBI, local hypertherm.,L2-3 Mel 120mg/m 2 , Eto 1800 mg/m 2 , 12 GyTBI

G-CSF DOC 21-2-95

Table 2. Patient data, primary multifocal disease. Patient Age at no initial diagnosis (years)

Date Site of initial diagnosis

20

31

3-88

Chest wall, BM, rib 6 + CESS-86, Sur/Rad 7 left 35 Gy

l.CR

29-11-88 PBSC

21

17

9-87

Pelvis, rib, L4

CESS-86, Rad 44.8 Gy

l.PR

2-5-88

AlloBM Mel 120mg/m2, Eto 60 mg/kg, 12 GyTBI

22

14

9-91

Scull, pelvis, liver, L4-5, S2, BM

CESS-91/Sur/Rad

1 CR

14-4-92

AlloBM Mel 120 mg/m 2 , Eto 1200 mg/m 2 , 12 GyTBI

G-CSF

DOD

27-6-92

23

15

12-90 Pelvis, rib 11 left, LI-3, CESS-91/Rad Th3-11, humerus, BM 31fr39Gy, Sur

l.CR

25-12-91 PBSC

Mel 120 mg/m 2 , Eto 1200 mg/m 2 , Carbo, 12 GyTBI

G-CSF, IL-2

DOD

10-7-92

24

12

2-91

5 rib left, humerus, pelvis, femura, tibiae, L5/S1, sternum

CESS-91/Sur

l.CR

11-2-92

AlloBM Mel 120 mg/m 2 , G-CSF, Eto 1200 mg/m 2 , TNF, Carbo 800 mg/m 2 , INF-y 12 GyTBI

DOD

20-6-93

25

13

4-92

Femur, tibia rib, spine, clavicula, sternum, scullpelvis, BM

CESS-86/Sur/Rad 32 Gy

1 CR

16-2-93

PBSC

Mel 120 mg/m 2 , G-CSF Eto 40 mg/kg, Carbo 800 mg/m 2 , 12 GyTBI

DOD

24-8-93

26

16

12-90 Tigh, lungs, lymph node

SIOP stage IV/ CWS-86/3 x Ifo + ActoD/Sur

l.CR

12-11-91 AlloBM Mell20mg/m 2 , Eto 1200 mg/m 2 , CPL 500 mg/m 2 , 12 GyTBI

DOC

24-11-91

27

18

9-88

CESS 86/Rad/Sur

l.PR

30-5-89

DOD

14-1-91

Os pubis, Os parietale, BM

Therapy of primary disease

Remission Date of status at graft transplant

Mode of Ablative therapy graft

PBSC

Mel 160mg/m2, Eto 60 mg/kg, 12 GyTBI

Mel 120 mg/m 2 , Eto 40 mg kg, CPL 1 g/m2

Cytokines

Outcome Date of 3-98 death or relapse

GM-CSF DOC

8-5-89

CR

G-CSF

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 5, 2016

16

1455 Table 2. (Continued). Date Site of initial diagnosis

Therapy of primary disease

Remission Date of status at graft transplant

Mode of graft

Ablative therapy

28

13

7-89

Hemithorax, ribs, bone marrow

CESS-86/Rad/Sur

1 CR

27-2-90

PBSC

29

10

6-90

Pelvis, lung

CESS-92, Sur

l.CR

18-12-90 PBSC

30

9

7-90

Fibula, bone marrow

CESS-92, Sur

l.CR

28-1-91

31

14

12-90

Femur, bone marrow

CESS-92, Sur

l.CR

32

20

4-88

Pelvis, scuU, femura, tibia, lymph node, BM

33

14

1-93

34

14

35

36

Cytokines

Outcome 3-98

Date of death or relapse

Mel 140mg/m 2 , Eto 60 mg/kg, CPL 1000 mg/m 2 , 12GyTBI

DOD

16-12-9

Mel 120 mg/m 2 , Eto 60 mg/kg, C P L l g / m 2 , 12Gy TBI

DOC

15-1-91

BM

Mel 120 mg/m 2 , Eto 40 mg/kg, CPL 1000 mg/m 2 , 12GyTBI

DOD

8-1-92

9-7-91

PBSC

Mel 140 mg/m 2 , Eto 60 mg/kg, CPL 1500 mg/m 2 , 12GyTBI

Secondary tumor (MDS), DOC

17-6-96

CESS-86, Rad 48 Gy, 1 PR Sur

1-11-88

BM + PBSC

Mel 140 mg/m 2 , Eto 50 mg/kg, CPL 3000 mg/m 2 , 12GyTBI

DOC

25-11-8

Pelvis, L5, clavicula, humerus, femura, BM

EICESS 92

1 CR

7-12-93

PBSC

Mel 120 mg/m 2 , Eto 60 mg/kg, 12GyTBI

G-CSF, 1L-2

Secondary tumor (hposarcoma), C R o f prim. a. sec. tin

6-93

Rib, pelvis, humerus, tibiae, scull, femura

EICESS-92, Sur, Rad 45 Gy

1 PR

25-1-94

PBSC

Mel 120 mg/m 2 , Eto 1800 mg/kg, 12GyTBI

G-CSF, IL-2

DOD

13-5-94

16

11-92

Fibula, tibia, femur, thalus, humerus, Th 7-8, L I , 2,4

EICESS 92, Rad 30-45 Gy

l.CR

1-3-94

PBSC

Mel 120 mg/m 2 , Eto 1800 mg/kg, 12GyTBI

G-CSF, IL2

Secondary tumor (MDS), DOC

17-7-97

19

12-93

Ribs, femura, humeri VAI-PAI (scand. pelvis, spine, scull, prot.) + IFO, tibiae, radii, EICESS-92/Rad sternum, fibulae, 56.8 Gy claviculae

l.PR

16-12-94 PBSC

Mel 120 mg/m 2 , Eto 1800 mg/m 2 , 12GyTBI

G-CSF, IL-2

DOD

24-5-95

G-CSF, EPO, IL-3

Abbreviations: Th - thoracic vertebra; L - lumbal vertebra; C - cervical vertebra; Sur - surgery; IFO - ifosfamide; CPL - carboplatin; CYC Cyclophosphamid; ADR - doxorubicin; VCR - vincristin; ETO - etoposid; ActoD - dactinomycin; CR - complete remission; PR - partial remission; auto - aulologous; allo - allogenic; BM - borne marrow; PBSC - peripheral blood-derived stem cells; Mel - melphalan; TBI - total body irradition; DOD - death of desease, DOC - death of complications; Cispl - cisplatin; Rad - Radiation; GM-CSF - Granulocytemacrophage-colony stimulating factor; G-CSF - Granulocyte-macrophage-colony stimulating factor.

nine CD34+ cells per ul was used as harvest criterion. CDA double lumen Hickman catheter was used for collection and a peripheral line for reinfusion. Collection of allogeneic (n = 10) bone marrow was performed on day of grafting. Involved compartment therapy Local treatment consisted of surgery (n = 8) or radiation (n - 12) or both surgery and radiation (n = 13). In one patient with primary multifocal disease and two patients with relapse there was no local therapy after the last event before myeloablative intensification. In the 21 patients with surgical treatment, resection of sites >200 ml at diagnosis (n = 16) or resection of residual lung metastases (n = 5) was performed preferably after the forth cycle of induction chemo-

therapy. Patients with primary metastases < 200 ml received only radiation therapy. In the 25 locally irradiated patients, fractionated involved compartment irradiation of tumor sites detected either byTc bone scan (n = 15) or by Tc bone scan and MRI (n = 10) was performed preferably parallel to the fifth and sixth cycle of chemotherapy with a target volume dose of 44.8 Gy (22.4 Gy per series, hyperfractionated in 2 x 1.6 Gy per day) adapted for the tolerance dose of the tissue. For the spine, the maximum dose was 40 Gy (i.e., 28 Gy local irradiation + 12 Gy total body irradiation). Myeloablative protocol The interval between the last cytotoxic chemotherapy and the start of the conditioning regimen was recommended to be at least four weeks

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 5, 2016

Patient Age at no. initial diagnosis (years)

1456

Immunotherapy Systemic recombinant DNA-derived Interleukin-2 (IL-2) (Proleukin*; EuroCetus, Frankfurt, Germany) was administered to 11 of 26 (42%)

Table 3. Risk factor distribution, local stage and extent if disease.* Kind of graft

Bone marrow involvement Bone metastases Single 2-5 >5

Lung metastases CR at transplant PR at transplant Age at initial dose (years) > 17 LM and B/ST Pelvic involvement BM and PI

Autologous (n = 26)

Allogeneic All (n = 10)

11 19 6 8 5 7 18 8 8 3 12 6

3 8 3 2 3 3 6 4 4 2 6 1

14 27 9 10 8 10 26 12 12 5 18 7

Abbreviations: PMD - primary multifocal diseases; CR - complete remission; PR - partial remission; LM + B/ST - lung and bone and/or soft tissue metastases; BM and PI - bone marrow and pelvic involvement. ° Distribution at time of diagnosis which led to transplantation. autologous graft recipients, starting preferably on day 56 after transplantation. IL-2 therapy was only performed in informed consent in patients in good condition and no evidence for relapse of disease. Three cycles of IL-2 were given at increasing doses from 6 (day 1), 9 (day 2) to 12 (day 3-5) x 106 IU/m2, with two weeks of rest between each cycle. In case of thrombocytopenia, IL-2 therapy was delayed until the platelet count was greater than 30,000/ul. In 3 of 11 (27%) patients IL-2 therapy was discontinued after 1-2 cycles because of severe complications or relapse.

Remission status before transplant The degree of remission after induction treatment (complete remission, CR; or partial remission, PR) was judged by the treating physician using clinical and radiographic criteria. Twenty-four of thirty-six (67%) patients were transplanted in CR, whereas twelve of thirty-six (33%) were transplanted in PR.

Protocol Outline for Treatment of Advanced Ewlng Tumors Btopty Tc-8one8can Total Body MRI CTLung ToM Body PET

Total Body MM CTLung Stem C«0 Harvest I & D Stem CeD Rncu*

VACA, VA1A, EVAIA

or

IV

IFO/ETO ±

other*

Surgery

I I 1I

V

Myeloablative Consolidation

VI

Interleukln 2

I I

Involved Compartment Irradiation Week

3

12

16

19

27

33

Figure 1. Current protocol outline for treatment of advanced Ewing tumors: six courses of induction chemotherapy with VACA (vincristine, adriamycine, carboplatin, actinomycine D), VAIA (vincristine, adriamycine, ifosfamide, actinomycine D), EVAIA (etoposide, vinenstine, ifosfamide, actinomycine D, adriamycine) or IFO/ETO ± others (ifosfamide, etoposide ± cyclophosphamide or carboplatin or cisplatin), accompanied by involved compartment therapy and staging procedures, followed by myeloablative consolidation according to the Hyper-ME (melphalan, etoposide) protocol and systemic IL-2 treatment. Abbreviations: MRI - magnetic resonance imaging; CT - computed tomography; PET - positron emission tomography.

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 5, 2016

and no longer than six weeks. Only single stem-cell transplantation procedures have been included in this analysis. The myeloablative intensification, termed Hyper-ME, was given to 26 of 36 (72%) patients, consisting of 12 Gy hyperfractionated total body irradiation (TB1) with 2 x 1.5 Gy at day -7 through day -4. In addition, fractionated high-dose melphalan was administered at day -7 through day -4 (30-45 mg/m2/h infusion per day) between the daily TB1 sessions. Etoposide (VP 16) was given at day -3 over four hours (40-60 mg/kg, maximum dose 1800 mg/m2/4 hours infusion per day). In 10 of 36 (28%) patients Carboplatin (3 doses of 300-1000 mg/m2 on days -4 to -2) was added to the regimen (Hyper-ME + Q. TBI was delivered using linear-accelerator or cobalt 60 beams. The dose specification point was the middle of the body at the umbilicus. Dose homogeneity in the entire target volume was in the range of ± 5% to ± 10% for anterior/posterior (ap) TBI The dose in the lungs was 8 Gy, since 1991 10 Gy. The lungs were shielded by ap transmission blocks. The shielded rib cast was boostered with electrons to 12 Gy. The biologic effective dose rate was 0.15-0.3 Gy/min. After three days of rest (i.e., >72 hours after start of VP 16) unmanipulated allogeneic BM was reinfused (Figure 1) with >3 x 108 nucleated cells per kg recipient body weight. Autologous BM or PBSC was reinfused (Figure 1) with J 4 x 104 colony-forming units-granulocyte macrophage (CFU-GM) per kg recipient body weight. Dose of CD34+ cells were >2 x 106 per kg body weight in patients receiving PBSC. After stem-cell transplantation, patients were treated with GM-CSF (n = 9 of 36, 25%) or G-CSF (n = 27 of 36, 75%) for enhancement of myeloid reconstitution starting at day 0. Treatment was reduced to 50% dosage, when absolute neutrophil counts (ANQ stayed above 1000/ul for three consecutive days and then tapered off in two further reduction steps. After allogeneic SCT patients received standard GVHD prophylaxis according to the Seattle protocol consisting of methotrexate (MTX) 15 mg/m2 on day 1 and 10 mg/m2 on days 3, 7 and II and daily cyclosporine A (CSA) starting at day —1. Dosage was adjusted to generate plasma levels of 300-400 ng/ml (monoclonal FPT, Abbott Diagnostics, Wiesbaden, Germany) beyond day 10. CSA administration was maintained up to day 100 post transplant and subsequently tapered according to GVHD severity. Patients younger than 10 years received MTX only.

1457 Statistical analysis

Results Analysis of events

36 patients treated wtth Hyper-ME±C unto 12/94

gntl CR 2/10 DOD 4/10 DOC 4/10 Autokigout graft CH 7/26 DOD 14/26 DOC 5/26 20

Risk factor analysis

00

SO

100

Figure 2. Event-free survival in 36 patients with advanced Ewing tumors (AET) after autologous or allogeneic stem-cell transplantation (SCT) treated according to the Hyper-ME protocol. Abbreviations: CR - complete remission; DOD - death of disease; DOC - death of complication; REL - relapse; auto - autologous; allo - allogeneic.

Advanced Ewing Tumors - Relapse Rate after HDT

Autologous vs. Allogeneic graft 36 patients treated with Hyper-ME±C until 12/94 1.0

\

,8 ,B

\

1 ^

allogeneic 0.42 ± 0.2 autologous 0.40 ± 0.1

2' (in 0

Death of complication (DOC) and death of disease (DOD) for all study patients Eighteen of thirty-six (50%) of transplanted patients suffered relapse or died of disease at a median of 12 months (range 6-20) after the last event before transplantation and 3 months (range 2-12) after transplantation. The median interval between relapse and death was 3.5 months (range 0.6-18). Nine of thirty-six (25%) died of treatment related toxicity at a median of 11 months (range 7-66) after the last event before transplantation and 3 months (range 0.4-59) after transplantation.

40

Month after last event before SCT

20

40

60

80

120

140

Month after last event before SCT Figure 3. Event-free interval in 36 patients with advanced Ewing tumors (AET) after autologous or allogeneic stem-cell transplantation (SCT) treated according to the Hyper-ME protocol. Abbreviations: see Figure 2.

Advanced Ewing Tumors

Age at Initial Diagnosis 36 patients treated with Hyper-ME±C untS 12/94 IJJ »

4

V

FKO.OOS

»9e<1 7 y 0J8±0.1

I

X

We analyzed the following factors that could possibly influence outcomes by comparison of EFS with the logrank and the Breslow test. Number of involved bones as detected by Tc bone scan (1 vs. 2-5 vs. more than 5 bones), degree of remission at time of SCT (CR vs. PR), type of graft (autologous vs. allogeneic graft; autologous BM vs. autologous unselected PBSQ, indication for SCT (relapse vs. primary multifocal disease), bone marrow infiltration (presence vs. absence), bone with concomitant lung disease (presence vs. absence), age at time of initial diagnosis (<17 years vs. > 17 years), pelvic involvement (presence vs. absence), involved compartment radiation (ICR) (no ICR vs. ICR of sites detected by Tc bone scan vs. ICR of sites detected by partial body

100

on

|oe»> 17y 0.00

Agt owr 17 y**ra CR 0/12 DOD 8/12 DOC 4/12 Afl. twlow 17 fan CR 9/24 DOO 10/24 DOC 5/24

Month after last event before SCT

Figure 4 Influence of age at initial diagnosis on EFS. Event-free survival in 36 patients with advanced Ewing tumors (AET) after stemcell transplantation (SCT) according to the Hyper-ME protocol. Abbreviations: see Figure 2.

- MRI vs. ICR of sites detected by total body - MRI) and histopathological diagnosis (ES vs. PNET). Statistical analysis showed, that only the following factor significantly influenced outcome: Age ^ 17 years at initial diagnosis significantly deteriorated outcome (P < 0.05) (Figure 4).

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 5, 2016

EFSfor all study patients As of August 1999 the probability of EFS for the 36 patients treated with Hyper-ME protocol was 0.24 ± 0.07. The median time of EFS for the entire group was 15 months (range 6-139) after the diagnosis that led to transplantation and 6 months (range 0.4-132) after transplantation. Nine of thirty-six (25%) of patients are surviving event-free at a median of 89 months (range 56-139) from diagnosis before transplant and 81 months (range 48-132) after transplantation.

Autologous vs. Allogeneic Graft

Pro bab ihty

The Kaplan-Meier method [31] was used to compute the probability of survival and EFS or event-free interval (EFI) as a function of time. In each case, intervals were calculated from date of transplantation until the date as event occurred or until patient follow-up was censored without event. For calculation of EFS, events were denned as either relapse (REL), death of complication or death of disease. For calculation of EFI, patients who died of a cause other than their underlying disease were considered to be free of event, which means events were defined as either death of disease or relapse. Comparisons between different groups of treatment were made using the log-rank test and the Breslow test. In patients who developed SMN the censored event was death of complication.

Advanced Ewing Tumors - EFS after HDT

1458 (a)

AET - EFS after HDT/Treated as Intended

Systemic IL-2 therapy 27 patterns I m D d wttti Hyper-ME, censored for patients with evsnt before day 120 post transplantation

1.0

•utotogoui (3CyctulL2) CR 5/8 DOC 1/8 DOO 2/8

I auto), with IL2 0 80*0.18 1 autot without IL2 0-22±0 13

I allog. without IL2 0.20±0.13 20

40

60

SO

100

'«

Month after last event before HDT

(wfthout IL2) CR 2/9 DOC 2/9 DOD 5/9

(b) AET - EFS after HDT/lnterrt to treat analysis

Systemic IL-2 Therapy 26 autologous graft recipients treated wtth Hyper-ME p=0.05

Autologous BM vs. autologous PBSC

In the autologous patient group {n = 26) the majority (21 of 26, 81%) were peripheral blood stem-cell (PBSC) 1 autol.wrthout IL-2 0.13±0 09 grafts; 5 of 26 (19%) of patients received autologous bone marrow (BM) grafts (Tables 1 and 2). 20 40 00 80 100 According to type of graft, the EFS was 0.21 ± 0.10 in Month after last event before HDT PBSC recipients and 0.40 ± 0.22 after autologous BM Figure 5. Influence of systemic IL-2 therapy on EFS. (a) Analysis 'treated as intended'; event-free survival (EFS) of 17 patients with AET transplantation. DOC occurred in 4 of 21 (19%) PBSC recipients and after autologous SCT with and without systemic IL-2 therapy compared to EFS of 10 patients after allogeneic SCT. Patients with event in 1 of 5 (20%) of the BM recipients. before day 120 posttransplantation have been censored, (b) Intent to Relapse or DOD was registered in 12 of 21 (57%) treat analysis; event-free survival of 26 patients with AET after autolunselected PBSC transplants and in 2 of 5 (40%) BM ogous SCT according to the Hyper-ME protocol. transplants. None of these differences reached statistical Abbreviations: see Figure 2. significance. Autologous vs. allogeneic grafts

IL-2 therapy

Ten of thirty-six (28%) patients underwent allogeneic BMT with HLA-matched sibling donors, twenty-six of thirty-six (72%) patients received autologous grafts. In the autologous patient cohort, 18 of 26 (69%) patients were transplanted in CR, 8 of 26 (31%) in PR. Of the 10 allogeneic transplant recipients, 6 of 10 (60%) were transplanted in CR and 4 of 10 (40%) in PR. The probability of EFS was 0.25 ± 0.09 after autologous transplantation and 0.20 ± 0.13 after allogeneic transplantation; even so the difference did not reach statistic significance (Figure 2). In fact the difference is so minimal that the number of patients required to show a statistic significance exceeds 150 [16]. Because of the low frequency of metastasized Ewing tumors it is very unlikely to reach this number. The median time of EFS for allogeneic transplantation was 11 months (range 7-129, confidence interval: 9-13), for autologous transplantation 17 months (range 6-139, confidence interval: 11-22) after the last event before HDT. Risk factor distribution was balanced in both patient groups (Table 3). Relapse and death of disease occurred in 14 of 36 (39%) of the autologous and in 4 of 10 (40%) of the allogeneic graft recipients. There was no difference in

Eight of twenty-six (31%) autologous graft recipients were treated with three cycles of systemic IL-2. The probability of EFS for this patient cohort was 0.60 ± 0.18 in comparison to 0.22 ± 0.14 in the 9 of 26 (35%) patients without IL-2 therapy. The latter were censored for patients with event before day 120 post transplant to avoid selection bias. The difference between both patient groups did not reach significance (Figure 5a and b). Three of thirty-six (8%) patients received only one or two cycles of IL-2 therapy. These patients were not considered for statistical analysis. Five of eight (63%) IL-2 treated patients survived event-free with a median time of EFS of seventy-one months (range 56-120 months). Two of eight (25%) suffered DOD. Of note, only one death of complication occurred in the subgroup of IL-2 treated patients. This patient developed myelodysplastic syndrome (Tables 1 and 2, patient no. 35). Toxicity and SMN In the 10 patients with allogeneic grafts, the incidence and extent of transplant related toxicity was more intense than in autologous graft recipients. Patient no. 26

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 5, 2016

I auto), wtth IL:2 0.44±0 1S

•utoi. wtth 1-3 CyctMlL2 c n 5/11 DOC 1/11 DOD 5/11 (utcri.arrthaut IL2 cn 2/15 DOC 4/13 DOD 9/15

relapse rate between allogeneic and autologous transplants as evidenced by the event-free interval (Figure 3). The type of marrow graft tended to influence the transplant related mortality (DOC, death of complication). Whereas 4 of 10 (40%) allogeneic graft recipients died of causes directly related to transplant, 5 of 26 (19%) autologous graft recipients died of procedure related toxicity (NS, P = 0.08, Fisher's exact test). Acute GVHD developed in four of eight, chronic GVHD (> day +100) in zero of four evaluable allogeneic graft recipients. Two allogeneic graft recipients were not evaluable for acute GVHD as they died before engraftment, six patients were not evaluable for chronic GVHD as they died before day 100. In two of four allogeneic patients who died of DOC, GVHD was present at time of death (Tables 1 and 2, patient nos 19 and 14).

1459

Discussion The prognosis of patients with localized Ewing tumor (ET) has been improved by multimodal strategy including surgery, radiation and chemotherapy. In contrast, the prognosis of patients with multifocal primary bone disease as well as with early (less than two years after diagnosis) or multiple relapses ET remains poor with conventional therapy. Patients with primary lung metastases as well as patients with late (more than two years after diagnosis) relapses have a better prognosis. Despite the curative potential of intensified HDTwith the Hyper-ME protocol [1], post transplant relapse remains the major problem in both autologous and allogeneic graft recipients. Thirty-six patients with primary multifocal Ewing's tumor or relapse were treated according to Hyper-ME protocol between November 1986 and December 1994. Nine of thirty-six of patients are still alive after a median follow-up of eighty-nine months from diagnosis and eighty-one months from transplantation. Nine of thirty-six patients died of toxicity and eighteen of thirty-six patients died of disease. This longterm follow-up report shows that about one fourth of patients with AET may achieve long-term survival. The occurrence of late relapses clearly affects the results. It

is obvious from that data, that long-term follow-up is imperative for evaluation of the therapeutic value of the different treatment strategies. Reports from other groups demonstrate that about one third of patients with ET may achieve intermediate term remission after transplantation. Ladenstein et al. [5] presented a retrospective analysis from 21 European transplant centers reported to the EBMT registry between 1982 and 1992 with an EFS rate of 27% at 5 years. As more than 20 different megatherapy strategies are considered in this analysis, it is difficult to interpret the results. Overall, they suggest a positive impact on survival by consolidation with myeloablative chemotherapy followed by stem cell transplantation in ET patients. In the review of EICESS data by Paulussen et al. [6] probability of EFS for patients with primary multifocal disease was 0.27 after four years. Megatherapy did not significantly influence the four year probability of EFS in patients with metastatic disease (0.23 vs. 0.28). The group with the worst prognosis was identified as patients with metastases to both the pulmonary and the skeletal system. In this subgroup of patients megatherapy in combination with lung irradiation significantly improved survival rates from 0% to 27%. Concerning the addition of TBI neither Ladenstein nor Paulussen reported significantly improved EFS in therapeutic regimens containing TBI in comparison to non-TBI regimens [5, 6]. There is still controversy about the substances used in the conditioning regimen. Atra et al. [2] reported encouraging results with a probability of EFS > 60% at two years after high-dose busulfan-melphalan and autologous SCT in AET patients. Unfortunately only 11 cases are enrolled in this analysis and 10 of 11 patients showed metastases only to the lung, which is reported to have a better prognosis than bone or bone marrow involvement [6]. Thus, the patient group is not representative for high-risk AET patients and long-term followup remains to be published. There are several groups that evaluated the efficiency of conventional chemotherapy without megatherapy. Sandoval et al. [35] from St. Jude's hospital report an overall survival of 35% at 4 years in 43 patients with primary metastases at diagnosis. Eleven of forty-three patients had multiple metastases. In this subgroup 4 of 11 survivors were reported, but the observation time was not specifically analyzed. Cangir et al. [36] observed an EFS of 30% at five years at the M.D. Anderson Cancer Center. However, this analysis did not differentiate between patients with lung disease and patients with multifocal bone disease. It is of interest to see the long-term follow-up data from both groups in order to evaluate the benefit of megatherapy in AET patients. Wessalowski et al. [17] analyzed 48 patients with Ewing's sarcoma and primary metastases. The probability of EFS was 18% after five years for all patients and 37% for patients with primary lung metastases only. The prognosis for patients with bone metastases was extremely poor. History of childhood cancer is associated with a 10-

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 5, 2016

had no engraftment and died of capillary leakage syndrome with renal failure and septic coagulopathy with pulmonary bleeding (day +12). Patient no. 18 developed systemic candidiasis refractory to treatment and died of massive gastrointestinal bleeding (day +24). Patient no. 19 died of progressive encephalitis with persistent hepatic GVHD (day +91). Patient no. 14 died of aspergillosis and multiorgan failure with persistent cutaneous graft- versus -host disease (GVHD) grade 2 (day +310). Amongst 26 patients with autologous grafts, patient no. 20 died after thrombocytopoietic graft failure (> day + 100), staphyloccocal sepsis, interstitial pneumonitis and respiratory failure precipitated by iatrogenic pulmonary bleeding (day +160). Patient no. 29 died of respiratory failure after pleural effusion with underlying capillary leakage syndrome (day +28). Patient no. 32 died of multiorgan failure after aspergillosis (day +24). Two further patients (nos 31 and 35) developed a secondary myelodysplastic syndrome. A third patient (no. 33) developed a liposarcoma of the right pelvis, which was located in the irradiation field (54 Gy) of the primary PNET. Patient no. 31 died during reinduction chemotherapy because of sepsis and multiorgan failure (day +1805). Patient (no. 35) died of sepsis and multiorgan failure on day +93 after second transplantation utilizing cord blood. Patient no. 33 underwent chemotherapy and surgical treatment for SMN and is currently (01/00) in remission for primary and secondary neoplasm. This patient is censored as event free in the Kaplan-Meier analysis since he could neither be censored as relapse or death of complication adhering to the criteria defined in 2.5.

1460 these factors were significantly skewed between both groups. It was apparent from many studies, that a GVT effect exists for various hematological malignancies like leukemia [7, 13], myelodysplastic syndrome [32] and myeloma [20, 21, 24, 26, 33] With regard to solid tumors to date there is no definitive proof of a clinical benefit of a GVT effect and even the existence of a GVT effect is still a matter of debate. Only a few case reports seem to show indirect evidence of a GVT effect for osteosarcoma [23], breast cancer [22, 25] and lymphoma [15-18]. The graft-verms-leukemia (GVL) effect in leukemia is evidenced by reduced leukemic relapse rates after allogeneic transplantation in comparison with autologous transplantation and is closely associated with the occurrence of GVHD [11, 12]. In our study, four of eight evaluable patients developed GVHD. Thus, lack of GVHD may also not account for lack of GVT in advanced ET. On the other hand, lack of demonstration of GVT in AET may just be due to the low number of evaluable patients with allogeneic grafts, because of higher transplant related mortality. GVHD can affect quality of life. In this study the two survivors of allogeneic transplant had no chronic GVHD and both had a Karnofsky Index of 100. Gene marking studies have demonstrated that transplanted tumor cells are found at relapse sites [14, 34] which led to the assumption that autologous transplantation of malignant cells might contribute to relapse. Therefore, it was hypothesized that allogeneic transplant procedures without tumor cell contamination might improve outcome by reduced relapse rate after PBSCT. In this study we cannot confirm this hypothesis, as relapse rate is identical in both patient groups. In our study, the only factor that significantly influenced outcome was age above 17 years. This finding is in accordance with the results from Stewart [37] who reports poor prognosis after high-dose melphalan ± TB1 in 13 patients over 16 years. Only 3 of 13 remained progression free at 25-108 months follow-up. Paulussen et al. [6] report that patients with 15 years of age or younger have substantially higher survival rates. Cangir et al. [36] found higher EFS rates in patients 10 years of age or younger. These findings might be explained by a different biology in tumors of elderly patients, as for example a lower replication rate of tumor cells. ET cells are sensitive to IL-2 induced effector cell lysis in vitro [38, 39]. In addition, ET are characterized by expression of oncofusion gene products [40] Chimeric oncoproteins can bind to HLA [41] and may induce an anti-tumor immune response. There is some evidence for a role of IL-2 in immunotherapy of hematological malignancies including lymphoma [42, 43]. In neuroblastoma, eradication of hepatic metastases was demonstrated with antibody-targeted IL-2 therapy [44]. In addition, there is evidence from preclinical in vivo studies that immunologic therapy has a potential role in the treatment of fusion gene expressing pediatric sarcoma [41, 45, 46]. In 11 of 26 (42%) of

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 5, 2016

20 times higher lifetime risk of second cancer compared with age matched controls [51]. Irradiation seems to be the therapeutic modality with the highest risk of SMN [52]. After radiotherapy of Hodgkin's disease SMN occurs in two-third of cases in the radiation fields [53]. Furthermore previous therapy with alkylating agents and epipodophylotoxins appears to have a high potential of SMN [54]. In this study 3 of 36 (8.3%) patients developed SMN. Two patients died of a myelodysplastic syndrome (MDS) possible caused by alkylating agents and one patient developed a liposarcoma in the pelvis. This liposarcoma was situated on the site of the irradiation field of the primary PNET in this region. The patient was currently in remission for primary and secondary neoplasm after chemotherapy for SMN. The high risk of SMN in Ewing's Sarcoma is well known. Horowitz et al. [49] reported a 679 times higher risk of carcinogenesis in the Ewing than in the normal population. Dunst et al. [47] published an incidence of second malignancies in patients treated in the CESS 81 and CESS 86 studies of 1.2% (8 of 674), acute myelogenic leukemia's (n = 4), MDS {n = 1) and sarcomas (n - 3). The cumulative risk of SMN was 0.7% after 5 years, 2.9% after 10 years and 4.7% after 15 years. All patients with secondary sarcomas had received radiotherapy. SMN caused less than 1% of all deaths within the first 10 years. Six of six hundred thirty-one (0.9%) patients registered in the EICESS 92 study have suffered SMN, two of these after myeloablative chemotherapy [48]. Long-term follow-up remains to be published. Higher incidence ranging from 9% to 17% with a cumulative risk after 10 years of 35% have been reported with heavy radiation treatment [55, 56]. Thus the higher rate of SMN in our study may be related to extensive multifocal bone disease with use of high doses of etoposide and extended radiation fields. The results of this study do not show that for patients with Advanced Ewing Tumor (AET) consolidated with Hyper-ME protocol the EFS is improved in allogeneic graft compared to autologous graft recipients. The likelihood of long-term survival following allogeneic transplantation for AET is influenced by a doubled although not significantly increased probability of treatment related mortality in allogeneic transplants. In addition, according to our results there is no difference in relapse rate between both patient groups and thus no evidence of an existing graft-versus-tumor (GVT) effect in AET patients. To our knowledge this is the first comparison between autologous and allogeneic transplant procedures in AET. To exclude any bias between the two analyzed patient groups, all factors thought to be associated with increased risk of relapse have been analyzed in both groups. These factors comprise number of involved bones, remission status at time of grafting, type of graft, histopathological subtype, primary multifocal disease vs. relapse, age at time of initial diagnosis, site of metastases (bone marrow involvement, pelvic involvement, bone with concomitant lung disease), systemic IL-2 therapy and involved compartment radiation. None of

1461

The results as well from individual centers as from cooperative groups have consistently reported poor prognosis for AET patients. Unfortunately there is still a wide diversification in the treatment strategies for AET patients and most reports suffer from small numbers of patients enrolled. Future therapeutic protocols in AET must overcome the inhomogeneity of AET treatment with a standardized treatment protocols in order to compile sufficient patient data for randomized studies with risk stratification at diagnosis.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

Acknowledgements This work was supported by grants from the Dr Mildred Scheel Stiftung fur Krebsforschung der Deutschen Krebshilfe (Wll/94/Bu2) and the Elterninitiative Kinderkrebsklinik e.V. The authors thank M. Schmitz for her help in data documentation. The authors are indebted to the medical and the nursing staff of the participating institutions for the provision of excellent patient care, which has been crucial for the achievement of the results reported here.

16.

17.

18.

19.

References 1. Burdach S, Jurgens H, Peters C et al. Myeloablative radiochemotherapy and hematopoietic stem-cell rescue in poor prognosis Ewing sarcoma. J Clin Oncol 1993; 11: 1482-8. 2. Atra A, Whelan JS, Calvagna V et al. High-dose busulphan/ melphalan with autologous stem-cell rescue in Ewing's sarcoma. Bone Marrow Transplant 1997; 20: 843-6. 3. Horowitz ME, Kinsella TJ, Wexler LH et al. Total-body irradiation and autologous bone marrow transplant in the treatment of high-risk Ewing's sarcoma and rhabdomyosarcoma. J Clin Oncol 1993; 11: 1911-8. 4. Hartmann O, Oberlin O, Beaujean F et al. Place de la chimiotherapie a hautes doses suivie d'autogreffe medullaire dans le

20.

21.

22.

23.

traitement des sarcomes d'Ewing metastatiques de l'enfant. Bull Cancer 1990; 77: 181-7. Ladenstein R, Lasset C, Pinkerton R et al. Impact of megatherapy in children with high-risk Ewing's tumours in complete remission: A report from the EBMT Solid Tumour Registry. Bone Marrow Transplant 1995; 15: 697-705. Paulussen M, Ahrens S, Burdach S et al. Primary metastatic (stage IV) Ewing tumor: Survival analysis of 171 patients from the EICESS studies. Ann Oncol 1998; 9: 275-81. Marmont AM. The graft-vers
Downloaded from http://annonc.oxfordjournals.org/ by guest on June 5, 2016

our autologous graft recipients 1-3 cycles of systemic IL-2 was administered after SCT. This additional immunological therapy could have hampered the detection of a GVT effect in allogeneic graft recipients. Of note, the 9 of 26 (35%) autologous patients without IL-2 therapy have the same dismal prognosis as allogeneic graft recipients (EFS 0.22 ± 0 . 1 3 vs. 0.20 + 0.13). Consequently improved results of autologous transplantation can be generated with immunoaugmentation by systemic IL-2 treatment. Unfortunately systemic IL-2 therapy is complicated by marked toxicity. Remarkably, in this study only 1 of 17 patients died of a complication (MDS), which was most likely not related to IL-2. Overall, the best treatment results were obtained in a subgroup of patients treated with autologous grafts and subsequent systemic IL-2 application. As we conclude from our recent update analysis, the occurrence of late relapses may alter the outcome considerably. Hence, long-term follow-up is to be awaited to evaluate the therapeutic efficiency of immunotherapy after stem-cell transplantation.

1462

24. 25. 26. 27.

28.

29.

31. 32.

33. 34.

35.

36. 37.

38.

39.

40.

41.

42. 43.

44.

45. 46.

47. 48. 49.

50. 51. 52. 53. 54. 55. 56.

of variable fusion proteins in the Ewing family of tumours. Embo J 1993; 12: 4481-7. Goletz TJ, Zhan S, Mackall CD et al. Tumor translocation fusion protein transcrition factors as targets for cancer vaccines to induce cytotoxic T cells against pediatric sarcoma. Comm. Abstract, Key Stone Symposia on Molecular and Cellular Biology, February 1-7, 1997. 1997; C2: Cellular Immunology and the Immunotherapy of Cancer III: 3. Toren A, Ackerstein A, Slavin S, Nagler A. Role of interleukin-2 in human hematological malignancies. Med Oncol 1995; 12: 177-86. Slavin S, Nagler A. Cytokine-mediated immunotherapy following autologous bone marrow transplantation in lymphoma and evidence of Interleukin-2-induced immunomodulation in allogeneic transplants. Cancer J Sci Am 1997; 3 (Suppl 1): S59. Pancook JD, Becker JC, Gillies SD, Reisfeld RA. Eradication of established hepatic human neuroblastoma metastases in mice with severe combined immunodeficiency by antibody-targeted interleukin-2. Cancer Immunol Immunother 1996; 42: 88-92. Dilloo D, Bacon K, Holden W et al. Combined chemokine and cytokine gene transfer enhances antitumor immunity. Nature Med 1996; 2: 1090-5. Dilloo D, Laws HJ, Hanenberg H et al. Induction of two distinct NK-cell populations, activated T cells and antineoplastic cytokines, by IL-2 therapy in children with solid tumors. Exp Hematol 1994; 22: 1081-8. Dunst J, Ahrens S, Paulussen M et al Second malignancies after treatment for Ewing's sarcoma: A report of the CESS-studies. Int J Radiat Oncol Biol Phys 1998; 42 (2): 379-84. Paulussen M, Ahrens S, Braun-Munzinger G et al. EICESS 92 (European Intergroup Cooperative Ewing's Sarcoma Study) preliminary results. KJin Padiatr. 1999; 211 (4): 276-83. Horowitz ME, Malawer MM, Shiao YW et al. Ewing sarcomafamily of tumors' Ewing's sarcoma of bone and soft tissue and the peripheral primitive neuroectodermal tumors. In Pizzo PA, Poplack DG (eds): Principles and Practice of Pediatric Oncology, third edition. Philadelphia, PA: Lippincott-Raven 1997; 831-61. Schmidt D, Harms D, Burdach S. Malignant peripheral neuroectodermal tumours of childhood and adolescence. Virchows Arch A Pathol Anat Histopathol 1985; 406 (3): 351-65. Mike V, Meadows AT, D'Angio GJ. Incidence of second malignant neoplasms in children: Results of an international study. Lancet 1982; 2 (8311): 1326-31. Tucker MA, D'Angio GJ, Boice JD Jr et al. Bone sarcomas linked to radiotherapy and chemotherapy in children. N Engl J Med 1987; 317 (10): 588-93. Tucker MA, Coleman CN, Cox RS et al. Risk of second cancers after treatment for Hodgkin's disease. N Engl J Med 1988; 318 (2): 76-81. Leone G, Mele L, Pulsoni A et al. The incidence of secondary leukemias. Haematologica 1999; 84 (10): 937^t5. Ehara S, Kattapuram SV, Egglin TK, Ewing's sarcoma. Radiographic pattern of healing and bony complications in patients with long-term survival. Cancer 1991; 68 (7): 1531-5. Strong LC, Herson J, Osborne BM et al. Risk of radiation-related subsequent malignant tumors in survivors of Ewing's sarcoma. J Natl Cancer Inst 1979; 62 (6): 1401-6.

Received 25 April 2000; accepted 9 August 2000. Correspondence to:

S. Burdach, MD, PhD Division of Pediatric Hematology/Oncology Children's Hospital Medical Center Martin Luther University Halle-Wittenberg 06097 Halle Germany E-mail: [email protected]

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 5, 2016

30.

marrow transfusion suppresses development of lung metastases in osteogenic sarcoma patients after radical surgery. Int J Cancer 1993; 54: 907-10. Tricot G, Vesole DH, Jagannath S et al. Graft-versus-myeloma effect: Proof of principle. Blood 1996; 87: 1196-8. Morecki S, Moshel Y, Gelfend Y et al. Induction of graft vs. tumor effect in a murine model of mammary adenocarcinoma. Int J Cancer 1997; 71: 59-63. Verdonck LF, Lokhorst HM, Dekker AW, Nieuwenhuis HK, Petersen EJ. Graft-vermr-myeloma effect in two cases (see comments). Lancet 1996; 347: 800-1. Milpied N, Fielding AK, Pearce RM et al. Allogeneic bone marrow transplant is not better than autologous transplant for patients with relapsed Hodgkin's disease. European Group for Blood and Bone Marrow Transplantation. J Clin Oncol 1996; 14: 1291-6. Ladenstein R, Lasset C, Hartmann O et al. Comparison of auto versus allografting as consolidation of primary treatments in advanced neuroblastoma over one year of age at diagnosis: Report from the European Group for Bone Marrow Transplantation. Bone Marrow Transplant 1994; 14: 31-46. Schmidt D, Herrmann C, Juergens H et al. Malignant peripheral neuroectodermal tumor and its necessary distinction from Ewing's sarcoma. Cancer 1991; 68: 2251-9. Ahrens S, Hoffmann C, Jabar S et al. Evaluation of prognostic factors in a tumor volume-adapted treatment strategy for localized Ewing sarcoma of bone: The CESS 86 experience. Med Pediatr Oncol 1999; 32: 186-95. Kaplan E, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457-81. Locatelli F, Zecca M, Niemeyer C et al. Role of allogeneic bone marrow transplantation for the treatment of myelodysplastic syndromes in childhood. The European Working Group on Childhood Myelodysplastic Syndrome (EWOG-MDS) and the Austria-Germany-Italy (AGI) Bone Marrow Transplantation Registry. Bone Marrow Transplant 1996; 2: 63-8. Gahrton G, Tura S, Ljungman P et al. Allogeneic bone marrow transplantation in multiple myeloma. European Group for Bone Marrow Transplantation. N Engl J Med 1991; 325: 1267-73. Rill DR, Santana VM, Roberts WM et al. Direct demonstration that autologous bone marrow transplantation for solid tumors can return a multiplicity of tumorigenic cells. Blood 1994; 84: 380-3. Sandoval C, Meyer WH, Parham MD et al. Outcome in 43 children presenting with metastatic Ewing sarcoma: The St. Jude Children's Research Hospital Experience, 1962-1992. Med Pediatr Oncol 1996; 26: 180-5. Cangir A, Vietti TJ, Gehan EA et al. Ewing's sarcoma metastatic at diagnosis. Results and comparisons of two intergroup Ewing's sarcoma studies. Cancer 1990; 66: 887-93. Stewart DA, Gyonyor E, Paterson AHG et al. High-dose melphalan ± total body irradiation and autologous hematopoietic stem-cell rescue for adult patients with Ewing's sarcoma or peripheral neuroectodermal tumor. Bone Marrow Transplant 1996; 18: 315-8. Engel BC, Laws HJ, Buttlies B et al. Induction of a CD3+/ CD56+ lymphocyte population following gene-therapy with transgenic IL-2 secreting fibroblasts in a child with peripheral neuroectodermal malignancy. Med Pediatr Oncol 1998; 31: 56-60. Laws HJ, Dilloo D, Hanenberg H et al. Immunstimulation durch Interleukin-2 nach autologer Stammzelltransplantation bei Kindern und Jugendlichen mit soliden Tumoren (IL-2 therapy post autologous stem-cell transplantation stimulates the production of tumor cytotoxic cytokines in children and adolescents with solid tumors). KJin Padiatr 1993; 205: 257-64. Zucman J, Melot T, Desmaze C et al. Combinatorial generation