JBJS | The Histological Response to Chemotherapy as a Predictor of the Oncological Outcome of Operative Treatment of Ewing Sarcoma*

The Journal of Bone & Joint Surgery, Volume 80, Issue 7

Articles | July 01, 1998

The Histological Response to Chemotherapy as a Predictor of the Oncological Outcome of Operative Treatment of Ewing Sarcoma*

JAY S. WUNDER, M.D.†; GABE PAULIAN, B.SC.‡; ANDREW G. HUVOS, M.D.‡; GLENN HELLER, PH.D.‡; PAUL A. MEYERS, M.D.‡; JOHN H. HEALEY, M.D.‡, NEW YORK, N.Y.

View Disclosures and Other Information

Investigation performed at the Departments of Surgery, Pathology, Pediatrics, and Epidemiology and Biostatistics, Memorial Sloan-Kettering Cancer Center, Affiliated with Cornell University Medical College, New York City

The Journal of Bone & Joint Surgery. 1998; 80:1020-33

5 Recommendations (Recommend) | 3 Comments | Saved by 3 Users Save Case

text A A A

Abstract

Seventy-four patients who had a Ewing sarcoma of bone were managed with preoperative and postoperative chemotherapy and operative resection, with or without postoperative irradiation. The primary objectives of the study were to determine the histological response to preoperative chemotherapy in terms of the percentage of tumor necrosis and to assess the relationship between the histological response and the oncological outcome.The minimum duration of follow-up of the surviving patients who were continuously free of disease was five years. Sections of each operative specimen were examined, and the histological response to chemotherapy was graded semiquantitatively. Grade I indicated necrosis of 50 per cent of the tumor or less; grade II, necrosis of more than 50 per cent but less than 90 per cent; grade III, necrosis of 90 to 99 per cent; and grade IV, necrosis of 100 per cent of the tumor. Of the seventy-four tumors, forty-four (59 per cent) were exquisitely sensitive to chemotherapy and had complete (grade-IV) or nearly complete (grade-III) necrosis. In contrast, fourteen tumors (19 per cent) had little or no response to chemotherapy (grade I) and sixteen (22 per cent) had a moderate degree of necrosis (grade II).The histological response to preoperative chemotherapy (p = 0.0001), followed by the size of the tumor (p = 0.001), were the most important predictors of event-free survival. At five years, the rate of event-free survival was zero of fourteen patients who had had a grade-I response, six of sixteen who had had a grade-II response, and thirty-seven (84 per cent) of forty-four who had had a grade-III or IV response. The risk of local recurrence was most strongly associated with the operative margins; there were only four local recurrences (6 per cent) after sixty-seven resections with negative margins. Local recurrence may also have been influenced by the histological response and the use of local radiation. There were no local recurrences after operative treatment of six tumors that had been associated with pathological fracture.The histological response to preoperative chemotherapy and the size of the primary tumor are the most important clinical predictors of the outcome of operative treatment of non-metastatic Ewing sarcoma. These indicators should be used to identify patients who are at high risk for metastasis as such patients may be candidates for more intensive or novel therapies.

Figures in this Article

Introduction

Introduction | Materials and Methods | Results | Discussion | References

The treatment of Ewing sarcoma of bone has evolved dramatically. The rates of long-term success after irradiation of the site of the primary tumor as the only treatment have ranged from 5 to 10 per cent because of the development of metastases⁴⁰. The use of intensive multiagent chemotherapy has been the most important advance in the treatment of Ewing sarcoma; in recent reports, the rates of disease-free survival have ranged from 50 to 60 per cent for patients who were managed in this manner^{6,14,15,37,45}. Originally used as adjuvant therapy, chemotherapy is currently indispensable as the first-line treatment because of its beneficial local as well as systemic effects^{26,39,42,44,53}.

Definitive local therapy for Ewing sarcoma is recognized as necessary, and until recently such treatment has involved high doses of radiation^{1,6,14,15,22,27,37,43}. However, with improved survival, problems associated with radiation therapy, including high rates of local recurrence^{1,14,15,22,27}, the risk of radiation-induced sarcoma^50,52, and tissue fibrosis affecting function^25,33, have become evident. Because of these problems, operative treatment has been investigated as an alternative that could allow a reduction in the dose of, or obviate the need for, local radiation therapy. Several non-randomized studies have suggested that the best way to achieve local control of the disease is through operative resection of the primary tumor, with or without low doses of radiation, and preoperative and postoperative chemotherapy^{2,14,15,26,44,47,54}.

The inclusion of resection in the protocol for the treatment of Ewing sarcoma provides an opportunity to examine the effect of preoperative chemotherapy on the primary tumor. In patients who have osteosarcoma, a tumor that usually is treated with preoperative chemotherapy, chemotherapy-induced tumor necrosis facilitates operative resection and the extent of the necrosis is a major prognostic indicator of both local and systemic recurrence^3,4,35,41,46. Jürgens et al.²⁶ as well as Picci et al.⁴² suggested that the histological response to preoperative chemotherapy may also be an important indicator of the systemic outcome for patients who have Ewing sarcoma. However, both of those studies had a relatively short duration of follow-up, patients who had metastatic disease were excluded from both studies, and only the study by Jürgens et al.²⁶ included patients who had a primary pelvic tumor.

Intensive multiagent chemotherapy remains the cornerstone of treatment for Ewing sarcoma. Over the years at our institution, we have increasingly performed operative resection with or without low doses of radiation, rather than using high doses of radiation alone, for local control of the tumor. In this report, we describe our experience with this type of treatment for Ewing sarcoma, with emphasis on the importance of the histological response to preoperative chemotherapy as a predictor of event-free survival and possibly of local recurrence.

*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.

†University Musculoskeletal Oncology Unit, Mount Sinai Hospital, 600 University Avenue, Suite 476, Toronto, Ontario M5G 1X5, Canada. E-mail address: wunder@mshri.on.ca.

‡Departments of Pediatrics (G. P. and P. A. M.), Pathology (A. G. H.), Epidemiology and Biostatistics (G. H.), and Surgery (J. H. H.), Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, N.Y. 10021. Please address requests for reprints to Dr. Healey. E-mail address for Dr. Healey: healeyj@mskcc.org.

*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.

†University Musculoskeletal Oncology Unit, Mount Sinai Hospital, 600 University Avenue, Suite 476, Toronto, Ontario M5G 1X5, Canada. E-mail address: wunder@mshri.on.ca.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1 Drawing showing the locations of the seventy-four Ewing sarcomas.

View Large | Download Slide(.ppt)
Add to My JBJS

Figs. 2-A through 2-D: Histological responses of Ewing sarcomas to preoperative chemotherapy. Figs. 2-A and 2-B: No necrosis (grade I) (x 40 and x 100, respectively).

View Large | Download Slide(.ppt)
Add to My JBJS

Figs. 2-A and 2-B: No necrosis (grade I) (x 40 and x 100, respectively).

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-C Almost complete necrosis (grade III) (x 40).

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-D Complete necrosis (grade IV) (x 40).

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 3 Graph showing the relationship between event-free survival and the histological response of the tumor to preoperative chemotherapy (p = 0.0001), with 95 per cent confidence intervals. Curve I = no necrosis (grade I), curve II = moderate necrosis (grade II), curve III = almost complete necrosis (grade III), and curve IV = complete necrosis (grade IV). The tick marks indicate the times of the latest follow-up evaluations. Censored indicates event-free survival.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 4 Graph showing the relationship between event-free survival and the number of cycles of preoperative chemotherapy (p = 0.08), with 95 per cent confidence intervals. The tick marks indicate the times of the latest follow-up evaluations. Censored indicates event-free survival.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 5 Graph showing the relationship between event-free survival and the size of the tumor (p = 0.001), with 95 per cent confidence intervals, for sixty-three patients. The tick marks indicate the times of the latest follow-up evaluations. Censored indicates event-free survival.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 6 Graph showing the relationship between event-free survival and the location of the primary tumor (p = 0.1), with 95 per cent confidence intervals. The tick marks indicate the times of the latest follow-up evaluations. Censored indicates event-free survival.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 7 Graph showing the relationship between event-free survival and the presence or absence of metastases at the time of the diagnosis, with 95 per cent confidence intervals. The tick marks indicate the times of the latest follow-up evaluations. Censored indicates event-free survival.

View Larger

TABLE I CHEMOTHERAPY PROTOCOLS

*The total dose of each agent that was administered over five complete cycles for a duration of forty-five weeks.

Agent	Dose* (mg/m² of skin)
T9 Protocol	T11 Protocol
Actinomycin D	7.5	6.0
Doxorubicin	480	480
Cyclophosphamide	19.5	18.0
Vincristine	37.5	37.5
Methotrexate	360	360
Bleomycin	150	150

Add to My JBJS

TABLE I CHEMOTHERAPY PROTOCOLS

*The total dose of each agent that was administered over five complete cycles for a duration of forty-five weeks.

Agent	Dose* (mg/m² of skin)
T9 Protocol	T11 Protocol
Actinomycin D	7.5	6.0
Doxorubicin	480	480
Cyclophosphamide	19.5	18.0
Vincristine	37.5	37.5
Methotrexate	360	360
Bleomycin	150	150

View Larger

TABLE II RELATIONSHIP BETWEEN THE HISTOLOGICAL RESPONSE AND THE NUMBER OF CYCLES OF PREOPERATIVE CHEMOTHERAPY*

*At five years.

Grade of Histological Response	1 Cycle		=2 Cycles
Total No. of Patients	No. of Patients Surviving Event-Free*	Total No. of Patients	No. of Patients Surviving Event-Free*
I	7	0	7	0
II	11	5	5	1
III or IV	24	23	20	14

Add to My JBJS

TABLE II RELATIONSHIP BETWEEN THE HISTOLOGICAL RESPONSE AND THE NUMBER OF CYCLES OF PREOPERATIVE CHEMOTHERAPY*

*At five years.

Grade of Histological Response	1 Cycle		=2 Cycles
Total No. of Patients	No. of Patients Surviving Event-Free*	Total No. of Patients	No. of Patients Surviving Event-Free*
I	7	0	7	0
II	11	5	5	1
III or IV	24	23	20	14

View Larger

TABLE III PROBABILITY OF SYSTEMIC RECURRENCE ON THE BASIS OF THE HISTOLOGICAL RESPONSE AND THE SIZE OF THE TUMOR

*The combination of a good histological response and a small size of the tumor was associated with the lowest risk of systemic recurrence (one [4 per cent] of twenty-five patients); these patients served as the reference group. The relative risk for each combination was determined on the basis of this combination.

Grade of Histological Response	Relative Risk
=8-cm Tumor	>8-cm Tumor
I	25	25
II	8.3	19.5
III or IV	1.0*	4.5

Add to My JBJS

TABLE III PROBABILITY OF SYSTEMIC RECURRENCE ON THE BASIS OF THE HISTOLOGICAL RESPONSE AND THE SIZE OF THE TUMOR

Grade of Histological Response	Relative Risk
=8-cm Tumor	>8-cm Tumor
I	25	25
II	8.3	19.5
III or IV	1.0*	4.5

View Larger

TABLE IV DATA ON THE SIX PATIENTS WHO HAD A PATHOLOGICAL FRACTURE

Case	Site of Fracture	Time of Fracture	Status of Fracture	Histological Response	Type of Operation	Local Irradiation	Local Recurrence	Status of Patient at Time of Latest Follow-up
1	Prox. part of femur	During preop. chemother.	Non-union	Poor (grade I)	Radical amputation	None	No	Died of disease at 16 mos.
2	Prox. part of femur	At time of presentation	Healed	Good (grade IV)	En bloc resection	None	No	Event-free survival at 60 mos.
3	Prox. part of humerus	At time of presentation	Healed	Good (grade III)	En bloc resection	3000 cGy	No	Event-free survival at 161 mos.
4	Prox. part of humerus	During preop. chemother.	Non-union	Poor (grade I)	Radical amputation	4400 cGy	No	Died of disease at 16 mos.
5	Fibula	At time of presentation	Healed	Poor (grade II)	En bloc resection	None	No	Died of disease at 32 mos.
6	Prox. part of radius	At time of presentation	Healed	Good (grade III)	Wide amputation	None	No	Event-free survival at 78 mos.

Add to My JBJS

TABLE IV DATA ON THE SIX PATIENTS WHO HAD A PATHOLOGICAL FRACTURE

Case	Site of Fracture	Time of Fracture	Status of Fracture	Histological Response	Type of Operation	Local Irradiation	Local Recurrence	Status of Patient at Time of Latest Follow-up
1	Prox. part of femur	During preop. chemother.	Non-union	Poor (grade I)	Radical amputation	None	No	Died of disease at 16 mos.
2	Prox. part of femur	At time of presentation	Healed	Good (grade IV)	En bloc resection	None	No	Event-free survival at 60 mos.
3	Prox. part of humerus	At time of presentation	Healed	Good (grade III)	En bloc resection	3000 cGy	No	Event-free survival at 161 mos.
4	Prox. part of humerus	During preop. chemother.	Non-union	Poor (grade I)	Radical amputation	4400 cGy	No	Died of disease at 16 mos.
5	Fibula	At time of presentation	Healed	Poor (grade II)	En bloc resection	None	No	Died of disease at 32 mos.
6	Prox. part of radius	At time of presentation	Healed	Good (grade III)	Wide amputation	None	No	Event-free survival at 78 mos.

Materials and Methods

Introduction | Materials and Methods | Results | Discussion | References

Patients

Inclusion in this study was restricted to patients who were first seen between 1978 and 1989 because all patients who were managed for Ewing sarcoma at our institution during this time-period received similar chemotherapy. Of the 113 patients who were seen because of Ewing sarcoma, five had an extra-osseous tumor, two had recurrent disease, and one was seen in consultation only and was managed elsewhere. Nineteen patients were managed with chemotherapy and irradiation without operative intervention, and six patients had only postoperative chemotherapy. Eighty patients had preoperative chemotherapy and operative resection. Six of these eighty patients had at least a portion of the radiation therapy before operative treatment; this possibly increased the extent of tissue necrosis and thus invalidated comparison with patients who had necrosis following chemotherapy alone. Exclusion of these six patients left seventy-four patients for the present analysis.

The minimum duration of follow-up of the patients who were alive and continuously free of disease was five years. The median age in the entire series at the time of the diagnosis was sixteen years (range, four to twenty-nine years). The thirty-two female patients had a slightly older median age than the forty-two male patients (seventeen compared with fifteen years). There were thirty-two axial and forty-two appendicular primary Ewing tumors (Fig. 1).

The initial evaluation of all patients included the recording of the medical history, physical examination, and hematological studies. Radiographs of the chest and the site of the primary tumor were made routinely. Local imaging included angiography; computerized tomography; and, in the latter part of the study, magnetic resonance imaging. Systemic staging was performed with use of total-body bone and gallium scans, conventional or computerized tomography scans of the chest, and bone-marrow biopsies.

The size of the tumor was estimated by measuring its largest available dimension on radiographic images that had been made before treatment or by determining the maximum size of the resected tumor with pathological examination. This information was available for sixty-three of the seventy-four patients. When the size of the tumor was determined according to the maximum dimension of the resected specimen, it usually represented the length of osseous involvement.

During treatment and for the first year after its completion, radiographs of the chest were made each month and total-body bone scans were performed every three months. During the next two years, radiographs of the chest were made every three months and bone scans were made every four months. Both examinations were then performed every six months for the following two years. After five years, the two studies were performed annually. Radiographs of the site of the primary tumor generally were made every three months for three years, then every six months until five years, and then annually.

Histological Studies

The histological diagnosis of Ewing sarcoma was based on the results of open biopsy or needle biopsy. Sections of tissue were stained with hematoxylin and eosin as well as with periodic acid-Schiff reagent, with or without diastase digestion. Electron microscopy often was used for detection of glycogen and neural markers. The tumors were differentiated from peripheral neuroectodermal tumors on the basis of these tests, a lack of rosettes or pseudorosettes, and neural differentiation^21,48. Cytogenetic and molecular analysis for the reciprocal translocations t(11;22) or t(21;22), and immunostaining for cell-surface proteins, including MIC2, were not available at the time of the study; only recently have they become part of the histopathological evaluation of small-round-cell tumors^8,16. However, specimens from thirty-six patients were available for reevaluation by immunohistochemical analysis. Each specimen was found to have MIC2 expression on the basis of positive staining with antibody 013^16,31. Eleven specimens previously were shown to have other alterations typical of Ewing sarcoma, including t(11;22) translocation, rearrangement of the EWS gene, and chimeric fusion transcripts^29-31.

Determination of the Percentage of Tumor Necrosis

Each resected specimen was bisected, and the perimeter of the tumor was defined grossly. The entire area of the bisected tumor was partitioned into blocks of tissue averaging 2.0 square centimeters en face. All areas of the specimen, including the medullary canal, residual cortical bone, and surrounding soft tissue, were included, and histological slides were prepared from these blocks. Ten to forty-five slides of material encompassing the entire tumor were used for histological analysis. The extent of necrosis was graded relative to the percentage of residual viable tumor in a semiquantitative manner, as described by Huvos for the grading of osteogenic sarcoma^24,46. An overall assessment, integrating the information obtained from the slides for each section (Figs. 2-A, 2-B, 2-C and 2-D), was used. A grade-I histological response was characterized by little or no necrosis (involving 50 per cent of the tumor or less); a grade-II response, by necrosis of more than 50 per cent but less than 90 per cent of the tumor; a grade-III response, by only scattered foci of viable tumor cells (necrosis of 90 to 99 per cent of the tumor); and a grade-IV response, by no viable tumor (100 per cent necrosis). The histological response to preoperative chemotherapy was determined retrospectively by the same pathologist in a blinded fashion.

Chemotherapy

All patients received intravenous chemotherapy according to the T9 protocol (ten patients) or the T11 protocol (sixty-four patients)⁴⁴. Both regimens consisted of five cycles of chemotherapy, administered over a forty-five-week period with use of analogous doses of all agents (Table I). Each cycle was divided into three phases, and each phase was administered for two days (the T9 protocol) or three days (the T11 protocol) approximately every three weeks. Because the two protocols were so similar and no differences in the clinical or histological responses were found, the protocols were analyzed together.

The operative treatment was timed to allow en bloc resection of the tumor through uninvolved tissue. The treatment strategy was to continue chemotherapy until a maximum response had been achieved, as determined by a clinical reduction in the size of the tumor and a decrease in symptoms at the time of the weekly examination as well as by the radiographic response at the sites of the primary tumor and the metastatic lesions. A pathological fracture that could not be adequately stabilized and did not have evidence of healing accelerated operative resection; two of six pathological fractures led to such an acceleration. The interval between the initiation of a single cycle of chemotherapy and the resection was approximately nine weeks.

Local Treatment

After administration of the preoperative multiagent chemotherapy, fifty-nine patients had an en bloc resection of the primary tumor and eight had an amputation. Although the operative margins were grossly negative for tumor in all patients who had an en bloc resection, three patients had positive margins on examination of the histological sections. Two of these patients subsequently had a reoperation that resulted in negative margins. One, who initially had had an excision of the proximal part of the fibula, subsequently had an above-the-knee amputation. The other patient, who had had a single positive margin at the ilium following an internal hemipelvectomy (iliosacral resection), had a repeat resection of the posterior portion of the ilium; the margins were negative after the repeat procedure. The third patient had positive margins after a sternectomy and was managed with postoperative irradiation. None of these three patients had a local recurrence. Seven patients who had a good clinical response to preoperative chemotherapy refused to have a wide resection and were managed with an intralesional excision. One of these patients subsequently had a planned below-the-knee amputation after complete regression of pulmonary metastases. All ten amputations resulted in negative margins of resection. The margins were considered to be negative if there was no tumor at the inked edge of the specimen either grossly or histologically.

Postoperatively, external-beam radiation was administered to the operative site of forty-eight patients: forty-one (72 per cent) of the fifty-seven patients who ultimately had negative margins after an en bloc resection, all six who had had only an intralesional excision, and one patient who had positive margins as seen histologically. The field of irradiation encompassed five-centimeter margins around the volume of the tumor before chemotherapy, and the median dose was 3000 centigray (range, 2340 to 6500 centigray). The patients who had had an intralesional excision received higher doses than did those who had had an en bloc resection (median, 5010 centigray [range, 3850 to 6500 centigray] compared with 3000 centigray [range, 2340 to 6000 centigray]; p = 0.0001). Radiation-induced sarcoma did not develop in any patient.

Treatment of Metastases at the Time of Presentation

Thirteen (18 per cent) of the seventy-four patients had evidence of systemic disease at the time of the diagnosis. Most of these patients were managed with irradiation of all areas of metastasis that had been detected at the initial, pretreatment assessment. Nine of the eleven patients who initially had pulmonary involvement received 1400 centigray to both lung fields^11,37. Of the two remaining patients, one was not managed with irradiation of the lungs. The other patient had a thoracotomy after resolution of radiographically detected metastases following preoperative chemotherapy; no disease was detected, and he did not receive radiation subsequently. Nevertheless, metastases reappeared at pulmonary and multiple osseous sites. Two patients had radiographic and histological evidence of a solitary osseous metastasis at the time of the diagnosis: one (who also was managed with irradiation because of pulmonary involvement) had administration of 3000 centigray to the involved bone, and the other had no specific treatment. One patient had involvement of the bone marrow at the time of the diagnosis. Restaging, performed before the definitive operative treatment, revealed that all metastatic lesions in these patients had resolved.

Statistical Analysis

The primary end point in this study was event-free survival, defined as the time from the start of chemotherapy to the most recent follow-up evaluation, local or systemic recurrence, or death unrelated to the tumor. The estimated event-free-survival probabilities were calculated with use of the product-limit method. The factors that were analyzed with regard to their potential importance in the prediction of event-free survival or local recurrence included the patient's age at the time of the diagnosis, the size and site of the primary tumor, the presence of metastases at the time of the diagnosis, the number of cycles of preoperative chemotherapy, the delay before chemotherapy was restarted after definitive resection, the type of operation and margins of resection, the use of local irradiation, and the histological response to preoperative chemotherapy. Comparisons of event-free survival were based on the standardized difference in the estimated five-year event-free probability of survival between groups. The more common comparison, based on a weighted log-rank statistic, was not used because of the obvious lack of proportional hazards between groups.

Results

Introduction | Materials and Methods | Results | Discussion | References

For the entire cohort of seventy-four patients, the rate of event-free survival was 81 per cent (sixty patients) at twelve months, 62 per cent (forty-six patients) at twenty-four months, and 58 per cent (forty-three patients) at five years. No patient had recurrent disease after thirty-four months. One patient died of cardiomyopathy at 141 months, with no evidence of tumor. The median duration of follow-up for the forty-two surviving patients was 132 months (range, sixty to 185 months). Twenty-five of the surviving patients were followed for more than 120 months.

Histological Response to Preoperative Chemotherapy

Histological analysis of the seventy-four primary tumors after preoperative chemotherapy revealed that thirty-one (42 per cent) had a complete response (grade IV—100 per cent necrosis) and thirteen (18 per cent) had a nearly complete response (grade III—necrosis of 90 to 99 per cent of the tumor). In contrast, fourteen tumors (19 per cent) demonstrated little or no response (grade I—necrosis of 50 per cent or less), and sixteen tumors (22 per cent) exhibited a moderate response (grade II—necrosis of more than 50 per cent but less than 90 per cent).

Factors Affecting Event-Free Survival

The grade of the histological response was strongly associated with event-free survival (p = 0.0001; Fig. 3). No significant difference in event-free survival could be detected between the patients who had a grade-III histological response and those who had a grade-IV response. Although the outcome associated with a grade-III response was better than that associated with a grade-IV response, this was because of the three deaths not related to the tumor (all verified by autopsy) in the grade-IV group. One patient died accidentally ten months after the diagnosis, after having completed the chemotherapy regimen. The other two deaths were related to the treatment. One patient died of Pneumocystis carinii pneumonia secondary to chemotherapy-induced immunosuppression toward the end of treatment. The other patient had doxorubicin-related cardiomyopathy and died of cardiac failure at 141 months. The difference in event-free survival between the patients who had a good (grade-III or IV) histological response and those who had a grade-I or II response was significant (p = 0.0001). Also, the patients who had grade-II necrosis had a significantly higher rate of event-free survival than did those who had grade-I necrosis (p = 0.007). The rate of event-free survival at five years was zero of fourteen patients who had grade-I necrosis, six of sixteen who had grade-II necrosis, and thirty-seven (84 per cent) of forty-four who had grade-III or IV necrosis.

Before the resection, forty-two patients received one cycle, twenty received two cycles, and twelve received at least three complete cycles of chemotherapy. The presence of metastatic disease at the time of the diagnosis was the only variable associated with an increased number of cycles of preoperative chemotherapy (p = 0.02, chi-square test). Patients who had metastatic disease received a median of two cycles of chemotherapy before the operation compared with a median of one cycle for those who had localized disease. There was a trend toward a lower rate of event-free survival with a longer duration of preoperative chemotherapy (p = 0.08; Fig. 4). Patients who received only one cycle of chemotherapy preoperatively appeared to have a better chance of survival than those who received at least two cycles.

When a higher number of cycles of preoperative chemotherapy had been used, the association between the histological response and the outcome diminished (Table II). At five years, the rate of event-free survival was 96 per cent (twenty-three of twenty-four) for the patients who had had a good (grade-III or IV) histological response after one cycle of chemotherapy, but it was only 70 per cent (fourteen of twenty) for the patients who had had a good response after at least two cycles of chemotherapy. The pattern of survival was similar for the patients who had had a grade-II response. No patient who had had a grade-I response survived, regardless of the duration of the preoperative chemotherapy. Exclusion of the thirteen patients who had metastatic disease at the time of presentation did not alter the predictive value of the histological response relative to the duration of preoperative chemotherapy.

Sixty-three tumors were classified into two groups according to whether the one-dimensional size was 8.0 centimeters (the median measurement) or less or greater than 8.0 centimeters (range of sizes, 4.0 to 25.0 centimeters). At five years, the rate of event-free survival was 79 per cent (twenty-seven of thirty-four) for the patients who had a small tumor compared with 38 per cent (eleven of twenty-nine) for those who had a large tumor (Fig. 5). The difference between these two groups was significant (p = 0.001).

The location of the tumor was not found to be significant, with the numbers available, with regard to event-free survival; however, the outcomes for the patients who had a distal appendicular lesion tended to be better than those for the patients who had a pelvic tumor, who in turn appeared to have a better prognosis than the patients who had either a proximal appendicular lesion or a lesion at another site, such as the scapula, clavicle, sternum, or rib (p = 0.1; Fig. 6). Although the distal lesions tended to be smaller than the proximal lesions, no significant relationship between location and one-dimensional size was detected, with the numbers available (p = 0.17, chi-square test).

The only patient or treatment-related variables that were found to predict event-free survival were the histological response of the primary tumor to preoperative chemotherapy (p = 0.0001) and the size of the primary tumor (p = 0.001). Combining these two factors, we developed a relative-risk table for the probability of systemic recurrence (Table III). Patients who had a small tumor and a good (grade-III or IV) histological response had the lowest rate of recurrence (one [4 per cent] of twenty-five) and were designated as the reference group, with relative risk being equal to 1.0. There was an increased risk of systemic recurrence in association with both a larger tumor and a lesser extent of tumor necrosis.

Outcomes for Patients Who Had Metastatic Disease

There was no apparent increase in the rate of survival of the patients who did not have metastases at the time of presentation (Fig. 7); however, the small number of patients who had metastases at the time of the diagnosis precluded proper comparison. All of the metastatic sites were restaged after preoperative chemotherapy, and all of the metastatic lesions were found to have been eradicated. Of the eleven patients who had pulmonary metastases at the time of the diagnosis, one subsequently had a pulmonary recurrence and two subsequently had osseous metastases. The tumor did not recur in either of the two patients who had had a solitary osseous metastasis at the time of presentation or in either of the two patients who had not received any specific local treatment for the initial metastatic disease, which involved the lungs in one and bone in the other. The patient who had bone-marrow involvement at the time of the diagnosis died of recurrent disseminated disease. The prognosis for the twelve patients who had pulmonary metastases or a solitary osseous metastasis (without involvement of the bone marrow), or both, at the time of presentation was associated with the histological response of the primary tumor to preoperative chemotherapy as well as with the size of the tumor, but prognosis was not related to the initial presence or absence of metastases.

Factors Affecting Local Recurrence

Of the seventy-four patients, six (8 per cent) had a local recurrence at a median of 20.5 months (range, twelve to twenty-six months) postoperatively. All six patients ultimately died of systemic disease. Four patients had a local recurrence and metastases concomitantly. One patient was managed for pulmonary metastases at sixteen months and had a local recurrence ten months later. Another patient had an amputation because of a local recurrence at twenty-five months, and metastases developed two years later.

The rate of local recurrence after only intralesional excision (two of six) was higher than that after only en bloc resection (four [7 per cent] of fifty-eight) or after amputation (zero of ten) (p = 0.07, Fisher exact test). Both patients who had a local recurrence after an intralesional excision had had a poor histological response to chemotherapy. In contrast, three of the four patients who did not have a recurrence after an intralesional excision had had a good histological response to chemotherapy. All six patients who had had an intralesional procedure had received radiation locally afterward. Radiation administered at a median dose of 5010 centigray failed to control local disease following an incomplete excision and a poor response to chemotherapy.

Of the fifty-seven patients who had negative margins after an en bloc resection, forty-one received radiation and sixteen did not. The likelihood of a patient receiving radiation locally was not related to the location or size of the tumor. In the group of patients who had negative margins after en bloc resection, two (5 per cent) of the forty-one who had been managed with radiation had a local recurrence compared with two (13 per cent) of the sixteen who had not been managed with radiation (relative risk, 2.6). Two (5 per cent) of the thirty-seven patients who had had a good histological response had a local recurrence compared with two (10 per cent) of the twenty patients who had had a poor histological response (relative risk, 1.9). There was a local recurrence in two of the seven patients who had had a poor histological response and did not have irradiation locally compared with two of the twenty-seven who had had a good histological response and did have such treatment (relative risk, 3.9).

Treatment of Pathological Fractures

As mentioned earlier, six patients had a pathological fracture (Table IV). Of the four patients who had had a fracture at the time of the diagnosis with healing during preoperative chemotherapy, three were managed with an en bloc resection and one, with a wide amputation¹². Three of these four patients had a good (grade-III or IV) histological response. In the remaining two patients, the fracture occurred during chemotherapy and did not heal subsequently; both patients were managed with a radical amputation¹², and both had little or no tumor necrosis (grade I). The margins of resection were negative in all six patients, and no patient had a local recurrence. The event-free survival for these six patients corresponded to the histological response of the primary tumor to chemotherapy.

Discussion

Introduction | Materials and Methods | Results | Discussion | References

Operative treatment for the local control of Ewing sarcoma provides an opportunity to examine the histological response of the primary tumor to preoperative chemotherapy and the relationship between the histological response and the oncological outcome. In patients who have osteosarcoma, the extent of tumor necrosis, as determined histologically, is the most important prognostic factor in the prediction of systemic disease after intravenous chemotherapy^3,4,35,41,46. We have found that chemotherapy-induced tumor necrosis is also the most important indicator of event-free survival for patients who have had operative treatment of Ewing sarcoma. Because spontaneous necrosis is slight in untreated tumors of both types, rarely exceeding 25 per cent in osteosarcomas⁴⁹ and 20 per cent in Ewing sarcomas⁹, it should not affect the measurement of chemotherapy-induced necrosis. A lesser extent of tumor necrosis following chemotherapy was associated with an increased risk of systemic recurrence (Fig. 3). Patients who had complete (grade-IV) necrosis or only scattered foci of viable tumor cells (grade-III necrosis) were considered to have had a good response to chemotherapy. At five years, the rate of event-free survival for these patients was superior to that for patients who had had either grade-I or grade-II necrosis (thirty-seven [84 per cent] of forty-four compared with zero of fourteen and six of sixteen, respectively).

When more than one cycle of chemotherapy had been given before the definitive operation, the association between a good histological response and survival was not as favorable as it was when only one cycle of preoperative chemotherapy had been administered (Table II). Similar findings have been reported in association with osteosarcoma³⁵. Whereas some tumors respond rapidly, less responsive tumors, which are associated with a poor outcome, require an increased amount of chemotherapy to achieve an adequate clinical response before resection. Indeed, the clinical response to initial chemotherapy has been associated with the outcome in patients who have Ewing sarcoma^26,38, although it has been less predictive than the histological response²⁶.

Magnetic resonance imaging has been investigated as a non-invasive method for assessing the response of Ewing sarcoma to preoperative chemotherapy. Most studies have shown that changes identified by static imaging, with or without gadolinium enhancement, are not useful for determining the response of the tumor^13,18,20,23. Magnetic resonance imaging has been found to overestimate as well as to underestimate the extent of residual disease. In particular, changes in the size of the tumor alone have not been shown to be strongly associated with the extent of tumor necrosis as determined histologically. More recent magnetic resonance imaging techniques, including dynamic and angiographic imaging, may provide a more accurate estimate of the extent of tumor necrosis following chemotherapy^13,17,32,53.

In the present study, the size of the tumor as measured in one dimension was found to be predictive of event-free survival, as has been documented in other series^1,15,22,51 (Fig. 5). These results were independent of the location of the tumor and the histological response, confirming previous findings^6,26. It is likely that the volume of the tumor may have been of even more predictive value^19,26,47. This would support the hypothesis of Goldie and Coldman that larger tumors are associated with a higher risk of an inadequate response to chemotherapy because of the presence or development of drug-resistant clones¹⁰.

Some studies have suggested that pelvic tumors are associated with a high risk of systemic recurrence^{2,14,15,27,37,44,45,47,54}; however, we found no significant differences in survival, with the numbers available, on the basis of the location of the tumor (Fig. 6). In particular, the rate of event-free survival of the patients who had a pelvic tumor was no different than that in the overall series. This finding may be explained by the extent of chemotherapy-induced tumor necrosis and the size of the primary tumors that were included in our study.

Patients who have metastatic Ewing sarcoma at the time of presentation generally have the worst prognosis^{5,7,27,34,47,54}. In one series, patients who had pulmonary metastases at the time of the diagnosis had a 42 per cent rate of event-free survival after five years compared with 0 per cent for the patients who had multifocal, osseous disease⁵. The prevalence of patients who had osseous metastases compared with that of patients who had pulmonary metastases in different studies may help to account for the discouraging prognosis associated with metastatic Ewing sarcoma at the time of the diagnosis. Although only a small number of patients had metastatic disease at the time of presentation in the current study, the outcomes for these patients were not significantly different than those for the patients who had localized disease (Fig. 7). In fact, the outcomes for the patients who had metastatic disease were strongly associated with the histological response and the size of the primary tumor. However, this was a select group of patients; none had multifocal osseous metastases, and all of the metastatic lesions completely resolved during preoperative chemotherapy. Therefore, complete eradication of primary pulmonary and possibly solitary osseous metastases, together with a good histological response to chemotherapy, may indicate a select group of patients who have a drug-sensitive tumor and a potential for event-free survival²⁸.

Of the seventy-four patients in the current study, six (8 per cent) had a local recurrence. An inadequate operative margin was the only factor that increased the risk of local recurrence (p = 0.07), as has been reported previously by others^{2,14,15,26,39,44,47,54}. Evidence also suggests that the histological response to chemotherapy, as shown in a recent study of osteosarcoma⁴¹, as well as postoperative local irradiation, may influence the risk of local recurrence after operative treatment. However, the risk of local recurrence of Ewing sarcoma remains even in patients who have negative margins of resection, a good histological response to chemotherapy, and local irradiation. In fact, a recent study demonstrated that the histological response and the use of radiation therapy had only a marginal effect on the risk of local recurrence of Ewing sarcoma after operative excision³⁹.

Local control with resection is possible even in patients who have a pathological fracture^44,54. Fracture-healing may be another indicator of a tumor that is sensitive to chemotherapy and that can be treated effectively and safely with a limb-salvage operation (Table IV). In the current study, none of the four patients in whom a pathological fracture had healed during chemotherapy, and neither of the two in whom it had not, had local recurrence after the resection. A fracture that does not heal during chemotherapy probably is associated with a tumor that is resistant to chemotherapy, and more extensive resection may be necessary to achieve adequate local control of the tumor. In five of the six patients who had a pathological fracture, the status of the fracture after preoperative chemotherapy was strongly associated with the histological response of the tumor and with the survival of the patient.

In summary, a poor histological response to preoperative chemotherapy and a large size are indicative of a tumor with a drug-resistant phenotype. The predictive values of these variables can be combined to improve risk stratification and to help physicians to decide among possible treatment options (Table III). Data from this study and others indicate that a tumor volume of more than 100 milliliters or a one-dimensional size of more than 8.0 centimeters is a strong negative risk factor that can be identified at the time of the diagnosis^{1,15,19,22,26,47,51}. A poor (grade-I or II) histological response (necrosis of less than 90 per cent of the tumor) to standard preoperative chemotherapy, as seen at the time of a resection, can be used to further identify high-risk patients who might have a better outcome if they receive alternative, intensive forms of chemotherapy. These treatment options might include myeloablative radiation and chemotherapy followed by bone-marrow transplantation^5,34 or regimens incorporating VP-16 (etoposide) and ifosfamide^28,36. The latter approach recently was reported to be successful for the treatment of osteosarcoma in patients who initially had had a poor response to chemotherapy³.

References

Introduction | Materials and Methods | Results | Discussion | References

Arai, Y.; Kun, L. E.; Brooks, M. T.; Fairclough, D. L.; Fontanesi, J.; Meyer, W. H.; Hayes, F. A.; Thompson, E.; and Rao, B. N.: Ewing's sarcoma: local tumor control and patterns of failure following limited-volume radiation therapy. Internat. J. Radiat. Oncol., Biol., Phys.,21: 1501-1508, 1991.211501 1991

Bacci, G.; Toni, A.; Avella, M.; Manfrini, M.; Sudanese, A.; Ciaroni, D.; Boriani, S.; Emiliani, E.; and Campanacci, M.: Long-term results in 144 localized Ewing's sarcoma patients treated with combined therapy. Cancer,63: 1477-1486, 1989.631477 1989 [PubMed]

Bacci, G.; Picci, P.; Ferrari, S.; Ruggieri, P.; Casadei, R.; Tienghi, A.; Brach del Prever, A.; Gherlinzoni, F.; Mercuri, M.; and Monti, C.: Primary chemotherapy and delayed surgery for nonmetastatic osteosarcoma of the extremities. Results in 164 patients preoperatively treated with high doses of methotrexate followed by cisplatin and doxorubicin. Cancer,72: 3227-3238, 1993.723227 1993 [PubMed]

Bacci, G.; Picci, P.; Pignatti, G.; De Cristofaro, R.; Dallari, D.; Avella, M.; Manfrini, M.; Marangolo, M.; Ferruzi, A.; Mercuri, M.; Ruggieri, P.; Biagini, R.; Capanna, R.; Ferrari, S.; Prasad, R.; and Campanacci, M.: Neoadjuvant chemotherapy for nonmetastatic osteosarcoma of the extremities. Clin. Orthop.,270: 87-98, 1991.27087 1991 [PubMed]

Burdach, S.; Jürgens, H.; Peters, C.; Nürnberger, W.; Mauz-Körholz, C.; Körholz, D.; Paulussen, M.; Pape, H.; Dilloo, D.; Koscielniak, E.; Gadner, H.; and Göbel, U.: Myeloablative radiochemotherapy and hematopoietic stem-cell rescue in poor-prognosis Ewing's sarcoma. J. Clin. Oncol.,11: 1482-1488, 1993.111482 1993 [PubMed]

Burgert, E. O., Jr.; Nesbit, M. E.; Garnsey, L. A.; Gehan, E. A.; Herrmann, J.; Vietti, T. J.; Cangir, A.; Tefft, M.; Evans, R.; Thomas, P.; Askin, F. B.; Kissane, J. M.; Pritchard, D. J.; Neff, J.; Makley, J. T.; and Gilula, L.: Multimodal therapy for the management of nonpelvic, localized Ewing's sarcoma of bone: Intergroup Study IESS-II. J. Clin. Oncol.,8: 1514-1524, 1990.81514 1990 [PubMed]

Cangir, A.; Vietti, T. J.; Gehan, E. A.; Burgert, E. O., Jr.; Thomas, P.; Tefft, M.; Nesbit, M. E.; Kissane, J.; and Pritchard, D.: Ewing's sarcoma metastatic at diagnosis. Results and comparisons of two intergroup Ewing's sarcoma studies. Cancer,66: 887-893, 1990.66887 1990 [PubMed]

Delattre, O.; Zucman, J.; Melot, T.; Garau, X. S.; Zucker, J. M.; Lenoir, G. M.; Ambros, P. F.; Sheer, D.; Turc-Carel, C.; Triche, T. J.; Aurias, A.; and Thomas, G.: The Ewing family of tumors—a subgroup of small-round-cell tumors defined by specific chimeric transcripts. New England J. Med.,331: 294-299, 1994.331294 1994

de Stefani, E.; Carzoglio, J.; Deneo Pellegrini, H.; Olivera, L.; Cendan, M.; and Kasdorf, H.: Ewing's sarcoma: value of tumor necrosis as a predictive factor. Bull. Cancer,71: 16-21, 1984.7116 1984 [PubMed]

DeVita, V. T., Jr.: The relationship between tumor mass and resistance to chemotherapy. Implications for surgical adjuvant treatment of cancer. Cancer,51: 1209-1220, 1983.511209 1983 [PubMed]

Dunst, J.; Paulussen, M.; and Jurgens, H.: Lung irradiation for Ewing's sarcoma with pulmonary metastases at diagnosis: results of the CESS-studies. Strahlenther. Onkol.,169: 621-623, 1993.169621 1993 [PubMed]

Enneking, W. F.; Spanier, S. S.; and Goodman, M. A.: A system for the surgical staging of musculoskeletal sarcoma. Clin. Orthop.,153: 106-120, 1980.153106 1980 [PubMed]

Erlemann, R.; Sciuk, J.; Bosse, A.; Ritter, J.; Kusnierz-Glaz, C. R.; Peters, P. E.; and Wuisman, P.: Response of osteosarcoma and Ewing sarcoma to preoperative chemotherapy: assessment with dynamic and static MR imaging and skeletal scintigraphy. Radiology,175: 791-796, 1990.175791 1990 [PubMed]

Evans, R. G.; Nesbit, M. E.; Gehan, E. A.; Garnsey, L. A.; Burgert, O., Jr.; Vietti, T. J.; Cangir, A.; Tefft, M.; Thomas, P.; Askin, F. B.; Kissane, J. M.; Pritchard, D. J.; Neff, J.; Makley, J. T.; and Gilula, L.: Multimodal therapy for the management of localized Ewing's sarcoma of pelvic and sacral bones: a report from the Second Intergroup Study. J. Clin. Oncol.,9: 1173-1180, 1991.91173 1991 [PubMed]

Evans, R.; Nesbit, M.; Askin, F.; Burgert, O.; Cangir, A.; Foulkes, M.; Gehan, E.; Gilula, L.; Kissane, J.; Makley, J.; Neff, J.; Perez, C.; Pritchard, D.; Tefft, M.; Thomas, P.; and Vietti, T.: Local recurrence, rate and sites of metastases, and time to relapse as a function of treatment regimen, size of primary and surgical history in 62 patients presenting with nonmetastatic Ewing's sarcoma of the pelvic bones. Internat. J. Radiat. Oncol., Biol., Phys.,11: 129-136, 1985.11129 1985

Fellinger, E. J.; Garin-Chesa, P.; Triche, T. J.; Huvos, A. G.; and Rettig, W. J.: Immunohistochemical analysis of Ewing's sarcoma cell surface antigen p30/32^MIC2. Am. J. Pathol.,139: 317-325, 1991.139317 1991 [PubMed]

Fletcher, B. D.; Hanna, S. L.; Fairclough, D. L.; and Gronemeyer, S. A.: Pediatric musculoskeletal tumors: use of dynamic, contrast-enhanced MR imaging to monitor response to chemotherapy. Radiology,184: 243-248, 1992.184243 1992 [PubMed]

Frouge, C.; Vanel, D.; Coffre, C.; Couanet, D.; Contesso, G.; and Sarrazin, D.: The role of magnetic resonance imaging in the evaluation of Ewing sarcoma. A report of 27 cases. Skel. Radiol.,17: 387-392, 1988.17387 1988

Gobel, V.; Jurgens, H.; Etspuler, G.; Kemperdick, H.; Jungblut, R. M.; Stienen, U.; and Gobel, U.: Prognostic significance of tumor volume in localized Ewing's sarcoma of bone in children and adolescents. J. Cancer Res. and Clin. Oncol.,113: 187-191, 1987.113187 1987

Golfieri, R.; Baddeley, H.; Pringle, J. S.; Leung, A. W.; Greco, A.; and Souhami, R.: MRI in primary bone tumors: therapeutic implications. European J. Radiol.,12: 201-207, 1991.12201 1991

Hartman, K. R.; Triche, T. J.; Kinsella, T. J.; and Miser, J. S.: Prognostic value of histopathology in Ewing's sarcoma. Long-term follow-up of distal extremity primary tumors. Cancer,67: 163-171, 1991.67163 1991 [PubMed]

Hayes, F. A.; Thompson, E. I.; Meyer, W. H.; Kun, L.; Parham, D.; Rao, B.; Kumar, M.; Hancock, M.; Parvey, L.; Magill, L.; and Hustu, H. O.: Therapy for localized Ewing's sarcoma of bone. J. Clin. Oncol.,7: 208-213, 1989.7208 1989 [PubMed]

Holscher, H. C.; Bloem, J. L.; Nooy, M. A.; Taminiau, A. H.; Eulderink, F.; and Hermans, J.: The value of MR imaging in monitoring the effect of chemotherapy on bone sarcomas. AJR: Am. J. Roentgenol.,154: 763-769, 1990.154763 1990 [PubMed]

Huvos, A. G.: Osteogenic sarcoma: pathologic assessment of preoperative (neoadjuvant) chemotherapy. In Bone Tumors: Diagnosis, Treatment, and Prognosis. Ed. 2, pp. 122-128. Philadelphia, W. B. Saunders, 1991.

Jentzsch, K.; Binder, H.; Cramer, H.; Glaubiger, D. L.; Kessler, R. M.; Bull, C.; Pomeroy, T. C.; and Gerber, N. L.: Leg function after radiotherapy for Ewing's sarcoma. Cancer,47: 1267-1278, 1981.471267 1981 [PubMed]

Jürgens, H.; Exner, U.; Gadner, H.; Harms, D.; Michaelis, J.; Sauer, R.; Treuner, J.; Voûte, T.; Winkelmann, W.; Winkler, K.; and Göbel, U.: Multidisciplinary treatment of primary Ewing's sarcoma of bone. A 6-year experience of a European Cooperative Trial. Cancer,61: 23-32, 1988.6123 1988 [PubMed]

Kinsella, T. J.; Miser, J. S.; Waller, B.; Venzon, D.; Glatstein, E.; Weaver-McClure, L.; and Horowitz, M. E.: Long-term follow-up of Ewing's sarcoma of bone treated with combined modality therapy. Internat. J. Radiat. Oncol., Biol., Phys.,20: 389-395, 1991.20389 1991

Kushner, B. H.; Meyers, P. A.; Gerald, W. L.; Healey, J. H.; La Quaglia, M. P.; Boland, P.; Wollner, N.; Casper, E. S.; Aledo, A.; Heller, G.; Schwartz, G. K.; Bonilla, M. A.; Lindsley, K. L.; Merchant, T. E.; Rosenfield, N. S.; Abramson, S. J.; and Cheung, N.-K. V.: Very-high-dose short-term chemotherapy for poor-risk peripheral primitive neuroectodermal tumors, including Ewing's sarcoma, in children and young adults. J. Clin. Oncol.,13: 2796-2804, 1995.132796 1995 [PubMed]

Ladanyi, M.: The emerging molecular genetics of sarcoma translocations. Diag. Molec. Pathol.,4: 162-173, 1995.4162 1995

Ladanyi, M.; Heinemann, F. S.; Huvos, A. G.; Rao, P. H.; Chen, Q. G.; and Jhanwar, S. C.: Neural differentiation in small round cell tumors of bone and soft tissue with the translocation t(11;22)(q24;q12): an immunohistochemical study of 11 cases. Human Pathol.,21: 1245-1251, 1990.211245 1990

Ladanyi, M.; Lewis, R.; Garin-Chesa, P.; Rettig, W. J.; Huvos, A. G.; Healey, J. H.; and Jhanwar, S. C.: EWS rearrangement in Ewing's sarcoma and peripheral neuroectodermal tumor. Molecular detection and correlation with cytogenetic analysis and MIC2 expression. Diag. Molec. Pathol.,2: 141-146, 1993.2141 1993

Lang, P.; Vahlensieck, M.; Matthay, K. K.; Johnston, J. O.; Rosenau, W.; Gooding, C. A.; and Genant, H. K.: Monitoring neovascularity as an indicator to response to chemotherapy in osteogenic and Ewing sarcoma using magnetic resonance angiography. Med. and Pediat. Oncol.,26: 329-333, 1996.26329 1996

Lewis, R. J.; Marcove, R. C.; and Rosen, G.: Ewing's sarcoma—functional effects of radiation therapy. J. Bone and Joint Surg.,59-A: 325-331, April 1977.59-A325 1977

Marcus, R. B., Jr.; Graham-Pole, J. R.; Springfield, D. S.; Fort, J. A.; Gross, S.; Mendenhall, N. P.; Elfenbein, G. J.; Weiner, R. S.; Enneking, W. F.; and Million, R. R.: High-risk Ewing's sarcoma: end-intensification using autologous bone marrow transplantation. Internat. J. Radiat. Oncol., Biol., Phys.,15: 53-59, 1988.1553 1988

Meyers, P. A.; Heller, G.; Healey, J.; Huvos, A.; Lane, J.; Marcove, R.; Applewhite, A.; Vlamis, V.; and Rosen, G.: Chemotherapy for nonmetastatic osteogenic sarcoma: the Memorial Sloan-Kettering experience. J. Clin. Oncol.,10: 5-15, 1992.105 1992 [PubMed]

Miser, J. S.; Kinsella, T. J.; Triche, T. J.; Tsokos, M.; Jarosinski, P.; Forquer, R.; Wesley, R.; and Magrath, I.: Ifosfamide with mesna uroprotection and etoposide: an effective regimen in the treatment of recurrent sarcomas and other tumors of children and young adults. J. Clin. Oncol.,5: 1191-1198, 1987.51191 1987 [PubMed]

Nesbit, M. E., Jr.; Gehan, E. A.; Burgert, E. O., Jr.; Vietti, T. J.; Cangir, A.; Tefft, M.; Evans, R.; Thomas, P.; Askin, F. B.; Kissane, J. M.; Pritchard, D. J.; Herrmann, J.; Neff, J.; Makley, J. T.; and Gilula, L.: 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.,8: 1664-1674, 1990.81664 1990 [PubMed]

Oberlin, O.; Patte, C.; Demeocq, F.; Lacombe, M. J.; Brunat-Mentigny, M.; Demaille, M. C.; Tron, P.; N'Guyen Bui, B.; and Lemerle, J.: The response to initial chemotherapy as a prognostic factor in localized Ewing's sarcoma. European J. Cancer and Clin. Oncol.,21: 463-467, 1985.21463 1985

Ozaki, T.; Hillman, A.; Hoffman, C.; Rube, C.; Blasius, S.; Dunst, J.; Jurgens, H.; and Winkelmann, W.: Significance of surgical margin on the prognosis of patients with Ewing's sarcoma. A report from the Cooperative Ewing's Sarcoma Study. Cancer,78: 892-900, 1996.78892 1996 [PubMed]

Phillips, R. F., and Higinbotham, N. L.: The curability of Ewing's endothelioma of bone in children. J. Pediat.,70: 391-397, 1967.70391 1967 [PubMed]

Picci, P.; Sangiorgi, L.; Rougraff, B. T.; Neff, J. R.; Casadei, R.; and Campanacci, M.: Relationship of chemotherapy-induced necrosis and surgical margins to local recurrence in osteosarcoma. J. Clin. Oncol.,12: 2699-2705, 1994.122699 1994 [PubMed]

Picci, P.; Rougraff, B. T.; Bacci, G.; Neff, J. R.; Sangiorgi, L.; Cazzola, A.; Baldini, N.; Ferrari, S.; Mercuri, M.; Ruggieri, P.; Caldora, P.; Benassi, M. S.; Fabbri, N.; Monti, C.; and Campanacci, M.: Prognostic significance of histopathologic response to chemotherapy in nonmetastatic Ewing's sarcoma of the extremities. J. Clin. Oncol.,11: 1763-1769, 1993.111763 1993 [PubMed]

Pizzo, P. A.; Horowitz, M. E.; Poplack, D. G.; Hays, D. M.; and Kun, L. E.: Solid tumors of childhood. In Cancer: Principles and Practice of Oncology, edited by V. T. DeVita, Jr., S. Hellman, and S. A. Rosenberg. Ed. 4, pp. 1738-1791. Philadelphia, J. B. Lippincott, 1993.

Rosen, G.; Caparros, B.; Nirenberg, A.; Marcove, R. C.; Huvos, A. G.; Kosloff, C.; Lane, J.; and Murphy, M. L.: Ewing's sarcoma: ten-year experience with adjuvant chemotherapy. Cancer,47: 2204-2213, 1981.472204 1981 [PubMed]

Rosen, G.; Wollner, N.; Tan, C.; Wu, S. J.; Hajdu, S. I.; Cham, W.; D'Angio, G. J.; and Murphy, M. L.: Disease-free survival in children with Ewing's sarcoma treated with radiation therapy and adjuvant four-drug sequential chemotherapy. Cancer,33: 384-393, 1974.33384 1974 [PubMed]

Rosen, G.; Caparros, B.; Huvos, A. G.; Kosloff, C.; Nirenberg, A.; Cacavio, A.; Marcove, R. C.; Lane, J. M.; Mehta, B.; and Urban, C.: Preoperative chemotherapy for osteogenic sarcoma: selection of postoperative adjuvant chemotherapy based on the response of the primary tumor to preoperative chemotherapy. Cancer,49: 1221-1230, 1982.491221 1982 [PubMed]

Sailer, S. L.; Harmon, D. C.; Mankin, H. J.; Truman, J. T.; and Suit, H. D.: Ewing's sarcoma: surgical resection as a prognostic factor. Internat. J. Radiat. Oncol., Biol., Phys.,15: 43-52, 1988.1543 1988

Schmidt, D.; Herrmann, C.; Jurgens, H.; and Harms, D.: Malignant peripheral neuroectodermal tumor and its necessary distinction from Ewing's sarcoma. A report from the Kiel Pediatric Tumor Registry. Cancer,68: 2251-2259, 1991.682251 1991 [PubMed]

Springfield, D. S.; Schakel, M. E., Jr.; and Spanier, S. S.: Spontaneous necrosis in osteosarcoma. Clin. Orthop.,263: 233-237, 1991.263233 1991 [PubMed]

Strong, L. C.; Herson, J.; Osborne, B. M.; and Sutow, W. W.: Risk of radiation-related subsequent malignant tumors in survivors of Ewing's sarcoma. J. Nat. Cancer Inst.,62: 1401-1406, 1979.621401 1979 [PubMed]

Thomas, P. R.; Foulkes, M. A.; Gilula, L. A.; Burgert, E. O.; Evans, R. G.; Kissane, J.; Nesbit, M. E.; Pritchard, D. J.; Tefft, M.; and Vietti, T. J.: Primary Ewing's sarcoma of the ribs. A report from the Intergroup Ewing's Sarcoma Study. Cancer,51: 1021-1027, 1983.511021 1983 [PubMed]

Tucker, M. A.; D'Angio, G. J.; Boice, J. D., Jr.; Strong, L. C.; Li, F. P.; Stovallo, M.; Stone, B. J.; Green, D. M.; Lombardi, F.; Newton, W.; Hoover, R. N.; and Fraumeni, J. F., Jr.: Bone sarcomas linked to radiotherapy and chemotherapy in children. New England J. Med.,317: 588-593, 1987.317588 1987

van der Woude, H. J.; Bloem, J. L.; Verstraete, K. L.; Taminiau, A. H.; Nooy, M. A.; and Hogendoorn, P. C.: Osteosarcoma and Ewing's sarcoma after neoadjuvant chemotherapy: value of dynamic MR imaging in detecting viable tumor before surgery. AJR: Am. J. Roentgenol.,165: 593-598, 1995.165593 1995 [PubMed]

Wilkins, R. M.; Pritchard, D. J.; Burgert, E. O., Jr.; and Unni, K. K.: Ewing's sarcoma of bone. Experience with 140 patients. Cancer,58: 2551-2555, 1986.582551 1986 [PubMed]

Topics

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1 Drawing showing the locations of the seventy-four Ewing sarcomas.

View Large | Download Slide(.ppt)
Add to My JBJS

Figs. 2-A through 2-D: Histological responses of Ewing sarcomas to preoperative chemotherapy. Figs. 2-A and 2-B: No necrosis (grade I) (x 40 and x 100, respectively).

View Large | Download Slide(.ppt)
Add to My JBJS

Figs. 2-A and 2-B: No necrosis (grade I) (x 40 and x 100, respectively).

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-C Almost complete necrosis (grade III) (x 40).

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-D Complete necrosis (grade IV) (x 40).

View Large | Download Slide(.ppt)
Add to My JBJS

View Larger

TABLE I CHEMOTHERAPY PROTOCOLS

*The total dose of each agent that was administered over five complete cycles for a duration of forty-five weeks.

Agent	Dose* (mg/m² of skin)
T9 Protocol	T11 Protocol
Actinomycin D	7.5	6.0
Doxorubicin	480	480
Cyclophosphamide	19.5	18.0
Vincristine	37.5	37.5
Methotrexate	360	360
Bleomycin	150	150

Add to My JBJS

TABLE I CHEMOTHERAPY PROTOCOLS

*The total dose of each agent that was administered over five complete cycles for a duration of forty-five weeks.

Agent	Dose* (mg/m² of skin)
T9 Protocol	T11 Protocol
Actinomycin D	7.5	6.0
Doxorubicin	480	480
Cyclophosphamide	19.5	18.0
Vincristine	37.5	37.5
Methotrexate	360	360
Bleomycin	150	150

View Larger

TABLE II RELATIONSHIP BETWEEN THE HISTOLOGICAL RESPONSE AND THE NUMBER OF CYCLES OF PREOPERATIVE CHEMOTHERAPY*

*At five years.

Grade of Histological Response	1 Cycle		=2 Cycles
Total No. of Patients	No. of Patients Surviving Event-Free*	Total No. of Patients	No. of Patients Surviving Event-Free*
I	7	0	7	0
II	11	5	5	1
III or IV	24	23	20	14

Add to My JBJS

TABLE II RELATIONSHIP BETWEEN THE HISTOLOGICAL RESPONSE AND THE NUMBER OF CYCLES OF PREOPERATIVE CHEMOTHERAPY*

*At five years.

Grade of Histological Response	1 Cycle		=2 Cycles
Total No. of Patients	No. of Patients Surviving Event-Free*	Total No. of Patients	No. of Patients Surviving Event-Free*
I	7	0	7	0
II	11	5	5	1
III or IV	24	23	20	14

View Larger

TABLE III PROBABILITY OF SYSTEMIC RECURRENCE ON THE BASIS OF THE HISTOLOGICAL RESPONSE AND THE SIZE OF THE TUMOR

Grade of Histological Response	Relative Risk
=8-cm Tumor	>8-cm Tumor
I	25	25
II	8.3	19.5
III or IV	1.0*	4.5

Add to My JBJS

TABLE III PROBABILITY OF SYSTEMIC RECURRENCE ON THE BASIS OF THE HISTOLOGICAL RESPONSE AND THE SIZE OF THE TUMOR

Grade of Histological Response	Relative Risk
=8-cm Tumor	>8-cm Tumor
I	25	25
II	8.3	19.5
III or IV	1.0*	4.5

View Larger

TABLE IV DATA ON THE SIX PATIENTS WHO HAD A PATHOLOGICAL FRACTURE

Case	Site of Fracture	Time of Fracture	Status of Fracture	Histological Response	Type of Operation	Local Irradiation	Local Recurrence	Status of Patient at Time of Latest Follow-up
1	Prox. part of femur	During preop. chemother.	Non-union	Poor (grade I)	Radical amputation	None	No	Died of disease at 16 mos.
2	Prox. part of femur	At time of presentation	Healed	Good (grade IV)	En bloc resection	None	No	Event-free survival at 60 mos.
3	Prox. part of humerus	At time of presentation	Healed	Good (grade III)	En bloc resection	3000 cGy	No	Event-free survival at 161 mos.
4	Prox. part of humerus	During preop. chemother.	Non-union	Poor (grade I)	Radical amputation	4400 cGy	No	Died of disease at 16 mos.
5	Fibula	At time of presentation	Healed	Poor (grade II)	En bloc resection	None	No	Died of disease at 32 mos.
6	Prox. part of radius	At time of presentation	Healed	Good (grade III)	Wide amputation	None	No	Event-free survival at 78 mos.

Add to My JBJS

TABLE IV DATA ON THE SIX PATIENTS WHO HAD A PATHOLOGICAL FRACTURE

Case	Site of Fracture	Time of Fracture	Status of Fracture	Histological Response	Type of Operation	Local Irradiation	Local Recurrence	Status of Patient at Time of Latest Follow-up
1	Prox. part of femur	During preop. chemother.	Non-union	Poor (grade I)	Radical amputation	None	No	Died of disease at 16 mos.
2	Prox. part of femur	At time of presentation	Healed	Good (grade IV)	En bloc resection	None	No	Event-free survival at 60 mos.
3	Prox. part of humerus	At time of presentation	Healed	Good (grade III)	En bloc resection	3000 cGy	No	Event-free survival at 161 mos.
4	Prox. part of humerus	During preop. chemother.	Non-union	Poor (grade I)	Radical amputation	4400 cGy	No	Died of disease at 16 mos.
5	Fibula	At time of presentation	Healed	Poor (grade II)	En bloc resection	None	No	Died of disease at 32 mos.
6	Prox. part of radius	At time of presentation	Healed	Good (grade III)	Wide amputation	None	No	Event-free survival at 78 mos.