Followup information was available for all patients (except for Patient 20, who was lost to followup after 40 months but who had no evidence of disease). Systemic and local control of disease were monitored carefully by routine clinical examination and appropriate radiographic studies. A local recurrence was tabulated for any patient who had recurrence in the original tumor bed.
The average length of followup for the surviving patients was 137 months (range, 40-276 months), and all surviving patients had no detectable disease at the time of the current report. The Musculoskeletal Tumor Society functional evaluation instrument4 and the Short Form-3618 were used to evaluate all available surviving patients.
Continuous data were analyzed using t tests. Categorical data were analyzed using chi square tests and Fisher's exact test. The survival experience of patients treated with surgery and those without surgery were compared with Kaplan-Meier survival curves.9
The estimated 5- and 10-year overall survival for these 50 patients was 44%. Table 5 shows the oncologic results of 37 patients without metastases at diagnosis. Sixteen of the 37 patients were treated without attempted surgical resection, and only three (18%) are alive. In contrast, 16 of the 21 (76%) patients without metastases who underwent surgical resection as part of their therapy are alive without evidence of disease. This difference in survival was statistically significant (p = 0.005). The survival curves of all patients and those patients with localized disease treated with surgery and those treated without surgery (radiation and chemotherapy alone) are shown with Kaplan-Meier survival curves in Figures 1 and 2, respectively. There were no differences between the surgical and nonsurgical groups in terms of patient age, size of tumor, patient gender, or stage of the disease. There were no discernible survival differences between patients with one pelvic bone or more than one pelvic bone involved with tumor. However, the addition of surgery to the subgroup of patients with more than one pelvic bone involved resulted in a statistically significant improvement in survival than in patients treated with chemotherapy and radiation therapy alone (p = 0.01).
Twenty-one patients without metastasis underwent surgical resection. Thirteen of the 21 patients had microscopically negative margins. Eight of these 21 patients had contaminated margins after an attempted wide resection. Eleven of the 13 patients with negative surgical margins are alive without evidence of disease, and five of the eight patients with contaminated margins are alive without evidence of disease; both groups have a minimum 57 months of followup. This difference in survival was not significant.
Local recurrence of disease was identified in seven of 37 (19%) patients who presented without metastases and is reported in Table 6. Two of the seven local recurrences were in patients who had combined chemotherapy, radiation therapy, and surgery, whereas the remaining five local recurrences were in the patients who underwent chemotherapy and radiation alone. There is no significant difference between the groups regarding the rate of local recurrence (p = 0.2). The average time to local recurrence was 9 months (range, 2-17 months), whereas the average time to death was 14 months in the 22 patients who died of their disease. No patient with local recurrence survived, but 21 of 30 (70%) patients who obtained local control of their disease were disease free at latest followup. This was statistically significant (p = 0.001).
Three of the 13 patients with metastases at the time of diagnosis currently are alive without evidence of disease 54, 87, and 276 months after diagnosis. Two of the three patients surviving after presentation with multiple bone and pulmonary metastases underwent resection of their primary disease, pulmonary wedge resection, and chemotherapy. The third patient was treated with local radiation and systemic chemotherapy. Two of the 28 patients in whom metastatic disease developed during therapy are alive, 151 and 89 months after diagnosis (121 and 37 months after relapse).
The study period of this series spanned 3 decades, which saw many improvements in the care of patients with cancer. There was a trend toward improved survival with successive decades of treatment, but this improvement was not statistically significant (p = 0.141). During the 1980s and 1990s, the surgical group achieved a 75% 5-year overall survival, whereas the nonsurgical group achieved only a 40% 5-year overall survival. This difference in survival was not statistically significant (p = 0.28). Figure 3 shows the improvement in survival between the earliest and more recent decades of treatment, and the improvement in survival between the surgical and nonsurgical groups treated during the 1980s and 1990s.
Functional evaluations of 16 of the 21 patients who were alive at the latest followup were used to compare patients who had surgical resection with those who had nonsurgical treatment (Table 7). The Musculoskeletal Tumor Society functional evaluation score average was 71% for the surgical group and 83% for the nonsurgical group. This difference was not statistically significant. The Short Form-36 functional evaluation is a multiitem scale that assesses eight health concepts. Overall, the nonsurgical group scored better in five categories (physical function, role physical, bodily pain, general health, and vitality), whereas the surgical group scored slightly better in the category of mental health. Although the numbers were small with only two surviving patients who did not have surgery, the general trend was an overall better functional outcome for patients in the nonsurgical group.
Surgery as part of the multidisciplinary treatment plan for patients with Ewing's sarcoma of the pelvis resulted in improved survival compared with survival of patients who were treated with radiation therapy and chemotherapy alone. Many previously reported studies that focused on Ewing's sarcoma of the pelvis show a trend toward improved local control and survival in patients treated with surgery.6,10,12,15
Several authors have been interested in a rationale for the improvement in oncologic outcome for patient's with Ewing's sarcoma who were treated with surgery. Picci et al12 reported their histologic experience in patients who had neoadjuvant chemotherapy without radiation. They found that 2/3 of patients had viable tumor present at the time of resection, despite receiving systemic chemotherapy, and 1/3 had macroscopic disease. Telles et al16 reported that autopsy specimens of patients with Ewing's sarcoma treated with radiation and chemotherapy also contained persistent viable cancer in nearly 2/3 of cases. It would seem logical that if surgery could remove the primary lesion, particularly in the central avascular region, it would enhance the local control that could be obtained with radiation alone. Thus, improved local control may be associated with improved survival.
Local control data in Ewing's sarcoma can be deceiving. The current study shows a statistically significant increase in survival in patients who achieved local control, but the study did not show a difference in local recurrence between the surgical and nonsurgical groups. It is possible that this study is underreporting subclinical recurrences in the early nonsurgical group, which had a short mean survival of 14 months; autopsies were not done routinely on these patients. However, it is unlikely that any local recurrence was missed in the patients who were treated with surgery who otherwise are free of disease at an average followup of 137 months.
Local recurrence was associated with poor oncologic outcome. Previously reported series generally have not reported the size of the pelvic primary tumor. It is possible that the patients who undergo surgery are those with a smaller primary tumor. Because patients with small primary Ewing's sarcoma may have a better prognosis, these surgical candidates with smaller masses may have had superior oncologic outcome, even if they had not had the surgery.14 There was no significant difference in the size of the tumor treated in the surgical and nonsurgical groups in the current series. In fact, the size of the tumors in the surgical group was slightly larger than those in the nonsurgical group.
The issue of adequate surgical margins in Ewing's sarcoma deserves comment. The small number of patients in the current study did not show a difference in survival based on the quality of surgical resection. The current approach to Ewing's sarcoma in the pelvis at the authors' institutions is to resect the entire lesion, including the soft tissue mass, based on the pretreatment studies. If a margin is contaminated or close, radiation is recommended. Debulking surgery is not recommended, although its virtue was not addressed by the current study.
Yang et al20 reported differences in oncologic outcome between patients who had two parts of the pelvic ring involved with the disease versus those with only one. Although the current study did not show a significant difference in survival between patients with two parts of the pelvic ring involved with the disease and those with only one, it did show a statistically significant improvement in survival between patients with two or more parts of pelvic ring involved with tumor who were treated with surgical resection versus those who were treated with chemotherapy and radiation alone (p = 0.01).
Functional outcome was assessed using two instruments: the Musculoskeletal Tumor Society's functional evaluation form and the Short Form-36. Both instruments revealed a better functional outcome for the nonsurgical group, but the difference was not statistically significant. The superior functional outcome seen with nonoperative treatment must be weighed against significantly improved 5- and 10-year survival of patients who had operative treatment.
The strength of this study is also its greatest weakness: long-term followup. This study spans 3 decades, in which advances in imaging studies, radiation therapy, and chemotherapy have improved the diagnosis and treatment of patients with Ewing's sarcoma. An internal control for the expected improvement in survival from these advances can be seen in a comparison of the nonsurgical groups of the 1960s and 1970s (19% 5-year survival) and the non-surgical group of the 1980s and 1990s (45% 5-year survival). This improvement in the rate of survival does not explain the much greater survival observed in the combined surgical and nonsurgical groups treated in the 1980s and 1990s (75% 5-year survival). Clearly, the patients treated in the 1980s and 1990s benefited from the increased incidence of surgical resection of their pelvic Ewing's sarcoma.
1. Bacci G, Toni A, Avella M, et al: Long-term results in 144 localized Ewing's sarcoma patients treated with combined therapy. Cancer 63:1477-1486, 1989.
2. Brown AP, Fixsen JA, Chir M, Plowman PN: Local control of Ewing's sarcoma: An analysis of 67 patients. Br J Radiol 60:261-268, 1987.
3. Capanna R, Toni A, Sudanese A, et al: Ewing's sarcoma of the pelvis. Int Orthop 14:57-61, 1990.
4. Enneking WF, Dunham W, Gebhardt MC, Malawer M, Pritchard DJ: A system for the evaluation of reconstructive procedures after surgical treatment of tumors of the musculoskeletal system. Clin Orthop 286:241-246, 1993.
5. Evan RG, Nesbit ME, Askin F, et al: Local recurrence rate and sites of metastases, and time to replace as a function of treatment regimen, size or primary and surgical history in 62 patients presenting with non-metastatic Ewing's sarcoma of the pelvic bones. Int J Radiat Oncol Biol Phys 11:129-136, 1985.
6. Evans RG, Nesbit ME, Gehan EA, et al: 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.
7. Falk S, Alpert M: Five year survival of patients with Ewing's sarcoma. Surg Gynecol Obstet 107:319-324, 1967.
8. Frassica FJ, Frassica DA, Pritchard DJ, et al: Ewing's sarcoma of the pelvis. J Bone Joint Surg 75A:1457-1465, 1993.
9. Kaplan EL, Meier P: Non-parametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958.
10. Li WK, Lane JM, Rosen G, et al: Pelvic Ewing's sarcoma. J Bone Joint Surg 65A:738-747, 1983.
11. O'Connor MI, Pritchard DJ: Ewing's sarcoma. Clin Orthop 262:78-87, 1991.
12. Picci P, Rougraff BT, Bacci G, et al: The prognostic significance of histopathologic chemotherapeutic response in the treatment of nonmetastatic Ewing's sarcoma of the extremities: The Instituto Rizzoli experience. J Clin Oncol 11:1763-1769, 1993.
13. Reinus WR, Gehan EA, Gilula LA, Nesbit M: Plain radiographic predictors of survival in treated Ewing's sarcoma. Skeletal Radiol 21:287-291, 1992.
14. Sailer SL, Harmon DC, Mankin HJ, Truman JT, Suit HD: Ewing's sarcoma: Surgical resection as a prognostic factor. Int J Radiat Oncol Biol Phys 15:43-52. 1988.
15. Scully SP, Temple HT, O'Keefe RJ, et al: Role of surgical resection in pelvic Ewing's sarcoma. J Clin Oncol 13:2336-2341, 1995.
16. Telles NC, Rabson AS, Pomeroy RC: Ewing's sarcoma: An autopsy study. Cancer 41:2321-2329, 1978.
17. Thomas IH, Cole WG, Waters KD, Menelaus MB: Function after partial pelvic resection for Ewing's sarcoma. J Bone Joint Surg 69B:271-275, 1987.
18. Ware JE, Sherbourne CD: The MOS 36-Item Short-Form health survey (SF-36). Conceptual framework and item selection. Med Care 30:473-483, 1992.
19. Wilkins RM, Pritchard DJ, Burget EO, Unni KK: Ewing's sarcoma of bone. Cancer 58:2551-2555, 1986.
© 2000 Lippincott Williams & Wilkins, Inc.
20. Yang RS, Eckardt JJ, Eilber FR, et al: Surgical indications for Ewing's sarcoma of the pelvis. Cancer 76:1388-1397, 1995.