Role of surgical resection in pelvic Ewing's sarcoma.

Abstract

PURPOSE The improved survival in patients with Ewing's sarcoma over the past two decades has placed increased importance on achievement of local disease control. Ewing's sarcoma that arises in the pelvis has been recognized to have a worse prognosis than that in the appendicular skeleton, and the role of surgical resection in these cases remains controversial. The current study attempts to identify a benefit to surgical resection in these patients. METHODS We retrospectively examined 39 patients who presented with Ewing's sarcoma in a pelvic location, all of whom were treated systemically with chemotherapy. Twenty patients received radiation only as a means of local control, and 19 underwent resection with or without radiation therapy. The patients were evaluated with end points of disease-free survival and overall survival for a minimum of 24 months and a mean of 58 months. RESULTS There was an even distribution among patients who underwent surgical resection for local control as compared with those who received only radiation therapy with respect to age, site, date of treatment, and stage of disease. Despite uncontrolled biases including tumor size and response to chemotherapy that would be expected to favor patients who undergo resection, surgery in addition to or in substitution for radiation therapy did not result in a statistically significant increase in disease-free survival or overall survival. Local disease control was comparable between those who underwent resection and those who did not: three patients in each group developed a local recurrence. CONCLUSION Currently, morbidity of surgical resection should be weighed against the efficacy and secondary complications of radiation therapy in the decision-making process for local disease control. The issue of whether overall survival and local disease control is improved in patients who undergo surgical resection remains controversial and may require a prospective randomized trial to be answered definitively.