BACKGROUND. Retroperitoneal sarcomas (RPSs) are rare tumors with poor survival rates due to difficult resectability and high local and distant recurrence rates. Preoperative radiation therapy appears to have dosimetric advantages to utilize the tumor as a tissue expander to limit exposure of small bowel to higher radiation doses. METHODS. Between June 1999 and December 2003, 16 consecutive patients with biopsy-proven RPS were treated with preoperative radiation with selective dose escalation. This included 45 grays (Gy) in 25 fractions to the entire tumor plus margin and a boost dose of 57.5 Gy to the volume predicted as high risk for positive surgical margins. Treatment toxicity and local control were evaluated prospectively as primary endpoints. The secondary goal was the theoretical calculation of future dose escalation and feasibility. Each patient underwent laparotomy. Tumor response was judged using computed tomography (CT) scan and by necrosis on final pathology. Theoretical treatment plans evaluated the potential for additional radiation dose escalation. RESULTS. All patients completed the radiation protocol. The most common acute side effects were nausea/vomiting, which affected 4 patients (25%), with only 1 patient requiring inpatient intravenous hydration. There was no severe late post-operative morbidity or mortality. Twelve tumors (75%) decreased in maximum dimension, with a median decrease of 9.4%. Fourteen of 16 patients (88%) underwent complete macroscopic resection. With a median follow-up of 28 months (range, 7-52 months), there were only 2 local recurrences. The actuarial 2-year local control rate was 80%. Theoretical treatment plans suggest that significant dose escalation (up to 80 Gy) may be possible. CONCLUSIONS. Preoperative radiation therapy with selective dose escalation to the margin at risk is tolerable and allows higher radiation dose to the volume judged to be at greatest risk for local tumor recurrence. © 2006 American Cancer Society.