Background: Despite achieving adequate local control with surgery and irradiation (RTX), distant metastasis is a frequent event in high-risk soft tissue sarcoma (STS). Outcome for this population is dismal. Neo-/adjuvant chemotherapy (nCTX, aCTX) has been applied to eradicate micrometastases and facilitate definite surgery. Here we report on the results of our non-randomized, phase II, NeoWTS trial (EudraCT 2004-002501-72). Methods: Patients with high-risk STS (size >5cm, deep/extracompart. localization, tumor grades II-III) were included. The protocol comprised 4 cycles of nCTX (EIA, etoposide 125 mg/m² iv day 1 and 4, ifosfamide 1500 mg/m² iv days 1–4, doxorubicin 50 mg/m² day 1, pegfilgrastim 6 mg sc day 5), definitive surgery with intraoperative RTX, aRTX and 4 adjuvant cycles of EIA. MRI scans were performed every 2 cycles and assessed according to RECIST criteria. Tumorregression was graded according to Salzer-Kuntschik. Results: Between 06/2005 and 03/2010 a total of 50 subjects (male=33, female=17, median age 50.1 years) were enrolled. STS were located in the upper extremity (n=4), lower extremity (n=31), trunk (n=11) and other loc. (n=4). Histological subtypes included liposarcoma (n=12), synovial sarcoma (n=9), NOS (n=9), MFH (n=8), leiomyosarcoma (n=5), and others (n=7) with tumor grades being II (n=21) and III (n=29). Response by RECIST criteria to nCTX was NED (n=3), PR (n=12), SD (n=31) and PD (n=4). Surgical status after nCTX was R0 (n=37), R1 (n=6) and R2 (n=2). Five patients did not undergo definitive surgery. Median follow-up was 30.5 months. Local recurrence occurred in 6 subjects. Distant metastasis was observed in 14 patients. Overall survival (OS) and disease-free survival (DFS) at 2 years was 83% and 63%, respectively. Median OS and DFS were not yet reached. Multivariant analysis failed to prove influence of resection status or grade of histological necrosis on OS or DFS. Severe toxicities included neutropenic fever (4/50), cardiac toxicity (2/50), and CNS toxicity (2/50) leading to CTX dose reductions in 4 subjects. Conclusions: The current protocol is feasible achieving local control rates, OS and DFS comparable to historical data on neo-/adjuvant CTX.