LBA5508 Background: CP is standard first-line chemotherapy for AOC. MITO-2 (Multicentre Italian Trials in Ovarian Cancer) is an academic randomized phase III study testing whether CLD is more effective than CP. Methods: AOC patients (pts), stage IC-IV, aged≤75, ECOG PS≤2, were randomized to CP (C AUC5 + P 175 mg/m2,d1q21) or to CLD (C AUC5 + LD 30 mg/m2,d1q21), both for 6 cycles. Primary endpoint is progression-free survival (PFS). Secondary endpoints are overall survival, objective response rate (ORR), toxicity and quality of life. To have 80% power in detecting a 0.80 HR in PFS, with 2-sided α error 0.05, 632 events are needed and 820 pts were planned. Activity was evaluated according to RECIST, toxicity according to NCI-CTC. Activity analysis was not blinded and no confirmation of response was required. Results: From January 2003 to November 2007, 820 pts were randomized, 410 to each arm. Median age was 57 yrs (range 21–77). Stage III (60%) and IV (22%) were prevalent. As of March 30, 2009, with a median follow-up of 35 months, 530 events for PFS and 269 deaths have been recorded. Six cycles were received by 86% and 80% of pts, with CP and CLD respectively. With complete data for 97% of pts, 290 pts were eligible for activity analysis (≥1 target lesion). ORR was 59% with CP and 57% with CLD (p=0.70). In 182 pts with non-target lesions only, complete response was 33% with CP and 29% with CLD (p=0.64). In 168 pts with elevated CA-125 only, CA-125 normalization was obtained in 83% with CP and 86% with CLD (p=0.56). Toxicity data are complete for 97% of pts. Statistically significant differences in haematological toxicity (CP vs CLD) were: anemia all grades 59% vs 68%, grade (G) 3–4 4% vs 10%; thrombocytopenia all grades 19% vs 48%, G3–4 2% vs 16%. Statistically significant differences in non haematological toxicity were: hair loss 63% vs 14%; skin toxicity all grades 6% vs 20%, G3–4 0 vs 2%; diarrhea all grades 13% vs 6%; stomatitis all grades 9% vs 20%; neurotoxicity all grades 47% vs 15%, G3–4 3% vs 0.3%. Conclusions: CLD as first-line treatment of AOC produced a similar activity, with a different toxicity profile, compared to CP. Required events are awaited for final PFS analysis. Partially supported by Schering-Plough. No significant financial relationships to disclose.