AbstractAbstract 2833 Introduction:No therapy with reliable, durable efficacy exists for patients with aggressive non-Hodgkin lymphoma (aNHL) who relapse following at least two lines of therapy. Pixantrone dimaleate (PIX) is a novel aza-anthracenedione, structurally similar to anthracyclines and mitoxantrone, and forms stable DNA adducts. Unlike anthracyclines and mitoxantrone, PIX does not bind iron, minimally promotes reactive oxygen species formation, and is substantially less cardiotoxic in preclinical models. On the basis of promising early clinical activity and acceptable safety, we conducted a phase 3 trial with PIX in patients with aNHL and ≥2 relapses. Patients and Methods:This randomized, multicenter, controlled, open-label study enrolled patients with aNHL (de novo or transformed) with ≥1 prior anthracycline-containing regimen and ≥2 relapses. Patients were randomized to PIX 85 mg/m2 on days 1, 8, and 15 of a 28-day cycle, for up to 6 cycles, or to investigator's choice of single-agent comparator (COMP): vinorelbine, oxaliplatin, ifosfamide, etoposide, mitoxantrone, or, in the US only, gemcitabine or rituximab. Both groups were followed for 18 months after last treatment. The primary endpoint, CR/CRu rate in ITT population, was assessed by an independent assessment panel. Other efficacy endpoints were overall response rate (ORR), duration of response, progression-free survival (PFS), and overall survival (OS). Final end-of-study results are reported here. Results:Planned enrollment was set at 320 patients; however, due to slow accrual, a total of 140 patients were randomized (n=70 per group) to this study. Median number of treatment cycles for PIX group was 4 vs 3 for COMP. Median duration of treatment for patients in the PIX group was about one month longer than in the COMP group (3.8 vs 2.6 months). The end-of-study CR/CRu rate was 24% (16% CR, 8%CRu) for PIX vs 7% (no CRs, 7% CRu only) for COMP (P = 0.009); ORR (CR/CRu/PR) was 40% vs 14% (P = 0.001). After treatment ended, 3 patients in the PIX group achieved CR with no subsequent therapy. Two of the 3 patients converted from SD to CR and 1 from PR to CRu. Median CR/CRu duration was 9.6 months for PIX vs 4.0 months for COMP (P = 0.081). For survival endpoints, there was a 40% improvement in PFS with a median of 5.3 months vs 2.6 months (HR = 0.60, log-rank P = 0.005) and a 21% improvement in OS with a median of 10.2 months for PIX vs 7.6 months for COMP (HR = 0.79, log-rank P = 0.251). Exploratory subgroup analyses of CR/CRu and ORR were consistently higher for PIX patients and suggest stronger efficacy with the subgroups sex (female), <3 prior regimens, no prior rituximab, and ≥1 yr from 1st to 2nd line therapy. During treatment, neutropenia and leukopenia were the most common (≥10%) grade 3/4 adverse events and the incidence of febrile neutropenia was 7.4% for PIX vs 3% for COMP. Percentage of patients with cardiac disorder SAEs was 8.8% for PIX vs 4.5% for COMP. There were 11 grade 1/2 and 2 grade 3 LVEF events for PIX vs 7 grade 1/2 for COMP. Conclusions:In this phase 3 study, patients treated with PIX achieved superior efficacy when compared with other agents, as assessed by CR/CRu rate, ORR, and PFS, and had a positive trend in OS. Pixantrone had a tolerable safety profile in heavily pretreated patients with relapsed aggressive NHL. Disclosures:Pettengell:Cell Therapeutics, Inc: Honoraria. Coiffier:Cell Therapeutics, Inc: Advisory Board. Schiller:Sunesis: Consultancy; Celgene, Genzyme, Millenium, CTI, Antisoma, Novartis: Research Funding. Rizzieri:Cell Therapeutics, Inc: Membership on an entity's Board of Directors or advisory committees. Cernohous:Cell Therapeutics, Inc: Employment. Wang:Cell Therapeutics, Inc: Employment. Singer:Cell Therapeutics, Inc: Employment, Membership on an entity's Board of Directors or advisory committees; DiaKine Therapeutics, Inc: Membership on an entity's Board of Directors or advisory committees.