Objectives: Anemia persists in up to 72% of patients (pts) with paroxysmal nocturnal hemoglobinuria (PNH) despite the inhibition of C5 complement protein with eculizumab (ECU). Pegcetacoplan, a targeted C3 inhibitor, has the potential to control both intravascular hemolysis (IVH) and extravascular hemolysis (EVH). PEGASUS, a phase 3, randomized, open-label, controlled trial (NCT03500549), assessed efficacy and safety of pegcetacoplan compared to ECU. Methods: Pts aged ≥ 18 years with PNH and hemoglobin <10.5g/dL (despite ≥ 3 months of ECU therapy) completed a 4-week run-in with pegcetacoplan plus ECU before 1:1 randomization to monotherapy with pegcetacoplan (n=41) or ECU (n=39). Primary endpoint: change in hemoglobin level from baseline (start of run-in period) to week 16. Secondary endpoints: hemoglobin normalization (defined as hemoglobin level ≥lower limit of normal range) in the absence of transfusions, transfusion avoidance, absolute reticulocyte count, lactate dehydrogenase (LDH), Functional Assessment of Chronic Illness Therapy-fatigue (FACIT-f) score, and adverse events (AEs). Hierarchical significance testing for secondary efficacy endpoints was gated on the success of the primary efficacy endpoint. Post hoc analyses included hemoglobin stabilization (defined as avoidance of a >1g/dL decrease from baseline) in the absence of transfusions. Results: Pegcetacoplan was superior to ECU for change in hemoglobin level at week 16, with an adjusted treatment difference of 3.84g/dL (p<0.0001). At week 16, a greater proportion of pts receiving pegcetacoplan achieved ≥ 2g/dL improvement in hemoglobin (61% vs 0%), hemoglobin normalization (34% vs 0%), and hemoglobin stabilization (85% vs 15%) in the absence of transfusion as compared with ECU. Non-inferiority was shown for transfusion avoidance (35/41 [85.4%] pegcetacoplan pts vs 6/39 [15.4%] ECU pts) and reticulocyte count, but not for LDH. FACIT-f score increased with pegcetacoplan (9.2 [1.61]) and decreased with ECU (-2.7 [2.82]), but was not tested for non-inferiority due to prespecified hierarchical testing. AEs occurred in 36/41 (87.8%) pts with pegcetacoplan and 34/39 (87.2%) pts with ECU; 7/41 (17.1%) and 6/39 (15.4%), respectively, had serious AEs. Most AEs were mild. AEs included injection site reactions (pegcetacoplan, 15/41 [36.6%]; ECU, 1/39 [2.6%] pts) and diarrhea (pegcetacoplan, 9/41 [22.0%]; ECU, 1/39 [2.6%]); infections were reported in 12/41 (29.3%) and 10/39 (25.6%) pts, respectively. Breakthrough hemolysis was reported in 4/41 (9.8%) pegcetacoplan pts and 9/39 (23.1%) ECU pts, leading to discontinuation in 3 pegcetacoplan pts. Discussion: In this phase 3 trial in pts with PNH, pegcetacoplan demonstrated superiority to ECU in hemoglobin level at week 16 with improved clinical outcomes including transfusion avoidance in most pts. The safety profile of pegcetacoplan was comparable to that of ECU. Conclusion: The efficacy of pegcetacoplan validates the prevention of extravascular as well as intravascular hemolysis in PNH, leading to a potential new therapeutic option. Funding: Apellis Pharmaceuticals.