S183: phase iii randomized, multicenter, open-label commodore 1 trial: comparison of crovalimab vs eculizumab in complement inhibitor-experienced patients with paroxysmal nocturnal hemoglobinuria (pnh)

Background: COMMODORE 1 (NCT04432584) is a global, randomized, open-label, multicenter Phase III trial evaluating the safety and efficacy of the C5 inhibitor (C5i) crovalimab (crova) vs eculizumab (ecu) in C5i-experienced patients (pts) with PNH. Crova enables rapid and sustained C5 inhibition with subcutaneous (SC) self-administration every 4 weeks (Q4W). Aims: To report the safety, pharmacokinetics (PK), pharmacodynamics (PD), and exploratory efficacy of crova in COMMODORE 1. Methods: Adult pts with PNH, weighing ≥40 kg, who had lactate dehydrogenase (LDH) ≤1.5× upper limit of normal (ULN), and were receiving 900 mg ecu every 2 weeks (Q2W) for ≥24 weeks were randomized 1:1 to crova or ecu. Randomization was stratified by history of packed red blood cell (pRBC) transfusion within 1 year prior to randomization (yes or no). Pts received a weight-based tiered crova regimen (Liu ASH 2022; #293) or 900 mg ecu. Regimens included loading doses, followed by maintenance dosing SC injection Q4W for crova and intravenous infusion Q2W for ecu. The primary study objective was safety, evaluated in pts who received ≥1 dose of study treatment. Exploratory efficacy endpoints, assessed in pts who completed ≥24 weeks of crova or ecu treatment, included the proportion of pts with transfusion avoidance (TA), hemolysis control (LDH ≤1.5×ULN), breakthrough hemolysis (BTH), and hemoglobin stabilization, as well as change from baseline (BL) in FACIT-Fatigue score. Safety and exploratory efficacy data were evaluated from BL up to Week 25. PK, PD and immunogenicity data were assessed from BL up to clinical cutoff. Data are presented from the primary analysis of the randomized arms. Results: The clinical data cutoff was Nov 16, 2022. Of 89 pts enrolled, 45 were randomized to crova and 44 to ecu; median (range) age was 42.0 (21–81) for crova and 49.0 (22–85) years for ecu. Mean LDH (×ULN) at baseline was 1.1 for crova and 1.0 for ecu; 23% and 25% had pRBC transfusions ≤1 year before screening, with a mean number of 1.6 and 2.3 units transfused, respectively. There were 86 safety-evaluable pts (44 crova-treated and 42 ecu-treated). Adverse events (AEs) occurred in 77% of pts treated with crova and 67% of pts treated with ecu, with Grade 3/4 AEs occurring in 18% and 2% of pts, respectively; this imbalance was not driven by a risk associated with crova. Serious AEs occurred in 14% of pts receiving crova and 2% receiving ecu; none were treatment related (Table). No meningococcal infections were reported; serious infections were reported in 7% of crova-treated and 2% of ecu-treated pts. No AEs led to withdrawal or death. 16% of crova-treated pts had drug-target-drug complex-mediated Type III hypersensitivity (T3H) events; most were mild/moderate and resolved with no change in crova treatment. The reported signs and symptoms of T3H were mainly skin and joint abnormalities (urticarial rashes, arthralgia, and vasculitis) with no renal involvement. Complete terminal complement activity inhibition, measured by liposome immunoassay and free C5 levels, was generally maintained in the crova arm. Overall, patients switching from ecu maintained disease control as supported by efficacy data presented in the Table. Summary/Conclusion: The safety profile of crova in C5i-experienced pts with PNH is consistent with that in C5i-naive pts, except for transient, self-limiting T3H events unique to pts switching between crova and ecu. The COMMODORE 1 data comparing crova vs ecu in C5i-experienced pts support the favorable benefit–risk profile of crova, including allowing for SC administration with the option to self-administer.Keywords: Phase III, Paroxysmal nocturnal hemoglobinuria (PNH), Complement, Randomized