Efficacy and safety of the C5 inhibitor crovalimab in complement inhibitor‐naive patients with PNH (COMMODORE 3): A multicenter, Phase 3, single‐arm study

The Phase 3 single‐arm COMMODORE 3 study (ClinicalTrials.gov, NCT04654468) evaluated efficacy and safety of crovalimab (novel C5 inhibitor) in complement inhibitor‐naive patients with paroxysmal nocturnal hemoglobinuria (PNH). COMMODORE 3 enrolled patients from five China centers. Eligible complement inhibitor‐naive patients with PNH were ≥12 years old, had lactate dehydrogenase (LDH) ≥2 × upper limit of normal (ULN), and had ≥4 transfusions of packed red blood cells within the prior 12 months. Patients received crovalimab loading doses (one intravenous, four subcutaneous) and subsequent every‐4‐weeks subcutaneous maintenance doses per weight‐based tiered‐dosing schedule. Co‐primary efficacy endpoints were mean proportion of patients with hemolysis control (LDH ≤1.5 × ULN) from Week (W)5 through W25 and difference in proportion of patients with transfusion avoidance from baseline through W25 versus within 24 weeks of prescreening in patients who had ≥1 crovalimab dose and ≥1 central LDH assessment after first dose. Between March 17 and August 24, 2021, 51 patients (15–58 years old) were enrolled; all received treatment. At primary analysis, both co‐primary efficacy endpoints were met. Estimated mean proportion of patients with hemolysis control was 78.7% (95% CI: 67.8–86.6). Difference between proportion of patients with transfusion avoidance from baseline through W25 (51.0%; n = 26) versus within 24 weeks of prescreening (0%) was statistically significant (p < .0001). No adverse events led to treatment discontinuation. One treatment‐unrelated death (subdural hematoma following a fall) occurred. In conclusion, crovalimab, with every‐4‐weeks subcutaneous dosing is efficacious and well tolerated in complement inhibitor‐naive patients with PNH.

transfusion avoidance from baseline through W25 versus within 24 weeks of prescreening in patients who had ≥1 crovalimab dose and ≥1 central LDH assessment after first dose. Between March 17 and August 24, 2021, 51 patients (15-58 years old) were enrolled; all received treatment. At primary analysis, both co-primary efficacy endpoints were met. Estimated mean proportion of patients with hemolysis control was 78.7% (95% CI: 67. 8-86.6). Difference between proportion of patients with transfusion avoidance from baseline through W25 (51.0%; n = 26) versus within 24 weeks of prescreening (0%) was statistically significant (p < .0001). No adverse events led to treatment discontinuation. One treatment-unrelated death (subdural hematoma following a fall) occurred. In conclusion, crovalimab, with every-4-weeks subcutaneous dosing is efficacious and well tolerated in complement inhibitor-naive patients with PNH.

| INTRODUCTION
Paroxysmal nocturnal hemoglobinuria (PNH) is a clonal hematological disease caused by an acquired somatic loss-of-function mutation in the gene encoding phosphatidylinositol glycan class A in hematopoietic stem cells. [1][2][3] Phosphatidylinositol glycan class A is involved in the synthesis of glycosylphosphatidylinositol anchor, a glycolipid that attaches many cell surface proteins to the cell membrane. Progeny of affected stem cells are deficient in all GPI-anchored proteins, including key complement regulatory proteins CD55 and CD59. [1][2][3][4] The complement system is an important part of the innate immune system, forming a first line of defense against pathogens. 5 Complement response is initiated through one of three pathways by a cascade of protein interactions that converge at C3 (proximal) and, subsequently, in a shared C5-based terminal pathway. 6 Deficiency of CD59 (which blocks membrane attack complex formation) and CD55 (which controls early complement activation through inhibiting C3 and C5 convertases) renders PNH erythrocytes vulnerable to complement-mediated lysis and leads to intravascular hemolysis-the hallmark of classic PNH-resulting in anemia, hemoglobinuria, and risk of potentially life-threatening thromboembolic events. 3,5 In patients with PNH, thrombosis, renal failure, severe infection, and evolution to myelodysplastic syndromes or acute myeloid leukemia are associated with greater mortality risk. 7,8 PNH is also associated with reduced quality of life and impaired working ability. 3,9 Bone marrow transplantation is currently the only curative treatment for PNH but is not offered as initial therapy and is infrequently performed due to its morbidity and mortality risks. 10,11 Complement system inhibition has proven to be successful in PNH treatment. Terminal complement inhibitors that prevent C5 activation, eculizumab and ravulizumab, [12][13][14] are the current standard of care for treatment of patients with PNH with symptomatic hemolysis or thrombosis in many countries. 15 Additionally, the C3-targeting proximal complement inhibitor, pegcetacoplan, 16,17 is approved in some countries. 15 Several complement factor inhibitors are being investigated in ongoing studies. 18 Despite significant advances in PNH treatment, a high unmet medical need remains as 34%-51% of patients treated with eculizumab continue to require regular transfusions, owing to underlying bone marrow dysfunction, extravascular hemolysis, and breakthrough hemolysis (BTH) episodes. 10,12,13,25 Hemolytic activity remained detectable in many patients during eculizumab treatment, which may be related to incomplete C5 blockage that is possibly attributable to underdosing. 26 Furthermore, in a small subset of patients, C5 polymorphisms can prevent binding of eculizumab or ravulizumab to C5.
One such polymorphism is c.2654G ! A (Arg885His), which is observed in approximately 3.5% of individuals of Japanese descent and less frequently in other ethnic groups. 27 The requirement for life-long intravenous (IV) infusions (e.g., every 2 weeks with eculizumab 12 ) poses a significant burden for patients and their caregivers, with a negative impact on quality of life and interference with work schedules. 3,9 Ravulizumab is approved for use as an IV infusion every 8 weeks 13,14 or subcutaneous (SC) infusion once weekly. 28 Crovalimab is a novel C5 inhibitor, engineered using the SMART-Ig recycling technology, with an every-4-weeks SC maintenance dosing regimen. 29 The SMART-Ig recycling technology combines isoelectric point, neonatal Fc receptor, and pH-dependent affinity engineering, resulting in efficient C5 binding, improved uptake of C5-bound crovalimab by endothelial cells, disposal of C5 in the endosome, and efficient neonatal Fc receptor-mediated recycling of crovalimab. In the Phase 1/2 global COMPOSER study, crovalimab showed rapid and sustained terminal complement activity inhibition in patients with PNH who were complement inhibitor-naive or switched from eculizumab. 29 Here, we report key efficacy, safety, and immunogenicity results from the Phase 3 COMMODORE 3 study in complement inhibitor-naive patients with PNH in China.

| Participants
Patients with PNH who were ≥12 years old and weighed ≥40 kg and had an LDH level of ≥2 Â the upper limit of normal (ULN) at enrollment and at least four transfusions of packed RBCs (pRBCs) during the 12 months prior to screening were enrolled. Key exclusion criteria included current or previous treatment with a complement inhibitor, history of allogeneic bone marrow transplantation, and known or suspected hereditary complement deficiency. All patients provided written informed consent to participate in the study.

| Procedures
All patients received crovalimab per a weight-based tiered-dosing schedule.
The loading doses comprised one IV infusion (weight <100 kg: 1000 mg; weight ≥100 kg: 1500 mg) on Day 1, followed by four once-weekly SC injections on Day 2 through Week 4 (340 mg for all weights). SC maintenance dosing (weight <100 kg: 680 mg; weight ≥100 kg: 1020 mg) was given every 4 weeks starting at Week 5. Patients who experienced clinical benefit after 24 weeks of treatment could continue crovalimab.
Patients had to be vaccinated against Neisseria meningitidis within 3 years or 7 days after initiating crovalimab treatment. Those vaccinated within 2 weeks before or after initiating treatment received appropriate prophylactic antibiotics from treatment initiation, continuing for ≥2 weeks after vaccination. No prophylactics for headache were defined in the protocol.
From Week 9, crovalimab could be self-administered as an SC injection by patients or caregivers without medical supervision, after in-clinic training and supervised self-administration.

| Outcomes
The two co-primary efficacy endpoints were: mean proportion of  Pharmacokinetics and pharmacodynamics of crovalimab were also assessed. Immunogenicity was examined by monitoring anti-drug antibodies (ADAs) at baseline and during the study duration. ADA responses were classified as transient or persistent. 30

| Statistical analyses
Planned enrollment in COMMODORE 3 was 50 patients, which was considered sufficient to preserve approximately 60% of the eculizumab effect on hemolysis control (LDH ≤1.5 Â ULN) that was observed in the ALXN1210-PNH-301 study, which compared ravulizumab and eculizumab in complement inhibitor-naive patients with PNH. 13 The co-primary efficacy endpoint of hemolysis control would be met if the lower bound of the 95% confidence interval (CI) for the estimated mean proportion of patients with hemolysis control was higher than the pre-defined success threshold of 60%. Furthermore, the study would meet the co-primary efficacy endpoint of transfusion avoidance if the difference in proportions of patients who achieve transfusion avoidance from baseline through Week 25 versus within 24 weeks prior to screening was statistically significant at the two-sided type I error level of 0.05 using a paired McNemar test.
All co-primary and secondary efficacy endpoints were analyzed in the primary analysis population, which includes all enrolled patients receiving at least one dose of crovalimab and having at least one central LDH level assessment after the first dose.
The co-primary efficacy endpoint of hemolysis control was analyzed using the generalized estimating equation with first-order autoregressive correlation structure. Sensitivity analyses were preplanned to assess the impact of missing LDH data due to COVID-19 travel restrictions on the robustness of the analysis results and to elucidate the impact of different covariance matrix structures on results.
The co-primary efficacy endpoint of transfusion avoidance was examined using an intra-patient comparison. The 95% CIs of proportions of transfusion avoidance were estimated by the Wilson method. The Newcombe method was used to calculate the 95% CI for the difference between the two proportions of patients achieving transfusion avoidance.   (Table S1).
One patient discontinued the study due to treatment-unrelated death at Week 22.
Median age at enrollment was 31 years (range 15-58); three patients were adolescents (12 to <18 years old; Table 1 Figure 1B). Sensitivity analyses yielded results that were consistent with those of the main analysis (Table S2).
Transfusion avoidance was reached in 51.0% (n = 26) of patients from baseline through Week 25, a statistically significant improvement (p < .0001) compared with the proportion of patients who were transfusion avoidant in the 24 weeks prior to screening (0%) (   Table S3). The safety summary and number of patients with at least one selected AE are presented in Table 3 and Table S4, respectively. No meningococcal infections were reported. None of the AEs led to treatment discontinuation.
One patient experienced a fatal subdural hematoma that was not treatment related, following a fall.
Most AEs were reflective of laboratory abnormalities that were either present at baseline or related to the underlying disease, relevant medical history, or concurrent medications (Table 3; Table S5).
Generally, the laboratory abnormalities that worsened from baseline   At baseline, one patient (2%) was ADA-positive (Table S6). Three adolescent patients weighing ≥40 kg were enrolled and treated with crovalimab. All three patients achieved rapid and sustained terminal complement activity inhibition, with no apparent differences compared with adult patients in terms of crovalimab concentration and terminal complement activity markers. Efficacy profiles measured by co-primary and secondary endpoints and safety profiles in these adolescent patients were consistent with those observed in adult patients.
Self-administration of crovalimab or administration by a caregiver was permitted from Week 9, after training and confirmation of proficiency by a healthcare professional. Between Weeks 9

PATIENT CONSENT STATEMENT
All patients provided written informed consent to participate in the study.