Background: Paroxysmal nocturnal hemoglobinuria (PNH) is a potentially life-threatening acquired hematopoietic stem-cell disorder, resulting in hematuria, anemia, and acquired thrombophilia. In the Phase I/II COMPOSER trial (NCT03157635), patients (pts) with PNH who were treatment-naive or switched from eculizumab were treated with crovalimab, a novel anti-C5 monoclonal antibody designed with Sequential Monoclonal Antibody Recycling Technology (Röth et al. Blood 2020).
Aims: To report the long-term efficacy, safety, pharmacokinetic (PK) and pharmacodynamic (PD) outcomes in pts with PNH treated with crovalimab in the open-label extension (OLE) period of the COMPOSER trial.
Methods: COMPOSER consists of four sequential parts followed by an OLE period. Pts with PNH who were treatment-naive or switched from eculizumab in Parts 2, 3, and 4, who completed the primary treatment period through Week 20 and derived benefit from crovalimab treatment, were eligible to enter the OLE (Röth et al. EHA 2019). Long-term efficacy, safety, PK and PD data are presented for the OLE. Efficacy endpoints include change in lactate dehydrogenase (LDH), transfusion avoidance and hemoglobin stabilization (both assessed in 24-week intervals), and breakthrough hemolysis (BTH). PK and PD endpoints include assessment of serum concentrations of crovalimab, complement activity by liposome immunoassay (LIA), and free C5 over time.
Results: Overall, 43 of 44 pts entered the OLE and at the clinical cutoff (Nov 1, 2021), 38 pts were ongoing.
Mean normalized LDH was maintained at ≤1.5 × upper limit of normal (ULN) during the OLE, and 80% to 100% of evaluable pts achieved LDH ≤1.5 × ULN at each visit (Figure). Transfusion avoidance was achieved in 83% to 92% of pts over each 24-week interval in the OLE. Similarly, hemoglobin stabilization remained relatively stable during the OLE, with 79% to 88% of pts achieving hemoglobin stabilization across 24-week intervals. There was a total of five BTH events during the OLE across all pts (none leading to withdrawal) for an overall BTH rate of 0.05 events per pt-year (95% confidence interval: 0.01, 0.11).
The incidence of adverse events (AEs) was reported cumulatively from baseline. Median treatment duration (range) for treatment-naive and switched pts on the study was 3.4 years (0.9-4.4) and 2.9 years (0.4-3.9), respectively. Any grade AEs occurred in 95% (42/44) of pts; none led to withdrawal from treatment. Severe AEs occurred in 17% (3/18) of treatment-naive pts and 19% (5/26) of switched pts. Serious AEs occurred in 32% (14/44) of pts (33% [6/18] in treatment-naive and 31% [8/26] in switched pts), of which 5% (2/44) were assessed as related to study treatment by the investigator. No meningococcal infections or deaths were reported.
Crovalimab concentrations were stable and within the expected range over time. Terminal complement activity by LIA was near the lower limit of quantification (10 U/mL) and free C5 levels were maintained at low levels (<0.001 g/L) in most pts. PK/PD relationships between free C5 or LIA with crovalimab confirmed that crovalimab levels of ≈100 μg/mL resulted in complete inhibition of terminal complement activity.
Image:
Summary/Conclusion: Treatment-naive and switched pts with PNH treated with crovalimab in the COMPOSER trial maintained disease control over long-term follow-up. No new safety signals were identified, and complete terminal complement inhibition was maintained during the OLE. These data support the continued clinical development of crovalimab in PNH.