Phase III COMMODORE 1 Trial: 2-Year Efficacy and Safety of Crovalimab in Patients with Paroxysmal Nocturnal Hemoglobinuria (PNH) Who Switched from Ravulizumab

Introduction

Crovalimab (crova; PiaSky®), a novel C5 inhibitor (C5i) allowing for maintenance subcutaneous (SC) injection every 4 wks (Q4W), was evaluated in the global, randomized, Phase III COMMODORE 1 and 2 trials (NCT04432584; NCT04434092). COMMODORE 2 demonstrated non-inferior efficacy of crova vs eculizumab (ecu) in C5i-naive patients (pts) (Röth Am J Hematol 2024). COMMODORE 1 showed similar safety and exploratory efficacy of crova vs ecu in randomized pts who switched from ecu (Scheinberg Am J Hematol 2024); this study also had a non-randomized arm with pts who switched from ravulizumab (ravu) to crova. Ravu, a C5i given IV every 8 wks, is becoming commonly used where available. Here, we report 2-y COMMODORE 1 data from the ravu switch cohort.

Methods

The ravu-pretreated cohort of the COMMODORE 1 non-randomized arm enrolled pts with PNH receiving ravu for ≥16 wks at study start and with lactate dehydrogenase (LDH) ≤2×upper limit of normal (ULN) at screening. All pts received a weight-based tiered dosing regimen of crova (loading doses and SC Q4W maintenance [self-administration was permitted]). The 24-wk primary treatment period was followed by an extension period.

Safety was evaluated from baseline (BL) up to the clinical cutoff date (CCOD). Efficacy was assessed in the primary treatment period from BL to Wk (W)25 and in the extension period from W25–97. Efficacy endpoints included the proportion of pts with hemolysis control (LDH≤1.5×ULN) at each visit, and proportions of pts with transfusion avoidance (TA), breakthrough hemolysis (BTH), and hemoglobin (hb) stabilization; data were reported from BL to W25 and over consecutive 24-wk intervals in the extension from W25–49, W49–73, and W73–97. Pts who discontinued before the end of each 24-wk interval due to lack of efficacy were conservatively imputed to have had a transfusion, had a BTH event, and not had hb stabilization in that interval. Mean Functional Assessment of Chronic Illness Therapy (FACIT)-Fatigue score was assessed at scheduled visits.

Results

27 pts were enrolled, with a median age of 45 y (range, 13–70). 59% were male and 41% had a history of aplastic anemia. At BL, mean LDH was 1.01×ULN (SD, 0.19) and mean hb level was 108.5 g/L (SD, 19.9); mean packed red blood cell units transfused ≤1 y before screening was 0.81 (SD, 2.43). 21 of 27 pts completed 24 wks of crova and entered the extension period. 6 pts discontinued the trial before W25; 5 due to pt withdrawal and 1 due to an adverse event (AE; sepsis). No pts discontinued due to lack of efficacy. At CCOD (Mar 12, 2024), all pts who entered the extension period continued crova. Median treatment duration was 96.1 wks (range, 4.1–148.1).

The proportion of pts with LDH≤1.5×ULN at each visit was 81–100% from BL–W25 and 95–100% from W25–97. From BL–W25, 70% of pts (19/27) achieved TA and 63% (17/27) achieved hb stabilization. From W25–97, across 24-wk intervals, 88–95% of pts achieved TA and 88–93% achieved hb stabilization. BTH occurred in 26% (7/27) of pts from BL–W25 and 0–10% of pts (2 pts) from W25–97, across 24-wk intervals; majority were singular events due to a concurrent complement-activating condition. Mean FACIT-Fatigue scores ranged from 31.2–37.0 from BL–W25 and 35.2–37.2 from W25–97.

Incidence of AEs per 100 pt years (95% CI) was 549.0 (482.5, 622.1) for all AEs, 96.0 (69.5, 129.3) for treatment-related AEs, 71.4 (48.9, 100.8) for Grade ≥3 AEs, and 58.0 (37.9, 85.0) for serious AEs. No deaths occurred.

Transient immune complex reactions (TICRs; one-time events that may be experienced by some pts when switching between crova and another C5i) occurred in 33% of pts (9/27). Common symptoms were rash, arthralgia, and myalgia (n=3 each), with no renal manifestations. Median time to 1^st TICR onset was 2.0 wks (range, 1.3–4.3) and median TICR duration was 1.7 wks (0.6–11.9). Many TICRs were mild/moderate, with no life-threatening or fatal events. All TICRs resolved with no change in crova treatment. No meningococcal infections occurred.

Conclusions

Most pts who switched from ravu to crova maintained hemolysis control through W25, consistent with COMMODORE 1 randomized data. Most pts maintained disease control during the extension period. These exploratory results should be interpreted with caution due to small sample size and inherent variability of underlying disease. No evidence suggests that the safety and efficacy profile of crova differs between pts who switched from ravu and pts who switched from ecu.