Efficacy and Safety of Pozelimab Plus Cemdisiran Vs Ravulizumab in Patients with Paroxysmal Nocturnal Hemoglobinuria Who Are Naïve to Complement Inhibition

Background: Paroxysmal nocturnal hemoglobinuria (PNH) is an ultra-rare, acquired, life-threatening disorder, characterized by hemolysis and an increased risk of thrombosis often co-existing with bone marrow insufficiency. Standard of care includes C5 inhibitors, given as chronic intravenous (IV) infusions. The combination of pozelimab and cemdisiran (combo), administered subcutaneously (SC) after one IV loading dose, is a novel approach currently being investigated for its ability to achieve complete, durable inhibition of terminal complement. Pozelimab is an investigational monoclonal antibody that binds to and inhibits activation of C5 while cemdisiran is a silencing RNA that reduces circulating C5. Here, we present results from an exploratory arm of a 26-week (wk), phase 3, open-label, active-controlled trial (NCT05133531) along with interim results of a follow-on open label extension (OLE) study (NCT05744921).

Aims/Methods: Treatment-naïve patients (pts) with PNH were randomized 1:1 (combo vs ravulizumab [ravu]) in the main trial. Combo pts received SC treatment every 4 wks (Q4W). From wk 2, ravu was administered IV every 8 wks per label. Both arms included an initial IV loading dose on day 1. The primary efficacy endpoint was the change in intravascular hemolysis assessed by lactate dehydrogenase (LDH).

All pts who enrolled in the 108-wk OLE received combo SC Q4W. Ravu pts entering the OLE transitioned to combo therapy over a period of 8 wks following completion of the main trial. The last ravu dose was given 8 wks prior to OLE start.

Results: Forty-eight pts were randomized (n=25 combo, n=23 ravu). Mean baseline LDH was 6.5x ULN in combo pts and 6.1x ULN in ravu pts. At wk 26, mean LDH was 0.8x ULN in combo pts and 1.2x ULN in ravu pts. From wks 8 – 26, LDH ≤1.5x ULN was maintained at every visit in 88% of combo pts vs 74% of ravu pts. The % of pts with LDH ≤1.5x ULN across each post-baseline visits ranged from 96–100% with combo vs 65–87% with ravu. Similarly, the % of pts with LDH ≤1x ULN ranged from 67–96% with combo vs 35–74% on ravu. Five of 23 pts treated with ravu and 1/25 on combo failed to achieve meaningful LDH control. Transfusion avoidance was defined as not requiring a red blood cell transfusion based on post-baseline hemoglobin values. A similar number of pts in each arm met criteria for transfusion avoidance (14 combo, 15 ravu). One pt in each arm had breakthrough hemolysis in the main study. Nineteen pts switched from ravu to combo in the OLE (data cut-off Feb 15, 2024), of whom 68% had an LDH ≤1.5x ULN at baseline. After switching to combo, >90% had an LDH ≤1.5x ULN at each visit. Four of 5 pts (80%) not achieving meaningful LDH control on ravu in the main trial achieved control with combo.

In the main trial, 84% of combo pts vs 87% of ravu pts had treatment-emergent adverse events (TEAEs); the most common was headache (7 pts [28%] combo vs 4 pts [17%] ravu). Two pts (8%) on combo had serious adverse events (SAEs): 1 with post-traumatic cellulitis that resolved with treatment, and 1 with fever, seizure, and hemolytic crisis within 1 wk of the first combo dose, and prior to achieving LDH control, that resolved on treatment; on day 130 of the study this pt had a fatal SAE of sepsis and disseminated intravascular coagulation. No SAEs were reported in patients who received ravu.

In the OLE, 13 (68.4%) pts had TEAEs with the most common being injection site reactions (3 pts,15.8%). Three (15.8%) pts had SAEs (1 pt with pyrexia; 1 pt with a urinary tract infection and a fall; and 1 pt with acute hemolysis secondary to infection and Klebsiella pneumonia; all recovered with treatment). No pt discontinued combo due to an adverse event. There were no type 3 hypersensitivity TEAEs due to large drug-target-drug (DTD) immune complexes nor fatal TEAEs.

Conclusion: The combination of pozelimab and cemdisiran led to robust control of LDH, with more pts achieving meaningful LDH control vs ravu. The safety profile is generally consistent with other approved C5 inhibitors. The potential transition regimen using cemdisiran to reduce C5 prior to administering a different C5 antibody appears effective at mitigating DTD reactions. Results support development of the combination of pozelimab and cemdisiran in PNH and other complement-mediated diseases.