Real-world analysis of ravulizumab safety and effectiveness in advanced age patients with paroxysmal nocturnal hemoglobinuria: Insights from the international PNH registry

Abstract

Abstract Background: Paroxysmal nocturnal hemoglobinuria (PNH) is a rare blood disorder characterized by uncontrolled terminal complement activation leading to intravascular hemolysis, major adverse vascular events (MAVEs, including thrombotic events [TEs]), and early mortality. Ravulizumab (RAV), a second-generation complement C5 inhibitor (C5i) derived from eculizumab (ECU), is the current standard of care for patients (pts) with PNH, where available. RAV's approval was supported by clinical trial data primarily involving adults aged <65 years (y), with limited evidence available for pts aged ≥65 y. The International PNH Registry (NCT01374360), the largest global database of pts with PNH, includes ~25% of patients on RAV aged ≥65 y at initiation. Objective: To compare baseline (BL) characteristics and RAV effectiveness in pts with advanced age (≥65 y) and younger adult pts (<65 y) enrolled in the registry. Safety outcomes and the impact of prior ECU exposure were also evaluated. Methods: Adults (≥18 y) enrolled in the registry (data cut-off: April 7, 2025),with ≥6 months of RAV treatment and ≥1 follow-up (FU) value for lactate dehydrogenase (LDH) or hemoglobin were stratified by age and treatment history. ECU-experienced pts had received ECU and discontinued <28 days before RAV initiation. C5i-naive pts had an LDH ratio ≥1.5 × upper limit of normal (ULN) at RAV initiation and ECU discontinuation ≥28 days before or no prior ECU. Changes in LDH ratio (least-square [LS] mean difference, adjusted for transfusion dependence and history of TEs at BL), transfusion rates, and rates of MAVEs (including TEs) and infections were assessed from BL to ≥6 months FU. Results: RAV data were available for 327 pts (≥65 y: n=81 [65 ECU-experienced, 16 C5i-naive], 55.6% male; <65 y: n=246 [212 ECU-experienced, 34 C5i-naive], 50.4% male). Most pts were White/Caucasian (65.2%), but 75.0% of C5i-naive pts aged ≥65 y were Asian and enrolled in Asian centers. Mean BL LDH ratio was higher in C5i-naive than ECU-experienced pts, but was similar in both age groups (C5i-naive: 5.6 × ULN; ECU-experienced: 1.2 × ULN). In C5i-naive pts, LDH ratio reduced to near normal levels (<1.5 × ULN) at 6 months of RAV, and was stable to last FU (LFU) (≥65 y: 1.1 × ULN; <65 y: 1.6 × ULN). ECU-experienced pts maintained an LDH ratio <1.5 × ULN through to LFU. No significant differences in LDH ratio change from BL to LFU were observed between age groups (LS mean difference [standard deviation]: ECU-experienced –0.07 [0.15], p=0.66; C5i-naive –0.51 [0.88], p=0.57). Transfusion independence increased with treatment in both age groups, but was lower in pts aged ≥65 y than those <65 y at BL (74.7% vs 83.3%) and LFU (79.1% vs 93.2%). Transfusion rate decreased in both age groups from BL to LFU (–0.7 units per person per y [PPPY], in both), and was higher in pts aged ≥65 y than those <65 y at BL (2.1 vs 1.1 units PPPY) and LFU (1.4 vs 0.4 units PPPY). History of MAVEs and TEs was more frequent in pts aged ≥65 y than those <65 y, but bone marrow disorder history was similar. During RAV treatment, MAVEs, TEs, and meningococcal infection rates were low in both age groups (≥65 y and <65 y, per 100 person-years: MAVEs 1.1 and 0.7; TEs 0.5 and 0.4; meningococcal infections 0.1 and no events, respectively). Overall, 9 pts aged ≥65 y (11.1%) and 11 pts aged <65 y (4.5%) discontinued RAV. Reasons were available for 7 and 9 pts aged ≥65 and <65 y, respectively, and included switching to another complement inhibitor (28.6% and 22.2%), switching to ECU (0.0% and 22.2%), physician decision (28.6% and 22.2%), lack of efficacy (0.0% and 11.1%), adverse events (0.0% and 11.1%), and death (42.9% and 11.1%). Causes of death in pts aged ≥65 y were gram-negative sepsis (n=1), acute myeloid leukemia and myelodysplastic syndrome (n=1), and unknown (n=1); in pts aged <65 y, metastatic renal cell carcinoma (n=1). Conclusion: RAV demonstrated effective control of PNH activity in C5i-naive and ECU-experienced pts in the real-world setting, regardless of age at initiation. LDH control and transfusion outcomes improved in both age groups, with no significant differences observed. Despite a greater proportion of pts aged ≥65 y having a history of MAVEs and TEs, the incidence of these complications during treatment was low. These findings further support the use of RAV as the first-line treatment for pts with PNH regardless of age or treatment experience, where available.