Long‐term effectiveness of eculizumab: Data from the International <scp>PNH</scp> Registry
Louis Terriou, Jong Wook Lee, Cecily Forsyth, Morag Griffin, Jeff Szer, Alexander Röth, Philippe Gustovic, Jesse Metzger, Ami Patel, Christopher J. Patriquin
Abstract
Abstract Objectives Data from the International PNH Registry (NCT01374360) were used to estimate the overall survival and first occurrence of thromboembolic events/major adverse vascular events (TEs/MAVEs) for eculizumab‐treated patients with paroxysmal nocturnal hemoglobinuria (PNH) compared with a contemporaneous untreated cohort. Methods Patients enrolled in the Registry from March 16, 2007, to February 14, 2022, were included. Treated patients received eculizumab for >35 days; untreated patients did not receive eculizumab at any time. Univariable and multivariable analyses were performed using a Cox proportional hazards regression model comparing eculizumab treatment periods to untreated periods and were adjusted for baseline covariates (e.g., high disease activity [HDA], transfusion dependency, and eculizumab treatment status). Results The analysis included 4118 patients. The univariable hazard ratio (HR) (95% CI) for mortality in eculizumab‐treated time versus untreated time was 0.51 (0.41–0.64; p < 0.0001). Significant baseline covariates included age, sex, history of bone marrow failure, ≥4 erythrocyte transfusions within 12 months before baseline, and an estimated glomerular filtration rate ≤ 60 mL/min/1.73 m 2 (all p < 0.0001). In the adjusted analysis, patients with baseline HDA had the greatest reduction in mortality risk (HR [95% CI], 0.51 [0.36–0.72]). Treated patients had approximately 60% reduction in TE/MAVE risk during treated versus untreated time (HR [95% CI]: TE: 0.40 [0.26–0.62], MAVE: 0.37 [0.26–0.54]; p < 0.0001). Conclusion Using data from the largest Registry of patients with PNH, with ≥14 years of overall follow‐up, we demonstrate that treatment with eculizumab conferred a 49% relative benefit in survival and an approximately 60% reduction in TE/MAVE risk.