Effectiveness of Iptacopan Versus C5 Inhibitors in Complement Inhibitor-Naive Patients With Paroxysmal Nocturnal Haemoglobinuria

Abstract

Background: Paroxysmal nocturnal haemoglobinuria (PNH) is characterised by haemolytic anaemia, bone marrow failure and thrombosis. The single-arm phase 3 APPOINT-PNH trial (NCT04820530) investigating iptacopan monotherapy in complement inhibitor-naive patients demonstrated significant haemoglobin concentration improvements.

Methods: We used target trial emulation to retrospectively predict outcomes if APPOINT-PNH trial patients had received C5 inhibitors instead of iptacopan. Estimates were derived from the real-world APPEX cohort treated with routine C5 inhibitors. The study used benchmarking and comparative effectiveness to evaluate the haematological response in APPOINT-PNH if patients had received C5 inhibitors. Treatment effect was estimated using propensity scores to model the probability of trial inclusion based on baseline covariates, followed by fitting an outcome model to the APPEX cohort.

Results: The analysis of 125 patients showed all estimated treatment effects (95% confidence interval) favoured iptacopan over C5 inhibitors: differences in the proportion of patients achieving haemoglobin increase from baseline of ≥ 2 g/dL, 68.2% (40.9-95.6); haemoglobin levels of ≥ 12 g/dL, 53.4% (31.4-75.3); transfusion independence, 38.8% (15.1-62.5); ratio of percent change from baseline in lactate dehydrogenase levels, 0.51 (0.40-0.67); change from baseline in reticulocytes, -75.5 × 10⁹/L (-106.9, -44.2).

Conclusions: Results indicate C5 inhibitor-naive patients with PNH may experience greater haematological response with iptacopan than with C5 inhibitors.

Trial registration: ClinicalTrials.gov identifier: NCT05842486.

Keywords: APPEX; APPOINT‐PNH; C5 inhibitors; iptacopan; paroxysmal nocturnal haemoglobinuria.

FIGURE 1

Love plot showing the balance in baseline covariates between the APPOINT‐PNH cohort and the APPEX cohort before and after weighting. The circles and triangles represent the unadjusted and adjusted variables used in the propensity score, respectively, and indicate differences between the retrospective cohort of patients and patients in the APPOINT‐PNH trial. The standardised mean differences for all variables were < 0.10, indicating a good balance in the two patient populations after adjustment. *Total number of units transfused in the prior 6 months. MAVE, major adverse vascular event; PNH, paroxysmal nocturnal haemoglobinuria; RBC, red blood cell.

FIGURE 2

Treatment outcomes for effectiveness with C5 inhibitors in the APPEX study and APPOINT‐PNH trial cohorts (overall analytical cohort [n = 125]). (A) Haemoglobin level increase of ≥ 2 g/dL from baseline in the absence of RBCTs, (B) Haemoglobin level of ≥ 12 g/dL in the absence of RBCTs, (C) transfusion avoidance, (D) percentage change from baseline in LDH levels, (E) change from baseline in reticulocyte count. CI, confidence interval; LDH, lactate dehydrogenase; PNH, paroxysmal nocturnal haemoglobinuria; RBCT, red blood cell transfusion.

FIGURE 3

Mean haemoglobin levels over time in the APPOINT‐PNH cohort versus the APPEX cohort. Data are shown as the mean ± SE values of the haemoglobin level at each visiting time. BL, baseline; n, number of patients with available data; PNH, paroxysmal nocturnal haemoglobinuria; SE, standard error.