Iptacopan is well tolerated in the long-term treatment of PNH.A pooled analysis presented at EHA 2026 demonstrates the long-term safety of the oral proximal complement inhibitor iptacopan in the treatment of paroxysmal nocturnal haemoglobinuria (PNH).
The safety data were pooled from three phase III* studies, two phase II** studies, and the rollover extension programme (PNH-REP). [N Engl J Med 2024;390:994-1008; EHA 2025, abstract S183; Lancet Haematol 2021;8:e344-e354; Blood Adv 2022;6:4450-4460; EHA 2025, abstract PF660; ASH 2025, abstract 3198; Lancet Haematol 2025;12:e414-e430]
Eligible participants were those who received iptacopan 200 mg at least once during any of the parent studies or the PNH-REP. Patients received iptacopan monotherapy 200 mg BID in the PNH-REP. The pooled population comprised 223 patients. [EHA 2026, abstract PF693]
Baseline was defined as day 1 of receiving iptacopan 200 mg BID in the parent study. Total iptacopan exposure was 673.8 patient-years (PY). Approximately 57 percent of participants were exposed to iptacopan for ≥3 years, and 15.2 percent for ≥4 years.
At data cutoff, 42.6 percent of participants had completed the PNH-REP, 47.5 percent remained on study, and 9.9 percent discontinued treatment due to AEs (1.3 percent), pregnancy (0.9 percent), and death (3.6 percent).
BTH, MAVEs
The proportions of participants who had breakthrough haemolysis (BTH) and serious BTH were 13 percent and 2.7 percent, respectively (occurrence rates [OccRs], 6.7 and 0.9 events/100 PY, respectively). All serious BTH events were recovered/resolved.
Eight major adverse vascular events (MAVEs) occurred in seven patients. These included transient ischemic attack (three patients); portal vein thrombosis and intestinal infarction (same patient); and unstable angina, thrombophlebitis/deep vein thrombosis, and lacunar infarction (one patient each). Four patients had recovered/resolved events, and two had not, while one had a fatal outcome.
All MAVEs occurred in patients with at least one cardiometabolic or cardiovascular risk factor, the most common being hypertension (n=5). None of the patients with MAVEs had BTH at any time.
Other outcomes
Approximately 95 percent of participants had ≥1 treatment-emergent adverse event (TEAE; OccR, 329.4 events/100 PY), and 30 percent had ≥1 serious TEAE (OccR, 19.4 events/100 PY), with the most common being infections (13 percent; OccR, 7 events/100 PY). The most common serious infection was bacterial pneumonia (2.7 percent).
Eight deaths were reported, but none of the events leading to death were deemed drug-related.
Overall, the mean total cholesterol was 4.1 mmol/L at baseline, 4.9 mmol/L at year 3, and 5 mmol/L at year 4.
In the pooled phase III cohort (n=188), the mean baseline HDL-C (1.4 mmol/L) remained stable through years 3 and 4 (mean changes from baseline [CFB], 0.0 and 0.1 mmol/L, respectively). The mean LDL-C was 2.1 mmol/L at baseline, with minimal increases at both timepoints (mean CFB, 0.8 mmol/L for both).
Of the nine pregnancies reported, two resulted in healthy live births, one in an ectopic pregnancy, and six in abortions (four elective and two spontaneous).
Consistent with prior evidence
Previous evidence shows that iptacopan was well tolerated, but long-term safety data remain limited. “In this pooled analysis of long-term safety data, the safety profile of iptacopan is consistent with that previously reported, with no new safety findings with up to 6 years of follow-up,” said the investigators, led by Dr Alexander Röth from the University of Duisburg–Essen, Germany.
In summary, treatment discontinuation rates due to AEs were low, serious BTH events and MAVEs were infrequent, and changes in LDL-C levels were minimal. These results show that iptacopan is well tolerated in the long-term treatment of PNH.